CN111126801A - Hierarchical analysis and evaluation system based on information entropy for equipment guarantee capability - Google Patents

Abstract

The invention belongs to the technical field of equipment test identification, and particularly relates to a hierarchical analysis and evaluation system based on information entropy for equipment guarantee capability. Compared with the prior art, the method has the advantages that the influence caused by the uncertainty of the test data of the equipment guarantee capability can not be ignored when the test evaluation is carried out, the uncertainty of the test data is quantitatively expressed, the credibility of the test data is obtained by increasing the data quantitative information when the hierarchical method evaluation is carried out, the corresponding uncertainty weight is given to the test data, the data quantitative information of the hierarchical analysis method is increased, and the evaluation accuracy can be effectively improved.

Description

Hierarchical analysis and evaluation system based on information entropy for equipment guarantee capability

Technical Field

The invention belongs to the technical field of equipment test identification, and particularly relates to a hierarchical analysis and evaluation system based on information entropy for equipment guarantee capability.

Background

In the equipment test evaluation, in the test evaluation method aiming at the equipment guarantee capability, the analytic hierarchy process is taken as a comprehensive evaluation method combining quantitative analysis method and qualitative analysis method, has the characteristics of simplicity, practicability and less required quantitative data information, and is suitable for system evaluation without structural characteristics and system evaluation of multiple targets, multiple criteria, multiple periods and the like. Then, the hierarchical analysis method has less quantitative data and more qualitative components, and is not easy to convince. Increasing quantitative data information of the analytic hierarchy process can effectively improve the reliability of an evaluation result, but how to increase quantitative mathematical information becomes a difficult point of research. When the method is applied to the engineering field, the input of equipment evaluation is test data, the test data has randomness, fuzziness, grayness and uncertainty due to the characteristics of multiple data types, nonlinear relation, different scales, high dimensionality and the like, and the influence caused by the uncertainty of the test data aiming at the equipment guarantee capability is not negligible when the test evaluation is carried out.

Disclosure of Invention

Technical problem to be solved

The technical problem to be solved by the invention is as follows: in the data-based test evaluation process aiming at the equipment guarantee capability, the data uncertainty is quantitatively expressed, the credibility of objective evaluation is increased, and the accuracy of the analytic hierarchy process is improved.

(II) technical scheme

In order to solve the technical problem, the invention provides a hierarchical analysis and evaluation system based on information entropy for equipment guarantee capability, which comprises: the device comprises an equipment guarantee requirement analysis module, an evaluation task decomposition module, a performance measurement acquisition module, a data source acquisition module, an uncertainty quantization operation module, an index weight acquisition module and an evaluation result acquisition module;

the equipment guarantee requirement analysis module is used for determining specific evaluation targets of fault detection, maintenance, reliability and guarantee aiming at the requirement of equipment guarantee in an equipment system and carrying out evaluation aiming at equipment guarantee capability of an equipment test;

the assessment task decomposition module is used for further decomposing specific assessment targets of fault detection, maintenance, reliability and security according to a tree analysis technology according to assessment task requirements of equipment guarantee capacity, decomposing the specific assessment targets according to a tree form of 'assessment task-target-performance metric-data source', decomposing key problems to the extent that actual data requirements and test measurement can be determined, and obtaining an equipment guarantee capacity assessment task tree diagram, wherein the specific assessment targets are as follows: fault detection capability, fault repair capability, reliability and security;

the performance measurement acquisition module is used for defining a specific evaluation target set as E-E according to the decomposition result of the equipment guarantee capability evaluation task tree diagram₁,E₂,E₃,E₄And (E) sequentially evaluating the performance metrics corresponding to specific evaluation targets by using a test model, wherein the test model is a test model, and the test model is a test model₁) M (fault detection capability) { fault detection rate, fault isolation rate, false alarm rate }, M (E)₂) M (failure repair capability) ═ average repair time }, M (E)₃) M (reliability) { mean time between failures, mean time between repairs }, M (E)₄) M (guaranteed) ═ M (guaranteed) { task rate executed, degree of availability };

the data source obtaining module is configured to determine, according to the performance metric corresponding to the evaluation target obtained by the performance metric obtaining module, a data source corresponding to each performance metric, and define a data source D (fault detection rate) ═ times }, a D (fault isolation rate) ═ times }, a D (false alarm rate) ═ times }, a D (average repair time) ═ time, times }, a D (average fault interval time) ═ time }, a D (average repair interval time) ═ time, times }, a D (execution task rate) ═ times }, and a D (availability) ═ times };

the uncertainty quantization operation module is used for carrying out uncertainty quantization calculation based on the information entropy on the data source obtained by the data source obtaining module to obtain an uncertainty matrix corresponding to the data source;

concept of information entropy: let P be { P ═ P_xX, X ∈ X } is a probability distribution over the domain of randomness uncertainty test data X, and the formula defining the uncertainty measure q (X) of the test data X is:

aiming at a data source D (fault detection rate), D (fault isolation rate), D (false alarm rate), D (average repair time), D (average fault interval time), D (average maintenance interval time), D (task execution rate) and D (usability), sequentially utilizing an uncertainty measure formula to calculate to obtain corresponding uncertainty measures Q (fault detection rate), Q (fault isolation rate), Q (false alarm rate), Q (average repair time), Q (average fault interval time), Q (average maintenance interval time), Q (task execution rate) and Q (usability), and forming an uncertainty matrix Q by the uncertainty measures; carrying out normalization processing on the uncertainty measurement matrix Q to obtain a normalized uncertainty measurement matrix Q', and further carrying out negation to obtain an uncertainty matrix U;

the formula of the normalization processing is as follows:

Q'＝{Q－min(Q)}/{max(Q)－min(Q)}

U＝1－Q'；

the index weight acquisition module is used for constructing a judgment matrix K with the performance metric meeting consistency conditions according to the performance metric corresponding to the specific evaluation target obtained by the performance metric acquisition module, normalizing each column of the judgment matrix by using a sum-product method, then summing the normalized data according to rows, carrying out normalization processing on the obtained result to obtain a weight matrix W of the performance metric, carrying out consistency check on the judgment matrix, and obtaining the final index weight W after the check;

the evaluation result obtaining module is configured to fuse the uncertainty matrix U obtained by the uncertainty quantization operation module and the final index weight W obtained by the index weight obtaining module to obtain a final weight matrix W' of the performance metric, as shown in the following formula:

W'＝U×W

and thus, the weight of the index determined by the equipment guarantee capability based on the analytic hierarchy process of the data information entropy is obtained, the size of the index is sorted, and the larger the weight corresponding to a specific index is, the more important the specific index is, and thus the final evaluation result is obtained.

(III) advantageous effects

Compared with the prior art, the method has the advantages that the influence caused by the uncertainty of the test data of the equipment guarantee capability can not be ignored when the test evaluation is carried out, the uncertainty of the test data is quantitatively expressed, the credibility of the test data is obtained by increasing the data quantitative information when the hierarchical method evaluation is carried out, the corresponding uncertainty weight is given to the test data, the data quantitative information of the hierarchical analysis method is increased, and the evaluation accuracy can be effectively improved.

Drawings

Fig. 1 is a schematic diagram provided by the technical scheme of the invention.

FIG. 2 is a schematic diagram of a tree analysis technique.

FIG. 3 is a tree analysis diagram of equipment security capability.

Detailed Description

In order to make the objects, contents, and advantages of the present invention clearer, the following detailed description of the embodiments of the present invention will be made in conjunction with the accompanying drawings and examples.

In order to solve the above technical problem, the present invention provides a hierarchical analysis and evaluation system based on information entropy for equipment security capability, as shown in fig. 1 to 3, the system includes: the device comprises an equipment guarantee requirement analysis module, an evaluation task decomposition module, a performance measurement acquisition module, a data source acquisition module, an uncertainty quantization operation module, an index weight acquisition module and an evaluation result acquisition module;

the equipment guarantee requirement analysis module is used for determining specific evaluation targets of fault detection, maintenance, reliability and guarantee aiming at the requirement of equipment guarantee in an equipment system and carrying out evaluation aiming at equipment guarantee capability of an equipment test;

the assessment task decomposition module is used for further decomposing specific assessment targets of fault detection, maintenance, reliability and security according to a tree analysis technology according to assessment task requirements of equipment guarantee capacity, decomposing the specific assessment targets according to a tree form of 'assessment task-target-performance metric-data source', decomposing key problems to the extent that actual data requirements and test measurement can be determined, and obtaining an equipment guarantee capacity assessment task tree diagram, wherein the specific assessment targets are as follows: fault detection capability, fault repair capability, reliability and security;

the performance measurement acquisition module is used for defining a specific evaluation target set as E-E according to the decomposition result of the equipment guarantee capability evaluation task tree diagram₁,E₂,E₃,E₄And (E) sequentially evaluating the performance metrics corresponding to specific evaluation targets by using a test model, wherein the test model is a test model, and the test model is a test model₁) M (fault detection capability) { fault detection rate, fault isolation rate, false alarm rate }, M (E)₂) M (failure repair capability) ═ average repair time }, M (E)₃) M (reliability) { mean time between failures, mean time between repairs }, M (E)₄) M (guaranteed) ═ M (guaranteed) { task rate executed, degree of availability };

the data source obtaining module is configured to determine, according to the performance metric corresponding to the evaluation target obtained by the performance metric obtaining module, a data source corresponding to each performance metric, and define a data source D (fault detection rate) ═ times }, a D (fault isolation rate) ═ times }, a D (false alarm rate) ═ times }, a D (average repair time) ═ time, times }, a D (average fault interval time) ═ time }, a D (average repair interval time) ═ time, times }, a D (execution task rate) ═ times }, and a D (availability) ═ times };

the uncertainty quantization operation module is used for carrying out uncertainty quantization calculation based on the information entropy on the data source obtained by the data source obtaining module to obtain an uncertainty matrix corresponding to the data source;

concept of information entropy: let P be { P ═ P_xX, X ∈ X } is a probability distribution over the domain of randomness uncertainty test data X, and the formula defining the uncertainty measure q (X) of the test data X is:

aiming at a data source D (fault detection rate), D (fault isolation rate), D (false alarm rate), D (average repair time), D (average fault interval time), D (average maintenance interval time), D (task execution rate) and D (usability), sequentially utilizing an uncertainty measure formula to calculate to obtain corresponding uncertainty measures Q (fault detection rate), Q (fault isolation rate), Q (false alarm rate), Q (average repair time), Q (average fault interval time), Q (average maintenance interval time), Q (task execution rate) and Q (usability), and forming an uncertainty matrix Q by the uncertainty measures; carrying out normalization processing on the uncertainty measurement matrix Q to obtain a normalized uncertainty measurement matrix Q', and further carrying out negation to obtain an uncertainty matrix U;

the formula of the normalization processing is as follows:

Q'＝{Q－min(Q)}/{max(Q)－min(Q)}

U＝1－Q'；

the index weight acquisition module is used for constructing a judgment matrix K with the performance metric meeting consistency conditions according to the performance metric corresponding to the specific evaluation target obtained by the performance metric acquisition module, normalizing each column of the judgment matrix by using a sum-product method, then summing the normalized data according to rows, carrying out normalization processing on the obtained result to obtain a weight matrix W of the performance metric, carrying out consistency check on the judgment matrix, and obtaining the final index weight W after the check;

the evaluation result obtaining module is configured to fuse the uncertainty matrix U obtained by the uncertainty quantization operation module and the final index weight W obtained by the index weight obtaining module to obtain a final weight matrix W' of the performance metric, as shown in the following formula:

W'＝U×W

and thus, the weight of the index determined by the equipment guarantee capability based on the analytic hierarchy process of the data information entropy is obtained, the size of the index is sorted, and the larger the weight corresponding to a specific index is, the more important the specific index is, and thus the final evaluation result is obtained.

In addition, the present invention also provides a hierarchical analysis and evaluation method based on information entropy for equipment security capability, as shown in fig. 1 to 3, the method includes the following steps:

step 1: specific evaluation targets of fault detection, maintenance, reliability and guarantee are determined according to requirements of equipment guarantee in an equipment system, and evaluation is carried out according to equipment guarantee capability of an equipment test;

step 2: according to the requirement of an assessment task of equipment guarantee capacity, specific assessment targets of fault detection, maintenance, reliability and guarantee are further decomposed according to a tree analysis technology, and the decomposition is carried out according to a tree form of 'assessment task-target-performance measurement-data source', so that key problems are decomposed to a degree that actual data requirements and test measurement can be determined, and an equipment guarantee capacity assessment task tree diagram is obtained, wherein the specific assessment targets are as follows: fault detection capability, fault repair capability, reliability and security;

and step 3: according to the decomposition result of the equipment guarantee capability evaluation task tree diagram, defining a specific evaluation target set as E ═ { E ═ E }₁,E₂,E₃,E₄And (E) sequentially evaluating the performance metrics corresponding to specific evaluation targets by using a test model, wherein the test model is a test model, and the test model is a test model₁) M (fault detection capability) { fault detection rate, fault isolation rate, false alarm rate }, M (E)₂) M (failure repair capability) ═ average repair time }, M (E)₃) M (reliability) { mean time between failures, mean time between repairs }, M (E)₄) M (guaranteed) ═ M (guaranteed) { task rate executed, degree of availability };

and 4, step 4: defining a data source corresponding to each performance metric according to the performance metric having the evaluation target obtained in step 3, defining a data source D (fault detection rate) ═ times }, D (fault isolation rate) ═ times }, D (false alarm rate) ═ times }, D (mean repair time) ═ time, times }, D (mean fault interval time) { time }, D (mean repair interval time) { time, times }, D (execution task rate) ═ times }, and D (availability) ═ times };

and 5: carrying out uncertainty quantitative calculation based on the information entropy on the data source obtained in the step 4 to obtain an uncertainty matrix corresponding to the data source;

concept of information entropy: let P be { P ═ P_xX, X ∈ X } is a probability distribution over the domain of randomness uncertainty test data X, and the formula defining the uncertainty measure q (X) of the test data X is:

aiming at a data source D (fault detection rate), D (fault isolation rate), D (false alarm rate), D (average repair time), D (average fault interval time), D (average maintenance interval time), D (task execution rate) and D (usability), sequentially utilizing an uncertainty measure formula to calculate to obtain corresponding uncertainty measures Q (fault detection rate), Q (fault isolation rate), Q (false alarm rate), Q (average repair time), Q (average fault interval time), Q (average maintenance interval time), Q (task execution rate) and Q (usability), and forming an uncertainty matrix Q by the uncertainty measures; carrying out normalization processing on the uncertainty measurement matrix Q to obtain a normalized uncertainty measurement matrix Q', and further carrying out negation to obtain an uncertainty matrix U;

the formula of the normalization processing is as follows:

Q'＝{Q－min(Q)}/{max(Q)－min(Q)}

U＝1－Q'；

step 6: constructing a judgment matrix K with the performance measurement meeting consistency conditions according to the performance measurement corresponding to the specific evaluation target obtained in the step 3, normalizing each column of the judgment matrix by using a sum-product method, then summing the normalized data according to rows, carrying out normalization processing on the obtained result to obtain a weight matrix W of the performance measurement, carrying out consistency test on the judgment matrix, and obtaining a final index weight W after the test;

and 7: fusing the uncertainty matrix U in the step 5 with the final index weight W obtained in the step 6 to obtain a final weight matrix W' of the performance metric, as shown in the following formula:

W'＝U×W

and thus, the weight of the index determined by the equipment guarantee capability based on the analytic hierarchy process of the data information entropy is obtained, the size of the index is sorted, and the larger the weight corresponding to a specific index is, the more important the specific index is, and thus the final evaluation result is obtained.

The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.

Claims (1)

1. An information entropy-based hierarchical analysis evaluation system for equipment securing capability, the system comprising: the device comprises an equipment guarantee requirement analysis module, an evaluation task decomposition module, a performance measurement acquisition module, a data source acquisition module, an uncertainty quantization operation module, an index weight acquisition module and an evaluation result acquisition module;

the equipment guarantee requirement analysis module is used for determining specific evaluation targets of fault detection, maintenance, reliability and guarantee aiming at the requirement of equipment guarantee in an equipment system and carrying out evaluation aiming at equipment guarantee capability of an equipment test;

the assessment task decomposition module is used for further decomposing specific assessment targets of fault detection, maintenance, reliability and security according to a tree analysis technology according to assessment task requirements of equipment guarantee capacity, decomposing the specific assessment targets according to a tree form of 'assessment task-target-performance metric-data source', decomposing key problems to the extent that actual data requirements and test measurement can be determined, and obtaining an equipment guarantee capacity assessment task tree diagram, wherein the specific assessment targets are as follows: fault detection capability, fault repair capability, reliability and security;

the performance measurement acquisition module is used for defining a specific evaluation target set asE＝{E₁,E₂,E₃,E₄And (E) sequentially evaluating the performance metrics corresponding to specific evaluation targets by using a test model, wherein the test model is a test model, and the test model is a test model₁) M (fault detection capability) { fault detection rate, fault isolation rate, false alarm rate }, M (E)₂) M (failure repair capability) ═ average repair time }, M (E)₃) M (reliability) { mean time between failures, mean time between repairs }, M (E)₄) M (guaranteed) ═ M (guaranteed) { task rate executed, degree of availability };

the data source obtaining module is configured to determine, according to the performance metric corresponding to the evaluation target obtained by the performance metric obtaining module, a data source corresponding to each performance metric, and define a data source D (fault detection rate) ═ times }, a D (fault isolation rate) ═ times }, a D (false alarm rate) ═ times }, a D (average repair time) ═ time, times }, a D (average fault interval time) ═ time }, a D (average repair interval time) ═ time, times }, a D (execution task rate) ═ times }, and a D (availability) ═ times };

the uncertainty quantization operation module is used for carrying out uncertainty quantization calculation based on the information entropy on the data source obtained by the data source obtaining module to obtain an uncertainty matrix corresponding to the data source;

concept of information entropy: let P be { P ═ P_xX, X ∈ X } is a probability distribution over the domain of randomness uncertainty test data X, and the formula defining the uncertainty measure q (X) of the test data X is:

aiming at a data source D (fault detection rate), D (fault isolation rate), D (false alarm rate), D (average repair time), D (average fault interval time), D (average maintenance interval time), D (task execution rate) and D (usability), sequentially utilizing an uncertainty measure formula to calculate to obtain corresponding uncertainty measures Q (fault detection rate), Q (fault isolation rate), Q (false alarm rate), Q (average repair time), Q (average fault interval time), Q (average maintenance interval time), Q (task execution rate) and Q (usability), and forming an uncertainty matrix Q by the uncertainty measures; carrying out normalization processing on the uncertainty measurement matrix Q to obtain a normalized uncertainty measurement matrix Q', and further carrying out negation to obtain an uncertainty matrix U;

the formula of the normalization processing is as follows:

Q'＝{Q－min(Q)}/{max(Q)－min(Q)}

U＝1－Q'；

the index weight acquisition module is used for constructing a judgment matrix K with the performance metric meeting consistency conditions according to the performance metric corresponding to the specific evaluation target obtained by the performance metric acquisition module, normalizing each column of the judgment matrix by using a sum-product method, then summing the normalized data according to rows, carrying out normalization processing on the obtained result to obtain a weight matrix W of the performance metric, carrying out consistency check on the judgment matrix, and obtaining the final index weight W after the check;

the evaluation result obtaining module is configured to fuse the uncertainty matrix U obtained by the uncertainty quantization operation module and the final index weight W obtained by the index weight obtaining module to obtain a final weight matrix W' of the performance metric, as shown in the following formula:

W'＝U×W

and thus, the weight of the index determined by the equipment guarantee capability based on the analytic hierarchy process of the data information entropy is obtained, the size of the index is sorted, and the larger the weight corresponding to a specific index is, the more important the specific index is, and thus the final evaluation result is obtained.

Images