Loading…
Attending this event?
View analytic

Log in to bookmark your favorites and sync them to your phone or calendar.

Monday, May 14
 

8:30am

Reliability Improvement Roadmap
This workshop introduces the strategies required for asset reliability improvement initiatives to attain a higher likelihood of success. The approach uses lecturettes, exercises, and group discussion.
Upon completion of the course, attendees will understand the nature of the challenge of initiating a reliability improvement program.
You Will Learn:
  • The nature of the real problem to be solved
  • How to develop a clearly defined current state and target state
  • Where your organization sits relative to best practice and why
  • Why you should improve asset reliability
  • What it will take to implement an asset reliability program
  • How to integrate your reliability improvement program with an Industrial Internet of Things (IIoT) initiative and assess your readiness for implementation
  • How T.F. Hudgins and Allied Reliability Group can assist in your reliability journey

Speakers
MA

Mike Aroney

Director of Operations Technical Services, Allied Reliability Group, Inc.
avatar for Preston Johnson

Preston Johnson

Allied Reliability Group, Inc.


Monday May 14, 2018 8:30am - 5:00pm
Cape May
 
Tuesday, May 15
 

7:30am

Breakfast
Tuesday May 15, 2018 7:30am - 8:30am
Ocean Grand Foyer

8:30am

Tuesday Plenary
Opening Remarks & Official Welcome
Amy Wynn, MFPT Executive Director
Chris Nemarich, MFPT Board of Directors Chairman

Keynote 1
Bob Randall
How can we bring the diagnostic advances of the past twenty years to the mainstream machine condition monitoring community?

Keynote 2
Steven Holland
Automotive PHM & the Role for Standards

Speakers
avatar for Steven W. Holland

Steven W. Holland

Research Fellow, GM Global R&D
Steven W. Holland is a Research Fellow at GM Global R&D and is currently responsible for technology strategy in Vehicle Health Management. He has been with GM for over 45 years and has held a wide variety of technical and executive positions in both R&D and Manufacturing Engineer... Read More →
avatar for Chris Nemarich

Chris Nemarich

Naval Sea Systems Command (NAVSEA)
avatar for Robert Randall

Robert Randall

Visiting Emeritus Professor, University of New South Wales
Bob Randall is a visiting Emeritus Professor in the School of Mechanical and Manufacturing Engineering at the University of New South Wales (UNSW), Sydney, Australia, which he joined as a Senior Lecturer in 1988. Prior to that, he worked for the Danish company Brüel & Kjær for... Read More →


Tuesday May 15, 2018 8:30am - 10:00am
Cape Hatteras

10:00am

AM Break
Tuesday May 15, 2018 10:00am - 10:30am
Ocean Grand Foyer

10:30am

10:30am

Multivariate Degradation Dependency Reliability Modeling Based on Copula Function
The degradation process of the complex system is affected by multiple failure mechanisms and brings about multiple degradations to the degradation process. At present, most studies considered the influence of a single degradation on the degradation process of the system, do not meet the actual situation of the degradation process of the system in a complex environment. Or consider multiple degradations, but it is assumed that the respective degradation is independent of each other. A few studies considered the dependency between multiple degradations and used multivariate normal distribution to characterize the dependency of multivariate performance degradations. However, the multivariate normal distribution reflects the linear dependency between variables, which is not applicable when there is a nonlinear dependency between variables. In this paper, we assumed that multiple failure mechanisms of the system correspond one-to-one to multiple degenerations in the degradation process. Firstly, the principal component analysis (PCA) is used to classify each of the two strong dependency degradations, and then the copula function is used to combine every two of the strong dependency degradations. The copula function is used iteratively until the multiple degradations are combined into a one-dimensional degradation. In the process of multivariate degradation, the dependency between multiple performance degradations of the same system may change over time due to changes in the working environment or the degradation rate of the system. In this paper, we considered the dependence of multivariate degradation varied over time. By constructing the time-varying model of Copula function to describe the change of the dependency between multivariate performance degradations with time, so as to establish a multivariate dependency degradation process model with time-varying dependency. The reliability model of system multivariate degradation dependency is established in two aspects: One is the structure of the dependency changes over time. One is the structure of the dependency does not change but the measure used to describe the degree of dependency changes over time. In the end, an example is used to verify the correctness of the model we established. In this paper, we study the degradation process of the system under the influence of multiple failure mechanisms and consider the time-dependent dependency of multiple degradations. This method avoids the difficulty of determining the exact failure mechanism of the system and the uncertainty associated with the multiple degenerations of time. It is of some significance for the study of the degradation process of the system which affected by multivariate degradation in complex environments.

Speakers
XS

Xiaotong Sun

Beihang University
GZ

Guangyan Zhao

Beihang University


Tuesday May 15, 2018 10:30am - 11:00am
Cape Fear

10:30am

Application of Advanced Modulation Maps for Diagnostics of Multi-Stage Epicyclic Gearboxes
Multi-stage epicyclic gearboxes are more commonly found in the industry due to their high ratios and compact sizes as desired features. Due to their complex construction, generated vibration signals could be very difficult to analyse due to the high number of modulating frequencies, which makes the monitoring of the technical condition challenging. For these reasons, researches have introduced different modulation maps that enable easier and more reliable analysis of the structure. Spectral Coherence Density, Modulation Intensity Distribution, and Instantaneous Circular Path Cycle Map are the most recently developed. In the scope of this work, these maps are presented and compared on the example of the experimental test rig containing two-stage epicyclic gearbox.

Speakers
ZD

Ziemowit Dworakowski

AGH University of Science and Technology
KD

Kajetan Dziedziech

AGH University of Science and Technology
AJ

Adam Jablonski

AGH University of Science and Technology, Department of Mechanical Engineering; AMC TECH


Tuesday May 15, 2018 10:30am - 11:00am
Cape Lookout

11:00am

Rotor Design Considerations to Prevent Impeller and Premature Bearing Failures in DWDI & SWSI Centrifugal Fans
Centrifugal Fans are subjected to blade-pass pulsation and mass imbalance forces as part of normal operation.  Fan impellers have several n-nodal diameter modes of natural frequency that can be sensitive to blade-pass pulsation forces.  Excitation of these modes can lead to catastrophic failure.  The principal flexural mode of a fan rotor is sensitive to mass imbalance force, and if excited, can result in amplified stresses in the shaft and amplified force transmission to the bearings.  In the case of SWSI fan rotors, where the 1-nodal diameter mode of the impeller couples with the flexural mode of the shaft, excitation of the rotor mode can lead to catastrophic failure of the impeller.

Speakers
RS

Robert Sayer

The Vibration Institute


Tuesday May 15, 2018 11:00am - 11:30am
Cape May

11:00am

Research on the Reliability Modeling Method of Complex Multi-state System based on UGF
The paper presents a new method based on Abstract syntax tree and the universal generating function(UGF) to reliability assessment for complex multi-state Flow system, which the system and its components can have different performance levels ranging from perfect functioning to complete failure. Generic generation function is an effective solution. It can solve the performance distribution of the system by combining the performance distribution of each component in the system and obtain the statistical value of the system related parameters. In the paper, we first introduced the basic theories in MSS modeling and analyzing, including universal generating function and stochastic process. Next we focus on the using of universal generating function to model and analyze the MSS. A Markov and semi-markov process approach was proposed to obtain the state probability. Then we proposed a MSS oriented modeling language and its analyzing algorithm. In the next part of the paper, a Monte-Carlo simulation method was proposed based on the universal generating function modeling method. In addition, for those multi-failure-modes system, a modeling and simulation approach based on dynamic fault tree was proposed. Finally, we developed some tools and software for the proposed approaches. They were preliminarily applied in the practical projects.

Speakers

Tuesday May 15, 2018 11:00am - 11:30am
Cape Fear

11:00am

11:30am

Crack Propagation in a Compact Tension Specimen Subjected to Gaussian Random Vibrations with Occasional Overloads
It is well known that cracks in structural components subjected to overloads manifest delayed growth for some period of time, slowly reverting to the initial rate on a  curve thereafter. Frequently, constant amplitude fatigue tests with occasional programmed overloads are used to demonstrate the phenomenon. However, in response to random loading in which the sequence of loads is uncertain, different growth rates result from different spectra, even if the distributions from which the loads are taken are identical.

This paper examines variation in the number of fatigue cycles necessary for a crack to pass through the plastic zone after a single overload. A Monte Carlo simulation generates a single initial overload on a compact tension specimen, creating a large plastic zone ahead of the crack. Following the overload both minimum and maximum cyclic loads are simulated from Gaussian distributions. The Forman equation is used to calculate linear crack growth, i.e., without any retardation, and the Generalized Willenborg retardation model is used to calculate the reduction in crack driving potential, KR. The Forman equation is then modified to use the effective stress intensity factor, ?Keff, and the effective load ratio, Reff, to calculate the reduced cyclic growth rate, which is then compared to the linear result. This procedure is repeated for each load cycle until the crack passes through the overload zone and the number of cycles is recorded. The entire process is then repeated a total of 1000 times.

Using standard hypothesis testing, the resulting crack growth data are analyzed to determine which distributions cannot be excluded from consideration at the 5% significance level. 95% two-sided confidence intervals are determined for the distribution parameters. The best candidate distributions are overlaid on the simulation histogram to provide graphical results. Finally, the first four statistical moments calculated from the simulation data are compared to those derived using the distribution's calculated parameters. It was found that lognormal and Birnbaum-Saunders distributions are both good fits.

The underlying basis of the Willenborg model is that large compressive residual stresses, which reduce the effect of applied tensile loads, exist in the vicinity of the crack tip after an overload. Finite Element Analysis was performed on a Ramberg-Osgood material using kinematic hardening, the von Mises yield criterion and an associated flow rule. The results, which verify the presence of a residual compressive stress distribution in the vicinity of the crack tip are presented and discussed.

Speakers
PL

Peter Liaw

University of Tennessee
JR

Julian Raphael

J R Technical Services, LLC


Tuesday May 15, 2018 11:30am - 12:00pm
Cape May

11:30am

Modeling and Simulation Analysis of Dual-Rotor Vibration System with Multiple Faults
Dual-rotor system is an important rotor form in rotating machinery like gas turbine engine. Its complex structure results in rich dynamical behaviors and more probability to failure. The modeling and simulation of the dual-rotor system can help to understand its dynamic characteristics and provide theoretical support for the design, operation and maintenance. In this paper, a dynamic model of a dual-rotor system with multiple rotor faults is established. The dual-rotor vibration model without any fault is built by finite element method, where the two shafts are connected by an inter-shaft bearing and nonlinear models of rolling element bearing and squeeze film damper are considered. The numerical integration method of Newmark-? is used to obtain the steady-state vibration response of the system. Then rotor faults are introduced to the system model, including unbalance, misalignment, looseness and rub-impact. The steady-state responses of single faults in the dual-rotor system are analyzed and typical fault features are obtained. Then coupling characteristics between different rotor faults, and the influences of the squeeze film dampers on the dynamic characteristics and fault features of the system are studied.

Speakers
DJ

Dongxiang Jiang

Tsinghua University
CL

Chao Liu

Tsinghua University
YY

Yizhou Yang

Tsinghua University
WY

Wenguang Yang

Tsinghua University


Tuesday May 15, 2018 11:30am - 12:00pm
Cape Fear

12:00pm

Lunch
Tuesday May 15, 2018 12:00pm - 1:30pm
Ocean Grand Foyer

1:30pm

Modeling Method for Creep-Fatigue Life of Alloy Steels Based on the Stress Relaxation Phenomenon
Numerous alloy steel engineering structures have to work under elevated temperature meanwhile suffering from mechanical fatigue loads in electric industry, nuclear industry and petrochemical industry. The life of these structures are shortened by the creep-fatigue interaction whose damage influence to the life has been a conundrum for years in engineering circle due to the complex creep-fatigue interaction micromechanism. To ensure the reliability and safety of these structures, it is essential to evaluate the creep-fatigue life of these structures. However, the existing life evaluation modeling methods are complicated and expansive requiring a lot of alloy performance data at elevated temperature. To solve these problems, the creep-fatigue damage is evaluated based on the stress relaxation phenomenon, meanwhile a valid modeling method of creep-fatigue life is promoted in this study.
Previous studies have shown that during the strain controlled creep-fatigue process, there will be a stress relaxation in the alloy which is similar to the test behavior under creep-tension load In this study, the stress relaxation behavior of alloy steel under elevated temperature with strain controlled fatigue loads are analyzed in detail. The creep-fatigue damage are described by the strain at certain temperature and the duration time at a certain tension force. With some simplification hypothesis, the creep rupture model are modified to describe the creep-fatigue life by modifying the model parameters.
The value of the modified model are determined by the Bayes updating method which regard the probability as a subjective reliability upon one thing. The posterior probability distribution is updated with the prior information and the likelihood function. And the updating process is finished through the WinBUSS which is an open source Bayes analysis software by fitting data with the Markov chain Monte Carlo Methods. In this study, the prior probability distribution are required from the creep rupture test and the likelihood function is derived from the modified creep-fatigue life model. In this study, the posterior probability distribution of the model parameter of the modified creep-fatigue life model are acquired by updating. The modified model are validated by using the creep-fatigue failure data from other thesis, ensuring the accuracy of the model. Thus, the creep-fatigue interaction damage to the alloy are evaluated based on the stress relaxation behavior during the strain controlled creep-fatigue test in this study. And, the creep-rupture life models are modified to evaluate the creep-fatigue life according to the creep-fatigue damage evaluation and some simplification hypothesis. And a more practical modeling method for creep-fatigue life using the Bayes updating is promoted for engineering application.

Speakers
WH

Weiwei Hu

Reliability and Systerm Engineering School
SL

Sufen Li

Reliability and Systerm Engineering School


Tuesday May 15, 2018 1:30pm - 2:00pm
Cape May

1:30pm

Degradation Signature Modeling
This paper presents models for complex systems or subsystems to transform conditioned-based data (CBD) signatures into functional-failure signatures (FFS) that are particularly amenable to processing by state-estimator prediction algorithms. A CBD-based approach overcomes certain problems associated with conventional approaches, such as model- or statistical-based methods, for producing prognostic information. Failure modes generate characteristic CBD signatures that are correlated to changes in value of a parameter, such as capacitance, as degradation progresses. Features are extracted from CBD signatures and transformed into fault-to-failure progression (FFP) signatures: a function of the feature data (FD). Then FFP signatures are transformed into degradationprogression signatures (DPS): a function of the change in the value of a parameter. A DPS, absent noise, has a characteristic linear curve: a straight line progressing from 0 (no degradation) to a value defined as a level of degradation at which the degrading component, and the assembly it is in, no longer functions within specification(s): functional failure occurs.

A DPS is then transformed into a functional-failure signature (FFS) for input to state-estimator prediction algorithms to support Prognosis for Health Monitoring/Management (PHM): (1) an FFS approaches an ideal straight-line transfer curve as noise is ameliorated and/or mitigated; (2) has negative values in the absence of degradation; (3) has positive values below 100 when there is degradation below a defined level of functional failure; and (4) has values at or above 100 when the level of degradation is at or above a level defined as functional failure. Even in the presence of noise and feedback effects, and even when the rate of degradation is nonlinear, an FFS is still a very linear transfer curve. Seven different families of signatures and models are presented to transform CBD-based signature data into FFS data, and when that data is so transformed and used, the estimation accuracy of prediction algorithms is greatly improved.

This paper presents a method to transform raw sensor data into data that is properly conditioned to be analyzed on a real time basis by anomaly detection and state estimator routines. This capability, in turn, provides a solid foundation that can detect and mitigate faults that occur in complex systems. An example of an electromechanical actuator will be analyzed using the algorithms discussed in this paper.

Speakers
DL

Douglas L. Goodman

Ridgetop Group, Inc.
JP

James P. Hofmeister

Ridgetop Group, Inc.
FS

Ferenc Szidarovszky

Ridgetop Group, Inc.


Tuesday May 15, 2018 1:30pm - 2:00pm
Cape Fear

2:00pm

Reliability Modeling of Degradation Process by Considering Natural Degradation Mutation and Impact Damage Recovery
Due to the changes in the external environment and the changes of internal mechanisms, some components of high-reliability products may have multi-stage failure processes under long-term operation. Some components of the degradation mechanism will show the phenomenon of multi-stage degradation, that is, a mutation occurs in a degraded trajectory at some time. This mutation is not caused by external shocks but due to the qualitative change caused by the degradation of natural degradation to a certain extent. Such as the expansion of the crack, this natural mutation point is called the change point, change point presents different degeneration characteristics. For example, the crack growth. This natural mutation point is called a change point, showing different degeneration characteristics after the change point. The traditional approach is to ignore the mutation point. Considered the mutation point is sick data, as an error directly discarded. However, this traditional approach obviously ignores the information of mutation point. The change point is not easy to be observed through the process of degradation. We can deduce it through statistical inference. In this paper, we determine the interval where the change point exists by using Schwarz information criterion (SIC) and then determine the change point estimate based on the minimum residual square sum criterion. The current study only considers that external impact has a certain impact on the degradation process, but does not take into account the abrupt change that is not caused by the external force when the natural degradation degenerates to a certain extent. In this paper, we consider the influence of natural degradation mutation on the degradation process and describe the degradation process more accurately. At present, the research on the reliability models based on system performance degradation mainly focus on the competing failure between natural degradation failure and impact failure, do not consider the effects of the impact time interval on the degradation process. A time interval in the degradation process is considered which called T. when the interval between the two shocks is greater than the time interval T, the damage caused by the impact is non-cumulative. When the interval between the two impacts is less than the time interval T, the damage caused by the impact is cumulative. The situation that the damage caused by the impact is non-cumulative when the interval between the two shocks is greater than the time interval T is called impact damage recovery. The natural degradation mutation and impact damage recovery in the degradation process is considered to improve the existing competitive failure models of natural degradation failure and impact failure.

Speakers
XS

Xiaotong Sun

Beihang University
GZ

Guangyan Zhao

Beihang University


Tuesday May 15, 2018 2:00pm - 2:30pm
Cape May

2:00pm

Crowdsourcing Machinery Diagnosis
Vibration analysis is an important part of predictive maintenance and is widely used to diagnose a wide range of machine component faults. One of the major limitations of vibration analysis as it stands today is that, in many cases, it still relies on experts to interpret vibration signatures manually. As such, it is at the mercy of human error and biases. To build an automatic machine learning system for vibration analysis a high-quality dataset is needed. While there are many ways to achieve this, one commonly used technique is to crowdsource data labeling.
In our paper, we will present the application of crowdsourcing for vibration analysis on common HVAC machinery.

Speakers

Tuesday May 15, 2018 2:00pm - 2:30pm
Cape Fear

2:30pm

Finite Element Simulation Research on Thermal Reliability of BGA Lead-Free Solder Joint
The paper selected the typical plastic ball grid array package devices, the thermal fatigue characteristics of lead-free solder joints were analyzed by finite element method. We selected Sn3.0Ag0.5Cu solder balls, using the viscoplastic material model to describe the mechanical constitutive relations of solders, and establish the finite element model of the device. The Coffin-Manson equation is used to analyze and predict the thermal fatigue life of the solder joints. Under the temperature cycle loads, the sensitivity of dwell time and ramp time on lead-free solder joints is considered. Considering that the actual height of the solder joint is not completely accurate, the influence on the device lifetime when the height of the solder joint deviates is analyzed, and the influence of different solder ball diameters is compared.

Speakers
LL

Lei Li

Beihang University, College of Reliability and Systems Engineering
ZL

Zhiqiang Li

Beihang University, College of Reliability and Systems Engineering
LQ

LELE QI

Beihang University, College of Reliability and Systems Engineering


Tuesday May 15, 2018 2:30pm - 3:00pm
Cape May

2:30pm

Optimized Testing Campaign and Creation of Onerous Baseline Data of Centrifugal Compressor - A Proposal to Equipment Manufacturer
The paper proposes an optimized testing agenda for centrifugal compressor with a compact manufacturing and testing schedule. As API mandates to conduct spin tests ( MRT ) of all compressors of similar deign and geometry , this paper proposes to conduct only one ASME PTC 10 modified Type 1 or Type 2 Full / Part  load test (as all test beds may not  have required hydrocarbon gas or power for type 1)  combining mechanical run test at vendor works among the full lot.

The  intention of the paper  is to classify the criticality of rotor in terms of rotor stability ratio and then deliberate on  the extent / type of tests taking account of OEM test bed capability and schedule of delivery of machines .To supplement the tests , the paper proposes to undertake extensive design audit activities in terms of rotor-dynamics , aerodynamics taking account of case histories of past failures. Assurance of dimensional repeatability in terms of metrology backed by latest methodology of fault identification of multi-layered manufacturing process with PQM ,which can avoid multiple tests of similar machines can  avoid multiple third party inspection at various stages.

In  No load spin test  and PTC10 type 2 tests ,  site conditions such as influence of piping loads , gas pressure / density , foundation dynamics are not replicated  . The above tests do not identify the region of incipient surge, torsional instability region as well . The paper proposes to introduce various instruments and sensors to detect the above instability region with some diagnostic flow charts for the extensive test .With the rotor-dynamic  data taken from the proposed test set up, it shall be easy to further enhance the base line data after site performance test .The same can be used for pre-alarm configuration based on zone wise amplitudes in vibration spectrum which can be very useful for reliability engineer engaged in diagnostic and prognostics. For critical machines located in hostile environment, anomaly detection may be carried out  with shape identification of plot / spectra.

Speakers
MB

Mantosh Bhattacharya

Petrofac International


Tuesday May 15, 2018 2:30pm - 3:00pm
Cape Fear

3:00pm

PM Break
Tuesday May 15, 2018 3:00pm - 3:30pm
Ocean Grand Foyer

3:30pm

A Cloud-Based Testbed for Standardized Comparison of Machinery Diagnostics Methods
Machinery diagnostic methods often perform well in the lab under specific idealized conditions on a particular test rig. However, their performance may not be as good for detecting and identifying failure on other types of machinery. In this way, many efforts in machinery diagnostics research yield results that are difficult to reproduce and to generalize. In this work, we present an architecture for general purpose excitation and loading of machine components in a testbed that can be replicated by others, or interrogated remotely by those who do not have the resources to produce their own system. A set of data structures is proposed for data collection so that a library of cases can be made available to practitioners who want to assess the performance of a technique using benchmark cases.

Speakers
avatar for Michael Lipsett

Michael Lipsett

University of Alberta


Tuesday May 15, 2018 3:30pm - 4:00pm
Cape Fear

3:30pm

New Possibilities of Redundant Data Transmission for Intelligent Sensor Networks
Condition monitoring systems (CMS) that are currently available offer many types of tools, such as stationary monitoring systems, portable on-site instrumentation, and finally wireless, autonomous systems. Nowadays, the technology related to electronics systems is developing rapidly resulting in the overall improvement within division of data processing (clock speed and number of processing cores), data transfer (speed and power efficiency of communication), and energy consumption. Unfortunately, this development is not being transferred into Condition Monitoring Systems, probably because the complexity of installation and cost of such systems is holding its manufacturers from introducing significant changes in its architecture. A potential point where all modern technologies will culminate is in a technology initiative that is already underway - this is the Industry 4.0. It leverages existing and emerging technologies to improve the efficiency, effectiveness, and service that need to be provided in order to be competitive in the future. Recently developed MEMS technology, ongoing trend to create smaller, more energy efficient electronics, followed by rapidly growing tendency for non-traditional devices being connected to the Internet (IoT) made it possible to design novel, small-size measurement units, which are capable of working autonomously as an individual device or in a set of connected devices, as a distributed system. Such a system might utilize ordinary smartphone device as a part in data transmission chain, as an end-user device that allows to monitor current state of machine or as a gateway that can transmit measurements further, into the cloud system.  Introducing redundant data transmission for such a device allows to create autonomous CMS, which can react to changes in machines operation and changes of the condition of the device itself. The redundant data transmission allows to develop different scenarios including priorities for: extending operation time on battery source, maximizing measurement rate or data transmission rate, prolonging the range of sensors network. Ultimately, these new possibilities might lead to self-learning distributed system and intelligent sensor networks.

Speakers
TB

Tomasz Barszcz

AGH University of Science and Technology, Department of Mechanical Engineering; AMC TECH
AJ

Adam Jablonski

AGH University of Science and Technology, Department of Mechanical Engineering; AMC TECH
WS

Wojciech Staszewski

AGH University of Science and Technology, Department of Mechanical Engineering; AMC TECH


Tuesday May 15, 2018 3:30pm - 4:00pm
Cape May

3:30pm

Generation of a Tachometer Signal from a Smart Vibration Sensor
The tachometer plays an important role in the quality of a vibration-based diagnostic. Because of the bandwidth limits of the machine controller, most machines have a slight change in speed over time. This changes in speed necessitate the resampling of the data, based on a tachometer signal, to facilitate shaft, gear and bearing analysis of the machine. This is because a change in shaft rate changes the spectral content of the signal, upon which vibration analysis is dependent.

Unfortunately, there may be cases, such as glandless pumps, where due to heat and pressure it is impractical or infeasible to install a tachometer sensor. In other situation, such as monitoring gas turbine engine, interfacing with the existing tachometer for the power turbine or compressor turbine, may change certification requirements (adding cost) or increase system cost and weight. These issues may make adding a tachometer impractical.

Using a novel, two step process, we were able to generate a high quality tachometer signal from the vibration data. The first uses an idealized bandpass filter to remove extraneous vibration signal such that the cyclic rate of the shaft can be constructed. The second step then removes any extraneous jitter in the tachometer signal. The resulting tachometer signal is usually of higher quality that achievable from a traditional tachometer signal. The tachometer signal is then used for the time synchronous average or time synchronous resampling algorithm, which is the basis for modern shaft, gear and bearing analysis.  We demonstrate the efficacy of the technique on two data sets, showing that the resulting condition indicators are indistinguishable from a system using a traditional tachometer signal as an input.

Speakers
DD

Dr. David He

University of Illinois at Chicago


Tuesday May 15, 2018 3:30pm - 4:00pm
Cape Lookout

4:00pm

Implementation Lessons Learned in the Condition Monitoring Industrial Internet of Things Era
Condition Monitoring, Condition Based Maintenance (CBM), and Reliability programs rely on a wide range of Big Data.  Implementation of a Condition Monitoring program in the Era of the Industrial Internet of Things (IIoT) requires automation of inspection data collection from sensors, existing systems, and from humans.  The planning stage of an IIoT implementation is critical to a successful CBM/IIoT implementation.  Planning elements include an understanding of asset failure modes, communications infrastructure, integration points of systems that must talk with one another, analytics that can make sense of all the data, training of personnel, and dashboards that display the metrics and key performance indicators and illustrate the business benefits achieved from the efforts.  

We are exposed to many CBM-IIoT efforts, most of which focus on the benefits of the efforts.  Understanding the expected benefits and goals of the CBM-IIoT effort is just one of the elements of a successful implementation.  Most of the effort lies in planning the implementation details of the CBM-IIoT system, where many of the details are obscure.  Starting with a small asset scope, thinking big with respect to business benefits, and scaling quickly is a trusted approach in the CBM-IIoT application space.  

This presentation introduces the planning steps for a CBM-IIoT implementation with examples and lessons learned.  Examples and lessons are drawn from a series of CBM-IIoT efforts, some that have struggled, and some that have been very successful.

Speakers
avatar for Preston Johnson

Preston Johnson

Allied Reliability Group, Inc.


Tuesday May 15, 2018 4:00pm - 4:30pm
Cape Fear

4:00pm

Improved RotoSense for Rolling Stock: Locomotives and Cars
This paper describes work related to improving an accelerometer-based sensor, RotoSense, used for monitoring rolling stock: the locomotives and cars used in trains. At the 2016 MFPT conference, the authors presented a paper, Accurate Vibration and Speed Measurement on Rotating Shafts using MEMS and IoT Single Wireless Triaxial Sensor,? capable of measuring shaft speeds of up to 5500 RPM: two example applications were described: (1) a helicopter gearbox, and (2) conditioning monitoring of a railroad track.
This paper describes subsequent improvements to that sensor, in terms of ruggedizing and signal quality, to meet requirements of a manufacturer of rolling stock. The sensor described in the previous paper was (and still is) the first to survive, intact, three days of testing at the National Test Track Center in Pueblo, Colorado, including a 10-hour, non-stop, 400- mile test run. Even so, a manufacturer of rolling stock wanted more ruggedizing and better signal quality. The rationale, the methods, and the results of those improvements are presented in the paper.

Speakers
DL

Douglas L. Goodman

Ridgetop Group, Inc.
JP

James P. Hofmeister

Ridgetop Group, Inc.
WP

Wyatt Pena

Ridgetop Group, Inc.
RW

Robert Wagoner

Ridgetop Group, Inc.


Tuesday May 15, 2018 4:00pm - 4:30pm
Cape May

4:00pm

Comparative Analysis of Bearing Health Monitoring Methods for Machine Tool Linear Axes
The study of rotating machinery ball bearing diagnostics and prognostics is quite mature and an abundance of methods/algorithms are available to perform these functions. However, extending these algorithms to other ball bearing applications is challenging and may not yield usable results. A linear axis testbed was used to study the ability of an inertial measurement unit to measure changes in geometric error motions. Faults were introduced on the recirculating ball bearings of two carriage trucks with increasing severity. The inertial measurement unit data was analyzed using a variety of methods proposed and used in the rotating machinery community, including auto-regressive filtering, self-adaptive noise cancellation, minimum entropy deconvolution, and spectral kurtosis. The results reveal a surprising ineffectiveness of the methods for bearing faults having low signal-to-noise ratio and/or weaker periodicity than faults in rotating machinery.

Speakers

Tuesday May 15, 2018 4:00pm - 4:30pm
Cape Lookout

4:30pm

Using a High Speed Video Camera System for Vibration Assessment
Speakers
WD

William D. Marscher

Mechanical Solutions, Inc.
CP

Chad Pasho

Mechanical Solutions, Inc.
MJ

Michael J. Platt

Mechanical Solutions, Inc.


Tuesday May 15, 2018 4:30pm - 5:00pm
Cape May

4:30pm

Confidence Assessment in Automated Classification of Nondestructive Evaluation (NDE) Data
Eddy current testing is a widely used non-destructive evaluation (NDE) technique used for identifying anomalies in steam generator (SG) tubes in nuclear power plants. Automated analysis of SG tube inspection data can provide considerable savings in cost, and improves the speed, accuracy and consistency of analysis. A typical analysis software uses various signal processing and machine learning algorithms based on the sensor type, defect location and defect category to classify the indications into non-flaw and flaw categories. In addition to achieving a high classification performance, a quantitative assessment of the reliability of classification results can be valuable in any automated signal classifier (ASC) system. Particularly in NDE systems, a confidence metric can be used as a measure of reliability associated with the classification results to incorporate self-evaluation feature in auto-analysis systems to monitor its own performance.
In ASC systems, reliability depends strongly on the underlying classification algorithm, which in turn depends on the adequacy and representativeness of training samples, noise in measurements or data quality, discriminatory property of signal features. This talk will present a confidence metric determination process for eddy current signal classification algorithms using SG tube inspection data. Derivation of confidence metrics is based on quantity of training data, distribution of signal attributes, multi-sensor inspection data and noise level in signal. By bootstrapping and weighting Bayes posterior probability with estimated noise distribution, effect of noise in NDE measurements is embedded in the resultant confidence measure. Further, classification results from previous inspection is incorporated to define prior probabilities in order to compute a comprehensive confidence measure of NDE data analysis systems.

Speakers
PB

Portia Banerjee

Nondestructive Evaluation Laboratory Dept. of Electrical and Computer Engineering Michigan State University
JB

Jim Benson

Electric Power Research Institute NDE Center
YD

Yiming Deng

Nondestructive Evaluation Laboratory Dept. of Electrical and Computer Engineering Michigan State University
ND

Nathan Driessen

Electric Power Research Institute NDE Center
LU

Lalita Udpa

Nondestructive Evaluation Laboratory Dept. of Electrical and Computer Engineering Michigan State University


Tuesday May 15, 2018 4:30pm - 5:00pm
Cape Lookout

5:00pm

6:30pm

Student Mingling Session
Tuesday May 15, 2018 6:30pm - 7:30pm
TBA
 
Wednesday, May 16
 

7:30am

Breakfast
Wednesday May 16, 2018 7:30am - 8:30am
Ocean Grand Foyer

8:30am

Plenary
Speakers
avatar for Edward Cuoco

Edward Cuoco

Vice President, ThingWorx Analytics
avatar for Mark Derriso

Mark Derriso

Core Technical Competency (CTC) Lead and Technical Advisor, Warfighter Interface Division, 711th Human Performance Wing, Air Force Research Laboratory
Mark Derriso, PhD, is an accomplished researcher with more than 27 years of experience in technology development in the areas of material/structural characterization, health monitoring systems, automatic controls, and autonomous systems. Dr. Derriso has broad experience in develo... Read More →


Wednesday May 16, 2018 8:30am - 10:00am
Cape Hatteras

10:00am

AM Break
Wednesday May 16, 2018 10:00am - 10:30am
Ocean Grand Foyer

10:30am

Sand Infiltration and CMAS Development within Dense Yttria-Stabilized Zirconia Pellets
Yttria-stabilized zirconia pellets were consolidated by pressureless sintering and spark plasma sintering.  The consolidated pellets varied in relative density from 40 to 99.5 % (as determined through Archimedes method) and were characterized via x-ray diffraction (XRD), optical microscopy, and scanning electron microscopy. Using a custom-built flame rig, each pellet was placed in intimate contact with AFRL-02 synthetic sand and exposed to an oxy-acetylene flame (surface temperature ~ 1300 ?C) for 15 minutes. Subsequent characterization results were compared to the corresponding pre-test data to draw conclusions on the degree of infiltration and calcia-magnesia-alumino-silicate (CMAS) formation. No significant chemical reactions were observed, but XRD results do show synthetic quartz formation, with a peak shift of 2? ~ 0.5? that may indicate compressive stress.  In addition to the normal routes of cracks and voids, significant infiltration along the grain boundaries of high density samples (99.5%) was also observed.

Speakers
AG

Anindya Ghoshal

US Army Research Laboratory
MM

Muthuvel Murugan

US Army Research Laboratory
AN

Andy Nieto

US Army Research Laboratory
MW

Michael Walock

US Army Research Laboratory


Wednesday May 16, 2018 10:30am - 11:00am
Cape Lookout

10:30am

An Application of Pattern Anomaly Detection Methods to Fleet-wide Asset Level Diagnostics
Centralized monitoring techniques have become more widely used as business demands and budgetary cuts for companies require streamlined operation and maintenance of a company?s assets. These assets may be located at a single site where the monitoring is taking place, or they may be located all over a state, country or the world. Local data collection with consolidated servers allows a central maintenance center to pool big data for fleet-wide monitoring purposes. Advanced pattern recognition (APR) software solutions have been on the forefront of managing big data for dealing with a multitude of assets.  APR techniques can provide evidence that a machine is not operating as expected, but the condition detected could indicate many possible underlying faults. The root cause may still be unknown.

Causal network analysis has been widely used in providing differential diagnosis in the medical field when a set of symptoms are known. This method is based on Bayesian probability which can handle uncertainty in the data, both input and output, and has a good theoretical foundation.  This paper discusses methods to utilize pattern anomalies as symptoms for a causal network to diagnose asset conditions and to mitigate failures for predictive maintenance programs.

Speakers
CK

Chance Kleineke

Engineering Consultants Group, Inc.
NM

Nilimb Misal

Engineering Consultants Group, Inc.
MS

Michael Santucci

Engineering Consultants Group, Inc.


Wednesday May 16, 2018 10:30am - 11:00am
Cape Fear

10:30am

Damage Detection and Monitoring
Sensors and damage detection devices are absolute necessity for all machineries and structures, be stationary or mobile, military or of civil applications. An understanding of the health of the system is paramount to mission success, cost reduction, lower downtime, longevity and endurance to operational service environment. In the past few decades numerous technological developments in the aspect of corrosion and wear sensing and monitoring systems have evolved which promise a better future for prognostic and diagnostic health management systems.  Some specific details of these developments will be presented.

Speakers
VA

Vinod Agarwala

Iron Pillar Consulting Engineers


Wednesday May 16, 2018 10:30am - 11:00am
Cape May

11:00am

Sand-phobic Thermal Barrier Coatings Materials and Processing for Life Extension of Power Generation Components in Austere Environments
Sand ingestion and chemo-mechanical attack of hot-section turbine components is a leading cause of engine maintenance and repair in modern rotorcraft operational environments. Here we summarize efforts to date in the research, development, test, and evaluation of novel materials and processes for Thermal Barrier Coatings (TBCs) which exhibit enhanced resistance to Calcia-Magnesia-Alumina-Silica (CMAS) attack. As part of the ongoing research effort, novel multilayer formulations and deposition methodologies were developed to provide samples for characterization, screening, and evaluation for quantitative and qualitative measures of CMAS resistance in coupon and representative configurations. US Army Research Laboratory (ARL) efforts, in collaboration with other Government partners, yielded a 30% improvement in CMAS accumulation in representative engine testing via novel multilayer TBC formulation and processing. Further investigations will leverage these materials and multilayer architectures for continued TBC advancement with planned forays into Environmental Barrier Coating (EBC) applications for Ceramic Matrix Composite (CMC) hot-section turbine hardware.

Speakers
BB

Blake Barnett

Army Research Labotory
RG

Robert Gamble

Bowhead Total Enterprise Solutions
MG

Mark Graybeal

Bowhead Total Enterprise Solutions
avatar for Marc Pepi

Marc Pepi

US Army Research Laboratory
JS

Jeffrey Swab

US Army Research Laboratory


Wednesday May 16, 2018 11:00am - 11:30am
Cape Lookout

11:00am

Autonomous Structural Health Monitoring via Low-Frequency, Self-Tuning Energy Harvesting Sensors
Commercial off-the-shelf (COTS) vibration energy harvesters are resonant devices that exhibit limited bandwidth. They operate in the ~kHz frequencies, far outside of everyday environment and processes. A small deviation of less than 5% from resonance would lower the power output by more than 50%. In response, Pyro-E has developed a new family of strain-amplified vibrational energy harvesters that overcome the limitations in bandwidth and robustness of existing devices. The novel design creates opportunities for monitoring bulk equipment with a vibrational signature below 50 Hz. Device performance coupled with field testing results will be presented for dissemination.

Speakers
KL

Kevin Lu

Pyro-E LLC


Wednesday May 16, 2018 11:00am - 11:30am
Cape May

11:30am

11:30am

A Sub-space Clustering Chart Using Heathy Data for Featureless Bearing Performance Degradation Assessment
The sensitivity and robustness of health index (HI) that are characteristics of machine condition may be subjected to the complex working conditions. Thus, for an intelligent health monitoring and management, it is critical to construct a systematic HI that can timely, automatically and reliably assess the machine performance without human experience and intervention. This paper proposes a subspace clustering HI using health reference data to automatically highlight the discriminative information of the monitored state compared to the health state. Different from the conventional HIs learned by empirically extraction from the raw feature sets, sub-space clustering HI aims to automatically describe the migration and variation of the condition clustering distribution by a series of two-class subspace models from the raw data. First, in the featureless process, a covariance-driven Hankel matrix is directly constructed from the raw time-domain signal and principal component analysis (PCA) is used to separate the feature sub-space and noise null-space. In this manner, a feature sub-space can be automatically mined in the Riemannian manifold space. Second, in the index construction process, the reference health subspaces (from healthy data) and the monitored subspaces (from monitored data) data are combined together to construct a referenced two-class model. Thus, a new spatial clustering HI with kernel operation is implemented to assess the current machine performance with discriminative features revealed. The effectiveness of the proposed subspace clustering HI for abnormal condition detection is evaluated experimentally on bearing test-beds, where mobile mapping mode with mobiles is tested. Furthermore, a novel sub-space clustering chart (SCC), CUSUM-based HI is developed to depict a comprehensibility of the real bearing performance degradation process. Compared to regular HIs, (e.g., root mean square, kurtosis), the proposed approach can provide a more accurate and reliable degradation assessment profile with an early fault occurrence alarm. The experimental results show the potential of the proposed sub-space clustering chart in bearing health performance degradation assessment.

Speakers
XD

Xiaoxi Ding

Chongqing University
QH

Qingbo He

University of Science and Technology of China
YS

Yimin Shao

Chongqing University


Wednesday May 16, 2018 11:30am - 12:00pm
Cape Fear

11:30am

Application of Sensor Design in Nitrogen Supply System for Space Station
China's space station will be built into a national science laboratory to carry out a large number of scientific experiments. The nitrogen loop system provides unified nitrogen resources for the scientific experiment cabinet in the space station. The safe and stable operation of nitrogen directly affects the operation efficiency of the Space Scientific Experimental Rack. In order to monitor the health status of nitrogen supply system and take into account the launch weight and interface resources, it is necessary to optimize the design of sensors for nitrogen supply system.

In order to save space as much as possible, the nitrogen loop system uses high pressure cylinders and brings potential safety problems. The security situation can be disposed of in a timely manner by monitoring components that affect safety; besides, margin design is used to ensure that accidents do not occur safety accidents. The design of enough sensors to monitor their status can ensure the timely location of faults. At the same time, the number and layout of sensors can be reasonably designed, which can modularized integration and facilitate maintenance.

There are two main functions of sensors in the nitrogen loop system. One is the ability to monitor the real-time state of the nitrogen system, the safety problem can be controlled in time, and the other is the location of the fault depending on the remote measurement of the sensor after the failure occurs. Based on these two purposes, the safety problems caused by the fault are analyzed by measurable coverage. At the same time, the test orientation analysis of the first order fault is carried out for the planned replaceable unit. The Fault Detection and Location about the nitrogen loop system is guaranteed by the rational application of the sensor.

Speakers
YD

Yu Dequan

Chinese Academy of Sciences


Wednesday May 16, 2018 11:30am - 12:00pm
Cape May

12:00pm

Lunch
Wednesday May 16, 2018 12:00pm - 1:30pm
Cape Charles

1:30pm

How can we bring the diagnostic and prognostics advances of the past twenty years to the mainstream machine condition monitoring community?
Speakers
avatar for Diego Galar

Diego Galar

Professor of Condition Monitoring Division of Operation and Maintenance Engineering, Luleå University of Technology
Dr. Diego Galar is Professor of Condition Monitoring in the Division of Operation and Maintenance Engineering at LTU, Luleå University of Technology where he is coordinating several H2020 projects related to different aspects of cyber physical systems, Industry 4.0, IoT or indus... Read More →


Wednesday May 16, 2018 1:30pm - 3:00pm
Cape Hatteras

3:00pm

PM Break
Wednesday May 16, 2018 3:00pm - 3:30pm
Ocean Grand Foyer

3:30pm

Haul Truck Final Drive Incipient Fault Detection Using On-board Sensors, Vibration Measurements and Oil Testing Data
Haul truck critical component fault detection, diagnostics and prognostics have always been a challenging issue for both reliability engineers and data scientists. Unplanned interruption can have high cost implications, hence a flexible maintenance strategy based on the needs and condition of the component is desired. The capability of early fault detection on critical haul truck components offers unparalleled advantages like improved equipment up-time, lower maintenance cost, and reduced safety risk. This article focuses on the incipient fault detection of the final drives of Cat 795 haul trucks. Complex nature of the haul truck final drive double planetary design makes it difficult to pick right maintenance strategy. In this paper, the authors utilize a variety sources of data such as onboard sensor data, vibration data and oil testing data to perform fault detection. Combined with state of the art machine learning algorithms, this research presents a viable fault detection solution for haul truck final drives. The robustness and effectiveness of the proposed solution are verified using historical data along with logged maintenance actions. In the future, this methodology will be integrated into a prescriptive maintenance matrix to offer work orders to the site technician in real time.

Speakers
GM

Gurpreet Mohaar

Barrick Gold
JZ

Junda Zhu

Barrick Gold


Wednesday May 16, 2018 3:30pm - 4:00pm
Cape Fear

3:30pm

Point of Need Manufacturing for Failed Components/repair
The DoD is interested in 'point of need' manufacturing as a means of being able to produce parts ?on-demand? in extreme environments such as on a forward operating base, or on a ship, to increase our operational readiness, and reduce our huge military logistics tail.  Additionally, research is being performed to determine whether recycled, reclaimed and/or indigenous materials can be utilized as feedstock for these components.  However, there are technical challenges that need to be overcome to fully achieve this capability in the future.  One such challenge is part quality, and whether such parts can provide a true replacement for standard parts.  This brief will discuss this and other challenges in more detail, and will outline the various related on-going efforts at the Army Research Laboratory.

Speakers
avatar for Marc Pepi

Marc Pepi

US Army Research Laboratory


Wednesday May 16, 2018 3:30pm - 4:00pm
Cape Lookout

3:30pm

Enabling Elements of an Embedded Smart Sensor for the Machine OEM
Emerging technologies are expanding the reach of Condition Monitoring to markets and applications not previously practical or economically feasible.  Advances in MEMS technology and next-generation MEMS accelerometers are enabling new opportunities in condition-based monitoring applications through improved performance and significant reductions in size.  These accelerometers are capable of quality vibration measurement, opening the door to highly integrated solutions, in physical form factors and power levels that enable the Industrial OEM and Machine Builders to embed specialized CBM sensing solutions inside the machine, not just mounting on the machine post-installation.  Additional flexibility to mount additional sensors closer to the sources can yield more insight into machine operation, increasing the value of the equipment.
To support these applications, careful consideration of the electrical and mechanical designs must be incorporated to ensure the fidelity of the sensor data is maintained for accurate and reliable information necessary to make critical maintenance decisions.  Sensor signal chains and solutions enabled by solid state electronics can integrate data processing capable of making decisions and generating condition indicators directly from the machine. This reduces the required bandwidth of the communication link, as well as the expense of data handling and memory management of traditional control systems.

This presentation will discuss the integration of MEMS accelerometers with solid state electronics signal processing and the potential for embedded approaches capable of extracting additional value and new insights from the machines.

Speakers
PS

Pete Sopcik

Analog Devices, Inc.
avatar for Ed Spence

Ed Spence

President and Founder, The Machine Instrumentation Group
Ed Spence is the President and Founder of The Machine Instrumentation Group, a collaborative network of product and service providers enabling machine OEMs to instrument their own equipment for condition monitoring. Network partner competencies include machine characterization an... Read More →


Wednesday May 16, 2018 3:30pm - 4:00pm
Cape May

4:00pm

Research on Dynamic Response of Planetary Gear Train with Spalling Fault
Planetary gear train(PGT) is widely used in a variety of fields like wind turbines, aircraft engines, etc, the failure often occurs in the PGT under working condition. The tooth surface spalling of the gear is one of the most common fault in PGT, which will seriously affect the reliability and safety of the mechanical transmission system, it may even cause serious incidents. However, most of the existing models reported in current literature only considered the spalling of gear train with fixed axes or forced on the PGT system under healthy condition, the researches of a fault planetary gear train are still insufficient, especially for the response characteristics of planetary gear train under the spalling condition, furthermore, the method of calculating the mesh stiffness of spalling gear is still have some area for improvement, aim at these shortcomings, a new method of calculating the mesh stiffness of spalling gear is proposed and an analysis on the relationship between the spalling fault excitation and the external vibration response characteristics of PGT system is studied, the transmission paths effect is also taking into consideration in this paper. The research can provide a theoretical basis for the early fault detection, diagnosis of the PGT system.

Speakers
YS

Yimin Shao

Chongqing University


Wednesday May 16, 2018 4:00pm - 4:30pm
Cape Fear

4:00pm

4:30pm

Degradation Modeling: From Condition-based Data Signatures into Functional Failure Signatures
This paper presents models to transform conditioned-based data (CBD) signatures into functional-failure signatures (FFS) that are particularly amenable to processing by prediction algorithms. CBD signatures comprise feature data (FD) that creates a signature that progresses from no degradation to degradation to a level of damage at which a component, and its assembly, no longer functions within operational specifications: functional failure occurs. Failure modes generate characteristic CBD signatures that are correlated to a change in value (dP) of a parameter (P0) as degradation progresses. Further, a feature signature can be transformed into a dimensionless ratio to create a fault-to-failure progression (FFP) signature: FFP = {f(FDi,FD0)g(dPi,P0)} that, when solved in terms of another ratio, creates a degradation progression signature (DPS): DPS =  {dPi/P0} = {f(FFPi)}, that absent noise is a linear straight-line transfer curve.
A DPS is easily transformed into a functional-failure signature (FFS) for input to prediction algorithms in support of Prognosis for Health Monitoring/Management (PHM): (1) an FFS approaches an ideal straight-line transfer curve as noise is ameliorated and/or mitigated; (2) has negative values in the absence of degradation; (3) has positive values below 100 when there is degradation below a defined level of functional failure; and (4) has values at or above 100 when the level of degradation is at or above a level defined as functional failure. Even in the presence of noise and feedback effects, and even when the rate of degradation is nonlinear, a DPS is still a very linear transfer curve. Seven different families of signatures and models are presented that are used to transform CBD-based signature data into FFS data.

Speakers
DL

Douglas L. Goodman

Ridgetop Group, Inc.
JP

James Pierson

Ridgetop Group
FS

Ferenc Szidarovszky

Ridgetop Group, Inc.


Wednesday May 16, 2018 4:30pm - 5:00pm
Cape Fear

4:30pm

4:30pm

6:00pm

Networking Reception At Virginia Aquarium & Marine Science Center
Wednesday May 16, 2018 6:00pm - 9:00pm
Virginia Aquarium & Marine Science Center 717 General Booth Blvd, Virginia Beach, VA 23451, USA
 
Thursday, May 17
 

7:30am

Breakfast
Thursday May 17, 2018 7:30am - 8:30am
Ocean Grand Foyer

8:30am

A Data-Driven Based Prognostics Approach to Predict the Performance Degradation of Rotary Seal
Rotating machinery is the most widely used mechanical equipment in oil and gas industry and keeping rotary drilling equipment in top operating condition is a key activity that helps maintain efficient process operations in oil and gas plants. The O-ring seal failure is one of the foremost causes of breakdown in rotary machinery, and such failure can be catastrophic, resulting in costly downtime and large expenses. Replacing a seal after its failure can be costly and at the same time an early tool replacement decision may lead to lower tool life utilization. Thus rotary seal performance degradation prediction is extremely important in condition-based maintenance to reduce the maintenance cost and improve the reliability. This paper proposes a data-driven prognostics approach based on Support Vector Machine (SVM) to achieve performance degradation prediction of rotary seals. Time domain, frequency domain and time-frequency domain feature extraction methods will be employed to extract features from the original vibration signals. Principal Component Analysis (PCA) will be employed to select relevant and sensitive features. Later, SVM model will be built and trained to predict rotary seals degradation process. Accelerated aging and testing is performed to validate the effectiveness of the approach proposed in this paper. The experimental datasets are generated from the test platform dedicated to test rotary seals.

Speakers
MR

Madhumitha Ramachandran

The University of Oklahoma


Thursday May 17, 2018 8:30am - 9:00am
Cape Lookout

8:30am

Fault Type Characteristics in Gears
Condition-Based Maintenance requires prediction of the remaining useful life (RUL) of each component of the system. One of the most effective CBM methods for rotating components is based on vibrations. Gear transmissions are widely used in industrial applications and are considered as critical monitoring components. This study focuses on fault detection and characterization in a one stage, spur gear transmission. Condition Indicators (CI) provide the ability to quantify significant information contained in the signatures into a specific value.
The prime objective of this research is to provide an infrastructure that enables characterization of defects and their severity. Today, the picture is unclear. Some of the current CIs can diagnose one type of fault at early stages yet cannot diagnose other types of faults. Establishing signal analysis capabilities is an important step for acquisition of qualified CIs that will be able to determine the type of a fault and foresee the gear failure. It is necessary to find methods that provide reliable and robust CIs.
In addition to experimental observations, a validated dynamic model of a gear system is utilized for the benefit of this research. The dynamic model can simulate the vibration signature of healthy and damaged gears with different types and sizes of faults.
In this study, three different common local faults are examined, i.e. spall, broken or missing tooth, and cracks at the tooth root. All three are thoroughly analyzed based on both simulations and experiments to find CIs that are sensitive to the existence and size of the fault. The ability to characterize the fault type is examined as well.

Speakers
JB

Jacob Bortman

Ben-Gurion University
ID

Ido Dadon

Ben-Gurion University
RK

Renata Klein

R.K. Diagnostics
NK

Niv Koren

Ben-Gurion University


Thursday May 17, 2018 8:30am - 9:00am
Cape Fear

8:30am

Failure Analysis of a Primary Reactor Cooling Pump Using Modal and Vibration Analysis
Four pumps redundantly supply primary cooling water to the reactor of the High Flux Isotope Reactor (HFIR) at the Oak Ridge National Laboratory (ORNL).  All four of the pumps have undergone maintenance during a recent scheduled down time. One pump, PU-1A, exhibited higher vibrations levels than the others during preliminary evaluation after the maintenance. The pump, being a safety class item, underwent further performance testing to rule out potential damage resulting from the higher vibration. Vibration analysis and modal analysis including steady state spectrum, operational deflection shape, run up transients, and modal impact have been utilized to identify failure modes that could contribute to the higher relative vibration levels. This report will cover the testing setup, methodology, analysis results, and maintenance suggestions.

Speakers
TH

Thomas Hazelwood

Oak Ridge National Laboratory
BV

Blake van Hoy

Oak Ridge National Laboratory


Thursday May 17, 2018 8:30am - 9:00am
Cape May

9:00am

Rotary Seal Performance Degradation Assessment Based on Vibration and Friction Torque Signal
The O-ring seal failure is one of the foremost causes of breakdown in rotary machinery, and such failure can be catastrophic, resulting in costly downtime and large expenses. The O-ring seal is widely used to retain fluid inside while preventing the passage of mud, dirt, dust, water, etc. Replacing a seal after its failure can be costly and at the same time an early tool replacement decision may lead to lower tool life utilization. Thus evaluating performance and severity diagnosis are very important for maintenance decision-making. In rotary equipment, seals are exposed to harsh environment including high temperature and contaminants. The prolonged exposure of rotary seals to such an environment gradually degrades the seal until it fails. In this study, accelerated aging in an aggressive fluid is performed to age the seals. Aged seals are then tested using the test platform dedicated to test rotary seals. The seals are tested starting from its nominal state (before aging) until failure and the seals performance are captured during its whole degradation process. Vibration signal and friction torque signal of different severity levels will be assessed in this study. Time domain, frequency domain and time-frequency domain features will be extracted from the raw vibration data. Then the extracted features will be examined for their sensitivity based on the different severity in seals. This study finding will show the relevant degradation features which is critical for detecting the incipient degradation and to demonstrate the whole process of degradation development.

Speakers
MR

Madhumitha Ramachandran

The University of Oklahoma


Thursday May 17, 2018 9:00am - 9:30am
Cape Lookout

9:00am

An Aircraft Level Fault Simulation System Design
With tight profit margin and intense market competition, reduce operating cost is key for airliners. This demand has well passed to aircraft OEM. Thus, maintenance support started to sold by flight hour. Faults of aircraft no longer just relate to  a product belongs to customer but also become directly link to aircraft manufacturers maintenance business.

Simulation is a common tool in Industrial 4.0 age. Traditionally, many fault simulation studies have been done for different aircraft systems such as landing gear, fuel system, actuation system, wing system and air condoning system. However, for a complex system as commercial aircraft, the problem gets worse. Integration and interact of systems as well as interaction of their faults cause problems. But few researchers will have resource and nor the need to consider an aircraft level fault simulation. On the other hand, when a new type of aircraft first come into market many false alarms is very common. So besides system level fault study, emphasis should be given on study of aircraft level fault interaction  for commercial aircraft OEM.

In this paper a design of aircraft level fault simulation is presented. Theoretical analysis of fault parameters is the beginning. Modelica language software is used to simulation connected subsystems and these subsystems are linked to an onboard maintenance computer which is common used current aircraft. With this basic structure built more studies could be done in different subsystem faults interaction and fault mitigation strategy as well as reduce false alarm for new designed aircraft.

Speakers
GL

Gao Limin

Beijing Aeronautical Science and Technology Research Institute Commercial Aircraft Corporation of China
CS

Chang Shuo

Beijing Aeronautical Science and Technology Research Institute Commercial Aircraft Corporation of China
WY

Wang Yi

Beijing Aeronautical Science and Technology Research Institute Commercial Aircraft Corporation of China
WZ

Wang Zhaobing

Beijing Aeronautical Science and Technology Research Institute Commercial Aircraft Corporation of China


Thursday May 17, 2018 9:00am - 9:30am
Cape Fear

9:00am

Comprehensive condition monitoring analysis for power plant boiler circulating pumps
The boiler circulating pumps (BCP) is an integral part of the power plant operations in both convention (coal fired) or nuclear plants. The BCP move super heated water under pressure to the steam generator. These pumps are glandless, megawatt induction power pumps. If the pump fails, the power plant must be removed from service to replace the pump. In many applications, these pumps serve power plant providing the base load power requirements of a community. The failure of a pump requires the operator to buy power from other generating plants at much higher rates.

Glandless BCP pose a difficult problem for automated analysis, as there is no way to introduce a tachometer to measure pump speed. Further, while a relatively simple machine, the failure modes include: out of balance/bearing wear that is best measured by vibration, and rotor bar/opens/shorts/eccentricity that is best measured by current. Further, these machines are asynchronous and typically very well balance, so that is difficult to determine shaft RPM from vibration.

This paper discusses the analysis of BCP using both vibration and current analysis methods and the processing needed to automate fault detection, diagnostics, and prognostics.

Speakers

Thursday May 17, 2018 9:00am - 9:30am
Cape May

9:30am

Single Mechanical Seal Vibration Condition Monitoring
Mechanical seals are devices that seal machines between rotating parts (shafts) and stationary parts (pump housing). Their main jobs are to prevent the pumped product from leaking into the environment and are manufactured as single or double seals. A single mechanical seal consists of two very flat surfaces that are pressed together by a spring and slide against each other. Between these two surfaces is a fluid film generated by the pumped product. This fluid film prevents the mechanical seal from touching the stationary ring. An absence of this fluid film (dry running of the pump) results in frictional heat and ultimate destruction of the mechanical seal.
Mechanical seals tend to leak a vapor from the high pressure side to the low pressure side. This fluid lubricates the seal faces and absorbs the heat generated from the associated friction, which crosses the seal faces as a liquid and vaporizes into the atmosphere. So, it is common practice to use a single mechanical seal if the pumped product poses little to no risk to the environment. In the same time the single mechanical seals still need to be controlled to prevent the mentioned leak problem.
This paper described the vibration based single mechanical seal condition monitoring technique and experimental results. We found that the high frequency vibration level in a range of 25 kHz to 50 kHz have a good correlation with the seal quality and could be used for there condition monitoring. The experimental results as vibration spectrums in the above frequency range from different single seals conditions are presented.

Speakers
GZ

George Zusman

President, Vibration Measurement Solutions, Inc


Thursday May 17, 2018 9:30am - 10:00am
Cape Lookout

9:30am

New Advances Associated with The Application of Advanced Health and Prognostics Management in Biologics Manufacturing
Keeping Health Factories Healthy with PHM
 
Advances in Prognostics and Health Management (PHM) technologies and platforms have proven invaluable for increasing reliability and availability across a multitude of manufacturing processes. For years, factories involving both continuous and discrete manufacturing operations have benefited from well-engineered PHM systems that focus on monitoring, assessing, optimizing, reconfiguring, and protecting the processes, the equipment, and the profits associated with the businesses. One relatively new kind of factory - those that manufacture biologics medicines  - is certainly no exception. Such factories are responsible for manufacturing ever increasing amounts of critical medicines that are improving lives across the planet. Although much of the equipment and the manufacturing processes are complex and unique, such “health” factories  - which demand virtually continuous availability and require utmost in protection from consequences of failure  - are perfect candidates for the assurances provided by modern PHM systems. This paper describes new advances and accomplishments associated with the application of advanced health and prognostics management to biologics manufacturing.

Speakers
MW

Mark Walker

Vice President of Engineering, D2K Technologies, LLC


Thursday May 17, 2018 9:30am - 10:00am
Cape Fear

9:30am

Pump Vibration Analytics Case Study and the Need for More Deployable Instrumentation
Emerging technologies are expanding the implementation of Condition Monitoring to markets and applications not previously practical or economically feasible. This trend is particularly observable as Industrial OEMs develop CBM programs for their own hardware. Data driven analytic approaches to CBM can result in the development of additional monitoring tools and expanded suite of condition indicators using both existing and new instrumentation, and overcoming limitations with traditional approaches developed for rotating/reciprocating equipment. Transferring learning from the lab to the field introduces challenges and obstacles to optimum deployment that opens the door to new, embedded sensor technologies. Bandwidth limitations in existing control systems drive the need for edge node processing or deployment of completely new CBM networks.
The paper will highlight an oil and gas pump condition monitoring case study, which used data driven analytics, but also faced several challenges with respect to the field instrumentation. We will suggest future work to overcome the deployment challenges with new technologies.

Speakers
DS

David Siegel

Predictronics Corp.
avatar for Ed Spence

Ed Spence

President and Founder, The Machine Instrumentation Group
Ed Spence is the President and Founder of The Machine Instrumentation Group, a collaborative network of product and service providers enabling machine OEMs to instrument their own equipment for condition monitoring. Network partner competencies include machine characterization an... Read More →


Thursday May 17, 2018 9:30am - 10:00am
Cape May

10:00am

AM Break
Thursday May 17, 2018 10:00am - 10:30am
Ocean Grand Foyer

10:30am

Slow Roll Run out Problem of Rotor Genuine or Benign
The issue of Rotor ?run-out? which causes high overall vibration (displacement) readings on rotor has always been a point of concern during spin test and site acceptance test of turbo-machines.  Rotor run-out is sub-divided to TIR (total indicated run out) commonly known as Mechanical run out, residual magnetism and electrical run out.
While API ( American Petroleum Institute ) standards have mandated a certain limit of Slow roll run out, the proposed paper discusses to find  which  of contributing factors of high run-out should be considered as genuine or benign during machinery health assessment and  monitoring as its main objective .
Proposed  paper differentiates  rotor bow and slow roll run out  , effect of run out during rotor balancing and  to point out the  lacunae of understanding exits at genuine point / stage of measurement. A case study is presented to explain the issue and delays caused for repeated repair works .The paper discusses the feasibility and overall technical effectiveness of eddy current probe, capacitance probe and other devices to address the issue.
The proposed paper suggests  the use of various diagnostic tools such as band and envelope spectrums for creating pre-alarm set up ,  using shape identification  algorithm in diagnostic system based on reference data captured by initial slow roll measurement with a detailed set up .

Speakers
MB

Mantosh Bhattacharya

Petrofac International


Thursday May 17, 2018 10:30am - 11:00am
Cape Fear

10:30am

Application of Piezoelectric MFC Sensors and Fibre Bragg Grating Sensors in Structural Health Monitoring of Composite Materials
Advanced composite materials have been integrated extensively into aircraft structures and have been emerged in civil infrastructure (e.g. bridges), in recent years. Composite materials are prone to initiation of hidden damage which makes it vital to detect damage at its onset. In this paper, first we introduce a procedure to fabricate structural carbon fiber reinforced polymer (CFRP) sheets at the university laboratory. Then, two structural health monitoring (SHM) systems will be developed to assess and monitor the performance of CFRP materials. The first SHM system is based on installing piezoelectric MFC sensors on CFRP sheets. These sensors are using guided lamb waves to detect possible damage in composites at its onset. The second SHM system will be developed using Fiber Bragg Grating (FBG) technology. This system will acquire signals from a fiber optic thermocouple and a fiber optic strain gauge. The damage will be introduced into CFRP sheets with an impact hammer. The damage detection will be performed by the two SHM systems for different damage severities. The pros and cons of each SHM system in composite damage detection will be investigated and the recommendations will be made for utilizing each system in real world composite applications.

Speakers
MA

Mohammad Azarbayejani

New Mexico Institute of Mining and Technology
CB

Chase Beckstead

New Mexico Institute of Mining and Technology
SL

Summer Little

New Mexico Institute of Mining and Technology


Thursday May 17, 2018 10:30am - 11:00am
Cape Lookout

10:30am

Pump Cavitation
Speakers
avatar for Tony Barlow

Tony Barlow

Condition Monitoring Analyst, Chevron
avatar for Suri Ganeriwala

Suri Ganeriwala

Spectra Quest, In


Thursday May 17, 2018 10:30am - 12:00pm
Cape May

11:00am

Grain Elevator Quality Assurance using Deep Convolutional Neural Network
The accidental mixing of different grain types at an elevator is a common and costly mistake resulting in either a total loss, or costly blending processes to correct. We present an automated and intelligent quality assurance system to prevent these incidents by using a deep convolutional neural for crop type classification deployed to an edge device.

Our research shows human crop type classification accuracy across eight different crop types is 99.7% for the best expert and as low as 83.9% for the worst novice with an overall accuracy of 96.2% for the entire group in lab experiments. It is anticipated that in real world conditions when faced with fatigue, distraction, varied skill level and negligence this number would reduce further. It is clear that a highly accurate classification system which is able to continuously monitor the process and warn the operator would help reduce incidents.

Deep convolutional neural networks have shown excellent performance at many complex image classification tasks and a number of popular architectures have emerged (Inception V3, Google and ResNet Microsoft). However these architectures are very computationally and memory expensive making them challenging to deploy to low cost rugged hardware needed for many industrial application spaces. In addition the typical process of retraining only the final layers of the network (a technique called Transfer Learning) does not allow for customizing the network architecture to reduce complexity and potentially achieving better performance.

To overcome these limitations we have constructed a custom deep convolutional neural network architecture consisting of five convolution layers and two fully connected layers. The trained model achieves an overall accuracy of 99.5% (surpassing the human classifiers) and a Kappa coefficient of 99.48% across eight different crop types with one ? five cultivars (varieties) per crop type. With an added voting algorithm based on N samples the number of misclassifications can be further reduced and add robustness. The resultant optimized model at only 800KB (weights and biases) represents a reduction of two orders of magnitude with respect to storage and memory requirements enabling it to be run in real-time at the edge requiring no network connectivity or communication with the cloud.

Speakers
JG

Josh Gelinske

Appareo Systems
NS

Nathan Schneck

Appareo Systems
JT

Jesse Trana

Appareo Systems


Thursday May 17, 2018 11:00am - 11:30am
Cape Fear

11:00am

Health and Usage Monitoring: Design of experiment & monitoring techniques
The following paper is a work in progress and is centered on the premise the Intelligent Transportation Systems (ITS), the self-driving automobile and the environment in which it operates will require the automobile to have an on-board Health and Usage Monitoring Systems (HUMS) to operate safely on roadways. HUMS are in their early stages of use to monitor vehicle components and critical structures to prevent routine and catastrophic failures. Vehicle health monitoring, prognostics, diagnostics and condition-based maintenance are the central tenants of a proposed HUMS for autonomous automobiles.  Over the past decade, autonomous vehicles have been studied and developed for a real-world application. Upon identifying the critical assets, it is essential to create a design of experiment contingency plan to understand monitoring techniques required to conduct safe operations in ITS. This paper identifies the techniques required to monitor those components through a development of a design of experiments for the critical assets.

Keywords: Health and Usage Monitoring Systems, Prognostics, Autonomous Systems, Automobile

Speakers
AP

Ankit Patel

Drexel University


Thursday May 17, 2018 11:00am - 11:30am
Cape Fear

11:00am

TBD
Speakers

Thursday May 17, 2018 11:00am - 11:30am
Cape Lookout

11:30am

System Reliability Modeling and Analysis of Alternating Work of Systems Based on Dynamic Fault Tree
Due to the large-scale and complication modern engineering system and the introduction of high-tech, system reliability has become the key to control the development of complex systems. As a key basic technology in implementing system reliability engineering, reliability analysis technology is facing several technical difficulties and application challenges caused by complex systems. In the actual complex engineering system, the system often has a variety of complex relationships and dynamic characteristics, such as the order of failure of components, or Alternating work of the systems. At present, some achievements have been made in the analysis of fault tree considering dynamic failure characteristics. However, there is still a lack of research in considering Alternating work of the system, so that the results obtained by conventional methods are not in accordance with the actual situation or even far apart.
The dynamic fault tree is a description model with powerful ability to describe dynamic systems. Therefore, this paper considers the establishment of a system reliability model based on dynamic fault tree. The dynamic fault tree includes the priority AND gate, the function related gate and the backup gate, which can describe part of the dynamic logic in the system, but can not describe alternating working of the logic of the system. Therefore, this article needs to expand the dynamic fault tree model, the method of extending the model is as follows: 1. Graphical description of the feature needs to be given based on graph theory and reliability theory; 2. Determine the input and output of the graph; 3. Give the definition of the graph and establish the logic gate that can describe the alternation of the system. After establishing the corresponding logic gates, the corresponding solution method of the established model is given. The solution method is modeled as follows: 1. According to the logic of this characteristic, the corresponding calculation flow chart is analyzed, and the flow chart of each step is transformed into the corresponding algorithm. For the specific problems involved, find the corresponding theoretical knowledge (such as conditional probability, total probability formula, Markov chain, etc.), and finally get the corresponding model solution method.
After expanding the dynamic fault tree model, the reliability model of the whole system is established based on the extended dynamic fault tree. When solving, the dynamic fault tree is usually transformed into Markov model. In this paper, the dynamic fault tree is first modularized to obtain independent static subtree and dynamic subtree, which are respectively solved by binary decision diagram method and Markov process method. Take a production system for example, using this method calculate the reliability of the production system.

Speakers
XS

Xiaotong Sun

Beihang University
JX

Jie Xuan

Beihang University
GZ

Guangyan Zhao

Beihang University


Thursday May 17, 2018 11:30am - 12:00pm
Cape Fear

11:30am

Using Optical Fiber Sensors for Detecting Natural Fault in Ball Bearing during Endurance Test
Rolling element bearings (REBs) are one of the most basic machine parts in rotating machinery. A defect in the REB can lead to a critical failure of the system. One of the common methods for monitoring the health of REBs is through vibrations analysis. This study examines a new method for monitoring rotating dynamic systems. In the new method, the monitoring is performed by analyzing the strains measured using an optical fiber Bragg grating (FBG) sensor. This sensor has many advantages over a standard accelerometer. It can be mounted close to the monitored bearing due to its small dimensions and flexibility. The proximity of the sensor to the bearing reduces the effect of the transmission path and improves the signal-to-noise ratio. These sensors are not influenced by electromagnetic interference and do not generate such interference themselves. Moreover, it is possible to put several sensors on a single fiber. Previous studies have shown that FBG sensors can not only detect artificially implanted defects in REBs but their readings can be also processed to determine the defects size and severity. In the current study, several endurance tests were conducted. The goal of these tests was to identify and monitor the initiation and propagation of natural defect in the REBs. After the REB was disassembled, the measured fault size was in excellent agreement with the FBG result. To conclude, by using FBG sensors, we were able to continuously monitor the severity of the defect in the bearings and to estimate the defect size at different stages of the experiment.

Speakers
HA

Hasib Alian

IAF, Israel Air Force
JB

Jacob Bortman

Ben-Gurion University
DG

Dmitri Gazizulin

Ben-Gurion University
RK

Renata Klein

R.K. Diagnostics
DS

Dana Shimoni

Ben-Gurion University


Thursday May 17, 2018 11:30am - 12:00pm
Cape Lookout

12:00pm

Awards Lunch
The MFPT Society recognizes individuals' and organizations' contributions and excellence in a number of different ways:
  • The Jack Frarey Memorial Award for Excellence awarded  for outstanding contributions to the failure prevention field
  • Fellows of the MFPT Society potential fellowships are considered annually, and elected where appropriate, because of their outstanding and devoted service to the MFPT Society over an extended period of time
  • The Henry and Sallie Pusey Best Paper Award awarded for the best paper submitted
  • The Hank Hegner Best Student Paper Award & The Runner Up Student Paper Award awarded for the best student papers submitted


Thursday May 17, 2018 12:00pm - 1:30pm
Cape Charles

1:30pm

Monitoring Faults in Belt-Driven Time-Varying Machinery
Many diagnostic approaches are not readily transferable to different scales of component size and severity of service. Usefulness in general application would require conducting ?clinical? trials, such as monitoring a broad range of machines where similar fault conditions appear during service (or are introduced artificially), or showing how a technique scales across different machines. In this work, normative assessment of classical machinery diagnostic techniques is done against statistical classification methods and regression using hybrid models incorporating both physics-based and data-driven modeling. A case study is presented for fault diagnostics in belt-driven machinery under time-varying load, for different failure modes and different machine configurations.

Speakers
avatar for Michael Lipsett

Michael Lipsett

University of Alberta
MR

Mohammad Riazi

University of Alberta


Thursday May 17, 2018 1:30pm - 2:00pm
Cape Fear

1:30pm

Arc Fault Monitoring
Speakers
avatar for Chris Nemarich

Chris Nemarich

Naval Sea Systems Command (NAVSEA)


Thursday May 17, 2018 1:30pm - 2:00pm
Cape May

1:30pm

Multiscale System for Rapid NDE of Advanced Composite Products
Key words: Nondestructive Evaluation, Condition Based Maintenance, Diagnostics and Prognostics
The use of advanced composite structures is increasing due to their light weight and associated fuel savings, improved fatigue performance and corrosion resistance. Due to their inherent anisotropy, composites pose new and unique challenges in non-destructive inspection (NDI). Improved defect detection and characterization along with rapid inspection of large areas is essential for safety of resulting components and structures.

Extensive work on NDI of aerospace composites has been investigated using several NDE techniques. The NDI modality employed is dependent on whether the composite material is carbon fiber or glass fiber based. In this paper, we investigate the use of a multi-scale approach to first rapidly detect presence of an anomalous region and then use a local NDE technique for characterizing the type and severity of damage. The NDE modality chosen is guided wave (GW) ultrasonic method with measurements taken using multiple transmit-receiver pairs of sensors. Transducer configurations can be adjusted to inspect the skins, bonded joints and bent regions. The inspection can be further enhanced using ultrasonic B-scans collected at orthogonal directions. The data will be analyzed using model based data interpretation algorithms that map the obtained scans onto an image of the inspected area.

Speakers
YD

Yiming Deng

Nondestructive Evaluation Laboratory Dept. of Electrical and Computer Engineering Michigan State University
MH

Mohamood Haq

Department of Civil and Environment Engineering, Michigan State University
OK

Oleksii Karpenko

NDE Laboratory, Department of Electrical and Computer Engineering, Michigan State University
LU

Lalita Udpa

Nondestructive Evaluation Laboratory Dept. of Electrical and Computer Engineering Michigan State University


Thursday May 17, 2018 1:30pm - 2:00pm
Cape Lookout

2:00pm

A Novel Approach for Stress Cycle Analysis based on Empirical Mode Decomposition
In real industrial stress/strain analysis applications, calculating equivalent constant amplitude cycles is important. Rain flow counting is a process to obtain equivalent constant amplitude cycles. The method is designed to count reversals in accordance with the material's stress/strain relationship including hysteresis loops. However, rain flow counting needs identify the peaks/valleys in the collected sensor signal and is sensitive to noise.  In this paper, a novel approach is developed to counting the fatigue cycles.  The approach first uses empirical mode decomposition method to decompose the signal adaptively. Then a systematic count method is developed to calculate the cycles based on the decomposed signal components. The effectiveness and the performance of the developed method are compared with the rain flow counting algorithm with simulated data with different level of noise added to the signal.

Speakers
GC

Gilbert Chahine

National Oilwell Varco
RL

Ruoyu Li

National Oilwell Varco
AM

Ali Marzban

National Oilwell Varco
JP

Jing Ping

National Oilwell Varco


Thursday May 17, 2018 2:00pm - 2:30pm
Cape Fear

2:00pm

The Application of the Condition-Based Maintenance for Lithium Battery in the Space Station
With the development of science and technology?the human race live in the space station will become a regular occurrence. Lithium battery as a energy product must be able to be repaired in time. Because when the lithium battery failed, there is a chain reaction caused by lack of electricity. especially the problems about security. With the development of sensors and integrated electronics, the state of lithium batteries can be monitored by signal acquisition circuit and sensors. The health state of lithium batteries can be judged by these characteristic variables.
Divider resistance and differential amplifier used to collect the total voltage and single voltage of lithium battery. To make sure that the acquired data for the differential amplifier is reliable, differential amplifier used aerospace grade components. The lithium battery charge and discharge current are collected by hall sensors. The collection of temperature data is realized by the thermistor. The information collected contains the digital quantity and the analog quantity.
In this paper, the fault characteristics of lithium battery are extracted to judge the local health state. The actual electrical characteristics of lithium batteries and method of judging the threshold of characteristic can be combined to realize determination of the maintenance time. The lithium battery maintenance is guided by using the best ways to achieve the best effect of maintenance at the best time.

Speakers
YD

Yu Dequan

Chinese Academy of Sciences


Thursday May 17, 2018 2:00pm - 2:30pm
Cape May

2:00pm

Use of Non Destructive techniques to analyse the fatigue of metals
Fatigue damage of steel structures has always been a safety issue as fatigue cracks occur at a lower stress than design stress. Timely identification of the damage is important to assure the safety operation. Therefore, using Non-Destructive Technique approaches such as acoustic emission and electrical resistivity to detect these damage initiation in metals are important. This paper presents the results on the study of fatigue-crack propagation of high strength metals using Modal AE along with Electrical Resistance (ER) and Digital Image Correlation (DIC) on test specimens with an edge notch. The results suggest a correlation of the acoustic emission events and potential drop along with DIC images in predicting the failure before occurring. It is anticipated that these Non-Destructive testing methods can be utilized as an efficient approach to inspect and monitor fatigued damaged structures.

Speakers
MK

Manigandan Kannan

University of Akron
GM

Gregory Morscher

University of Akron
MP

Michael Presby

University of Akron
SS

Sulochana Shrestha

University of Akron
YS

Yogesh Singh

University of Akron


Thursday May 17, 2018 2:00pm - 2:30pm
Cape Lookout

2:30pm

Prediction of vibration responses of a dual-rotor-blade-casing system with blade-tip rubbing
Based on finite element (FE) method, a dynamic model of a dual-rotor-blade-casing coupling system is established. The bearings, bladed rotors, disks and casing are simulated using spring-damping element, beam element, shell element and beam element suitable for curved beam separately (the casing stiffness and damping using spring-damping element). The blade-casing rub-impact is simulated using point-point contact elements; here the corresponding nodes of the blades and the casing are identified as the contact points. These contact elements describe the coupling of the rotor and the casing by the augmented Lagrangian method. Complicated vibration responses of the coupled system are analyzed by time-domain waveform, amplitude spectrum and orbit.

Speakers
DJ

Dongxiang Jiang

Tsinghua University
NW

Nanfei Wang

Tsinghua University


Thursday May 17, 2018 2:30pm - 3:00pm
Cape Fear

2:30pm

The Magnetic Analysis of Electrical Motors
Diagnosing electrical motors based on magnetic field signatures is a subject of interest in academic research, but has yet to be adopted widely.  In this paper, we will present an innovative IoT monitoring system based on magnetic field detection and analysis.  
The system itself is composed of magnetic sensors that measure radial and axial stray fields emitted from the motor. Our analysis shows that these fields correlate with air-gap magnetic fields, and therefore, provide valuable information on the motor's condition. A careful analysis of a magnetic field?s time signal and its spectra may provide indications of various electrical and mechanical faults. These faults may be less obvious when using conventional methods of fault detection such as vibration analysis. Additionally, we will present a novel approach for energy analysis that is based on the vibration, magnetic, and temperature signature of the motor.

Speakers

Thursday May 17, 2018 2:30pm - 3:00pm
Cape May

2:30pm

Particle Filtering techniques for Prognosis of Fatigue delamination Growth using multi-sensory NDE methods
Key words: Nondestructive Evaluation, Condition Based Maintenance, Diagnostics and Prognostics
With increasing use of fiber reinforced polymer (FRP) composites in several industries such as aviation, automotive and construction, effective diagnosis and prognosis of composite structures have become an extremely critical task in recent years. The primary goal of prognosis is to accurately predict failure threshold using inspection data from early stages of degradation. Despite outstanding qualities such as light-weight, high specific stiffness and strength, components made of composite materials are often vulnerable to damages caused due to fatigue or external impacts which compromise their performance and hence propel the need of their periodic inspection by robust non-destructive evaluation (NDE) techniques. Overall accurate health prognosis is critical for condition-based-maintenance (CBM) and for reducing life-cycle costs by taking full advantage of the remaining useful life (RUL) of a component.

In this paper, growth of delamination caused by fatigue loading in GFRP Mode I composite samples was periodically inspected by optical transmission scanning (OTS) after regular intervals of cyclic loading. In addition to OTS, measurements were obtained using guided waves (GW) which were excited and sensed by surface-mounted piezoelectric sensors (PZT). An integrated prognostics framework for estimation of damage propagation was proposed which utilizes physical model based on Paris law and CBM data obtained from both NDE techniques. A Bayesian method based on particle filtering was implemented to estimate model parameters using damage-sensitive features extracted from the two sets of NDE measurements. Material and model uncertainties were taken into account during update of model parameters and RUL computation. Prediction from both methods were compared against ground truth observed directly using a digital camera image of cross-section of delaminated sample. Results demonstrating the prognostic capabilities of the two NDE methods on inspection of GFRP composites will be presented.

Speakers
PB

Portia Banerjee

Nondestructive Evaluation Laboratory Dept. of Electrical and Computer Engineering Michigan State University
YD

Yiming Deng

Nondestructive Evaluation Laboratory Dept. of Electrical and Computer Engineering Michigan State University
MH

Mohamood Haq

Department of Civil and Environment Engineering, Michigan State University
RP

Rajendra Prasath Palanisamy

Department of Civil and Environment Engineering, Michigan State University
LU

Lalita Udpa

Nondestructive Evaluation Laboratory Dept. of Electrical and Computer Engineering Michigan State University


Thursday May 17, 2018 2:30pm - 3:00pm
Cape Lookout