Equipment health management system function evaluation method based on variable weight fuzzy comprehensive evaluation

Abstract

The invention discloses an equipment health management system function evaluation method based on variable weight fuzzy comprehensive evaluation, and relates to the equipment health management system evaluation field, and the method comprises the steps: analyzing the function demands of an equipment health management system, respectively selecting qualitative indexes and quantitative indexes, the method comprises the following steps: constructing a function evaluation index system of two sets of equipment health management systems, constructing a hierarchical evaluation model, and obtaining initial weights of indexes in an index layer and a factor layer; correcting the initial weight of each index by setting a state variable weight function; and finally, based on a fuzzy comprehensive evaluation method, performing function evaluation on the equipment health management system in combination with the corrected index weight set to obtain a corresponding evaluation grade. According to the method, a perfect function evaluation index system of the equipment health management system is constructed through meticulous index screening, a scientific comprehensive evaluation model is constructed by adopting a fuzzy analytic hierarchy process and considering a variable weight principle, and the functionality of the equipment health management system can be effectively evaluated.

Description

Technical Field

[0001] The present invention relates to the field of equipment health management system evaluation, and more specifically to an equipment health management system function evaluation method based on variable weight fuzzy comprehensive evaluation. Background Art

[0002] With the continuous updating of current equipment and the continuous injection of high technology, the equipment is becoming more and more complex, and the high integration, high intelligence and high efficiency are increasing day by day. The requirements for equipment reliability, maintenance and support are also getting higher and higher. "Real-time, dynamic, fast and accurate" is the new theme of equipment support today. In recent years, in order to enhance equipment reliability and reduce the logistical support burden for equipment and equipment, health management technology has been widely used in equipment on land, sea, and in the air, and has achieved very significant application results. Engineering applications and technical analysis show that the health management system can greatly reduce the maintenance and support costs by reducing the demand for support resources such as spare parts, support equipment, and maintenance manpower; it can shorten the maintenance time and improve the functional integrity of equipment by reducing maintenance, especially the number of unplanned maintenance; it can reduce the risks caused by failures during the mission through health perception and improve the success rate of the mission.

[0003] Based on the significant benefits of the above health management system in equipment applications, it can be foreseen that with the rapid development of equipment health management technology around the world, health management systems will be applied to more and more equipment to improve the overall functional benefits of equipment. However, how to evaluate its functionality has become an important issue. A reasonable and effective comprehensive evaluation method can greatly improve the scientific nature of the health management system in equipment applications and provide valuable technical references for the design and finalization of the health management system. At present, there is neither a mature evaluation index system nor a perfect comprehensive evaluation method for equipment health management systems. Summary of the invention

[0004] In view of this, the present invention provides an equipment health management system function evaluation method based on variable weight fuzzy comprehensive evaluation, which can reasonably and effectively evaluate the functionality of the equipment health management system; at the same time, the method constructs two sets of equipment health management system function evaluation index systems; at the same time, the weight set is obtained by considering the hierarchical analysis method with variable weights, and the evaluation result is calculated by combining the fuzzy comprehensive evaluation method. This method can achieve the maximum integration of all evaluation index elements and ensure that the evaluation results are scientific and reasonable, thereby realizing high-precision comprehensive evaluation of the equipment health management system function.

[0005] In order to achieve the above object, the present invention adopts the following technical solution:

[0006] A method for evaluating the function of an equipment health management system based on variable weight fuzzy comprehensive evaluation comprises the following steps;

[0007] S1. By analyzing the functional requirements of the equipment health management system, qualitative indicators and quantitative indicators are selected respectively, and a functional evaluation indicator system of the equipment health management system considering qualitative indicators and a functional evaluation indicator system of the equipment health management system considering quantitative indicators are constructed;

[0008] S2. Based on the hierarchical analysis method, a hierarchical model for the functional evaluation of the equipment health management system is constructed and the initial weights of the indicators in the indicator layer and the factor layer are obtained. The establishment of the indicator layer selects the corresponding functional evaluation indicator system of the equipment health management system considering qualitative indicators or the functional evaluation indicator system of the equipment health management system considering quantitative indicators by considering the difficulty of obtaining the indicators of the evaluation object and the required accuracy of the evaluation results;

[0009] S3. Based on the variable weight principle, the initial weights of various indicators are modified by setting the state variable weight function;

[0010] S4. Based on the fuzzy comprehensive evaluation method, combined with the revised indicator weight set, the equipment health management system is evaluated for its function and the corresponding evaluation level is obtained.

[0011] Furthermore, the S1 specifically includes:

[0012] Analyze the functional requirements of the equipment health management system and break it down into six sub-functions, including: data collection and processing function, health status assessment function, fault diagnosis function, prediction function, decision support function and human-computer interaction function;

[0013] Select qualitative indicators and quantitative indicators respectively to construct a functional evaluation index system for the equipment health management system. The functional evaluation index system for the equipment health management system considering qualitative indicators includes: the integrity, consistency, accuracy and timeliness of data acquisition and processing functions, the rationality, consistency, accuracy and timeliness of health status assessment functions, the perfection, accuracy, timeliness and robustness of fault diagnosis functions, the perfection, accuracy, timeliness and robustness of prediction functions, the comprehensiveness, timeliness and effectiveness of decision-making support functions, and the human-computer friendliness and effectiveness of human-computer interaction functions.

[0014] The functional evaluation index system of the equipment health management system that considers quantitative indicators includes: the coverage rate of monitoring parameters of data acquisition and processing functions, the compliance rate of data encoding format, data sampling accuracy and data sampling frequency; the health status level, status assessment time and confidence of the health status assessment function; the coverage of the fault mode standard library, fault detection time, fault detection rate, missed alarm rate, false alarm rate, fault isolation rate and fault isolation time of the fault diagnosis function; the prediction range, relative accuracy, prediction confidence, prediction stability, prediction false alarm rate, prediction sensitivity and effective prediction endpoint of the prediction function; the task completion rate, decision response time and solution quality of the decision-making support function; and the user learning curve and task completion time of the human-computer interaction function.

[0015] Further, the S2 includes:

[0016] Construct a hierarchical model for the functional evaluation of the equipment health management system, including the target layer, factor layer, and indicator layer. The target layer is set as the functionality of the equipment health management system. The factor layer is set as six factors: data collection and processing function, health status assessment function, fault diagnosis function, prediction function, decision support function, and human-computer interaction function. The indicator layer selects the corresponding functional evaluation indicator system of the equipment health management system considering qualitative indicators or the functional evaluation indicator system of the equipment health management system considering quantitative indicators as the indicator layer indicators by considering the difficulty of obtaining indicators and the required accuracy of the evaluation results;

[0017] The analytic hierarchy process is used to determine the initial weights of each single evaluation indicator in the indicator layer and factor layer. Specifically, the expert scoring method is used and the 0.1-0.9 scaling method is selected to determine the element priority relationship and obtain the indicator priority relationship matrix P;

[0018] The indicator priority relationship matrix P is transformed into a consistent priority relationship matrix P′ that meets the unit, complementarity and consistency. The transformation formula is calculated as follows:

[0019]

[0020] P′ uo =(P u -P o ) / 2(n-1)+0.5

[0021] Where P u and P o are the priority scores of indicator u and indicator o in the indicator priority relationship matrix P, respectively. uk is the priority relationship value between index u and index k in the index priority relationship matrix P, p ok is the priority relationship value between index o and index k in the index priority relationship matrix P, P′ uois the matrix element in the consistent priority relationship matrix P′, indicating the consistent priority relationship value between index u and index o, and n is the number of indexes in the index priority relationship matrix;

[0022] Calculate the initial weight of the indicator under the constant weight state, take the weight resolution parameter β = (n-1) / 2, and the weight of each indicator is calculated as follows:

[0023]

[0024] In the formula, ω r is the initial weight of index r under constant weight state, P′ rk It is a matrix element in the consistent priority relationship matrix P′, representing the consistent priority relationship value between index r and index k.

[0025] Furthermore, in S3, based on the variable weight principle, the initial weights of various indicators are corrected by setting a state variable weight function, including:

[0026] The variable weight theory is introduced to correct the constant weight ω obtained by applying the hierarchical analysis method, and the corrected variable weight ω* is obtained.

[0027] Among them, the modified indicator weight set is ω*(X)=(ω 1 *(X),ω 2 *(X),…ω n *(X)), which is represented by:

[0028]

[0029] Where X = (x 1 , x 2 , …x n ) represents different indicator data, ω=(ω 1 ,ω 2 ,…ω n ) is the initial weight under the constant weight state, S(X) is the state variable weight function, S h (X) and ω h are the state variable weight function value and weight value of indicator h respectively, and n is the number of indicators in the indicator priority relationship matrix;

[0030] The state variable weight function S(X) is expressed as:

[0031]

[0032] Among them, a, b, c, d, e are parameters in [0, 1], a is called the negative level, b is the passing level, c is the incentive level, d is the adjustment level, and e is the ratio of the incentive to the penalty when ω = 1 / n.

[0033] Furthermore, in S4, based on the fuzzy comprehensive evaluation method, combined with the modified indicator weight set ω*, the function evaluation of the equipment health management system is performed, including:

[0034] Establish an evaluation comment set: used to classify the functionality of the equipment health management system and specify the evaluation accuracy;

[0035] Normalize each indicator: used to map indicators with different dimensions and thresholds to the same [0,1] interval;

[0036] Determine the membership function: used to determine the membership of different indicators to each subset in the evaluation review set;

[0037] Calculate the fuzzy evaluation matrix: used to represent the membership results of each indicator in each factor layer;

[0038] Combined weight set: used to distinguish the importance of the indicator's impact on the evaluation results;

[0039] Calculate the comprehensive fuzzy evaluation matrix: combine the weight set and the fuzzy evaluation matrix and select the appropriate fuzzy synthesis operator to perform fuzzy operations to obtain the comprehensive fuzzy evaluation matrix for outputting the evaluation results.

[0040] Furthermore, the evaluation comment set V is set to: V = {excellent, good, average, poor}.

[0041] Furthermore, the normalization process adopts the following processing methods: For the normalization process of qualitative indicators, a scoring table is set up to standardize the determination of scores for different indicators. For the normalization process of quantitative indicators, according to the meaning of quantitative indicators and their impact on the functionality of the health management system, they are divided into positive indicators and negative indicators, and the following formulas are used for normalization calculation respectively;

[0042] For positive indicators, the larger the value, the better the functionality of the equipment health management system. The calculation method is as follows:

[0043]

[0044] In the formula, x is the index value after normalization, x i is the obtained indicator measurement value, x o is the optimal value of the indicator (generally given based on expert scoring), x alarm It is the compliance threshold (generally the worst value required by the contract).

[0045] For negative indicators, the smaller the value, the better the functionality of the equipment health management system. The calculation method is as follows:

[0046]

[0047] In the formula, x is the index value after normalization, x i is the obtained indicator measurement value, x o is the optimal value of the indicator (generally given based on expert scoring), x alarm It is the compliance threshold (generally the worst value required by the contract).

[0048] Furthermore, the membership function is expressed as A = {A V1 (x),A V2 (x),A V3 (x),A V4 (x)}, which is used to determine the degree to which the value of each indicator in the indicator layer belongs to the four levels in the evaluation set V = {excellent, good, average, poor}. Further, the specific expression of the membership function is as follows:

[0049] When considering qualitative indicators as the indicator set, taking into account the subjectivity and diversity of the indicators, the semi-trapezoidal and semi-triangular membership function is used for representation, and the function expression is:

[0050]

[0051] When considering quantitative indicators as the indicator set, taking into account the objectivity of the indicators, a semi-trapezoidal and semi-ridge membership function is used for characterization. The function expression is:

[0052]

[0053]

[0054] in, and There are four piecewise functions, which respectively represent the degree to which the indicators belong to excellent, good, average, and poor.

[0055] Furthermore, the membership of each indicator in the evaluation set V is obtained to belong to four levels respectively. The four membership results of each indicator constitute the fuzzy evaluation matrix R of the indicator. i = {r 1 ,r 2 ,r 3 ,r 4}={A V1 (x),A V2 (x),A V3 (x),A V4 (x)}, the fuzzy evaluation matrix R of the factor layer is obtained by combining the index layer membership matrix of the same type of factors. The factor layer contains "data collection and processing type X 1 ","Health status assessment X 2 ","Fault diagnosis category X 3 ","Prediction class X4 ","Decision-making assistance X 5 " and "Human-Computer Interaction X 6 ", and then we can get the six-factor fuzzy evaluation matrix R Xi :

[0056]

[0057] In the formula, i is the factor number, which takes values ​​as [1, 2, 3, 4, 5, 6], and j is the number of indicators in each factor layer.

[0058] Furthermore, the six-factor fuzzy evaluation matrix Rx 1 , Rx 2 , Rx 3 , Rx 4 , Rx 5 , Rx 6 By performing fuzzy operation with the modified variable weight ω*(X), the comprehensive evaluation matrix Bxi of the indicator layer can be obtained, which contains the elements b ij The calculation is as follows:

[0059]

[0060] In the formula, ω ij and r ij They represent the corrected weight value and membership result of indicator j in factor i, respectively. i is the factor number, and its value is [1, 2, 3, 4, 5, 6]. j is the number of indicators in each factor layer.

[0061] Get the comprehensive evaluation matrix Rx of the six indicator layers 1 , Rx 2 , Rx 3 , Rx 4 , Rx 5 , Rx 6 After that, the comprehensive fuzzy evaluation matrix B of the equipment health management system function can be calculated by combining the factor layer weights:

[0062]

[0063] In the formula, {ω X1 ,ω X2 ,ω X3 ,ω X4 ,ω X5 ,ω X6} is the factor layer weight set, It is the factor layer fuzzy evaluation matrix after combining the six index layer comprehensive evaluation matrices. 1 、V 2 、V 3 、V 4} are the degrees to which the function of a certain equipment health management system of the evaluation object belongs to the four levels of V = {excellent, good, general, poor}. According to the principle of maximum degree of subordination, the level with the highest proportion is the current functional evaluation of the equipment health management system.

[0064] Furthermore, in said S4, by considering the actual development of the equipment health management system, the functional level of the evaluation object can be determined first, and then a comprehensive functional evaluation can be performed on it to reflect the actual functional status of the evaluation object. The evaluation results can be expressed as: Level III - excellent;

[0065] Among them, the functional level of the equipment health management system is set to four levels. Level I corresponds to the equipment health management system which only has the fault diagnosis function; Level II corresponds to the equipment health management system which has three functions including fault diagnosis, data collection and processing, and human-computer interaction; Level III corresponds to the equipment health management system which has four functions including fault diagnosis, data collection and processing, human-computer interaction, and prediction; Level IV corresponds to the equipment health management system which has all six functions including fault diagnosis, data collection and processing, human-computer interaction, prediction, health status assessment, and decision support.

[0066] It can be seen from the above technical solutions that, compared with the prior art, the technology of the present invention can realize effective and reasonable comprehensive function evaluation of the health management system while improving the evaluation index system of the health management system. It has the following beneficial effects:

[0067] 1. By analyzing typical equipment health management systems and health management related technical standards at home and abroad, the functions of the health management system are broken down into six sub-functions, including (1) data collection and processing function (2) health status assessment function (3) fault diagnosis function (4) prediction function (5) decision support function (6) human-computer interaction function, which ensures the accuracy and rationality of the functional division and helps to ensure the accuracy of the evaluation results.

[0068] 2. By referring to the relevant evaluation indicators of health management technology at home and abroad and analyzing the functional requirements of the equipment health management system, the functional evaluation indicator system of the equipment health management system has been improved to the maximum extent. At the same time, considering the difficulty of obtaining indicators, the evaluation indicator system considering qualitative indicators and the evaluation indicator system considering quantitative indicators were constructed respectively to ensure the practicality of the indicator system.

[0069] 3. Based on the characteristics of the fuzzy hierarchical analysis method, it is very suitable for accurate analysis and evaluation of targets with hierarchical structures, different evaluation index priorities, and indicators with a certain degree of fuzziness. Therefore, this model is very suitable for the comprehensive evaluation of equipment health management systems.

[0070] 4. Since the evaluation results may not truly reflect the functionality of the equipment health management system due to the small weight of a certain evaluation indicator in the constant weight evaluation, this method further corrects the reliability of the evaluation results by combining the variable weight principle, especially the reasonable setting of the state variable weight function. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0072] Figure 1 A comprehensive evaluation flow chart of an equipment health management system based on a variable weight fuzzy comprehensive evaluation method provided by the present invention;

[0073] Figure 2 A diagram of the function evaluation index system of the equipment health management system provided by the present invention;

[0074] Figure 3 The comprehensive fuzzy evaluation model of the equipment health management system provided by the present invention;

[0075] Figure 4 This is a functional level definition diagram of the health management system provided by the present invention.

[0076] Figure 5 This is a functional hierarchical model of the equipment health management system provided by the present invention.

[0077] Figure 6 The semi-trapezoidal and semi-triangular membership function images provided by the present invention.

[0078] Figure 7 The semi-trapezoidal and semi-ridge type membership function image provided by the present invention. DETAILED DESCRIPTION

[0079] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0080] Embodiment 1;

[0081] like Figure 1 and Figure 2As shown, this embodiment provides a method for evaluating the function of an equipment health management system based on variable weight fuzzy comprehensive evaluation, comprising the following steps:

[0082] S1. By analyzing the functional requirements of the equipment health management system, qualitative indicators and quantitative indicators are selected respectively, and a functional evaluation indicator system of the equipment health management system considering qualitative indicators and a functional evaluation indicator system of the equipment health management system considering quantitative indicators are constructed;

[0083] S2. Based on the hierarchical analysis method, a hierarchical model for the functional evaluation of the equipment health management system is constructed and the initial weights of the indicators in the indicator layer and the factor layer are obtained. The establishment of the indicator layer selects the corresponding functional evaluation indicator system of the equipment health management system considering qualitative indicators or the functional evaluation indicator system of the equipment health management system considering quantitative indicators by considering the difficulty of obtaining the indicators of the evaluation object and the required accuracy of the evaluation results;

[0084] S3. Based on the variable weight principle, the initial weights of various indicators are modified by setting the state variable weight function;

[0085] S4. Based on the fuzzy comprehensive evaluation method, combined with the revised indicator weight set, the equipment health management system is evaluated for its function and the corresponding evaluation level is obtained.

[0086] This method can effectively evaluate the functionality of the equipment health management system through accurate indicator selection, application of scientific comprehensive evaluation methods, and reasonable membership function and state variable weight function settings, thereby providing valuable reference for the development and verification of the health management system.

[0087] The above steps and related technical features are further described in detail below:

[0088] In this embodiment S1, by deeply analyzing the functional requirements of typical equipment health management systems at home and abroad, and referring to domestic and foreign standards related to health management technology, the health management system is disassembled into six sub-functions, including: (1) data collection and processing function (2) health status assessment function (3) fault diagnosis function (4) prediction function (5) decision support function (6) human-computer interaction function;

[0089] In this step, qualitative and quantitative indicators were selected to construct two sets of functional evaluation indicator systems for equipment health management systems.

[0090] Among them, the functional evaluation index system of the equipment health management system that considers qualitative indicators includes: the integrity, consistency, accuracy, and timeliness of data acquisition and processing functions, the rationality, consistency, accuracy, and timeliness of health status assessment functions, the perfection, accuracy, timeliness, and robustness of fault diagnosis functions, the perfection, accuracy, timeliness, and robustness of prediction functions, the comprehensiveness, timeliness, and effectiveness of decision-making support functions, and the human-computer friendliness and effectiveness of human-computer interaction functions.

[0091] The functional evaluation index system of the equipment health management system that considers quantitative indicators includes the coverage rate of monitoring parameters of data acquisition and processing functions, the compliance rate of data encoding format, data sampling accuracy, and data sampling frequency; the health status level, status assessment time, and confidence level of the health status assessment function; the coverage of the fault mode standard library, fault detection time, fault detection rate, underreporting rate, false alarm rate, fault isolation rate, and fault isolation time of the fault diagnosis function; the prediction range, relative accuracy, prediction confidence, prediction stability, prediction false alarm rate, prediction sensitivity, and effective prediction endpoint of the prediction function; the task completion rate, decision response time, and solution quality of the decision-making support function; and the user learning curve and task completion time of the human-computer interaction function.

[0092] In this embodiment S2, based on the hierarchical analysis method, a hierarchical evaluation model is constructed and the initial weights of each indicator in the indicator layer and the factor layer are obtained. Among them, the hierarchical evaluation model of the equipment health management system function includes a target layer, a factor layer, and an indicator layer. The target layer is set to the functionality of the equipment health management system. The factor layer is set to six factors, namely, data acquisition and processing function, health status assessment function, fault diagnosis function, prediction function, decision support function, and human-computer interaction function. The indicator layer selects the corresponding qualitative / quantitative evaluation indicator system as the indicator layer indicator by considering the difficulty of indicator acquisition and the required accuracy of the evaluation results;

[0093] The analytic hierarchy process is used to determine the initial weights of each single evaluation indicator in the indicator layer and factor layer. Specifically, the expert scoring method is used and the 0.1-0.9 scaling method is selected to determine the element priority relationship and obtain the indicator priority relationship matrix P.

[0094] The indicator priority relationship matrix P is transformed into a consistent priority relationship matrix P′ that meets the unit, complementarity and consistency. The transformation formula is calculated as follows:

[0095]

[0096] P′ uo =(P u -P o ) / 2(n-1)+0.5

[0097] Where Pu and P o are the priority scores of indicator u and indicator o in the indicator priority relationship matrix P, respectively. uk is the priority relationship value between index u and index k in the index priority relationship matrix P, p ok is the priority relationship value between index o and index k in the index priority relationship matrix P, P′ uo is the matrix element in the consistent priority relationship matrix P′, indicating the consistent priority relationship value between index u and index o, and n is the number of indexes in the index priority relationship matrix;

[0098] Calculate the initial weight of the indicator under the constant weight state, take the weight resolution parameter β = (n-1) / 2, and the weight of each indicator is calculated as follows:

[0099]

[0100] In the formula, ω r is the initial weight of index r under constant weight state, P′ rk is a matrix element in the consistent priority relationship matrix P′, representing the consistent priority relationship value between index r and index k, and n is the number of indexes in the index priority relationship matrix.

[0101] In this embodiment S3, the variable weight theory is introduced to correct the constant weight ω obtained by applying the hierarchical analysis method to obtain the corrected variable weight ω*.

[0102] Among them, the modified indicator weight set is ω*(X)=(ω 1 *(X),ω 2 *(,…ω n *(X)), which is represented by:

[0103]

[0104] Where X = (x 1 , x 2 , …x n ) represents different indicator data, ω=(ω 1 ,ω 2 ,…ω n ) is the initial weight under the constant weight state, S(X) is the state variable weight function, S h (X) and ω h are the state variable weight function value and weight value of indicator h respectively, and n is the number of indicators in the indicator priority relationship matrix;

[0105] Among them, the state variable weight function S(X) is expressed as:

[0106]

[0107] Where a, b, c, d, and e are parameters in [0, 1]. a is called the negative level, b is the passing level, c is the incentive level, d is the adjustment level, and e is the ratio of the incentive to the penalty when ω = 1 / n.

[0108] In this embodiment S4, based on the fuzzy comprehensive evaluation method, combined with the modified indicator weight set, the equipment health management system is evaluated by setting the comment set, normalizing the indicators, constructing the membership function, selecting the fuzzy synthesis operator, calculating the fuzzy comprehensive evaluation matrix, etc., to obtain the corresponding evaluation level. Figure 3 As shown, the specific process of the method includes:

[0109] (1) Determine the functional level of the evaluation object:

[0110] By considering the actual development of the equipment health management system, the functional level of the evaluation object can be determined first, and then a comprehensive functional evaluation can be conducted to reflect the actual functional status of the evaluation object. For example, the evaluation results can be expressed as: Level III - Excellent;

[0111] Among them, the functional level of the equipment health management system is set to four levels. Level I corresponds to the equipment health management system which only has the fault diagnosis function; Level II corresponds to the equipment health management system which has three functions including fault diagnosis, data collection and processing, and human-computer interaction; Level III corresponds to the equipment health management system which has four functions including fault diagnosis, data collection and processing, human-computer interaction, and prediction; Level IV corresponds to the equipment health management system which has all six functions including fault diagnosis, data collection and processing, human-computer interaction, prediction, health status assessment, and decision support.

[0112] By analyzing the functional attributes of the evaluation object, its functional level is determined, and the corresponding functional evaluation indicators are selected within the corresponding level for subsequent evaluation calculations.

[0113] (2) Establish a functional hierarchical model of the equipment health management system, such as Figure 5 As shown:

[0114] The system is divided into three layers of progressive models, namely, the target layer, the factor layer, and the indicator layer. Among them, the target layer is set as the functionality of the equipment health management system, the factor layer is set as six items, namely, data collection and processing function, health status assessment function, fault diagnosis function, prediction function, decision support function, and human-computer interaction function. The indicator layer selects all indicators in the comprehensive function evaluation indicator system of the equipment health management system that considers qualitative / quantitative indicators as the indicator layer indicators based on the difficulty of obtaining the indicators of the evaluation object.

[0115] (3) Determine the evaluation criteria for the equipment health management system function:

[0116] It is used to divide the evaluation level of the evaluation system and stipulate the evaluation accuracy. Specifically, the evaluation comment set is set to V = {excellent, good, average, poor} to clearly divide the functionality of the equipment health management system.

[0117] (4) Index normalization:

[0118] It is used to map indicators of different dimensions and thresholds to the same [0,1] interval. Among them, a scoring table is set for the normalization of qualitative indicators, and the properties corresponding to different indicator scores are standardized. For the normalization of quantitative indicators, according to their meaning and impact on the functionality of the health management system, they are divided into positive indicators and negative indicators, and the following formulas are used for normalization calculation respectively;

[0119] For positive indicators, the larger the value, the better the functionality of the equipment health management system. The calculation method is as follows:

[0120]

[0121] In the formula, x is the index value after normalization, x i is the obtained indicator measurement value, x o is the optimal value of the indicator (generally given based on expert scoring), x alarm It is the compliance threshold (generally the worst value required by the contract).

[0122] For negative indicators, the smaller the value, the better the functionality of the equipment health management system. The calculation method is as follows:

[0123]

[0124] In the formula, x is the index value after normalization, x i is the obtained indicator measurement value, x o is the optimal value of the indicator (generally given based on expert scoring), x alarm It is the compliance threshold (generally the worst value required by the contract).

[0125] (5) Determine the index membership function:

[0126] The evaluation of the function / performance of the equipment health management system requires consideration of multiple indicators, which leads to a certain ambiguity in the evaluation results. Therefore, it is necessary to apply a membership function to determine the degree of membership of different indicators to each subset of the review set. Specifically, the membership function is expressed as A = {A V1 (x),A V2 (x),A V3 (x),A V4(x)}, which is used to determine the degree to which the value of each indicator in the indicator layer belongs to the four levels in the evaluation set V = {excellent, good, average, poor}. The specific expression of the membership function is as follows:

[0127] like Figure 6 As shown in the figure, when considering qualitative indicators as the indicator set, taking into account the subjectivity and diversity of the indicators, the semi-trapezoidal and semi-triangular membership function is used for characterization, and the function expression is:

[0128]

[0129] like Figure 7 As shown in the figure, when considering quantitative indicators as the indicator set, taking into account the objectivity of the indicators, a semi-trapezoidal and semi-ridge membership function is used for characterization, and the function expression is:

[0130]

[0131]

[0132] in, and There are four piecewise functions, which respectively represent the degree to which the indicators belong to excellent, good, average, and poor.

[0133] (6) Establish the fuzzy evaluation matrix at the indicator level:

[0134] It is used to represent the membership results of each indicator in each factor layer. By calculating the score value of the indicator, the membership of each indicator in the evaluation set V is obtained, and then the fuzzy evaluation matrix can be constructed. The four membership results of each indicator constitute the fuzzy evaluation matrix of the indicator.

[0135] R i = {r 1 ,r 2 ,r 3 ,r 4}={A V1 (x),A V2 (x),A V3 (x),A V4 (x)}, the indicator layer membership matrix of the same type of factors is combined to obtain the fuzzy evaluation matrix R of the factor layer.

[0136] is the characterization factor u i For the subset of comments v j The degree of membership r ij , construct the fuzzy relationship matrix:

[0137]

[0138] In the formula, i is the factor number, which takes values ​​as [1, 2, 3, 4, 5, 6], and j is the number of indicators in each factor layer.

[0139] (7) Calculate the comprehensive fuzzy evaluation matrix of the index layer:

[0140] Combining the variable weight set and the fuzzy evaluation matrix of the index layer, fuzzy operation is performed to obtain the comprehensive fuzzy evaluation matrix of the index layer. Specifically, the six factor layer fuzzy evaluation matrices Rx 1 , Rx 2 , Rx 3 , Rx 4 , Rx 5 , Rx 6 By performing fuzzy operation with the modified variable weight ω*(X), the comprehensive evaluation matrix Bxi of the indicator layer can be obtained, which contains the elements b ij The calculation is as follows:

[0141]

[0142] In the formula, ω ij and r ij They respectively represent the modified weight value and membership result of indicator j in factor i, i is the factor number, and its value is [1, 2, 3, 4, 4, 5, 6], and j is the number of indicators in each factor layer.

[0143] (8) Calculate the comprehensive fuzzy evaluation matrix of the equipment health management system function:

[0144] Get the comprehensive evaluation matrix Rx of the six indicator layers 1 , Rx 2 , Rx 3 , Rx 4 , Rx 5 , Rx 6 After that, the comprehensive fuzzy evaluation matrix B of the equipment health management system function can be calculated by combining the factor layer weights:

[0145]

[0146] In the formula, {ω X1 ,ω X2 ,ω X3 ,ω X4 ,ω X5 ,ω X6} is the factor layer weight set, It is the factor layer fuzzy evaluation matrix after combining the six index layer comprehensive evaluation matrices. 1 、V 2 、V 3 、V 4} respectively represent the degree to which the function of the health management system of a certain equipment under evaluation belongs to the four levels of V = {excellent, good, average, poor}.

[0147] (9) Output evaluation results

[0148] According to the maximum membership principle, in {V 1 、V 2 、V 3 、V 4 The highest level in the table is the current functional evaluation of the equipment health management system within its functional level.

[0149] Embodiment 2;

[0150] In this embodiment, a proposed equipment health management system function evaluation method based on variable weight fuzzy comprehensive evaluation is applied to comprehensively evaluate the function of a special vehicle onboard health management system, which is further described in detail below;

[0151] S4.1. Determine the functional level of the evaluation object

[0152] Analyze the function settings of the on-board health management system for special vehicles, which has the following functions:

[0153] (1) Data collection and processing functions:

[0154] ① On-board data storage function. Stores the bus information of the entire vehicle and the analog signal acquisition information. According to system requirements, it can record 1 week of working data and 100,000 fault code information (including additional maintenance data).

[0155] ②Data download function: The bus data files and fault code data files stored internally can be quickly exported in sections through the CAN or USB port, and the acquired data information can be used for vehicle performance analysis at the same time.

[0156] (2) Fault diagnosis function:

[0157] According to the planning requirements of the information system fault diagnosis system, each subsystem needs to upload fault or health codes through the bus when an abnormal situation occurs. The handheld terminal in the health monitoring system will collect special vehicle data and fault codes, and can draw working curves according to the requirements of the vehicle subsystem or special test, and record and query problems for analysis and processing by passengers or maintenance personnel. During the test process, the fault code data can be continuously improved according to the problems exposed, which improves the user's ability to handle vehicle problems.

[0158] (3) Human-computer interaction function:

[0159] Push fault diagnosis and alarm information in the form of fault codes on the display and control terminal.

[0160] By analyzing the functional attributes of the evaluation object, refer to Figure 4 According to the functional level classification of equipment health management system, the functionality of the evaluation object is at level II, so further comprehensive functional evaluation of the evaluation object is carried out within the scope of level II.

[0161] S4.2. Establish a hierarchical model for functional evaluation of evaluation objects

[0162] The evaluation object is divided into a three-layer progressive model of target layer, factor layer and indicator layer. Among them, the target layer is set as the functionality of the equipment health management system, the factor layer is set as data acquisition and processing function, fault diagnosis function and human-computer interaction function, and the indicator layer is based on the difficulty of obtaining the indicators of the evaluation object. The three qualitative indicators of data acquisition and processing function, fault diagnosis function and human-computer interaction function in the equipment health management system function evaluation index system considering qualitative indicators are selected as the indicator layer indicators.

[0163] S4.3. Determine the evaluation criteria for the evaluation object:

[0164] The evaluation comment set of the evaluation object is set to V = {excellent, good, average, poor} to clearly distinguish the functionality of the on-board health management system of the special vehicle.

[0165] S4.4, Index Normalization:

[0166] Combining the monitoring data and test data of the on-board health management system of special vehicles, the qualitative indicators can be scored and evaluated according to the qualitative indicator scoring criteria, and the normalized values ​​of the qualitative indicators of the three factor layers, including data acquisition and processing function, fault diagnosis function, and human-computer interaction function, are obtained as follows:

[0167] Table 1 Normalized values ​​of data acquisition and processing function indicators

[0168]

[0169] Table 2 Normalized values ​​of fault diagnosis function indicators

[0170]

[0171] Table 3 Normalized values ​​of human-computer interaction function indicators

[0172]

[0173] S4.5. Determine the index membership function:

[0174] Choose Figure 6 The semi-trapezoidal and semi-triangular membership functions shown are used to characterize the membership of the indicators.

[0175] S4.6, Calculation of fuzzy evaluation matrix at indicator layer:

[0176] Substitute the data acquisition and processing function index value, fault diagnosis function index value, and human-computer interaction function index value into the above membership function in turn for calculation, and obtain the fuzzy evaluation matrix Rx of the index layer. 1 , Rx 2 , Rx 3 .

[0177]

[0178] S4.7. Consistent priority relationship matrix and constant weight calculation:

[0179] Select the 0.1-0.9 scaling method, and have professional experts compare the importance of factors at the same level to obtain the priority relationship matrix P of the target layer. x and the factor layer priority relationship matrix P x1 , P x2 , P x3 .

[0180]

[0181] Convert the indicator priority relationship matrix P into a consistent priority relationship matrix

[0182]

[0183] Calculate the weights of each layer:

[0184] (1) Weight of target layer X ωx = [0.30, 0.52, 0.18]

[0185] (2) Data collection and processing function X 1 The weight ω X1 =[0.29,0.19,0.32,0.20]

[0186] (3) Fault diagnosis function X 2 The weight ω X2 =[0.24,0.20,0.28,0.28]

[0187] (4) Human-computer interaction function X 3 The weight ω X3 =[0.60,0.40]

[0188] S4.8. Calculation of weight set correction considering variable weights:

[0189] The variable weight theory is introduced to modify the constant weight ω obtained by applying the hierarchical analysis method to obtain the variable weight ω*, so as to make the evaluation result flexible and reasonable.

[0190] The normalized weight matrix considering variable weights is recorded as ω*(X)=(ω 1 *(X),ω 2 *(X),…ω n *(X)), which is represented by:

[0191]

[0192] Where X = (x 1 , x 2 , …x n ) represents different indicator data, ω=(ω 1 ,ω 2 ,…ω n ) is the initial weight under the constant weight state, S(X) is the state variable weight function, S h (X) and ω h are the state variable weight function value and weight value of indicator h respectively, and n is the number of indicators in the indicator priority relationship matrix;

[0193] Set the state change function S(X) to be:

[0194]

[0195] Among them, a, b, c, d, e are parameters within [0, 1], a is called the negative level, b is the passing level, c is the incentive level, d is the adjustment level, and e is the ratio of the magnitude of incentive to punishment when ω = 1 / n. According to the functional characteristics of the on-board health management system for special vehicles, a = 0.4, b = 0.7, c = 0.9, d = 0.2, e = 0.8 are taken.

[0196] Through the variable weight operation, the variable weight set is obtained as follows:

[0197] (1) Weight set of target layer X

[0198] (2) Data collection and processing function X 1 The variable weight set

[0199] (3) Fault diagnosis function X 2 The variable weight set

[0200] (4) Human-computer interaction function X 3 The variable weight set

[0201] S4.9, comprehensive fuzzy evaluation matrix calculation:

[0202] Through the above steps, the three-factor layer fuzzy evaluation matrix Rx is obtained 1 , Rx 2 , Rx3 And the weight set ω* of each layer considering variable weights. Further, combining the index membership and the modified variable weights, the factor layer comprehensive fuzzy evaluation matrix Bx can be obtained: 1 , Bx 2 , Bx 3 :

[0203] B X1 =-[0,0,0.035,0.965]

[0204] B X2 =[0,0,0.23,0.77]

[0205] B X3 =[0.47,0.16,0.09,0.28]

[0206] Combining the factor layer comprehensive fuzzy evaluation matrix and the factor layer weight, we can calculate:

[0207] B=[0.08,0.03,0.15,0.74]

[0208] According to the principle of maximum membership, the degree to which the special vehicle onboard health management system belongs to the four states of {poor, general, good, excellent} is judged, and the "excellent" rating with the highest proportion is the current functionality. Combined with the functional level division of the special vehicle onboard health management system, the evaluation result is: Level II-Excellent.

[0209] This evaluation model not only takes into account the various functional indicators of the health management system, but also effectively ensures the accuracy and effectiveness of the evaluation model. In view of the lack of a complete comprehensive evaluation method for health management systems, this method can effectively evaluate the functionality of equipment health management systems, and thus provide a reference for the development and verification of health management systems.

[0210] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0211] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for evaluating the function of an equipment health management system based on variable weight fuzzy comprehensive evaluation, characterized in that: The steps include: S1. By analyzing the functional requirements of the equipment health management system, qualitative indicators and quantitative indicators are selected respectively, and a functional evaluation indicator system of the equipment health management system considering qualitative indicators and a functional evaluation indicator system of the equipment health management system considering quantitative indicators are constructed; S2. Based on the hierarchical analysis method, a hierarchical model for the functional evaluation of the equipment health management system is constructed and the initial weights of the indicators in the indicator layer and the factor layer are obtained. The establishment of the indicator layer selects the corresponding functional evaluation indicator system of the equipment health management system considering qualitative indicators or the functional evaluation indicator system of the equipment health management system considering quantitative indicators by considering the difficulty of obtaining the indicators of the evaluation object and the required accuracy of the evaluation results; S3. Based on the variable weight principle, the initial weights of various indicators are modified by setting the state variable weight function; S4. Based on the fuzzy comprehensive evaluation method, combined with the revised indicator weight set, the equipment health management system is evaluated for its function and the corresponding evaluation level is obtained.

2. According to claim 1, a method for evaluating the function of an equipment health management system based on variable weight fuzzy comprehensive evaluation is characterized in that: The S1 specifically includes: Analyze the functional requirements of the equipment health management system and break it down into six sub-functions, including: data collection and processing function, health status assessment function, fault diagnosis function, prediction function, decision support function and human-computer interaction function; Select qualitative indicators and quantitative indicators respectively to construct a functional evaluation index system for the equipment health management system. The functional evaluation index system for the equipment health management system considering qualitative indicators includes: the integrity, consistency, accuracy and timeliness of data acquisition and processing functions, the rationality, consistency, accuracy and timeliness of health status assessment functions, the perfection, accuracy, timeliness and robustness of fault diagnosis functions, the perfection, accuracy, timeliness and robustness of prediction functions, the comprehensiveness, timeliness and effectiveness of decision-making support functions, and the human-computer friendliness and effectiveness of human-computer interaction functions. The functional evaluation index system of the equipment health management system that considers quantitative indicators includes: the coverage rate of monitoring parameters of data acquisition and processing functions, the compliance rate of data encoding format, data sampling accuracy and data sampling frequency; the health status level, status assessment time and confidence of the health status assessment function; the coverage of the fault mode standard library, fault detection time, fault detection rate, missed alarm rate, false alarm rate, fault isolation rate and fault isolation time of the fault diagnosis function; the prediction range, relative accuracy, prediction confidence, prediction stability, prediction false alarm rate, prediction sensitivity and effective prediction endpoint of the prediction function; the task completion rate, decision response time and solution quality of the decision-making support function; and the user learning curve and task completion time of the human-computer interaction function.

3. The equipment health management system function evaluation method based on variable weight fuzzy comprehensive evaluation according to claim 2 is characterized in that: The S2 includes: Construct a hierarchical model for the functional evaluation of the equipment health management system, including the target layer, factor layer and indicator layer, where the target layer is set as the functionality of the equipment health management system, the factor layer is set as the data collection and processing function, health status assessment function, fault diagnosis function, prediction function, decision support function and human-computer interaction function, and the indicator layer selects the corresponding functional evaluation indicator system of the equipment health management system considering qualitative indicators or the functional evaluation indicator system of the equipment health management system considering quantitative indicators as the indicator layer indicators by considering the difficulty of indicator acquisition and the required accuracy of the evaluation results; The analytic hierarchy process is used to determine the initial weights of each indicator in the indicator layer and factor layer. Specifically, the expert scoring method and the 0.1-0.9 scaling method are used to determine the element priority relationship and obtain the indicator priority relationship matrix P; The indicator priority relationship matrix P is transformed into a consistent priority relationship matrix P′ that meets the unit, complementarity and consistency. The transformation formula is calculated as follows: P′ uo =(P u -P o ) / 2(n-1)+0.5 Where P u and P o are the priority scores of indicator u and indicator o in the indicator priority relationship matrix P, respectively. uk is the priority relationship value between index u and index k in the index priority relationship matrix P, p ck is the priority relationship value between index o and index k in the index priority relationship matrix P, P′ uo is the matrix element in the consistent priority relationship matrix P′, indicating the consistent priority relationship value between index u and index o, and n is the number of indexes in the index priority relationship matrix; Calculate the initial weight of the indicator under the constant weight state, take the weight resolution parameter β = (n-1) / 2, and the weight of each indicator is calculated as follows: In the formula, ω r is the initial weight of index r in the constant weight state, P rk It is a matrix element in the consistent priority relationship matrix P′, representing the consistent priority relationship value between index r and index k.

4. The equipment health management system function evaluation method based on variable weight fuzzy comprehensive evaluation according to claim 1 is characterized in that: In S3, the modified indicator weight set is ω * (X)=(ω1*(X),ω2*(X),…ω n *(X)), which is represented by: where X = (x1, x2, ... x n ) represents different indicator data, ω=(ω1,ω2,…ω n ) is the initial weight under the constant weight state, S(X) is the state variable weight function, S h (X) and ω h are the state variable weight function value and weight value of indicator h respectively, and n is the number of indicators in the indicator priority relationship matrix; The state variable weight function S(X) is expressed as: Among them, a, b, c, d, e are parameters in [0, 1], a is called the negative level, b is the passing level, c is the incentive level, d is the adjustment level, and e is the ratio of the incentive to the penalty when ω = 1 / n.

5. The equipment health management system function evaluation method based on variable weight fuzzy comprehensive evaluation according to claim 4 is characterized in that: In S4, based on the fuzzy comprehensive evaluation method, combined with the revised indicator weight set, the function evaluation of the equipment health management system is performed, including: Establish an evaluation comment set: used to classify the functionality of the equipment health management system and specify the evaluation accuracy; Normalize each indicator: used to map indicators with different dimensions and thresholds to the same [0, 1] interval; Determine the membership function: used to determine the membership of different indicators to each subset in the evaluation review set; Calculate the fuzzy evaluation matrix: used to represent the membership results of each indicator in each factor layer; Combined weight set: used to distinguish the importance of the indicator's impact on the evaluation results; Calculate the comprehensive fuzzy evaluation matrix: combine the weight set and the fuzzy evaluation matrix and select the appropriate fuzzy synthesis operator to perform fuzzy operations to obtain the comprehensive fuzzy evaluation matrix for outputting the evaluation results.

6. The equipment health management system function evaluation method based on variable weight fuzzy comprehensive evaluation according to claim 5 is characterized in that: The evaluation comment set V is set as: V = {excellent, good, average, poor}.

7. The equipment health management system function evaluation method based on variable weight fuzzy comprehensive evaluation according to claim 5 is characterized in that: The normalization process adopts the following processing method: For the normalization of qualitative indicators, a scoring table is set up to standardize the determination of scores for different indicators. For the normalization of quantitative indicators, according to their meaning and impact on the functionality of the health management system, they are divided into positive indicators and negative indicators, and the following formulas are used for normalization calculation respectively; For the positive indicator, the calculation is as follows: For negative indicators, the calculation is as follows: In the formula, x is the index value after normalization, x i is the obtained indicator measurement value, x o is the optimal value of the indicator, x alarm is the threshold value of reaching the standard.

8. The equipment health management system function evaluation method based on variable weight fuzzy comprehensive evaluation according to claim 5 is characterized in that: The membership function is expressed as A = {A V1 (x), A V2 (x), A v3 (x), A V4 (x)} is used to determine the degree to which the value of each indicator in the indicator layer belongs to the four levels in the evaluation comment set V = {excellent, good, average, poor}. The specific expression of the membership function is as follows: When considering qualitative indicators as the indicator set, taking into account the subjectivity and diversity of the indicators, the semi-trapezoidal and semi-triangular membership function is used for characterization. The function expression is: When considering quantitative indicators as the indicator set, taking into account the objectivity of the indicators, a semi-trapezoidal and semi-ridge membership function is used for characterization. The function expression is: in, and There are four piecewise functions, which respectively represent the degree to which the indicators belong to excellent, good, average, and poor.

9. The equipment health management system function evaluation method based on variable weight fuzzy comprehensive evaluation according to claim 8 is characterized in that: The fuzzy evaluation matrix is ​​expressed as R, and the specific calculation process is: Obtain the membership of each indicator in the evaluation comment set V to four levels. The four membership results of each indicator constitute the fuzzy evaluation matrix R of the indicator. i ={r1, r2, r3, r4} = {A V1 (x), A V2 (x), A V3 (x), A V4 (x)}, the index layer membership matrix of the same type of factors is combined to obtain the fuzzy evaluation matrix R of the factor layer. The factor layer includes data acquisition and processing class X1, health status assessment class X2, fault diagnosis class X3, prediction class X4, auxiliary decision class X5 and human-computer interaction class X5, and then the six factor layer fuzzy evaluation matrices R are obtained. Xi : In the formula, i is the factor number, which takes values ​​as [1, 2, 3, 4, 5, 6], and j is the number of indicators in each factor layer.

10. The equipment health management system function evaluation method based on variable weight fuzzy comprehensive evaluation according to claim 9 is characterized in that: The calculation process of the comprehensive fuzzy evaluation matrix is ​​as follows: The fuzzy evaluation matrices Rx1, Rx2, Rx3, Rx4, Rx5, Rx6 corresponding to the six factor layers of data acquisition and processing X1, health status assessment X2, fault diagnosis X3, prediction X4, auxiliary decision-making X5 and human-computer interaction X5 are respectively x6 Fuzzy operation is performed with the modified variable weight ω*(X) to obtain the index layer comprehensive evaluation matrix Bxi, which contains the element b ij , calculated as follows: In the formula, ω ij and r ij They represent the corrected weight value and membership result of indicator j in factor i respectively; After obtaining the six indicator-level comprehensive evaluation matrices Rx1, Rx2, Rx3, Rx4, Rx5, and Rx6, the comprehensive fuzzy evaluation matrix B of the equipment health management system function is obtained by combining the factor-level weight calculation: Wherein, {V1, V2, V3, V4} respectively represent the degree to which the function of a certain equipment health management system of the evaluation object belongs to the four levels of V = {excellent, good, general, poor}. According to the principle of maximum degree of subordination, the level with the highest proportion is the current functional evaluation of the equipment health management system.

Images