CN113887452A - Fault diagnosis optimization method based on correlation matrix - Google Patents

Abstract

The invention discloses a fault diagnosis optimization method based on a correlation matrix, which comprises the following steps: 1) establishing a multi-signal flow graph model according to a target system to be diagnosed and a test scheme; 2) generating a correlation matrix of the fault and the test according to the multi-signal flow graph model; 3) judging whether a fault redundancy test exists according to the generated correlation matrix, and removing the redundancy test; wherein the redundancy test is that the same indistinguishable columns exist in the matrix; 4) and constructing fault diagnosis strategies under different search width and depth combinations by adopting an information gain-based method, comparing average test costs under different search widths and depths, and optimally designing the fault diagnosis strategies. The invention can improve the searching efficiency and precision of the fault diagnosis test sequence.

Description

Fault diagnosis optimization method based on correlation matrix

Technical Field

The invention relates to a fault diagnosis technology, in particular to a fault diagnosis optimization method based on a correlation matrix.

Background

With the development of science and technology, the performance of a large-scale system is continuously improved, and meanwhile, the complexity of the system is increased, so that the difficulty of fault diagnosis of the system is increased day by day, the diagnosis cost is more expensive, and therefore, the research and optimization of a fault diagnosis algorithm of the system are more and more urgent.

The purpose of the integrated diagnostics is to detect and isolate all known or average faults in the system, at the lowest cost to meet the mission requirements of the system. A system with good testability can greatly reduce the time required by fault detection and isolation, thereby obviously shortening the maintenance time, and reducing the corresponding skill requirement on maintenance personnel, thereby achieving the purposes of improving the reliability of the system and reducing the life cycle cost. The testability analysis design is developed in the analysis design stage, so that a large amount of modeling time can be saved, and the system design efficiency is improved. The system which is designed completely and is in operation is upgraded and modified and maintained daily, and testability analysis is also needed to obtain an optimal test scheme for guiding maintenance work, so that timely, accurate and low-cost maintenance work is realized.

At present, algorithms for generating fault diagnosis strategies include a dynamic programming algorithm, a greedy algorithm, an ant colony algorithm, a genetic algorithm, and the like. The existing fault strategy algorithm has certain defects, and can be roughly expressed as follows:

1) dynamic programming algorithms are not suitable for complex systems.

2) The greedy algorithm can only solve the local optimal solution of the system, and cannot guarantee the global optimal solution.

3) The ant colony algorithm and the genetic algorithm cannot guarantee that a global optimal solution can be obtained certainly, and meanwhile, the convergence speed is low in the later stage of the algorithm, the algorithm is prone to falling into a local optimal solution and the like.

Disclosure of Invention

The technical problem to be solved by the invention is to provide a fault diagnosis optimization method based on a correlation matrix aiming at the defects in the prior art.

The technical scheme adopted by the invention for solving the technical problems is as follows: a fault diagnosis optimization method based on a correlation matrix comprises the following steps:

1) establishing a multi-signal flow graph model according to a target system to be diagnosed and a test scheme, and completing the description of the correlation between faults possibly occurring in each component of the system and the test; the multi-signal flowsheet model comprises:

the multi-signal flowsheet model includes the following elements:

1.1) limited set of m +1 system faults X ═ X₀,x₁,……,x_mIn which x₀Indicating a no fault condition, x_i，1<i<m, representing one of m possible fault states in the system;

1.2) probability of occurrence of failure p ═ { p ═ p₀,p₁……,p_m}；

1.3) test point TP ═ TP₁,TP₂,……，TP_qEach test point at least contains one test t_i；

1.4) test set t ═ t₁,…，t_j…,t_nWhere the cost of the corresponding test is C ═ C₁,c₂,……,c_n}；

2) Generating a correlation matrix of the fault and the test according to the multi-signal flow graph model; matrix element d of the correlation matrix_ijIs a Boolean variable, if fault x_iCan be tested t_jDetected, that order d_ijIs 1, otherwise is 0;

3) judging whether a fault redundancy test exists according to the generated correlation matrix, and if the fault redundancy test exists, rejecting the redundancy test; wherein the redundancy test is that the same indistinguishable columns exist in the matrix;

4) and constructing fault diagnosis strategies under different search width and depth combinations by adopting an information gain-based method, comparing average test costs under different search widths and depths, and optimally designing the fault diagnosis strategies.

According to the scheme, the step 4) is as follows:

4.1) calculating the average information increment of each test;

selecting a current fault state fuzzy set X, and sequentially selecting a test t from a test set t_jFor each test t_jThe original fault state fuzzy set will be divided into two new OR nodes x_jpAnd x_jfRespectively corresponding to "test passed" and "test failed"; dividing the fuzzy fault set according to the test result until all faults are isolated or all tests are used; at test t_jThen, the system obtains the information gain as follows:

wherein p (x) is ∑_xi∈xp(x_i)；x＝xjp∪xjf；

In the formula, IG (X; t)_j) Indicates the passage of the test t_jIncrement of information of the later set X, c_jTo test t_jThe cost of (a); equation (4) is called an information increment heuristic function;

4.2) the searching algorithm established according to the information gain heuristic function comprises the following steps:

step 4.2.1) firstly, establishing a new set Z, wherein the set only comprises a root node S, and the S is a complete fuzzy node; creating an empty graph G;

step 4.2.2) repeat the steps listed below until the set Z is empty; then, the test sequence in the graph G is used as a diagnosis tree to identify a fault source;

step 4.2.2.1) extracting an OR node q from the set Z, and adding the node q into the graph G; if q is not a terminal node, then use each test t in the available test set_jNode qThe division into pass and fail sets: x is the number of_qjpAnd x_qif；x_qjpAnd x_qifEstablishing a new set Y and an empty graph G 'for the direct subsequent OR nodes of the node q, and adding all the direct subsequent OR nodes of the node q into the graph G'; if q is a terminal node, continuing to select the next OR node in the set Z; repeating the following steps until the test depth of the graph G' is l, wherein l is a test depth parameter defined in advance;

step 4.2.2.1.1) extracting an OR node r from the set Y; if the node r is the target node, continuing to process the next OR node in the set Y; otherwise, calculating the unit cost information increment of each available test on the node according to the step 4.1);

step 4.2.2.1.2) for node r, selecting the test with the largest increment of test information per unit cost, if a plurality of test values are the same, selecting the test with the smallest index, then adding the test into graph G ', dividing node r into passing and failing subset fuzzy nodes through the test, and respectively adding the obtained subset nodes into set Y and graph G';

step 4.2.2.2) calculating the average test cost of each direct subsequent OR node of the node q according to the generated fault diagnosis strategy stored in the graph G';

step 4.2.2.3) calculate each available test t for node q using the following formula_j∈T_qAverage test cost of (d):

wherein,

p(x_qjf)＝1-p(x_qjp)；

wherein, h (x)_qjp) And h (x)_qif) Test pass and test fail subsets x, respectively_qjpAnd x_qifGenerated diagnostic strategyAveraging the test costs;

step 4.2.2.4) sorting the tests based on the average test cost to select n best tests t^*∈T_q(ii) a If the values of several tests are the same, then the test with the smallest index is selected to be t^*Added to graph G while adding t^*The subsets of pass and fail resulting from the pass and fail results, respectively, are appended to set Z and graph G, generating a diagnostic tree.

The invention has the following beneficial effects:

the invention combines the information gain heuristic function with the diagnosis strategy search algorithm, carries out iterative update, increases the search width and depth, constructs an approximate optimal diagnosis strategy, can reduce the average test cost, is suitable for a complex system and can ensure the overall optimal solution;

through result analysis, the influence trend of different search widths and depths on the average test cost can be obtained, and the search process of the test sequence is guided and optimized, so that the search efficiency and the accuracy are improved.

Drawings

The invention will be further described with reference to the accompanying drawings and examples, in which:

FIG. 1 is a flow chart of a method of an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

As shown in fig. 1, a fault diagnosis optimization method based on a correlation matrix includes the following steps:

1) establishing a multi-signal flow diagram model according to a related structure, a function and a test scheme of a system to be diagnosed, wherein the content of the multi-signal flow diagram model comprises a system component, a signal set related to a fault module, a limited set of test points and a test priority set, and the description of a related relation between faults and tests possibly occurring in each component of the system is completed;

establishing a multi-signal flow graph model according to a target system, wherein the multi-signal flow graph model consists of the following elements:

a finite set of m +1 system faults X ═ X₀,x₁,……,x_mIn which x₀Indicates a no fault condition, xi (1)<i<m) represents one of m possible fault conditions in the system;

2. probability of occurrence of failure p ═ p₀,p₁……,p_m}；

3. Test point TP ═ { TP ═ TP₁,TP₂,……TP_pEach test point at least comprises a test t;

4. test t ═ t₀,t₁,……,t_nWhere the cost of the test is C ═ C₁,c₂,……,c_nJudging by the required time, the manpower requirement or other economic factors;

5. set of independent function signals S ═ S₁,s₂,……,s_kFor representing observed effects in the system, each system state modifying a function signal SX (x)_i) Each test can detect a set of functional signals ST (t)_j)。

2) After a multi-signal flow graph model is established, obtaining first-order correlation between a fault and a test according to whether a connection relation exists between a fault node and a test node in the model; judging the connection relation between the fault node and the fault node, obtaining high-order correlation by first-order correlation recursion, and recording a fault-test correlation matrix as follows:

wherein the matrix element d_mnTypically a boolean variable, if fault x_iCan be tested t_jDetected, that order d_ijIs 1, otherwise is 0.

And judging whether a redundancy test exists according to the generated correlation matrix, if so, eliminating the redundancy test, and reserving one test, thereby realizing the optimization of the matrix.

And adopting a search algorithm based on information gain to the optimized matrix, and constructing fault diagnosis strategies under different search width and depth combinations.

3) Judging whether a fault redundancy test exists according to the generated correlation matrix, and if the fault redundancy test exists, rejecting the redundancy test; wherein the redundancy test is that the same indistinguishable columns exist in the matrix;

4) and constructing fault diagnosis strategies under different search width and depth combinations by adopting an information gain-based method, comparing average test costs under different search widths and depths, and optimally designing the fault diagnosis strategies.

The step 4) is as follows:

4.1) calculating the average information increment of each test;

the heuristic function of information gain is to find the average information increment of each test, take the value as the standard for judging the test quality, select the fuzzy set X of the current fault state to be isolated, and select the test t from the test set t in turn_jFor each test t_jThe original fault state fuzzy set will be divided into two new OR nodes x_jpAnd x_jfRespectively corresponding to "test passed" and "test failed"; dividing the fuzzy fault set according to the test result until all faults are isolated or all tests are used; at test t_jThen, the system obtains the information gain as follows:

wherein p (x) is ∑_xi∈xp(x_i)；x＝xjp∪xjf；

The value of the information entropy reflects the degree of the failure mode that causes the system failure,

in the formula, IG (X; t)_j) Indicates the passage of the test t_jIncrement of information of the later set X, c_jTo test t_jThe cost of (a); equation (4) is called an information increment heuristic function;

4.2) determining according to an information gain heuristic function, wherein the searching algorithm comprises the following steps:

step 4.2.1) firstly, establishing a new set Z, wherein the set only comprises a root node S, and the S is a complete fuzzy node; creating an empty graph G;

step 4.2.2) repeat the steps listed below until the set Z is empty; then, the test sequence in the graph G is used as a diagnosis tree to identify a fault source;

step 4.2.2.1) extracting an OR node q from the set Z, and adding the node q into the graph G; if q is not a terminal node, then use each test t in the available test set_jThe nodes q are divided into two sets, pass and fail: x is the number of_qjpAnd x_qif；x_qjpAnd x_qifEstablishing a new set Y and an empty graph G 'for the direct subsequent OR nodes of the node q, and adding all the direct subsequent OR nodes of the node q into the graph G'; if q is a terminal node, continuing to select the next OR node in the set Z; repeating the following steps until the test depth of the graph G' is l, wherein l is a test depth parameter defined in advance;

step 4.2.2.1.1) extracting an OR node r from the set Y; if the node r is the target node, continuing to process the next OR node in the set Y; otherwise, calculating the unit cost information increment of each available test on the node according to the step 4.1);

step 4.2.2.1.2) for node r, selecting the test with the largest increment of test information per unit cost, if a plurality of test values are the same, selecting the test with the smallest index, then adding the test into graph G ', dividing node r into passing and failing subset fuzzy nodes through the test, and respectively adding the obtained subset nodes into set Y and graph G';

step 4.2.2.2) calculating the average test cost of each direct subsequent OR node of the node q according to the generated fault diagnosis strategy stored in the graph G';

step 4.2.2.3) calculate each available test t for node q using the following formula_j∈T_qAverage test cost of (d):

wherein,

p(x_qjf)＝1-p(x_qjp)；

wherein, h (x)_qjp) And h (x)_qif) Test pass and test fail subsets x, respectively_qjpAnd x_qifAverage test cost of the generated diagnostic strategy;

step 4.2.2.4) sorting the tests based on the average test cost to select n best tests t^*∈T_q(ii) a If the values of several tests are the same, then the test with the smallest index is selected to be t^*Added to graph G while adding t^*The subsets of pass and fail resulting from the pass and fail results, respectively, are appended to set Z and graph G, generating a diagnostic tree.

Through the steps, diagnosis strategies with different search widths and depth combined structures are compared, so that lower average test cost can be obtained, and the method has guiding significance on the search process of the test sequence.

It will be understood that modifications and variations can be made by persons skilled in the art in light of the above teachings and all such modifications and variations are intended to be included within the scope of the invention as defined in the appended claims.

Claims (2)

1. A fault diagnosis optimization method based on a correlation matrix is characterized by comprising the following steps:

1) establishing a multi-signal flow graph model according to a target system to be diagnosed and a test scheme, and completing the description of the correlation between faults possibly occurring in each component of the system and the test; the multi-signal flowsheet model comprises:

the multi-signal flowsheet model includes the following elements:

1.1) limited set of m +1 system faults X ═ X₀,x₁,……,x_mIn which x₀Indicating a no fault condition, x_i，1<i<m, representing one of m possible fault states in the system;

1.2) probability of occurrence of failure p ═ { p ═ p₀,p₁……,p_m}；

1.3) test point TP ═ TP₁,TP₂,……，TP_qEach test point at least contains one test t_i；

1.4) test set t ═ t₁,…，t_j…,t_nWhere the cost of the corresponding test is C ═ C₁,c₂,……,c_n}；

2) Generating a correlation matrix of the fault and the test according to the multi-signal flow graph model; matrix element d of the correlation matrix_ijIs a Boolean variable, if fault x_iCan be tested t_jDetected, that order d_ijIs 1, otherwise is 0;

3) judging whether a fault redundancy test exists according to the generated correlation matrix, and if the fault redundancy test exists, rejecting the redundancy test; wherein the redundancy test is that the same indistinguishable columns exist in the matrix;

4) and constructing fault diagnosis strategies under different search width and depth combinations by adopting an information gain-based method, comparing average test costs under different search widths and depths, and optimally designing the fault diagnosis strategies.

2. The correlation matrix-based fault diagnosis optimization method according to claim 1, wherein the step 4) is specifically as follows:

4.1) calculating the average information increment of each test;

selecting a current fault state fuzzy set X, and sequentially selecting a test t from a test set t_jFor each test t_jThe original fault state fuzzy set will be divided into two new OR nodes x_jpAnd x_jfRespectively corresponding to "test passed" and "test failed"; dividing the fuzzy fault set according to the test result until all faults are isolated or all tests are used; at test t_jThen, the system obtains the information gain as follows:

wherein p (x) is ∑_xi∈xp(x_i)；x＝xjp∪xjf；

In the formula, IG (X; t)_j) Indicates the passage of the test t_jIncrement of information of the later set X, c_jTo test t_jThe cost of (a); equation (4) is called an information increment heuristic function;

4.2) the searching algorithm established according to the information gain heuristic function comprises the following steps:

step 4.2.1) firstly, establishing a new set Z, wherein the set only comprises a root node S, and the S is a complete fuzzy node; creating an empty graph G;

step 4.2.2) repeat the steps listed below until the set Z is empty; then, the test sequence in the graph G is used as a diagnosis tree to identify a fault source;

step 4.2.2.1) extracting an OR node q from the set Z, and adding the node q into the graph G; if q is not a terminal node, then use each test t in the available test set_jThe nodes q are divided into two sets, pass and fail: x is the number of_qjpAnd x_qif；x_qjpAnd x_qifEstablishing a new set Y and an empty graph G 'for the direct subsequent OR nodes of the node q, and adding all the direct subsequent OR nodes of the node q into the graph G'; if q is a terminal node, continuing to select the next OR node in the set Z; repeating the following steps until the test depth of the graph G' is l, wherein l is a test depth parameter defined in advance;

step 4.2.2.1.1) extracting an OR node r from the set Y; if the node r is the target node, continuing to process the next OR node in the set Y; otherwise, calculating the unit cost information increment of each available test on the node according to the step 4.1);

step 4.2.2.1.2) for node r, selecting the test with the largest increment of test information per unit cost, if a plurality of test values are the same, selecting the test with the smallest index, then adding the test into graph G ', dividing node r into passing and failing subset fuzzy nodes through the test, and respectively adding the obtained subset nodes into set Y and graph G';

step 4.2.2.2) calculating the average test cost of each direct subsequent OR node of the node q according to the generated fault diagnosis strategy stored in the graph G';

step 4.2.2.3) calculate each available test t for node q using the following formula_j∈T_qAverage test cost of (d):

wherein,

p(x_qjf)＝1-p(x_qjp)；

wherein, h (x)_qjp) And h (x)_qif) Test pass and test fail subsets x, respectively_qjpAnd x_qifAverage test cost of the generated diagnostic strategy;

step 4.2.2.4) sorting the tests based on the average test cost to select n best tests t^*∈T_q(ii) a If the values of several tests are the same, then the test with the smallest index is selected to be t^*Added to graph G while adding t^*The subsets of pass and fail resulting from the pass and fail results, respectively, are appended to set Z and graph G, generating a diagnostic tree.

Images