Lee IMS Intro 2009.ppt - Center for Intelligent Maintenance

17
INTRODUCTION

Transcript of Lee IMS Intro 2009.ppt - Center for Intelligent Maintenance

I N T R O D U C T I O N
The National Science Foundation Industry/University Cooperative Research Center for Intelligent Maintenance Systems
NSF I/UCRC Since 2000
513.556.3412 513.556.3390 imscenter.net
513.556.3412 513.556.3390 imscenter.net
THE IMS CENTER The IMS Center is a leading NSF Industry/University Cooperative Research Center (I/UCRC) in the area of Prognostics
and Health Management (PHM). The Center has over twelve years of experience in developing and delivering PHM
solutions for a wide-range of applications. IMS Center’s mission is to enable products and systems to achieve and
sustain near-zero breakdown performance, and transform maintenance data into useful information for improved
productivity and asset life-cycle utilization. Since its inception, the Center has conducted over 100 successful industry
and NSF supported projects, and has attracted over 80 members from all across the globe. The IMS Center was recently
identified as the most economically impactful I/UCRC in NSF’s recent study titled Measuring the Economic Impacts of
the NSF Industry/University Cooperative Research Centers Program: A Feasibility Study. According to this study, the
Center delivered its members $846.7 Million in combined benefits over the last ten years.
IMS CENTER MEMBERSHIP There are two types of IMS Membership: full and affiliate. Full members are companies or organizations with more than
500 employees and pay a membership fee of $40,000 annually; affiliate members have less than 500 employees and
pay $12,000 annually. All members enjoy over a 30 to 1 leveraging ratio, and the opportunity to develop specific research
projects with the center, based on their interests. Membership is the highest level of collaboration.
SELECTED MEMBERSHIP BENEFITS The following represent some of the many benefits of being a member of the IMS Center:
• All IMS members will have non-exclusive and royalty
free licensing rights in using all technologies and
information developed by the Center.
• The university will waive the overhead for the
membership as the contribution to the Center.
• Company members can receive the leveraged
research results (at least 15:1 ratio) from its
membership investment.
• Company members can share the best practices and
experiences gained from the IMS Center’s testbeds
and develop partnerships.
• The Center will work with the Company to develop a
separate research contract and IP agreement based
on companys interests.
and can hire and recruit experienced IMS researchers
with great impacts.
these projects can be executed by either a research
team at the Center or with dedicated full-time on-
side researchers working at company site and assist
companies to implement IMS technologies.
NSF I/UCRC Since 2000 3
MAXIMIZING IMS MEMBERSHIP The IMS Center actively engages with member companies, satellite centers and partner institutions to advance research
and broaden the knowledge in the area of PHM. Listed below are opportunities for collaboration that the Center pursues.
Core Membership Research Projects - These projects target areas that are central furthering the IMS Center’s overall
mission and that are of interest to the majority of the Center’s members. These areas are identified by IMS researchers
and the Center’s members. Core Research Projects are presented to the members at the Industrial Advisory Board (IAB)
Meetings, held every six months. Members at these meetings have the opportunity to indicate their interest or support for
these projects, and can give targeted feedback to researchers to help guide these projects to mutually beneficial results.
Feasibility Studies - Prior to a sponsored project, a feasibility study may need to be conducted, the purpose of which is
to determine the specific needs of the member, as well as the technologies required to meet those needs. Such studies
are funded by membership funds, with the ultimate result being a member-specific sponsored project.
Member-specific Sponsored Project - When a member company has a specific research goal, or its needs go beyond
existing technologies, a member-specific sponsored project is generated. Specific intellectual property terms can be
decided upon based on the sponsors input, and in keeping with the terms of the IMS membership agreement, and the
policies of the IMS Center Site’s host institution.
Technical Assistance Agreement - A Technical Assistance Agreement (TAA) can be generated for situations in which
a member has a specific issue that can be addressed using existing technologies. For such projects, no new IP will be
created, though some customization work may be required. TAAs work well for projects involving consultation, training,
test-bed validation, existing tool deployment, and any other project involving the application of existing core technologies.
Internships - As an alternative to developing a member-specific project or technical assistance agreement, an IMS
member has the opportunity to host a researcher from the Center at their facility. Such internships are common, and
serve as an excellent way to promote collaboration, as well as to share information, experiences, technologies, etc.
The work conducted by an IMS researcher while on internship is owned solely by the host member.
Technical Training - The IMS Center offers training courses for engineers from member organizations in the use of
its prognostics methods and its Watchdog Agent Toolbox. These courses can be tailored to the interests and level of
experience of the attendees. Such courses can run from 3 days to 3 weeks, depending on the level of detail required.
Corporate Training - The IMS Center provides training for engineers and executives in Dominant Innovation: a tool for
helping organizations to identify and develop value-added services to achieve improved productivity and performance.
Particpation in this training can transform an existing business into a smart product service business.
Joint Proposal Writing - Many opportunities exist for IMS members for joint proposal writing; this is especially true for
small companies (SBIRs, STTRs, etc.) and research institutes.
NSF I/UCRC Since 2000 4
The Watchdog Agent®-enabled monitoring process begins with the identification of critical equipment. Sensors and
data acquisition systems accumulate raw performance data from this equipment. Features are then extracted using
signal processing tools. These features are then analyzed using appropriate tools specialized for health and performance
assessment. When the performance of the monitored equipment falls below a defined threshold, fault diagnosis tools can
determine the specific fault type and location. If data from previous operations exists, performance degradation can be
further analyzed by prediction tools to determine when potential failures will occur, thus making the Watchdog Agent®
a transformational technology for predicting and preventing failures for worry-free uptime.
WATCHDOG AGENT® TOOLS
CORE TECHNOLOGY: THE WATCHDOG AGENT® The Watchdog Agent® is a collection of intelligent software tools developed by the IMS Center that can be customized
for monitoring equipment and systems in many diverse applications. Monitoring in this regard can refer to health or
condition assessment, fault detection and performance prediction, among others. Monitoring results can be used in an
enterprise resource planning system to ensure the appropriate preventative actions are taken before failures can occur,
optimizing maintenance scheduling and resulting in extensive cost savings.
SIGNAL PROCESSING & FEATURE EXTRACTION
Frequency Domain Analysis Principal Component Analysis
Time-frequency Analysis Expert Extracted Features
PERFORMANCE PREDICTION
HEALTH ASSESSMENT
Self-organizing Maps Auto-Associative NN
Self-organizing Maps Hidden Markov Model
The Watchdog Agent® can ensure that the correct preventative actions are taken before failures can occur, resulting in significant cost savings.
NSF I/UCRC Since 2000 5
CLOUD-BASED PROGNOSTICS The IMS Center has developed a cloud-based solution to monitor the health of industrial robots using torque signals.
This system is designed to be scalable, and to require little to no additional instrumentation. Beyond industrial robots, this
solution could be deployed for many applications that involve a large number and/or widely distributed assets. A similar
solution for bearing health monitoring has also been developed that integrates the Center’s virtual bearing concept.
M1 C-320-GNC
C-320-GNC
HYDRAULIC COOLANT
BLADE GEARBOX
GEARBOX HYRDAULIC
NSF I/UCRC Since 2000 6
MOBILE HEALTH MONITORING APPLICATIONS The development of the Center’s cloud-based pronositcs platform has enabled the creation of the Center’s first mobile
app-based health monitoring solution. The IMS Center’s Mobile Health Monitoring Application can provide users with
flexibility as well as functionality. This App presents a global view of the user’s facility or production line while also allowing
the user to drill down to see the health status of each individual machine and each component being monitored.
100%
LOGIN
100%
BLADE
GEARBOX
GEARBOX
IMS Center M6 C-1000-GNC
0.92 6/13
1m ago1m ago
100%9:41 AM
IMS CENTER APP DEMO With this App, personnel can have access to the current status of the assets within
their facility, and can receive critical notifications of pending failures regardless of time
or location. This enables preventative actions to be taken before costly failures—failures
that otherwise could be overlooked—can occur. This video shows a demonstration
version of the Center’s App in action using data acquired from one of the Center’s own
in-house test beds.
NSF I/UCRC Since 2000 7
PATENT TITLE PATENT NUMBER & INDUSTRY
Methods for Prognosing Mechanical Systems
This Patent was awarded for novel methods for predicting when a failure will occur
in a mechanical system based on extracting features from measurement data and
selecting a prediction model based on the degradation status of the mechanical
system and a reinforcement learning model.
US – 8,301,406
This patent was awarded for novel methods for predicting the remaining-useful-life
of a wind turbine or turbine component based on the performance of wind turbines
within a cluster of turbines. These clusters are established based on performance
metrics within a turbine farm, as well as environmental conditions, age, historical
information, etc.
US – 13/674,200
Mechanical Fleet Systems
Methods and Systems for Energy Prognosis
This patent was awarded for novel methods for determining and predicting the
health condition and operating performance of a system based on non-intrusive
power consumption measurements.
Manufacturing and Industrial
Assets and Systems
Method and System for Electric Vehicle Battery Prognostics and Health Management
This patent was awarded for a novel system for managing the mobility of an
electrically-powered vehicle. This system is based upon inputs from a network of
sensors that provide details about the operating condition of the vehicle as well as
the ambient and environmental conditions. The estimates and prediction of mobility
is displayed for the user for improved decision making.
US – 20120296512
INT – PCT/US12/35136
Electric Vehicles and
Autonomously Guided Vehicles
Method and System for Prognostics & Health Management Based on Cloud Computing
This provisional patent was filed based on a novel method for assessing and pre-
dicting health and performance of a system utilizing advances in cloud computing,
cyber-physical systems, modeling and prognostics.
US – 61/509,945
Quick-test Method for Battery State of Charge Estimation
A provisional patent application has been submitted for a novel method for rapidly
testing and determining the initial quality and potential life-cycle of a battery.
Provisional Patent
CENTER IMPACTS MEMBER IMPACTS BROADER IMPACTS
Through working with ALSTOM, IMS
researchers have gained insight into
the challenges and opportunities in
railway industry and have shared
their knowledge in Prognostics and
Health Management with ALSTOM
IMS has also integrated the analysis
functions into ALSTOM’s software
platform, significantly improving the
Smart Maintenance Capabilities of
The approach developed on this
project has the potential to not
only save ALSTOM significant
safety. Through publications and
community, further increasing its
potential for creating impacts.
AMD SEMICONDUCTORS 2006 - 2008
has enabled researchers to expand
the applicability of IMS technologies
and acquire additional research
opportunities in the semiconductor
SiSoc and Applied Materials.
able to incorporate the Center’s
technologies into its production lines.
In addition, AMD hired one of the
Center’s most gifted researchers and
ISMI was able to generate interest in
IMS work among its members.
The semiconductor industry is
economy. The Center’s work in this
area will continue to help the US
maintain its competitive edge, bring
about new innovations and realize
substantial cost-savings.
Through collaborations with COSEN,
develop, test, validate and refine
its approach to cloud-based
saw test bed. The Center has also
constructed an App for machine
health monitoring for machine health
monitoring that continues to be
developed for additional applications.
developed by the Center for its
band saws, and from the platform
created to deliver those tools
via the cloud and the develop
App. Cosen continues working
on commercializing this work
customers.
mobile App development conducted
additional platforms, and to provide
solutions that can be accessed from
virtually any location.
CENTER IMPACTS MEMBER IMPACTS BROADER IMPACTS
Through collaborations with GE
beds equipped with cutting-edge
environment understanding the role
of research in industry.
helps in making GE’s competitive
advantages sustainable in global
been hired by GE business and
research centers.
example for transferring scientific
production benefits. In addition,
many research developments and
patented, broadening the impact
GOODYEAR TIRE MONITORING 2010 - PRESENT
CENTER IMPACTS MEMBER IMPACTS BROADER IMPACTS
Collaborating with Goodyear
general pressure-temperature model
for similar components. This data
provides IMS with an opportunity to
develop model based diagnostics
and prognostics. Truck fleet
more than one hundred tires which
further enable statistical modeling for
detecting abnormal tires.
prediction methods developed by
managers and drivers optimize truck
tire maintenance actions, schedule
assign tasks to different trucks based
on their tire conditions. Implemented
in the call center of the company,
this method can reduce maintenance
cost, extend average usage life of
tires, and reduce unexpected truck
breakdown caused by tires.
The tire condition assessment
and anomaly detection algorithms
truck engines, wind turbines, and
EV batteries. This method is suitable
for large machine fleets where data
from each machine may be limited
due to sensor availability and/or
connectivity.
HARLEY-DAVIDSON MANUFACTURING 2001 - 2008
This collaborative effort resulted
Agent® prognostics platform into
an existing industrial monitoring
system. A standardized approach
has been established based on this
work, giving the Center the ability to
rapidly deploy its technologies on
multiple platforms.
Agent® platform made possible
real-time health monitoring and
prognostics for multiple networked
cost effective production line at
Harley-Davidson, as indicated in
be…Predictable in Lean Tools for
Maintenance & Reliability magazine.
to most of large scale factories with
a number of machine tools and
similar machines. The techniques
enterprise asset management by
accurate machine health information.
CENTER IMPACTS MEMBER IMPACTS BROADER IMPACTS
The advance techniques developed
in collaboration with Komatsu
the application of prognostics
develop new fleet-based approaches
that have delivered additional
The developed techniques through
have given Komatsu the ability to
transform data into information
technologies and methods in
Komatsu’s business units.
learn more recent prognostic efforts
made by their global competitors.
The methodology developed can be
leveraged in many industries with
widely distributed assets, such as
automotive, heavy-duty construction
among others.
PROJECT HIGHLIGHTS
NATIONAL INSTRUMENTS
In collaboration with National
Instruments, the IMS Center
Watchdog Agent® Prognostics
offer its tools, but has also provided
researchers with the experience
used by its customers to create
health monitoring solutions. This
addition customers to its LabVIEW
platform, and to its hardware
solutions that utilize this platform.
The development of the
and otherwise to access the Center’s
technologies. In addition, this toolkit
provides IMS Center members a
commercial version of the Center’s
tools for use in their operation and in
projects with the Center.
2009 - PRESENT
Through this effort, the Center
established a successful partnership
University of Cincinnati. Nanoworld
multidisciplinary collaboration.
Hannifin with the ability to provide
commercial products and services
monitoring of their customers
has the potential to be expanded
to additional business units within
Parker Hannifin.
by Parker Hannifin and the research
team at the University of Cincinnati,
which included multi-disciplinary
and Nanoworld Laboratory. This
for the masters thesis of one of the
members of this research team.
PROJECT HIGHLIGHTS
SIEMENS MACHINERY 2007 - 2009
Under the collaboration with
developed a reconfigurable “Plug-
n-Prognose” black box which
for practitioners to assess and
predict the performance of critical
components. A patent was awarded
for this unique technology.
Center, Siemens was able to develop
a new service for its customers,
strengthening SIEMENS product in
SIEMENS hired Dr. Leo Liao, one of
the Center’s most gifted researchers.
The patented technology provides
prognostics information for rotary
TOYOTA MANUFACTURING 2005 - 2008
The Toyota facility (TMMK) in
Lexington, KY has provided the
Center with several real-world test
beds involving such critical assets as
compressor and industrial robots, in
order to validate Watchdog Agent®
tools and algorithms. The company
setup a robot test cell within their
production floor, and allocated two
robots for experimentation that were
used by IMS researchers.
With the IMS collaboration,
compressors with the surge
protection and avoidance capability.
This functionality allowed the
expenses annually. With regard to
the robot project, production floor
health maps were developed in order
to determine production bottlenecks
to prioritize maintenance activities.
consumption due to degrading/
investigated. With the use of
Watchdog Agent® tools, such
identified and controlled.
WIND TURBINES WIND ENERGY 2008 - PRESENT
CENTER IMPACTS MEMBER IMPACTS BROADER IMPACTS
Through collaborations with
wind turbines from both in-field and
test beds. This has allowed for the
establishment of a systematic health
assessment and fault diagnosis
demonstration platforms for turbine
farm level monitoring.
production efficiency and operation
optimization of maintenance and
approach can be used to reduce
maintenance costs, increase uptime,
catastrophic failures of turbines.
The multi-regime modeling approach
systems with dynamic working
vibration analysis can be applied to
other rotary machines and the fleet-
based prognosis can be applied to
build the baseline model of machine
fleets for criticality prioritization and
collaborative diagnosis based on
CENTER IMPACTS MEMBER IMPACTS BROADER IMPACTS
Through collaborations with its
a Smart Battery sysem by integrating
the intelligent tools to customize
battery performance management.
hybrid and electric vehicle industries.
The integration of Smart Battery
with a Battery Management System
can enhance capabilities of EV
manufacturers to diagnose abnormal
pack, and predict the remaining
useful life based on driving mode and
environment conditions.
batteries from robots and AGVs to
electronic devices in a number of
industries such as aerospace and
consumer electronics.
systems for acquisition of various
health and activity related data,
the IMS Center is developing
information-driven patient profiles
clinical decision making and resource
scheduling for patients.
term care facility, will be able to
predict when residents need various
equipment and resources. The Daniel
Drake Medical Center for Post-
Acute Care, a rehabilitation facility,
will be able to make more informed
decisions and predict recovery rates.
By implementing the Watchdog
can be drawn about how to predict
dynamic patient health, which can
later be applied to other dynamic
systems.
NORTH AMERICA 21st Century Systems Caterpillar GCWW ITW SCK
Advantech Chevron GE Aviation Johnson Controls Siemens TTB
API Cisco Genex Kistler Spirit Aero
AMD Coherix GM LAM Research Toyota
AMP Electric Vehicles Daimler-Chrysler Goodyear McKinsey & Co. United Technologies
Applied Materials Eaton Harley-Davidson Montronix USPS
ARL EDAptive Honeywell National Instruments We Energies
Avetec Emerson / Fisher HRL P&G Woodward
Boeing ETAS Idaho Natl Lab Prometec TechSolve
BorgWarner Festo Ingersoll Rand Parker Hannifin
Bosch Ford Intel Pratt Whitney
Bosch Rexroth FTI Inteligistics Rockwell
SOUTH AMERICA SENAI-CETA
Baoshan Steel Dongling Komatsu OMRON Shanghai Electric
Beijing Shenzhou Hitachi Korea Aerospace Univ. PMC Sinovel
CEI HIWIN MIRDC Samsung Tongtai
CEPREI Huazhong NC Mitsubishi SANY Toshiba
China Ship IDEAS (III) National Taiwan Univ. ServTech Zhen Ding
Cosen Saws
EUROPE ALSTOM FMTC FORCAM KONE Tekniker
INTERNATIONAL MEMBERS AND COLLABORATORS Over the past twelve years, the IMS Center has worked with over 80 members from over 10 countries, including the US,
Taiwan, Japan, China, Germany, Brazil, Finland, Spain, Belgium and France. In addition, the Center has collaborated with
research institutions and organizations throughout the world. These collaborations give the Center a global perspective,
which further broadens the potential applicability of its solutions, and enhances the development of the Center’s
researchers and staff.
INTERNATIONAL COLLABORATORS
COUNTRY INSTITUTION
M O
Brazil Universidade Federal do Rio Grande do Sul X X
Canada University of Toronto X
China BUAA X X
China China University of Petroleum-Beijing X
China North China Electrical Power University X
China Shanghai Jiao Tong University X X X X
China Tsinghua University X
Egypt Nile University X
Germany Technische Universität Berlin X X
Germany University of Magdeburg X
Hongkong City University of Hongkong X
Hongkong Polytechnic University of Hongkong X
Japan University of Tokyo X
Norway University of Stavenger X
Singapore SIMTech – Satellite Center X X
Sweden Lulea University X X
Switzerland École Polytechnique Fédérale de Lausanne X X
Taiwan Industrial Technology Research Institute X X
Taiwan National Taiwan University X
Taiwan National Taiwan University of S&T X
Taiwan National Yunling University of S&T X X
Taiwan Precision Machinery R&D Center X X
UK University of Manchester X
UK Cranfield University X
Jay Lee
University of Cincinnati
Intelligent Maintenance Systems (IMS)