CN112464565B - Equipment fault early warning method integrating intelligent modeling and fuzzy rules - Google Patents

Equipment fault early warning method integrating intelligent modeling and fuzzy rules Download PDF

Info

Publication number
CN112464565B
CN112464565B CN202011382067.8A CN202011382067A CN112464565B CN 112464565 B CN112464565 B CN 112464565B CN 202011382067 A CN202011382067 A CN 202011382067A CN 112464565 B CN112464565 B CN 112464565B
Authority
CN
China
Prior art keywords
fault
early warning
regression
parameter
symptom
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.)
Active
Application number
CN202011382067.8A
Other languages
Chinese (zh)
Other versions
CN112464565A (en
Inventor
李刚
仇晨光
曹帅
王亚欧
于国强
陈鑫
陈波
郑建勇
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.)
Southeast University
State Grid Jiangsu Electric Power Co Ltd
Jiangsu Fangtian Power Technology Co Ltd
Original Assignee
Southeast University
State Grid Jiangsu Electric Power Co Ltd
Jiangsu Fangtian Power Technology Co Ltd
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 Southeast University, State Grid Jiangsu Electric Power Co Ltd, Jiangsu Fangtian Power Technology Co Ltd filed Critical Southeast University
Priority to CN202011382067.8A priority Critical patent/CN112464565B/en
Publication of CN112464565A publication Critical patent/CN112464565A/en
Application granted granted Critical
Publication of CN112464565B publication Critical patent/CN112464565B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/048—Activation functions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0635—Risk analysis of enterprise or organisation activities
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/20—Administration of product repair or maintenance
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/08—Thermal analysis or thermal optimisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
    • G06F2218/12—Classification; Matching
    • G06F2218/16—Classification; Matching by matching signal segments
    • G06F2218/20—Classification; Matching by matching signal segments by applying autoregressive analysis

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Human Resources & Organizations (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • Evolutionary Computation (AREA)
  • Strategic Management (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Business, Economics & Management (AREA)
  • General Health & Medical Sciences (AREA)
  • Tourism & Hospitality (AREA)
  • Marketing (AREA)
  • General Engineering & Computer Science (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Software Systems (AREA)
  • Mathematical Physics (AREA)
  • Quality & Reliability (AREA)
  • Computational Linguistics (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Biophysics (AREA)
  • Computing Systems (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Educational Administration (AREA)
  • Development Economics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • Primary Health Care (AREA)
  • Testing And Monitoring For Control Systems (AREA)

Abstract

The invention discloses an equipment fault early warning method, system and storage medium fusing intelligent modeling and fuzzy rules, wherein the method comprises the following specific steps: establishing a four-pipe leakage fault expert knowledge base; (2) constructing a membership function based on the distance function; (3) Obtaining a parameter symptom value and perfecting a fuzzy membership function; (4) acquiring an early warning signal; and (5) calculating the membership degree of the early warning signal and judging the fault state. According to the method, on the premise of early warning, the specific fault mode is judged according to the expert database and the fuzzy rule, and the accuracy and the effectiveness of equipment fault early warning are improved.

Description

Equipment fault early warning method integrating intelligent modeling and fuzzy rules
Technical Field
The invention relates to a fuzzy membership function identification method, in particular to an equipment fault early warning method integrating intelligent modeling and fuzzy rules.
Background
In recent years, with continuous deep innovation of the power market, the connection between the two sides of the power grid is tighter, various requirements on the generator set are also improved continuously, especially the deep peak shaving of the thermal power unit breaks through the operation space of the former generator set, and the uncertainty of the operation of the generator set is greatly improved. The power grid and power generation enterprises expect that the unit can keep continuous and stable operation, the accident occurrence rate is reduced, according to statistics, the four-pipe leakage of the boiler accounts for nearly 50% of the unit accidents and more than 60% of the boiler accidents, and the timely and accurate diagnosis of the four-pipe leakage can create very favorable conditions for the safe and stable operation of the unit. At present, operator pictures of a DCS (distributed control system) of a unit generally provide fixed value alarming and trip protection functions. However, in most cases, when the unit is operated to give an alarm, the four-pipe leakage fault has been seriously degraded, and the site can only be processed by applying for emergency shutdown with reduced load, which results in great loss to power generation enterprises and certain influence on the power grid. Therefore, in the early stage of occurrence of a failure, the failure state is determined in advance by technical means, and the focus of attention of researchers in the field is focused.
The operation data is the only source for obtaining the operation state of the unit, the hidden state information in the data is deeply mined, and a feasible and effective path is provided for the research of the unit state early warning technology. Zhang Wei deduces a reference value by mining fuzzy association rules so as to judge the state of the auxiliary machine of the fluidized bed; a learner extracts representative data to establish an early warning model of the fan, and provides a decline index for state evaluation; a learner establishes a rule base by using an association mining technology and is used for judging the forming process of equipment faults; a learner establishes a multivariate state estimation early warning model of the unit pulverizing system and provides a clustering method for constructing a state matrix; the improved fuzzy C mean value is provided by scholars for equipment fault diagnosis in the thermal slow-changing process; a learner researches a modeling method of multivariate state estimation based on density clustering, and provides a sliding window deviation degree mode for out-of-limit judgment; a learner provides a fault early warning method based on power analysis and a neural network, which is used for diagnosing a wind turbine; other studies related to early warning methods have also emerged, such as correlation analysis, neural networks, gaussian models, etc.
After the fault early warning signal is obtained, the specific fault mode is judged according to the expert database and the fuzzy rule: relevant documents and scientific research results for concluding the leakage faults of the four pipes of the boiler are summarized, and a fault diagnosis expert knowledge base for the leakage of the four pipes of the boiler is listed according to distance type fuzzy judgment rules. Meanwhile, according to standard deviations of all parameters obtained by statistical analysis of historical data, fuzzification is carried out on the parameters according to the result of comparison between regression errors of the intelligent model and 5 times of the standard deviations, and the membership degree of the current state which belongs to each leakage mode is further calculated according to a fuzzy rule so as to judge the current fault state. On the basis of a certain 1000MW thermal power simulation system, normal working conditions and related leakage faults are simulated respectively to verify the method provided by the invention.
Disclosure of Invention
The technical problems to be solved by the invention are as follows: the equipment fault early warning method can judge a specific fault mode according to an expert database and a fuzzy rule.
The technical scheme adopted by the invention is as follows: an equipment fault early warning method fusing intelligent modeling and fuzzy rules comprises the following steps:
(1) Establishing a four-pipe leakage fault expert knowledge base;
(2) Constructing a membership function based on the distance function;
(3) Obtaining a parameter symptom value and perfecting a fuzzy membership function;
(4) Acquiring an early warning signal;
(5) And calculating the membership degree of the early warning signal and judging the fault state.
Specifically, the step (1) comprises the following steps:
(11) Counting related characteristic parameters of the leakage of four pipes of the boiler;
(12) Obtaining the change relation between each symptom parameter and the fault type;
(13) And establishing a fault expert knowledge base.
Specifically, the step (2) includes:
(21) Selecting a suitable distance function as a basis;
(22) And constructing a membership function.
Further, the step (3) includes:
(31) According to the regression error of the extreme learning machine, a parameter symptom value solving method is provided;
(32) And (5) performing parameter fuzzification and perfecting a fuzzy membership function.
Specifically, the step (4) includes:
(41) Establishing a fault early warning model based on a multiple regression estimation method;
(42) And performing unit model simulation to obtain a fault early warning signal.
Specifically, the step (5) includes:
(51) Receiving an early warning signal;
(52) Calculating the membership between the early warning signal and various faults according to the constructed fuzzy membership function;
(53) And judging the current fault state according to the calculation result.
The invention achieves the following beneficial effects:
compared with other data modeling methods, the extreme learning machine has unique advantages in the aspects of model reliability and accuracy, the result obtained by fuzzy recognition is consistent with the simulated fault mode, and the accuracy and the effectiveness of equipment fault early warning are improved.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention;
FIG. 2 is a schematic diagram of health indicator trends before and after a fault.
Detailed Description
The technical scheme of the invention is further explained by combining the drawings and the specific embodiments.
As shown in fig. 1, an equipment fault early warning method fusing intelligent modeling and fuzzy rules includes the following steps:
step one, establishing a four-pipe leakage fault expert knowledge base. The fault knowledge base is the basis of fault identification, describes mode relations among various fault categories and symptom parameters, summarizes characteristic parameters related to four-pipe leakage of the boiler, and indicates the change relations among the symptom parameters and the fault types through thermodynamic calculation. And according to a five-value symptom set description method, integrating the expert knowledge of each document to obtain a fault expert knowledge base.
And step two, constructing a membership function based on the distance function. And selecting a proper distance function as a basis to construct a membership function.
The fault identification is actually based on a fault diagnosis expert database, and the current fault state is judged and identified according to a certain rule. The invention provides a novel fuzzy membership degree judging method according to regression errors of an extreme learning machine, which adopts a distance function:
Figure GDA0003798138100000041
in the formula d j (u 0 ,u j ) For a fault u to be identified 0 And typical failure mode u j The smaller the value of the distance between the two, the greater the probability of such a fault. z is a radical of j Fault symptom value for ith symptom parameter, z ij Symptom value of ith symptom parameter under jth typical fault, membership function
Figure GDA0003798138100000047
Comprises the following steps:
Figure GDA0003798138100000042
wherein
D＝max(d j (u 0 ,u j )) (3)
It is obvious that, as shown in formula (2), the greater the degree of membership is close to 1, the greater the probability of such a failure.
And step three, obtaining a parameter symptom value and perfecting a fuzzy membership function. And (3) according to the regression error of the extreme learning machine, providing a parameter symptom value solving method, performing parameter fuzzification, and perfecting a fuzzy membership function. The invention provides a parameter symptom value z according to the regression error of the extreme learning machine j The method comprises the following steps:
Figure GDA0003798138100000043
in the formula
Figure GDA0003798138100000044
Is the regression error of the extreme learning machine, sigma is the standard deviation of the corresponding parameters, is obtained through historical data statistics,
Figure GDA0003798138100000045
is a measured value of a parameter that is,
Figure GDA0003798138100000046
is a regression estimate of the parameter;
and step four, acquiring an early warning signal. The multiple regression estimation method is to establish an autoregressive model of multiple parameters based on the potential correlation among the parameters so as to diagnose whether the potential correlation of the parameters changes.
The present invention uses a regression model of the extreme learning machine, but all similar autoregressive models are applicable to the process of the present invention.
Given a training sample set X T The excitation output function of the hidden layer node is G (a) i ,b i ,X T ) Selecting the number of nodes of the hidden layer as L, excitingThe excitation function adopts a sigmoidal function, and the process of the extreme learning machine algorithm is as follows:
(1) randomly generating parameter values (a) of the hidden node function k ,b k ),k＝1,…,L；
(2) Calculating a hidden layer output matrix H;
(3) computing output weight vectors
Figure GDA0003798138100000051
Figure GDA0003798138100000052
(4) Calculating the regression value X of the parameter R ：
Figure GDA0003798138100000053
Figure GDA0003798138100000054
D is the generalized inverse of matrix H, D = H T (HH T ) -1 ，
Figure GDA0003798138100000055
In order to be a weight vector, the weight vector,
Figure GDA0003798138100000056
under the normal operation condition, the deviation between the regression value and the measured value of the fault early warning model is within a normal random error range; if the operation of the equipment deviates from the normal operation condition, the regression value and the measured value of the fault early warning model have obvious deviation and exceed the reasonable random error range, the regression relationship among the parameters is considered to be changed, and an early warning signal is given.
And step five, calculating the membership degree of the early warning signal and judging the fault state.
The simulation of normal operation working conditions and various fault working conditions is carried out by utilizing a certain 1000MW thermal power generating unit simulation system in China, and the simulation is used for verifying the effectiveness of the method for diagnosing the leakage fault of the four pipes of the boiler.
The specific unit simulation process comprises the following steps:
(1) simulating the normal running state of the unit under the working condition of 50% -100% load;
(2) and respectively simulating four types of leakage fault states under the full-load working condition for verifying the effectiveness of the diagnosis method.
The unit simulation system adopted by the invention is subjected to repeated debugging and authentication, and the dynamic characteristic of the unit simulation system and the dynamic characteristic of a field unit are approximately 1: the relation 1 was used as a supply model for the simulation machine match in the national and provincial industries, and the simulation system is used for setting four-pipe leakage faults of different types respectively to perform model simulation so as to obtain the change curves of various typical symptom parameters.
When the early warning signal appears, a specific fault type needs to be further diagnosed, and effective information is provided for timely processing. Firstly, according to regression deviation, a symptom value of a parameter is obtained, then, the membership of each fault is obtained by utilizing an expert knowledge base and a fuzzy membership function, and therefore a specific fault mode is determined.
The utility model provides an equipment trouble early warning of integration intelligent modeling and fuzzy rule, includes following program module:
expert knowledge base module: establishing a four-pipe leakage fault expert knowledge base;
a membership function module: constructing a membership function based on the distance function;
a symptom parameter module: obtaining a parameter symptom value and perfecting a fuzzy membership function;
the early warning signal module: acquiring an early warning signal;
a fault state module: and calculating the membership degree of the early warning signal and judging the fault state.
A storage medium of an equipment fault early warning system fusing intelligent modeling and fuzzy rules is characterized by storing the following program modules:
expert knowledge base module: establishing a four-pipe leakage fault expert knowledge base;
a membership function module: constructing a membership function based on the distance function;
a symptom parameter module: obtaining a parameter symptom value and perfecting a fuzzy membership function;
the early warning signal module: acquiring an early warning signal;
a fault status module: and calculating the membership degree of the early warning signal and judging the fault state.
Example 1
Model parameters:
the simulation of normal operation working conditions and various fault working conditions is carried out by utilizing a certain 1000MW thermal power generating unit simulation system in China, and the simulation is used for verifying the effectiveness of the method for diagnosing the leakage fault of the four pipes of the boiler. The specific unit simulation process comprises the following steps: (1) simulating the normal running state of the unit under the 50% -100% load working condition, and verifying the validity of the regression model; (2) and respectively simulating four types of leakage fault states under the full-load working condition for verifying the effectiveness of the diagnosis method.
And establishing a four-pipe leakage fault expert knowledge base. The fault knowledge base is the basis of fault identification, describes mode relations among various fault categories and sign parameters, summarizes the feature parameters related to the four-pipe leakage of the boiler according to the past documents, and indicates the change relations among the sign parameters and the fault types through thermal calculation. The invention adopts a five-value symptom set to describe the change characteristic of the symptom parameters under the four-tube leakage fault:
Figure GDA0003798138100000071
the five-value symptom set description mode shown in the formula (10) can express the change of the parameters in the positive direction and the negative direction, also considers the change degree, and is more suitable for being applied to complex fault diagnosis.
And (3) according to a five-value symptom set description method, integrating expert knowledge described in each document to obtain a fault expert knowledge base shown in tables 1 and 2.
TABLE 1 four-tube set of leakage symptom parameters
Figure GDA0003798138100000072
TABLE 2 four-tube leakage expert knowledge base
Figure GDA0003798138100000081
The fault identification is actually based on a fault diagnosis expert database, and the current fault state is judged and identified according to a certain rule. The invention provides a novel fuzzy membership degree judging method according to regression errors of an extreme learning machine, which adopts a distance function:
Figure GDA0003798138100000082
in the formula d j (u 0 ,u j ) For a fault u to be identified 0 And typical failure mode u j The smaller the distance between, obviously, the more likely it is that a fault of this type will occur. z is a radical of j A fault symptom value, z, for the ith symptom parameter ij Symptom value of ith symptom parameter under jth typical fault. The membership function is:
Figure GDA0003798138100000083
wherein
D＝max(d j (u 0 ,u j )) (3)
It is obvious that, as shown in formula (2), the greater the degree of membership is close to 1, the greater the probability of such a failure.
The invention provides a parameter symptom value z according to the regression error of the extreme learning machine j The method comprises the following steps:
Figure GDA0003798138100000091
in the formula
Figure GDA0003798138100000092
And the sigma is the standard deviation of the corresponding parameters and is obtained through historical data statistics.
Collecting data corresponding to various faults, carrying out regression estimation on various symptom parameters by using the established multiple regression model, and fitting a calculation formula of a health degree index zt representing the whole running state of the unit, wherein the calculation formula comprises the following steps:
Figure GDA0003798138100000093
in the formula, p is the number of sign parameters related to the leakage fault, and obviously, the closer the index is to 1, the more accurate the regression model is, and the more normal the running state of the unit is. If the health degree index value is remarkably decreased in trend at a certain time, the abnormal operation state of the unit is shown, and fig. 2 shows the change of the health degree index after the leakage fault of the superheater on the side A.
After the multivariate parameter regression model is carried out, a health degree index zt is obtained by calculation according to a regression value and a measured value and a formula (6), the trend of the zt reflects the running state of the modeling object, and once the trend of the form is reduced, the state is abnormal, at the moment, a fault diagnosis mechanism is triggered, namely, the fuzzy diagnosis process is carried out.
As shown in fig. 2, the solid curve represents the variation trend of the health degree, and the chain line is the limit value of the health degree warning, which is generally obtained by statistical analysis under normal conditions. It can be seen that, under the normal working condition of the unit, the health degree is maintained at a relatively stable level, after a fault occurs, the health degree has a trend of being remarkably reduced, the health degree reaches an early warning limit value in the 26 th second, namely, the health degree early warning occurs after the fault occurs for 6 seconds, and the steam temperature overtemperature warning is given only by DCS about 90S, obviously, the sensitivity of an early warning signal to micro degradation is relatively high, and the early warning effect on the fault is achieved in advance.
When the early warning signal appears, a specific fault type needs to be further diagnosed, and effective information is provided for timely processing. Firstly, a symptom value of a parameter is obtained according to regression deviation, and then the membership degree of each fault is obtained by utilizing an expert knowledge base and a fuzzy membership function, so that a specific fault mode is determined. And table 3 is the calculation result of the membership degree of each type of fault after the fault.
TABLE 3 membership calculation for four-tube leak diagnostics
Figure GDA0003798138100000101
As shown in table 3, membership values for 3 time points were calculated using a four-tube leakage expert knowledge base and fuzzy membership. It can be seen that, along with the occurrence of the fault and the increase of the degradation, the membership degree of the current fault mode to u3 is obviously increased and gradually tends to 1, after the early warning signal occurs, the fact that the fault of leakage of the superheater occurs can be judged through the membership degree value, and the more the certainty is, the more the correctness of the judgment of the early warning fault is verified by the membership degree calculation result of the 30S.
Under the premise of early warning, the invention judges the specific failure mode according to the expert database and the fuzzy rule: relevant documents and scientific research results for concluding the leakage faults of the four pipes of the boiler are summarized, and a fault diagnosis expert knowledge base for the leakage of the four pipes of the boiler is listed according to distance type fuzzy judgment rules. Meanwhile, according to standard deviations of all parameters obtained by statistical analysis of historical data, fuzzification is carried out on the parameters according to the result of comparison between regression errors of the intelligent model and 5 times of the standard deviations, and the membership degree of the current state which belongs to each leakage mode is further calculated according to a fuzzy rule so as to judge the current fault state. The 1000MW unit simulation system is used for simulating the fault characteristics of the operation condition and four-pipe leakage of the unit, the method provided by the invention is verified, and the result shows that the result obtained by fuzzy recognition is consistent with the simulated fault mode, thereby demonstrating the correctness and effectiveness of the method provided by the invention.

Claims (5)

1. An equipment fault early warning method integrating intelligent modeling and fuzzy rules is characterized by comprising the following steps:
(1) Establishing a four-tube leakage fault expert knowledge base, which specifically comprises the following steps:
(11) Counting related characteristic parameters of the leakage of four pipes of the boiler;
(12) Obtaining the change relation between each symptom parameter and the fault type;
(13) Establishing a fault expert knowledge base;
(2) Constructing a membership function based on the distance function;
(3) Obtaining a parameter symptom value, perfecting a fuzzy membership function, and specifically comprising the following steps:
(31) According to the regression error of the extreme learning machine, a parameter symptom value solving method is provided;
(32) Fuzzification of parameters is carried out to perfect fuzzy membership function,
parameter symptom value z i The method for obtaining comprises the following steps:
Figure FDA0003782655350000011
in the formula
Figure FDA0003782655350000012
Is the regression error of the extreme learning machine, sigma is the standard deviation of the corresponding parameters, is obtained through historical data statistics,
Figure FDA0003782655350000013
is a measured value of a parameter that is,
Figure FDA0003782655350000014
is a regression estimate of the parameter;
collecting data corresponding to various faults, carrying out regression estimation on various symptom parameters by using the established multivariate regression model, and fitting a health degree index calculation formula representing the whole running state of the unit into the formula:
Figure FDA0003782655350000015
wherein p is the number of symptom parameters related to the leakage fault;
(4) Acquiring an early warning signal, specifically comprising:
(41) Establishing a fault early warning model based on a multiple regression estimation method;
(42) Performing unit model simulation to acquire a fault early warning signal, wherein the specific process is as follows:
under the normal operation condition, the deviation between the regression value and the measured value of the fault early warning model is within a normal random error range; if the operation of the equipment deviates from the normal operation condition, the regression value and the measured value of the fault early warning model have obvious deviation and exceed the reasonable random error range, the regression relationship among the parameters is considered to be changed, and an early warning signal is given;
(5) Calculating the membership degree of the early warning signal, and judging the fault state, wherein the specific process comprises the following steps:
when the early warning signal appears, the specific fault type is further diagnosed, firstly, a symptom value of a parameter is obtained according to the regression deviation, then, the membership degree of each fault is obtained by utilizing an expert knowledge base and a fuzzy membership function, and therefore, the specific fault mode is determined.
2. The equipment fault early warning method integrating intelligent modeling and fuzzy rules according to claim 1, wherein the method comprises the following steps: the step (2) specifically comprises:
(21) Selecting a suitable distance function as a basis;
(22) And constructing a membership function.
3. The equipment fault early warning method fusing intelligent modeling and fuzzy rules according to claim 2, characterized in that: the distance function is:
Figure FDA0003782655350000021
in the formula d j (u 0 ,u j ) For a fault u to be identified 0 And typical failure mode u j Distance between, z ij The symptom value of the ith symptom parameter under the jth typical fault;
membership function
Figure FDA0003782655350000022
Comprises the following steps:
Figure FDA0003782655350000023
wherein D = max (D) j (u 0 ,u j ))(3)。
4. The utility model provides an equipment trouble early warning system who fuses intelligent modeling and fuzzy rule which characterized in that: the method comprises the following program modules:
the expert knowledge base module: establishing a four-tube leakage fault expert knowledge base, which specifically comprises the following steps:
(11) Counting related characteristic parameters of the leakage of four pipes of the boiler;
(12) Obtaining the change relation between each symptom parameter and the fault type;
(13) Establishing a fault expert knowledge base;
a membership function module: constructing a membership function based on the distance function;
a symptom parameter module: obtaining a parameter symptom value, perfecting a fuzzy membership function, and specifically comprising the following steps:
(31) According to the regression error of the extreme learning machine, a parameter symptom value solving method is provided;
(32) Fuzzification of parameters is carried out to perfect fuzzy membership function,
parameter symptom value z j The method comprises the following steps:
Figure FDA0003782655350000031
in the formula
Figure FDA0003782655350000032
Is the regression error of the extreme learning machine, sigma is the standard deviation of the corresponding parameters, is obtained through historical data statistics,
Figure FDA0003782655350000033
is a measured value of a parameter that is,
Figure FDA0003782655350000034
is a regression estimate of the parameter;
collecting data corresponding to various faults, carrying out regression estimation on various symptom parameters by using the established multivariate regression model, and fitting a health degree index calculation formula representing the whole running state of the unit into the formula:
Figure FDA0003782655350000035
wherein p is the number of symptom parameters related to the leakage fault;
the early warning signal module: acquiring an early warning signal, specifically comprising:
(41) Establishing a fault early warning model based on a multiple regression estimation method;
(42) Performing unit model simulation to acquire a fault early warning signal, wherein the specific process is as follows:
under the normal operation condition, the deviation between the regression value and the measured value of the fault early warning model is within the normal random error range; if the operation of the equipment deviates from the normal operation condition, the regression value and the measured value of the fault early warning model have obvious deviation and exceed the reasonable random error range, the regression relationship among the parameters is considered to be changed, and an early warning signal is given;
a fault state module: calculating the membership degree of the early warning signal, and judging the fault state, wherein the specific process comprises the following steps:
when the early warning signal appears, a specific fault type is further diagnosed, firstly, a symptom value of a parameter is obtained according to the regression deviation, then, the membership degree of each fault is obtained by utilizing an expert knowledge base and a fuzzy membership function, and therefore a specific fault mode is determined.
5. A storage medium of an equipment fault early warning system fusing intelligent modeling and fuzzy rules is characterized by storing the following program modules:
expert knowledge base module: establishing a four-tube leakage fault expert knowledge base, which specifically comprises the following steps:
(11) Counting related characteristic parameters of the leakage of four pipes of the boiler;
(12) Obtaining the change relation between each symptom parameter and the fault type;
(13) Establishing a fault expert knowledge base;
a membership function module: constructing a membership function based on the distance function;
a symptom parameter module: obtaining a parameter symptom value, perfecting a fuzzy membership function, and specifically comprising the following steps:
(31) According to the regression error of the extreme learning machine, a parameter symptom value solving method is provided;
(32) Fuzzification of parameters is carried out to perfect fuzzy membership function,
parameter symptom value z i The method comprises the following steps:
Figure FDA0003782655350000041
in the formula
Figure FDA0003782655350000042
Is the regression error of the extreme learning machine, sigma is the standard deviation of the corresponding parameters, is obtained through historical data statistics,
Figure FDA0003782655350000043
is a measured value of a parameter that is,
Figure FDA0003782655350000044
is a regression estimate of the parameter;
collecting data corresponding to various faults, carrying out regression estimation on various symptom parameters by using the established multivariate regression model, and fitting a health degree index calculation formula representing the whole running state of the unit into the formula:
Figure FDA0003782655350000045
wherein p is the number of symptom parameters related to the leakage fault;
the early warning signal module: acquiring an early warning signal, specifically comprising:
(41) Establishing a fault early warning model based on a multiple regression estimation method;
(42) Performing unit model simulation to acquire a fault early warning signal, wherein the specific process is as follows:
under the normal operation condition, the deviation between the regression value and the measured value of the fault early warning model is within the normal random error range; if the operation of the equipment deviates from the normal operation condition, the regression value and the measured value of the fault early warning model have obvious deviation and exceed the reasonable random error range, the regression relationship among the parameters is considered to be changed, and an early warning signal is given;
a fault state module: calculating the membership degree of the early warning signal, and judging the fault state, wherein the specific process comprises the following steps:
when the early warning signal appears, the specific fault type is further diagnosed, firstly, a symptom value of a parameter is obtained according to the regression deviation, then, the membership degree of each fault is obtained by utilizing an expert knowledge base and a fuzzy membership function, and therefore, the specific fault mode is determined.
CN202011382067.8A 2020-12-01 2020-12-01 Equipment fault early warning method integrating intelligent modeling and fuzzy rules Active CN112464565B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011382067.8A CN112464565B (en) 2020-12-01 2020-12-01 Equipment fault early warning method integrating intelligent modeling and fuzzy rules

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011382067.8A CN112464565B (en) 2020-12-01 2020-12-01 Equipment fault early warning method integrating intelligent modeling and fuzzy rules

Publications (2)

Publication Number Publication Date
CN112464565A CN112464565A (en) 2021-03-09
CN112464565B true CN112464565B (en) 2022-10-04

Family

ID=74805002

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011382067.8A Active CN112464565B (en) 2020-12-01 2020-12-01 Equipment fault early warning method integrating intelligent modeling and fuzzy rules

Country Status (1)

Country Link
CN (1) CN112464565B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114054517B (en) * 2021-11-19 2023-09-15 南京邮电大学 Fault detection method for emulsion injection device of single-frame cold rolling mill
CN115468718B (en) * 2022-08-24 2024-06-14 大连海事大学 Diagnosis method for leakage fault of ship wind wing rotary hydraulic system
CN117150879B (en) * 2023-07-14 2024-04-09 华能国际电力股份有限公司上海石洞口第二电厂 Super-heater overtemperature early warning method and device based on fuzzy comprehensive evaluation model
CN117131946A (en) * 2023-08-30 2023-11-28 广东电网有限责任公司 Energy storage battery fault early warning method and device, electronic equipment and medium

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108644752A (en) * 2018-05-11 2018-10-12 中国神华能源股份有限公司 Method, apparatus and machine readable storage medium for analyzing four main tubes of boiler leakage
CN110377001A (en) * 2019-06-04 2019-10-25 上海华电奉贤热电有限公司 Industrial equipment intelligent Fault Diagnose Systems and method based on data fusion

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108644752A (en) * 2018-05-11 2018-10-12 中国神华能源股份有限公司 Method, apparatus and machine readable storage medium for analyzing four main tubes of boiler leakage
CN110377001A (en) * 2019-06-04 2019-10-25 上海华电奉贤热电有限公司 Industrial equipment intelligent Fault Diagnose Systems and method based on data fusion

Also Published As

Publication number Publication date
CN112464565A (en) 2021-03-09

Similar Documents

Publication Publication Date Title
CN112464565B (en) Equipment fault early warning method integrating intelligent modeling and fuzzy rules
Cai et al. Bayesian networks in fault diagnosis
CN112597696B (en) Boiler four-pipe leakage early warning method based on extreme learning machine principle
Gao et al. A novel fault identification method for photovoltaic array via convolutional neural network and residual gated recurrent unit
CN106154182A (en) A kind of based on the lithium battery method for diagnosing faults improving D S evidence theory
CN113011010A (en) Boiler fault diagnosis method and diagnosis system based on structural mechanism and operation data
CN112821424B (en) Power system frequency response analysis method based on data-model fusion driving
Cao et al. PCA-SVM method with sliding window for online fault diagnosis of a small pressurized water reactor
Zhou A fuzzy petri-net approach for fault analysis considering factor influences
CN116205265A (en) Power grid fault diagnosis method and device based on deep neural network
CN111445037A (en) Steam turbine fault diagnosis method, system and equipment based on evidence theory
CN110108457A (en) A kind of First air machine shaft temperature diagnostic method based on gravitation neural network
CN112257335B (en) Oil-immersed transformer fault diagnosis method combining PNN and SVM
CN111563685B (en) Power generation equipment state early warning method based on auto-associative kernel regression algorithm
CN114548701B (en) Full-measurement-point-oriented coupling structure analysis and estimation process early warning method and system
Razavi-Far et al. Neuro-fuzzy based fault diagnosis of a steam generator
Xu et al. Study of fault diagnosis based on probabilistic neural network for turbine generator unit
Yun et al. Fault diagnosis of wind turbine pitch system based on multiblock KPCA algorithm
Han et al. Gated Recurrent Unit Neural Network and Decision Tree for Offshore Nuclear Power Plant'S Fault Diagnosis
Gang et al. Fault warning method based on extreme learning regression and fuzzy reasoning
Xiong et al. Research on health condition assessment method for spacecraft power control system based on SVM and cloud model
Dai et al. A Novel Fault Diagnosis Method Integrating Reconstruction-based Principal Component Analysis with Probabilistic Neural Network on Condenser
Hines et al. Application of monitoring and prognostics to small modular reactors
Liu et al. Bearing Fault Diagnosis Method Based on Fuzzy Petri Nets with Adam Optimization
Guohua et al. Research on Fault Diagnosis Based on Hierarchical Signed Directed Graph for Nuclear Power Plants

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant