GB2277151A - Machine monitoring using neural network - Google Patents

Machine monitoring using neural network Download PDF

Info

Publication number
GB2277151A
GB2277151A GB9307096A GB9307096A GB2277151A GB 2277151 A GB2277151 A GB 2277151A GB 9307096 A GB9307096 A GB 9307096A GB 9307096 A GB9307096 A GB 9307096A GB 2277151 A GB2277151 A GB 2277151A
Authority
GB
United Kingdom
Prior art keywords
data
vibration
machine
system described
neural network
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
GB9307096A
Other versions
GB2277151B (en
GB9307096D0 (en
Inventor
T J Harris
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Brunel University
Original Assignee
Brunel University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Brunel University filed Critical Brunel University
Priority to GB9307096A priority Critical patent/GB2277151B/en
Publication of GB9307096D0 publication Critical patent/GB9307096D0/en
Publication of GB2277151A publication Critical patent/GB2277151A/en
Application granted granted Critical
Publication of GB2277151B publication Critical patent/GB2277151B/en
Anticipated expiration legal-status Critical
Expired - Fee Related legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Programme-control systems
    • G05B19/02—Programme-control systems electric
    • G05B19/18—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form
    • G05B19/406—Numerical control [NC], i.e. automatically operating machines, in particular machine tools, e.g. in a manufacturing environment, so as to execute positioning, movement or co-ordinated operations by means of programme data in numerical form characterised by monitoring or safety
    • G05B19/4063—Monitoring general control system
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M13/00—Testing of machine parts
    • G01M13/02—Gearings; Transmission mechanisms
    • G01M13/028—Acoustic or vibration analysis
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/33—Director till display
    • G05B2219/33027—Artificial neural network controller
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/33—Director till display
    • G05B2219/33037—Learn parameters of network offline, not while controlling system
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/37—Measurements
    • G05B2219/37122—Signal analyser
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/37—Measurements
    • G05B2219/37434—Measuring vibration of machine or workpiece or tool

Abstract

An automatic machine health monitoring system which combines the use of vibration analysis and self organising map neural networks to facilitate fault detection and diagnosis with training examples taken only from the machine when in a good condition. Component specification data is used to determine a set of key frequencies, the amplitudes of which are used as the input parameters of the self organising map network. The networks outputs are in the form of individual distance values from each of the key frequencies. The control computer uses this information to detect and diagnose faults in the machine under examination. <IMAGE>

Description

NEURAL NETWORK BASED MACHINE HEALTH MONITORING SYSTEM This invention relates to the use of neural networks in the field of machine health monitoring.
Machine health monitoring or condition monitoring of machinery is commonly carried out by vibration analysis. Vibration detecting probes are attached or placed on machine casings and the vibrations detected converted to spectral information and analysed by experts. The condition of the machine can be determined by the overall vibration level and by the level of particular vibration frequencies. These measures are taken against readings made when the machine was known to be in good condition. Analysis of individual frequency levels can by used to diagnose the fault since within the machine different components will vibrate at different frequencies. Hence using a database of component specifications and a reading of the machine speed it is possible to isolate the cause of a particular vibration.Conversely it is possible to use the specification and speed information to select key vibration frequencies to examine in order to check the condition of indi vidual components.
A variety of neural network systems have been suggested and researched to carry out the condition monitoring and vibration analysis tasks automatically. The most suitable neural network architecture being the Self Organising Map or Kohonen Network.
This network type can use frequency values as its input and train to either output a general condition value similar to the overall vibration level, or if given examples cf known faults can detect the onset of that fault.
Fault identification using a neural network has the disadvantage that examples of fault conditions are required for training. This training data can only be collected if the machine to be monitored develops a fault since it is very often highly undesirable or impossible to specially run machinery with faults in order to collect example data.
According to the present invention there is provided a system which combines the use of a database of component specifications with a self organising map network to facilitate the automatic fault diagnosis from vibration spectra with only training examples taken from the vibration of the machine when known to be in a good condition.
The invention is now described with reference to the accompanying drawing in which: Figure 1 shows a block diagram of the main system components Referring to figure 1 the analysis system comprises a database of either component specifications within the machine to be monitored from which a set of key frequencies (as described above) can be established by the control computer with reference to the machines running speed, or a set of previously established key frequencies for the machine to be monitored. The relative amplitudes of the established key frequencies are used by the self organising map network as input dimensions.
During the networks training process vibration data is collected by the data collection system (shown in figure 1) and spectrum analysis carried out on it. A measure of the machines running speed is also collected. The control computer uses the key frequencies derived from the database and the measure of running speed to select parts of the vibration spectrum to use as a multi dimensional input vector to the self organising map network.
Training data is collected and used in this manner while the machine is run in its normal operating conditions. This can include and variations in load, running speed and other operating parameters. The network is trained using this data.
The self organising map network is configured to provide an output which consists of individual distance measures for each dimension, from the multi dimensional input vector, described above, to the nearest/winning neuron in the network.
When monitoring the machine, these networks outputs (described above) are passed to the control software and provide a measure of normality for each of the key frequencies in the spectrum. The control software using this information can not only detect a fault but diagnose the fault.

Claims (10)

1 A machine health monitoring system which combines the use of vibration analysis techniques and self organising map neural networks to perform fault detection and diagnosis.
2 The use of a tachometer and vibration sensors as the input devices from machinery subject to vibration to the system described in claim 1.
3 The use of vibration sensors as the input device from machinery subject to vibration, whose operating speed is constant, to the system described in claim 1.
4 The use of a separate data collection and storage system with data transfer to the analysis system described in claims 1, 2 and 3 by some telemetric link.
5 The use of dedicated data collection it data transfer to the analysis system described in claims i, 2and
6 The system illustrated in figure 1 to implement claim 1.
7 Emulations of the system described in claims l to 6 inclusive.
Amendments to the claims have been filed as follows
8. A method of monitoring machine health which includes the steps of sensing the normal running of a healthy machine at a given speed, storming data derived by said sensing, and using said data in a neural network to perform fault detection and diagnosis.
9. A method as claimed in claim 8, wherein said data concerns vibration amplitudes, including the step of analysing the data to determine key frequencies and amplitudes for use by the neural network.
10. A method as in claim 8 or 9 wherein data is collected at a number of different speeds.
GB9307096A 1993-04-05 1993-04-05 Neural network based machine health monitoring system Expired - Fee Related GB2277151B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
GB9307096A GB2277151B (en) 1993-04-05 1993-04-05 Neural network based machine health monitoring system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
GB9307096A GB2277151B (en) 1993-04-05 1993-04-05 Neural network based machine health monitoring system

Publications (3)

Publication Number Publication Date
GB9307096D0 GB9307096D0 (en) 1993-05-26
GB2277151A true GB2277151A (en) 1994-10-19
GB2277151B GB2277151B (en) 1997-06-25

Family

ID=10733369

Family Applications (1)

Application Number Title Priority Date Filing Date
GB9307096A Expired - Fee Related GB2277151B (en) 1993-04-05 1993-04-05 Neural network based machine health monitoring system

Country Status (1)

Country Link
GB (1) GB2277151B (en)

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0829809A1 (en) * 1996-09-16 1998-03-18 ABBPATENT GmbH Method for process-visualization
EP0897155A1 (en) * 1997-08-13 1999-02-17 ABBPATENT GmbH Method for controlling processes
WO2002003041A1 (en) * 2000-07-05 2002-01-10 Rolls-Royce Plc Monitoring the health of a power plant
US6999884B2 (en) 2003-01-10 2006-02-14 Oxford Biosignals Limited Bearing anomaly detection and location
EP1906282A3 (en) * 2006-09-29 2008-10-01 Matsushita Electric Works, Ltd. Device for estimating machining dimension of machine tool
CN102721941A (en) * 2012-06-20 2012-10-10 北京航空航天大学 Method for fusing and diagnosing fault information of circuit of electric meter on basis of SOM (self-organized mapping) and D-S (Dempster-Shafer) theories
CN102854006A (en) * 2011-06-22 2013-01-02 霍尼韦尔国际公司 Severity analysis apparatus and method for shafts of rotating machinery
CN103267652A (en) * 2013-05-24 2013-08-28 北京工业大学 Intelligent online diagnosis method for early failures of equipment
US8601322B2 (en) 2005-10-25 2013-12-03 The Trustees Of Columbia University In The City Of New York Methods, media, and systems for detecting anomalous program executions
US8694833B2 (en) 2006-10-30 2014-04-08 The Trustees Of Columbia University In The City Of New York Methods, media, and systems for detecting an anomalous sequence of function calls
CN103852255A (en) * 2013-11-15 2014-06-11 北京能高自动化技术股份有限公司 Typical transmission fault intelligent diagnosis method based on neural network wind generating set
EP3073337A1 (en) * 2015-03-24 2016-09-28 Fabian Sacharowitz Actuating device for fittings with measurement and processing device for the fittings impact sound
CN106198005A (en) * 2016-08-30 2016-12-07 北京首钢自动化信息技术有限公司 A kind of multiaxis shakes the method that monitoring point blower fan Shaft chattering fluctuation judges and processes
CN107576491A (en) * 2017-10-18 2018-01-12 河海大学 A kind of breaker mechanical fault recognition method
CN107704712A (en) * 2017-10-31 2018-02-16 郑州恩普特科技股份有限公司 Mechanical failure diagnostic method and system based on vector spectrum feature extraction
CN109901537A (en) * 2019-03-18 2019-06-18 北京大通惠德科技有限公司 Mechanical equipment method for monitoring operation states and system for edge calculations side
CN110940518A (en) * 2019-11-27 2020-03-31 中北大学 Aerospace transmission mechanism analysis method based on fault data

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2007022454A2 (en) 2005-08-18 2007-02-22 The Trustees Of Columbia University In The City Of New York Systems, methods, and media protecting a digital data processing device from attack
US9495541B2 (en) 2011-09-15 2016-11-15 The Trustees Of Columbia University In The City Of New York Detecting return-oriented programming payloads by evaluating data for a gadget address space address and determining whether operations associated with instructions beginning at the address indicate a return-oriented programming payload

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE4012278A1 (en) * 1989-04-14 1990-10-18 Hitachi Ltd Diagnostic system for installations with special operating conditions - uses neural network model for efficient, objective diagnosis
JPH03154896A (en) * 1989-11-13 1991-07-02 Toshiba Corp Diagnostic system for abnormality of recirculation pump of nuclear reactor
EP0569994A2 (en) * 1992-05-14 1993-11-18 Mitsubishi Jukogyo Kabushiki Kaisha Vibration detection and reduction system and vibration sensors for use in micro-gravity environment

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE4012278A1 (en) * 1989-04-14 1990-10-18 Hitachi Ltd Diagnostic system for installations with special operating conditions - uses neural network model for efficient, objective diagnosis
JPH03154896A (en) * 1989-11-13 1991-07-02 Toshiba Corp Diagnostic system for abnormality of recirculation pump of nuclear reactor
EP0569994A2 (en) * 1992-05-14 1993-11-18 Mitsubishi Jukogyo Kabushiki Kaisha Vibration detection and reduction system and vibration sensors for use in micro-gravity environment

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
WPI Accession No90-321797/43 & DE4012278A (HITA) 18/10/90 (see abstract) *
WPI Accession No91-235535/32 & JP3154896A (Toke) 2/7/91 (see abstract) *

Cited By (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0829809A1 (en) * 1996-09-16 1998-03-18 ABBPATENT GmbH Method for process-visualization
EP0897155A1 (en) * 1997-08-13 1999-02-17 ABBPATENT GmbH Method for controlling processes
US6314413B1 (en) 1997-08-13 2001-11-06 Abb Patent Gmbh Method for controlling process events using neural network
WO2002003041A1 (en) * 2000-07-05 2002-01-10 Rolls-Royce Plc Monitoring the health of a power plant
US6928370B2 (en) 2000-07-05 2005-08-09 Oxford Biosignals Limited Health monitoring
US6999884B2 (en) 2003-01-10 2006-02-14 Oxford Biosignals Limited Bearing anomaly detection and location
US7191073B2 (en) 2003-01-10 2007-03-13 Oxford Biosignals Limited Bearing anomaly detection and location
US8601322B2 (en) 2005-10-25 2013-12-03 The Trustees Of Columbia University In The City Of New York Methods, media, and systems for detecting anomalous program executions
US7778724B2 (en) 2006-09-29 2010-08-17 Panasonic Electric Works Co., Ltd. Device for estimating machining dimension of machine tool
EP1906282A3 (en) * 2006-09-29 2008-10-01 Matsushita Electric Works, Ltd. Device for estimating machining dimension of machine tool
US8694833B2 (en) 2006-10-30 2014-04-08 The Trustees Of Columbia University In The City Of New York Methods, media, and systems for detecting an anomalous sequence of function calls
CN102854006A (en) * 2011-06-22 2013-01-02 霍尼韦尔国际公司 Severity analysis apparatus and method for shafts of rotating machinery
CN102721941B (en) * 2012-06-20 2014-08-20 北京航空航天大学 Method for fusing and diagnosing fault information of circuit of electric meter on basis of SOM (self-organized mapping) and D-S (Dempster-Shafer) theories
CN102721941A (en) * 2012-06-20 2012-10-10 北京航空航天大学 Method for fusing and diagnosing fault information of circuit of electric meter on basis of SOM (self-organized mapping) and D-S (Dempster-Shafer) theories
CN103267652B (en) * 2013-05-24 2015-05-20 北京工业大学 Intelligent online diagnosis method for early failures of equipment
CN103267652A (en) * 2013-05-24 2013-08-28 北京工业大学 Intelligent online diagnosis method for early failures of equipment
CN103852255A (en) * 2013-11-15 2014-06-11 北京能高自动化技术股份有限公司 Typical transmission fault intelligent diagnosis method based on neural network wind generating set
EP3073337A1 (en) * 2015-03-24 2016-09-28 Fabian Sacharowitz Actuating device for fittings with measurement and processing device for the fittings impact sound
CN106198005A (en) * 2016-08-30 2016-12-07 北京首钢自动化信息技术有限公司 A kind of multiaxis shakes the method that monitoring point blower fan Shaft chattering fluctuation judges and processes
CN106198005B (en) * 2016-08-30 2018-08-03 北京首钢自动化信息技术有限公司 A kind of multiaxis monitoring point wind turbine Shaft chattering fluctuation that shakes judges method with processing
CN107576491A (en) * 2017-10-18 2018-01-12 河海大学 A kind of breaker mechanical fault recognition method
CN107704712A (en) * 2017-10-31 2018-02-16 郑州恩普特科技股份有限公司 Mechanical failure diagnostic method and system based on vector spectrum feature extraction
CN109901537A (en) * 2019-03-18 2019-06-18 北京大通惠德科技有限公司 Mechanical equipment method for monitoring operation states and system for edge calculations side
CN110940518A (en) * 2019-11-27 2020-03-31 中北大学 Aerospace transmission mechanism analysis method based on fault data
CN110940518B (en) * 2019-11-27 2021-08-24 中北大学 Aerospace transmission mechanism analysis method based on fault data

Also Published As

Publication number Publication date
GB2277151B (en) 1997-06-25
GB9307096D0 (en) 1993-05-26

Similar Documents

Publication Publication Date Title
GB2277151A (en) Machine monitoring using neural network
Betta et al. A DSP-based FFT-analyzer for the fault diagnosis of rotating machine based on vibration analysis
US6053047A (en) Determining faults in multiple bearings using one vibration sensor
US8285498B2 (en) Method for detection and automatic identification of damage to rolling bearings
US4435770A (en) Vibration diagnosing method and apparatus for a rotary machine
Olsson et al. Fault diagnosis in industry using sensor readings and case-based reasoning
JP6557110B2 (en) Condition diagnosis apparatus and program
JPH10176949A (en) Apparatus and method for testing mechanical component using neural network processing vibration data analysis
CN108363952A (en) Diagnostic device
WO2006120550A2 (en) Vehicle vibration analyzer
Harris A Kohonen SOM based, machine health monitoring system which enables diagnosis of faults not seen in the training set
ATE301318T1 (en) SYSTEM, DEVICE AND METHOD FOR DIAGNOSING A FLOW SYSTEM
CN107291475B (en) Universal PHM application configuration method and device
CN111964909A (en) Rolling bearing operation state detection method, fault diagnosis method and system
CN110553789A (en) state detection method and device of piezoresistive pressure sensor and brake system
De Almeida et al. New technique for evaluation of global vibration levels in rolling bearings
Betta et al. A DSP-based FFT-analyzer for the fault diagnosis of rotating machine based on vibration analysis
CN109241838A (en) Speed changer quality evaluating method based on psychologic acoustics objective parameter
CN115855463A (en) Rotary mechanical equipment fault detection method and device
KR20210006832A (en) Method and apparatus for machine fault diagnosis
CN113392874A (en) Rail vehicle abnormal state diagnosis method and device and terminal equipment
TW202045898A (en) Analysis apparatus, analysis method, and non-transitory computer readable storage medium
Leith et al. Real time expert system for identifying rotor faults and mechanical influences in induction motor phase current
CN112594356B (en) High-speed railway and subway gear box operation monitoring and fault diagnosis system
CN110411730A (en) A kind of rotating equipment failures judgment method, system and readable storage medium storing program for executing

Legal Events

Date Code Title Description
PCNP Patent ceased through non-payment of renewal fee

Effective date: 20020405