CN116224950A - Intelligent fault diagnosis method and system for self-organizing reconstruction of unmanned production line - Google Patents
Intelligent fault diagnosis method and system for self-organizing reconstruction of unmanned production line Download PDFInfo
- Publication number
- CN116224950A CN116224950A CN202211103784.1A CN202211103784A CN116224950A CN 116224950 A CN116224950 A CN 116224950A CN 202211103784 A CN202211103784 A CN 202211103784A CN 116224950 A CN116224950 A CN 116224950A
- Authority
- CN
- China
- Prior art keywords
- rule
- base
- representing
- ebrb
- fault diagnosis
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 88
- 238000003745 diagnosis Methods 0.000 title claims abstract description 69
- 238000004519 manufacturing process Methods 0.000 title claims abstract description 27
- 238000012360 testing method Methods 0.000 claims abstract description 40
- 238000005457 optimization Methods 0.000 claims abstract description 34
- 230000008569 process Effects 0.000 claims abstract description 32
- 238000012549 training Methods 0.000 claims abstract description 29
- 230000009467 reduction Effects 0.000 claims abstract description 18
- 230000004913 activation Effects 0.000 claims description 27
- 230000002776 aggregation Effects 0.000 claims description 15
- 238000004220 aggregation Methods 0.000 claims description 15
- 238000009826 distribution Methods 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 238000003860 storage Methods 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 2
- 238000000605 extraction Methods 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 description 14
- 238000006243 chemical reaction Methods 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 229910001220 stainless steel Inorganic materials 0.000 description 2
- 239000010935 stainless steel Substances 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000004931 aggregating effect Effects 0.000 description 1
- 125000003275 alpha amino acid group Chemical group 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 235000014594 pastries Nutrition 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
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
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0259—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the response to fault detection
- G05B23/0262—Confirmation of fault detection, e.g. extra checks to confirm that a failure has indeed occurred
-
- 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
- G05B2219/20—Pc systems
- G05B2219/24—Pc safety
- G05B2219/24065—Real time diagnostics
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides an intelligent fault diagnosis method and system for self-organizing reconstruction of an unmanned production line, and relates to the technical field of fault diagnosis of industrial equipment. The invention processes the operation data of the equipment through the plurality of base EBRB subsystems, outputs a plurality of reasoning results, integrates the information of the plurality of reasoning results, judges whether the equipment has faults or not, and outputs the fault type. The base EBRB subsystem training process includes: acquiring a data set, constructing an extended confidence rule base based on a training set in the data set, constructing a multi-objective optimization model, and optimizing the extended confidence rule base based on the multi-objective optimization model to obtain a rule base after reduction, and the relative weight of the rule and the weight of the precondition attribute; and constructing and testing a plurality of base EBRB subsystems based on the weights. According to the method, the problem of more inconsistent rules and redundant rules caused by larger rule base scale is solved through rule reduction, the inconsistency of the rule base obtained after the rule reduction is reduced is lower, and the fault diagnosis precision is improved.
Description
Technical Field
The invention relates to the technical field of fault diagnosis of industrial equipment, in particular to an intelligent fault diagnosis method and system for self-organizing reconstruction of an unmanned production line.
Background
With the tremendous progress of technologies such as big data, artificial intelligence, industrial internet and information technology, the manufacturing industry has also entered the informatization and intelligence "industry 4.0" era. The mechanical equipment represented by the high-speed rail, the aircraft carrier, the shield machine and the wind motor has the characteristics of more and more complex structure, larger scale and higher intelligent degree. In the unmanned production line of an intelligent factory, efficient operation of equipment is an important guarantee for collaborative scheduling of the production line. In the operation process of the equipment, the service life of the equipment can be effectively prolonged, the production efficiency can be improved, the operation condition of the equipment can be effectively judged according to the fault diagnosis result, and the equipment can be conveniently and timely maintained.
In order to further improve the efficiency of fault maintenance of workshop processing equipment facing an unmanned production line, a fault diagnosis technology is provided, and the technology monitors the running state of the equipment so as to identify the fault of the equipment. The core of the fault diagnosis technology is the identification of the fault state of the equipment, the traditional fault diagnosis technology realizes the fault diagnosis of the equipment mainly by simply comparing the extracted fault characteristic signals with the characteristic information of a database, and along with the development of science and technology and information technology, the function of the equipment is more and more complex, various fault information can possibly appear, and the fault type is difficult to accurately position. Therefore, based on the traditional fault diagnosis technology, the existing fault diagnosis technology is a rule base constructed based on rules converted from data, the technology introduces a belief structure into descriptions of precondition attributes, so that an extended confidence rule system can process uncertain information more effectively than a confidence rule system, and the method effectively solves the problem of combined explosion existing in the confidence rule system.
However, the existing fault diagnosis technology has lower accuracy of the activation rule and lower diagnosis precision due to more inconsistent rules and redundant rules in an extended confidence rule base obtained by data conversion.
Disclosure of Invention
(one) solving the technical problems
Aiming at the defects of the prior art, the invention provides an intelligent fault diagnosis method and system for self-organizing reconstruction of an unmanned production line, which solve the technical problem of lower diagnosis precision of the existing fault diagnosis technology.
(II) technical scheme
In order to achieve the above purpose, the invention is realized by the following technical scheme:
in a first aspect, the invention provides an intelligent fault diagnosis method for self-organizing reconstruction of an unmanned production line, which processes operation data of equipment through a plurality of base EBRB subsystems trained in advance, outputs a plurality of reasoning results, performs information aggregation on the plurality of reasoning results, judges whether the equipment has faults, and outputs fault types; the training process of the pre-trained multiple base EBRB subsystems comprises the following steps:
s1, acquiring a data set, and constructing an extended confidence rule base based on a training set in the data set, wherein the data set comprises equipment operation data and fault types;
s2, constructing a multi-objective optimization model, and optimizing and expanding a confidence rule base based on the multi-objective optimization model to obtain a rule base after reduction, and the relative weight of the rule and the weight of the precondition attribute;
s3, constructing a plurality of base EBRB subsystems according to the reduced rule base, the relative weights of the rules and the weights of the precondition attributes;
s4, testing the plurality of base EBRB subsystems through a testing set in the data set.
Preferably, the multi-objective optimization model includes:
MIN E(δ i ,θ k ) (4)
MIN In(τ k ) (5)
0≤τ k ≤1,k=1,…,L TR (6)
0≤δ i ≤1,i=1,…,M (7)
0≤θ k ≤1,k=1,…,L TR (9)
wherein , representing a system reasoning result; />Representing belief distribution obtained by converting the true diagnosis result of the data, wherein n=1, … and N; τ k R represents k Index of whether to decrease, if τ k <0.5 then means pruned the rule, otherwise the rule is preserved; in (τ) k ) Representing the degree of inconsistency of the constructed rule base; delta i Representing the weight of the ith precondition attribute in the extended confidence rule base; θ k Representing the relative weight of the kth extended confidence rule; l (L) TR Representing the number of rules in the training dataset.
Preferably, the method for calculating the inconsistency degree of the rule base includes:
the inconsistent degree of the rule base is obtained by accumulatively calculating the inconsistent degree of each rule, wherein the inconsistent degree of each rule is determined by the precondition attribute and the result attribute similarity among the rules; the similarity of belief distribution between two rule precondition attributes is expressed as SRA, and the similarity of belief distribution between result attributes is expressed as SRC, and the calculation process is as follows:
and then calculating the consistency degree and the inconsistency degree between the two rules:
Cons(R t ,R k )=1–(SRA(t,k)-SRC(t,k)) 2 (12)
according to the above formula, the degree of inconsistency of the confidence rule base is:
wherein ,representing the reference value A used in the kth rule i,j Description precondition attribute U i And satisfy the belief degree of Representing the reference value D used in the kth rule n Belief describing the result attribute D and satisfyingL TR A number of bars representing rules in the training dataset; r is R k Representing the kth expanded belief rule; r is R t Representing the t-th expanded belief rule.
Preferably, the S4 includes:
s401, testing a plurality of base EBRB subsystems through a testing set in a data set, and outputting a testing result;
s402, the analysis evidence reasoning method is used for information aggregation of the test results.
Preferably, the step S401 includes:
in the p-th base EBRB subsystem system, random extraction is performed from a rule base obtained by rule reductionRules are used to build a rule base for the base EBRB, which is then based on the belief structure of the input test dataPrecondition attribute U for the kth alternative activation rule i The individual matching degree between the two is calculated as follows:
wherein ,precondition attribute U representing the kth alternative activation rule i Described as reference value A i,j Belief of alpha i,j Representing input data +.>Described as reference value A i,j Is a confidence level of (1);
and calculating the activation weight of the kth alternative activation rule by using the obtained individual matching degree, wherein the calculation process is as follows:
wherein ,wk Activation weights representing the kth alternative activation rule and should satisfy 0.ltoreq.w k ≤1,k=1,..., If w k =0, then represents the input data +.>The kth alternative activation rule is not activated, otherwise the kth alternative activation rule is activated.
Preferably, the step S402 includes:
the activated rules are clustered based on an analytical evidence reasoning method, and the method comprises the following steps:
the result obtained by the aggregation of the evidence reasoning algorithm is expressed as
wherein ,n=1,…,N,p=1,…,P,/>expressing the reasoning result of the p-th base EBRB subsystem, namely a test result;
the reasoning result obtained by the P-base EBRB subsystem system is thatp=1, …, P; the reasoning results of the P base EBRB subsystems are aggregated by utilizing the analytic evidence reasoning method in the formula (17) to obtain +.>n=1, …, N, wherein the normalized inference accuracy of the P-base EBRB subsystems is the weight of each inference result in the information aggregation process, and the calculation formula is as follows:
suppose D n The nth level representing the classification problem, the output classification result of the integrated extended confidence rule system is as follows:
In a second aspect, the invention provides an intelligent fault diagnosis system for self-organizing reconstruction of an unmanned production line, wherein in the equipment fault diagnosis system, operation data of equipment are processed through a plurality of base EBRB subsystems trained in advance, a plurality of reasoning results are output, information is gathered on the plurality of reasoning results, whether the equipment has faults or not is judged, and a fault type is output; wherein, a plurality of base EBRB subsystems trained in advance include:
the data acquisition module is used for acquiring a data set, and constructing an extended confidence rule base based on a training set in the data set, wherein the data set comprises equipment operation data and fault types;
the reduction module is used for constructing a multi-objective optimization model, optimizing and expanding the confidence rule base based on the multi-objective optimization model to obtain a reduced rule base, and the relative weight of the rule and the weight of the precondition attribute;
the construction module is used for constructing a plurality of base EBRB subsystems according to the reduced rule base, the relative weights of the rules and the weights of the premise attributes;
and the test module is used for testing the plurality of base EBRB subsystems through a test set in the data set.
Preferably, the multi-objective optimization model includes:
MIN E(δ i ,θ k ) (4)
MIN In(τ k ) (5)
0≤τ k ≤1,k=1,…,L TR (6)
0≤θ k ≤1,k=1,…,L TR (9)
wherein , representing a system reasoning result; />Representing belief distribution obtained by converting the true diagnosis result of the data, wherein n=1, … and N; τ k R represents k Index of whether to decrease, if τ k <0.5 then means pruned the rule, otherwise the rule is preserved; in (τ) k ) Representing the degree of inconsistency of the constructed rule base; delta i Representing the weight of the ith precondition attribute in the extended confidence rule base; θ k Representing the relative weight of the kth extended confidence rule; l (L) TR Representing the number of rules in the training dataset.
In a third aspect, the present invention provides a computer-readable storage medium storing a computer program for intelligent fault diagnosis for unmanned line-oriented self-organizing reconstruction, wherein the computer program causes a computer to execute the intelligent fault diagnosis method for unmanned line-oriented self-organizing reconstruction as described above.
In a fourth aspect, the present invention provides an electronic device comprising:
one or more processors;
a memory; and
one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs comprising an intelligent fault diagnosis method for performing the unmanned line-oriented self-organizing reconstruction as described above.
(III) beneficial effects
The invention provides an intelligent fault diagnosis method and system for self-organizing reconstruction of an unmanned production line. Compared with the prior art, the method has the following beneficial effects:
the invention processes the operation data of the equipment through a plurality of pre-trained base EBRB subsystems, outputs a plurality of reasoning results, integrates the information of the reasoning results, judges whether the equipment has faults or not, and outputs the fault type; the training process of the pre-trained multiple base EBRB subsystems comprises the following steps: acquiring a data set, and constructing an extended confidence rule base based on a training set in the data set, wherein the data set comprises equipment operation data and fault types; constructing a multi-objective optimization model, and optimizing and expanding a confidence rule base based on the multi-objective optimization model to obtain a rule base after reduction, and the relative weight of the rule and the weight of the precondition attribute; constructing a plurality of base EBRB subsystems according to the reduced rule base, the relative weights of the rules and the weights of the precondition attributes; the plurality of base EBRB subsystems are tested by a test set in the dataset. According to the method, the problem of more inconsistent rules and redundant rules caused by larger rule base scale is solved through rule reduction, the inconsistency of the rule base obtained after the rule reduction is reduced is lower, and the fault diagnosis precision is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a block diagram of a training process for a base EBRB subsystem.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention are clearly and completely described, and it is obvious that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
According to the intelligent fault diagnosis method and system for self-organizing reconstruction of the unmanned production line, the technical problem that the existing fault diagnosis technology is low in diagnosis precision is solved, and the diagnosis precision and efficiency of fault diagnosis are improved.
The technical scheme in the embodiment of the application aims to solve the technical problems, and the overall thought is as follows:
in order to solve the problems in the prior art, the embodiment of the invention provides a novel intelligent fault diagnosis method based on an extended confidence rule base method.
In order to better understand the above technical solutions, the following detailed description will refer to the accompanying drawings and specific embodiments.
The embodiment of the invention provides an intelligent fault diagnosis method for self-organizing reconstruction of an unmanned production line, which processes operation data of equipment through a plurality of pre-trained base EBRB subsystems (EBRB, extended belief rule base and an extended confidence rule base), outputs a plurality of reasoning results, performs information aggregation on the plurality of reasoning results, judges whether the equipment has faults, and outputs fault types; the training process of the pre-trained multiple base EBRB subsystems is shown in fig. 1, and includes:
s1, acquiring a data set, and constructing an extended confidence rule base based on a training set in the data set, wherein the data set comprises equipment operation data and fault types;
s2, constructing a multi-objective optimization model, and optimizing and expanding a confidence rule base based on the multi-objective optimization model to obtain a rule base after reduction, and the relative weight of the rule and the weight of the precondition attribute;
s3, constructing a plurality of base EBRB subsystems according to the reduced rule base, the relative weights of the rules and the weights of the precondition attributes;
s4, testing the plurality of base EBRB subsystems through a testing set in the data set.
According to the embodiment of the invention, the problem of more inconsistent rules and redundant rules caused by larger rule base scale is solved through rule reduction, the inconsistency of the rule base obtained after the rule reduction is lower, and the fault diagnosis precision is improved.
The following details the individual steps:
in step S1, a dataset is acquired, and an extended confidence rule base is constructed based on a training set in the dataset, wherein the dataset comprises equipment operation data and fault types. The specific implementation process is as follows:
wherein the device operational data is based on the diagnostic indicator. The device operation data is uploaded to the computer through an interface, wireless communication or the like.
There are differences between diagnostic indicators and fault types of different devices, such as in the case of fault diagnosis for stainless steel blades, the dataset contains 27 diagnostic indicators and 7 fault types, including path, z_scratch, k_ Scatch, stains, dirtiness, bumps, and other_faults. The data set is collected from an open source industrial defect data set.
Randomly decimating data set L TR The bar data is used as training data set, and L is remained TE The bar data is used as test data.
Based on L in training dataset TR Stripe data, extended confidence rule base comprising L TR Bar extension confidence rule r= { R k ,k=1,…,L TR ) M precondition attributes U i (i=1, …, M), and one result attribute D. By J i Reference value A i,j (j=1,…,J i ) Description precondition attribute U i In addition, N reference values D n (n=1, …, N) describes the result attribute D. The weight of the ith precondition attribute in the extended confidence rule base is expressed as delta i (0<δ i 1) or less, the relative weights of the kth extended confidence ruleWeight is theta k (0<θ k And is less than or equal to 1). Kth extended belief rule R k The expression of (2) is as follows:
in the formula (1), the amino acid sequence of the formula (1),representing the reference value A used in the kth rule i,j Description precondition attribute U i Is satisfied with +.> Reference value D for representation n Belief describing the result attribute D and satisfyingIf->Then the result of the kth rule is indicated to be complete, otherwise incomplete.
The following describes the training data x using the equivalent transformation technique l ={x l,1 ,…,x l,M Conversion to U i Reference value A i,j (j=1,…,J i ) The belief distribution process:
S(x l,i )={(A i,j ,α i,j );j=1,…,J i },i=1,…,M. (2)
wherein :
α i,k =0,for k=1,…,J i and k≠j,j+1 (3)
u(A i,j ) Representing the reference value A i,j Utility value. Training data x l Corresponding y l (l=1,…,L TR ) Conversion to reference value D n Process for believing and x l The conversion process of (2) is the same and will not be described in detail herein.
In step S2, a multi-objective optimization model is constructed, and the confidence rule base is optimized and expanded based on the multi-objective optimization model, so that the rule base after reduction, the relative weight of the rule and the weight of the precondition attribute are obtained. The specific implementation process is as follows:
based on the constructed extended confidence rule base, the structural optimization and the parameter optimization of the rule base are simultaneously carried out by constructing a multi-objective optimization model. The rule base structure optimization mainly reduces redundant rules in the rule base, and the parameter optimization mainly objectively determines attribute weights and rule weights in the optimized rule base. Eliminating redundant rules may increase the degree of rule base inconsistency, and therefore, both system accuracy and rule base inconsistency should be considered in optimizing rule base structure. Therefore, the diagnosis precision of the system and the inconsistency degree of the rule base are used as two optimization targets of an optimization model, and the optimization model is as follows.
MIN E(δ i ,θ k ) (4)
MIN In(τ k ) (5)
0≤τ k ≤1,k=1,…,L TR (6)
0≤δ i ≤1,i=1,…,M (7)
0≤θ k ≤1,k=1,…,L TR (9)
wherein ,representing the result of system reasoning, < >>And representing belief distribution obtained by converting the real diagnosis result of the data. τ k R represents k Index of whether to decrease, if τ k <0.5 then means that the rule is pruned, otherwise the rule is preserved. In (τ) k ) Representing the degree of inconsistency of the constructed rule base, the rule base degree of inconsistency measures the procedure as follows.
The degree of inconsistency of the rule base is calculated by accumulating the degree of inconsistency of each rule, wherein the degree of inconsistency of each rule is determined by precondition attributes among rules and similarity of result attributes. The similarity of belief distribution between two rule precondition attributes is expressed as SRA, and the similarity of belief distribution between result attributes is expressed as SRC, and the calculation process is as follows:
the degree of consistency and the degree of inconsistency between the two rules are then calculated.
Cons(R t ,R k )=1–(SRA(t,k)-SRC(t,k)) 2 (12)
According to the above formula, the degree of inconsistency of the confidence rule base is:
through the steps, the structure and parameters of the rule base are optimized, and the rule base after reduction, the relative weight of the rule and the weight of the precondition attribute are obtained. Wherein the optimization of the parameters refers to: and optimizing two weight parameters, namely the relative weight and the precondition attribute weight of the rules in the rule base. The optimization of the structure means: and optimizing the rules in the rule base, deleting the rules with lower redundancy and consistency, and optimizing the rule base structure.
In step S3, a plurality of base EBRB subsystems are constructed from the reduced rule base, and the relative weights of the rules and the weights of the precondition attributes. The specific implementation process is as follows:
the reduced rule base and the optimal relevant parameters are obtained by solving an optimization model, P base EBRB subsystems (P=5 in the embodiment of the invention) are constructed, and the rules in the base EBRB subsystems are obtained by randomly extracting 90% of the rules in the reduced rule base, namely, each base EBRB subsystem is provided withAnd (5) bar rules. Wherein->The number of rules contained in the rule base after the subtraction is reduced.
In step S4, the plurality of base EBRB subsystems are tested by a test set in the data set.
S401, testing the plurality of base EBRB subsystems through a testing set in the data set, and outputting a testing result. The specific implementation process is as follows:
after the base EBRB subsystem is built in the P (p=1, …, P) base EBRB subsystem system, data is input based on the belief structure of the input test dataPrecondition attribute U for the kth alternative activation rule i The individual degree of matching between can be calculated as follows
wherein ,precondition attribute U representing the kth alternative activation rule i Described as reference value A i,j Belief of alpha i,j Representing input data +.>Described as reference value A i,j Is a confidence level of (1).
The activation weight of the kth alternative activation rule is then calculated using the individual degree of matching that has been obtained, as follows
wherein ,wk Activation weights representing the kth alternative activation rule and should satisfy 0.ltoreq.w k and />If w k =0, which means that the input data +.>The kth alternative activation rule is not activated, otherwise the kth alternative activation rule is activated.
S402, information aggregation is carried out on the test results. The method specifically comprises the following steps:
the activated rules are clustered based on an analytical evidence reasoning method, and the method comprises the following steps:
the result obtained by the aggregation of the evidence reasoning algorithm is expressed as
wherein ,n=1,…,N,p=1,…,P,/>and (5) expressing the reasoning result of the p-th base EBRB subsystem, namely the test result.
The reasoning result obtained by the P-base EBRB subsystems is thatp=1, …, P. And (3) aggregating the reasoning results of the P base EBRB subsystems by using the analytical evidence reasoning method in the formula (17), wherein the normalized reasoning precision of the P base EBRB subsystems is the weight of each reasoning result in the information aggregation process. The reasoning result obtained by using the analytical evidence reasoning method is as follows:
suppose D n The nth level representing the classification problem, the output classification result of the integrated extended confidence rule system is as follows:
the current operation data of the equipment are processed through the plurality of base EBRB subsystems, a plurality of reasoning results are output, information aggregation is carried out on the plurality of reasoning results to obtain detection results, whether the equipment has faults or not is judged, if the detection results indicate that the equipment has faults, an alarm device sends an alarm signal, the equipment stops operating, and maintenance staff is waited to overhaul. The procedure is similar to the test procedure of step S4, and will not be described here again.
The following is a specific data to verify the embodiments of the present invention:
aiming at the equipment fault diagnosis problem, the invention utilizes the fault diagnosis data set of the stainless steel blade in the UCI public data set to verify the effectiveness of the method. In this fault diagnosis dataset, there are 27 diagnostic indices, which introduce 7 fault types, including path, z_scratch, k_ Scatch, stains, dirtiness, bumps, and other_faults, as shown in table 1. The data set has 1941 pieces of data, 80% of data, namely 1553 pieces of data, are randomly extracted to serve as training data, and the rest 388 pieces of data serve as test data.
Table 1 introduction to diagnostic index
Each index is described by 5 reference values (VL, L, M, H, VH) based on the maximum and minimum values of the index in table 1, as shown in table 2 in particular.
Table 2 index reference definition
The training data is converted into rules based on the reference value definitions of the indicators in table 2, taking one piece of training data as an example, provided that the indicators are (42, 50, 270900, 270944, 267, 17, 44, 24220, 76, 108, 1687,1,0, 80,0.0498,0.2415,0.1818,0.0047,0.4706,1,1,2.4265,0.9031,1.6435,0.8182, -0.2913,0.5822) and the fault type is path.
R 1 :IF U 1 is{(VL,0.9015),(L,0.0985),(M,0),(H,0),(VH,0)}∧U 2 {(VL,0.8923),(L,0.1077),(M,0),(H,0),(VH,0)}∧U 3 {(VL,0.9186),(L,0.0814),(M,0),(H,0),(VH,0)}∧U 4 {(VL,0.9186),(L,0.0814),(M,0),(H,0),(VH,0)}∧U 5 {(VL,0.9931),(L,0.0069),(M,0),(H,0),(VH,0)}∧U 6 {(VL,0.9943),(L,0.0057),(M,0),(H,0),(VH,0)}∧U 7 {(VL,0.9905),(L,0.0095),(M,0),(H,0),(VH,0)}∧U 8 {(VL,0.9917),(L,0.0083),(M,0),(H,0),(VH,0)}∧U 9 {(VL,0),(L,0.5025),(M,0.4975),(H,0),(VH,0)}∧U 10 {(VL,0),(L,0.6852),(M,0.3148),(H,0),(VH,0)}∧U 11 {(VL,0),(L,0),(M,0),(H,0.7549),(VH,0.2451)}∧U 12 {(VL,0),(L,0),(M,0),(H,0),(VH,1)}∧U 13 {(VL,1),(L,0),(M,0),(H,0),(VH,0)}∧U 14 {(VL,0.38462),(L,0.6154),(M,0),(H,0),(VH,0)}∧U 15 {(VL,0.7998),(L,0.2002),(M,0),(H,0),(VH,0)}∧U 16 {(VL,0),(L,0.9766),(M,0.0234),(H,0),(VH,0)}∧U 17 {(VL,0.3002),(L,0.6998),(M,0),(H,0),(VH,0)}∧U 18 {(VL,0.9854),(L,0.0146),(M,0),(H,0),(VH,0)}∧U 19 {(VL,0),(L,0.1485),(M,0.8515),(H,0),(VH,0)}∧U 20 {(VL,0),(L,0),(M,0),(H,0),(VH,1)}∧U 21 {(VL,0),(L,0),(M,0),(H,0),(VH,1)}∧U 22 {(VL,0),(L,0.2588),(M,0.7412),(H,0),(VH,0)}∧U 23 {(VL,0.1315),(L,0.8685),(M,0),(H,0),(VH,0)}∧U 24 {(VL,0),(L,0.4563),(M,0.5437),(H,0),(VH,0)}∧U 25 {(VL,0),(L,0),(M,0),(H,0.3500),(VH,0.65)}∧U 26 {(VL,0),(L,0.2752),(M,0.7248),(H,0),(VH,0)}∧U 27 {(VL,0),(L,0),(M,0.8969),(H,0.1031),(VH,0)}
THEN D is{(Pastry,1),(Z_Scratch,0),(K_Scatch,0),(Stains,0),(Dirtiness,0),(Bumps,0),(Other_Faults,0)}
And solving the constructed multi-objective optimization model by utilizing the training data, so as to optimize the rule base structure, delete inconsistent and redundant rules, and optimize parameters to obtain rule weights and attribute weights of a new rule base to perfect rule base parameter information. And constructing a plurality of base rule bases based on the new rule base, obtaining a plurality of reasoning results by utilizing the plurality of base rule bases according to the test data, and gathering the plurality of results to obtain final output. The fault diagnosis results obtained by this verification process are shown in table 3.
The verification proves that the embodiment of the invention has high accuracy.
The comparison result of the method proposed by the embodiment of the invention and the original method (extended confidence rule base C-EBRB) is shown in Table 4.
Table 4 results of comparison with the original method
The prediction accuracy calculation method in the above table is as follows
wherein ,representing a system reasoning result; />Representing the belief distribution resulting from conversion of the true diagnostic result of the data, where n=1, …, N. Since the inference results outputted from the EBRB system are information in the form of belief distribution, in order to calculate the performance of the two methods in fault diagnosis more accurately, the similarity between the inference results in the form of calculation of the belief distribution is adopted as accuracy.
The embodiment of the invention also provides an intelligent fault diagnosis system for self-organizing reconstruction of the unmanned production line, wherein the system processes the operation data of the equipment through a plurality of base EBRB subsystems trained in advance, outputs a plurality of reasoning results, performs information aggregation on the plurality of reasoning results, judges whether the equipment has faults, and outputs the fault type; wherein, a plurality of base EBRB subsystems trained in advance include:
the data acquisition module is used for acquiring a data set, and constructing an extended confidence rule base based on a training set in the data set, wherein the data set comprises equipment operation data and fault types;
the reduction module is used for constructing a multi-objective optimization model, optimizing and expanding the confidence rule base based on the multi-objective optimization model to obtain a reduced rule base, and the relative weight of the rule and the weight of the precondition attribute;
the construction module is used for constructing a plurality of base EBRB subsystems according to the reduced rule base, the relative weights of the rules and the weights of the premise attributes;
and the test module is used for testing the plurality of base EBRB subsystems through a test set in the data set.
It may be understood that, the intelligent fault diagnosis system for self-organizing and reconstructing an unmanned production line provided by the embodiment of the invention corresponds to the intelligent fault diagnosis method for self-organizing and reconstructing an unmanned production line, and the explanation, the examples, the beneficial effects and the like of the relevant content can refer to the corresponding content in the intelligent fault diagnosis method for self-organizing and reconstructing an unmanned production line, which is not repeated here.
The embodiment of the invention also provides a computer readable storage medium which stores a computer program for intelligent fault diagnosis of the unmanned line-oriented self-organizing reconstruction, wherein the computer program enables a computer to execute the intelligent fault diagnosis method of the unmanned line-oriented self-organizing reconstruction.
The embodiment of the invention also provides electronic equipment, which comprises:
one or more processors;
a memory; and
one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs comprising an intelligent fault diagnosis method for performing the unmanned line-oriented self-organizing reconstruction as described above.
In summary, compared with the prior art, the method has the following beneficial effects:
1. according to the embodiment of the invention, the problem of more inconsistent rules and redundant rules caused by larger rule base scale is solved through rule reduction, the inconsistency of the rule base obtained after the rule reduction is lower, and the fault diagnosis precision is improved.
2. The embodiment of the invention solves the problem of subjective determination of the rule weight and the attribute weight, further objectively calculates the activation weight of the rule and determines the activation rule, and improves the accuracy of fault diagnosis.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. An intelligent fault diagnosis method for self-organizing reconstruction of an unmanned production line is characterized in that the method processes operation data of equipment through a plurality of base EBRB subsystems trained in advance, outputs a plurality of reasoning results, performs information aggregation on the plurality of reasoning results, judges whether the equipment has faults, and outputs fault types; the training process of the pre-trained multiple base EBRB subsystems comprises the following steps:
s1, acquiring a data set, and constructing an extended confidence rule base based on a training set in the data set, wherein the data set comprises equipment operation data and fault types;
s2, constructing a multi-objective optimization model, and optimizing and expanding a confidence rule base based on the multi-objective optimization model to obtain a rule base after reduction, and the relative weight of the rule and the weight of the precondition attribute;
s3, constructing a plurality of base EBRB subsystems according to the reduced rule base, the relative weights of the rules and the weights of the precondition attributes;
s4, testing the plurality of base EBRB subsystems through a testing set in the data set.
2. The intelligent fault diagnosis method for self-organizing reconstruction of an unmanned production line according to claim 1, wherein the multi-objective optimization model comprises:
MIN E(δ i ,θ k ) (4)
MIN In(τ k ) (5)
0≤τ k ≤1,k=1,…,L TR (6)
0≤δ i ≤1,i=1,…,M (7)
0≤θ k ≤1,k=1,…,L TR (9)
wherein , representing a system reasoning result; />Representing belief distribution obtained by converting the true diagnosis result of the data, wherein n=1, … and N; τ k R represents k Index of whether to decrease, if τ k <0.5 then means pruned the rule, otherwise the rule is preserved; in (τ) k ) Representing the degree of inconsistency of the constructed rule base; delta i Representing the weight of the ith precondition attribute in the extended confidence rule base; θ k Representing the relative weight of the kth extended confidence rule; l (L) TR Representing the number of rules in the training dataset.
3. The intelligent fault diagnosis method for self-organizing reconstruction of an unmanned production line according to claim 2, wherein the calculation method for the inconsistency degree of the rule base comprises the following steps:
the inconsistent degree of the rule base is obtained by accumulatively calculating the inconsistent degree of each rule, wherein the inconsistent degree of each rule is determined by the precondition attribute and the result attribute similarity among the rules; the similarity of belief distribution between two rule precondition attributes is expressed as SRA, and the similarity of belief distribution between result attributes is expressed as SRC, and the calculation process is as follows:
and then calculating the consistency degree and the inconsistency degree between the two rules:
Cons(R t ,R k )=1–(SRA(t,k)-SRC(t,k)) 2 (12)
according to the above formula, the degree of inconsistency of the confidence rule base is:
wherein ,representing the reference value A used in the kth rule i,j Description precondition attribute U i And satisfy the belief degree of Representing the reference value D used in the kth rule n Belief describing the result attribute D and satisfyingL TR A number of bars representing rules in the training dataset; r is R k Representing the kth expanded belief rule; r is R t Representing the t-th expanded belief rule.
4. The intelligent fault diagnosis method for self-organizing reconstruction of an unmanned production line according to claim 1, wherein S4 comprises:
s401, testing a plurality of base EBRB subsystems through a testing set in a data set, and outputting a testing result;
s402, the analysis evidence reasoning method is used for information aggregation of the test results.
5. The intelligent fault diagnosis method for self-organizing reconstruction of an unmanned production line according to claim 4, wherein the step S401 comprises:
in the p-th base EBRB subsystem system, random extraction is performed from a rule base obtained by rule reductionRules are used to build a rule base for the base EBRB, which is then based on the belief structure of the input test dataPrecondition attribute U for the kth alternative activation rule i The individual matching degree between the two is calculated as follows:
wherein ,precondition attribute U representing the kth alternative activation rule i Described as reference value A i,j Belief of alpha i,j Representing input data +.>Described as reference value A i,j Is a confidence level of (1);
and calculating the activation weight of the kth alternative activation rule by using the obtained individual matching degree, wherein the calculation process is as follows:
6. The intelligent fault diagnosis method for self-organizing reconstruction of an unmanned production line according to claim 4, wherein S402 comprises:
the activated rules are clustered based on an analytical evidence reasoning method, and the method comprises the following steps:
the result obtained by the aggregation of the evidence reasoning algorithm is expressed as
p-base EBThe RB subsystem system obtains the reasoning result that The reasoning results of the P base EBRB subsystems are aggregated by utilizing the analytic evidence reasoning method in the formula (17) to obtainThe normalized reasoning precision of the P base EBRB subsystems is the weight of each reasoning result in the information aggregation process, and the calculation formula is as follows:
suppose D n The nth level representing the classification problem, the output classification result of the integrated extended confidence rule system is as follows:
7. An intelligent fault diagnosis system for self-organizing reconstruction of an unmanned production line is characterized in that operation data of equipment are processed through a plurality of base EBRB subsystems trained in advance in the equipment fault diagnosis system, a plurality of reasoning results are output, information aggregation is carried out on the plurality of reasoning results, whether the equipment has faults or not is judged, and the fault type is output; wherein, a plurality of base EBRB subsystems trained in advance include:
the data acquisition module is used for acquiring a data set, and constructing an extended confidence rule base based on a training set in the data set, wherein the data set comprises equipment operation data and fault types;
the reduction module is used for constructing a multi-objective optimization model, optimizing and expanding the confidence rule base based on the multi-objective optimization model to obtain a reduced rule base, and the relative weight of the rule and the weight of the precondition attribute;
the construction module is used for constructing a plurality of base EBRB subsystems according to the reduced rule base, the relative weights of the rules and the weights of the premise attributes;
and the test module is used for testing the plurality of base EBRB subsystems through a test set in the data set.
8. The intelligent fault diagnosis system for self-organizing reconstruction of an unmanned production line of claim 7, wherein the multi-objective optimization model comprises:
MIN E(δ i ,θ k ) (4)
MIN In(τ k ) (5)
0≤τ k ≤1,k=1,…,L TR (6)
0≤δ i ≤1,i=1,…,M (7)
0≤θ k ≤1,k=1,…,L TR (9)
wherein , representing a system reasoning result; />Representing belief distribution obtained by converting the true diagnosis result of the data, wherein n=1, … and N; τ k R represents k Index of whether to decrease, if τ k <0.5 then means pruned the rule, otherwise the rule is preserved; in (τ) k ) Representing the degree of inconsistency of the constructed rule base; delta i Representing the weight of the ith precondition attribute in the extended confidence rule base; θ k Representing the relative weight of the kth extended confidence rule; l (L) TR Representing the number of rules in the training dataset.
9. A computer-readable storage medium storing a computer program for intelligent fault diagnosis for unmanned line-oriented self-organizing reconstruction, wherein the computer program causes a computer to execute the unmanned line-oriented self-organizing reconstruction intelligent fault diagnosis method according to any one of claims 1 to 6.
10. An electronic device, comprising:
one or more processors;
a memory; and
one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs comprising an intelligent fault diagnosis method for performing the unmanned line-oriented self-organizing reconstruction of any of claims 1-6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211103784.1A CN116224950A (en) | 2022-09-09 | 2022-09-09 | Intelligent fault diagnosis method and system for self-organizing reconstruction of unmanned production line |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211103784.1A CN116224950A (en) | 2022-09-09 | 2022-09-09 | Intelligent fault diagnosis method and system for self-organizing reconstruction of unmanned production line |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116224950A true CN116224950A (en) | 2023-06-06 |
Family
ID=86587992
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211103784.1A Pending CN116224950A (en) | 2022-09-09 | 2022-09-09 | Intelligent fault diagnosis method and system for self-organizing reconstruction of unmanned production line |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116224950A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117389155A (en) * | 2023-12-07 | 2024-01-12 | 西北工业大学 | Self-adaptive fault detection method and system for unmanned aerial vehicle cluster |
-
2022
- 2022-09-09 CN CN202211103784.1A patent/CN116224950A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117389155A (en) * | 2023-12-07 | 2024-01-12 | 西北工业大学 | Self-adaptive fault detection method and system for unmanned aerial vehicle cluster |
CN117389155B (en) * | 2023-12-07 | 2024-04-09 | 西北工业大学 | Self-adaptive fault detection method and system for unmanned aerial vehicle cluster |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108320043B (en) | Power distribution network equipment state diagnosis and prediction method based on electric power big data | |
CN110703057B (en) | Power equipment partial discharge diagnosis method based on data enhancement and neural network | |
CN110929847A (en) | Converter transformer fault diagnosis method based on deep convolutional neural network | |
CN108304567B (en) | Method and system for identifying working condition mode and classifying data of high-voltage transformer | |
CN105930629B (en) | A kind of on-line fault diagnosis method based on magnanimity service data | |
CN111340248A (en) | Transformer fault diagnosis method and system based on intelligent integration algorithm | |
CN113866642A (en) | Lithium ion battery fault diagnosis method based on gradient lifting tree | |
CN110458039A (en) | A kind of construction method of industrial process fault diagnosis model and its application | |
CN114997342B (en) | SCR fault diagnosis method, device, equipment and storage medium | |
CN111639850A (en) | Quality evaluation method and system for multi-source heterogeneous data | |
CN116956215A (en) | Fault diagnosis method and system for transmission system | |
CN116224950A (en) | Intelligent fault diagnosis method and system for self-organizing reconstruction of unmanned production line | |
CN115587290A (en) | Aero-engine fault diagnosis method based on variational self-coding generation countermeasure network | |
CN118114118A (en) | CNDT-based fault diagnosis method for typical weapon equipment | |
CN116070140B (en) | Power distribution substation safe operation state monitoring system and method | |
CN113340547A (en) | Winch vibration diagnosis method based on improved lumped mean-square decomposition | |
CN116910677A (en) | Industrial instrument fault diagnosis method and system | |
CN112001530A (en) | Predictive maintenance method and system for transformer oil chromatography online monitoring device | |
CN116244600A (en) | Method, system and equipment for constructing GIS intermittent discharge mode identification model | |
CN114137915B (en) | Fault diagnosis method for industrial equipment | |
CN116522241A (en) | Intelligent ammeter reliability assessment method based on survival analysis | |
CN113889274B (en) | Method and device for constructing risk prediction model of autism spectrum disorder | |
CN112069621B (en) | Method for predicting residual service life of rolling bearing based on linear reliability index | |
CN114252266A (en) | Rolling bearing performance degradation evaluation method based on DBN-SVDD model | |
Cao et al. | Bearing remaining useful life prediction based on optimized support vector regression model with denoising technique |
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 |