Residual risk analysis method for complex structure of equipment based on quantitative evaluation

Abstract

Provided is a residual risk analysis method for complex equipment structure based on quantitative evaluation, comprising the following steps: constructing a structure failure mode information system; obtaining expert utility score of each risk control measure in reducing risk occurrence probability and severity, and constructing a risk control utility evaluation information system; calculating opinion distance between two experts to obtain an expert evaluation opinion distance matrix; calculating conflict degree delta of each expert with the expert group; calculating expert weight; obtaining total utility of each risk control measure through weighted summation and arithmetic square; calculating residual risk of each failure mode; and calculating total utility and increment ratio of the risk control measure. The present application simultaneously considers the problems of difficulty in quantifying residual risk of equipment structure failure mode in structure residual risk and experience conflict between experts, and determines an expert weight calculation method based on experience conflict and a residual risk evaluation calculation method and process on the premise of design stage FMEA result.

Description

Technical Field

[0001] This invention belongs to the field of residual risk assessment of equipment structures, and involves an expert weight calculation method based on experience conflict and a calculation method and process for residual risk assessment. Specifically, it relates to a method for analyzing residual risk of complex equipment structures based on quantitative assessment. Background Technology

[0002] Residual risk refers to the risk that persists even after risk control measures have been implemented, based on the inherent structural risks identified in the Failure Mode and Effects Analysis (FMEA) during the design phase. Currently, the understanding of residual structural risks during the design phase is inadequate. High-intensity, high-load use of equipment leads to operating environments that are more severe than anticipated during the design phase, making residual structural risks more likely to cause safety issues and representing a significant safety hazard that cannot be ignored during the equipment's service life. Therefore, considering the difficulty in quantitatively assessing residual structural risks, it is necessary to explore quantitative evaluation methods for risk control measures to quantitatively assess residual structural risks based on known inherent risks. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention provides a method for residual risk analysis of complex equipment structures based on quantitative assessment, specifically including the following steps:

[0004] Step 1: Based on the existing equipment structural design FMEA results, construct a structural failure mode information system SFMIS = (FM, A, R, F), where FM is the set of objects composed of all structural failure modes in FMEA, A is the set of structural failure mode attributes, and A = {RL, MeD, MeU}, where RL is the inherent risk level, which is a comprehensive measure of severity S and probability of occurrence P, MeD is the design improvement measure, and MeU is the use compensation measure;

[0005] R is the set of attribute value ranges, that is, the set of all values ​​that all failure modes can take under each attribute;

[0006] F = {f a :FM→R a |a∈A} is a binary relation between failure modes and attributes, where f a R represents a binary relation mapping function. a Let f(x,a) represent the set of values ​​for attribute a, where a is an attribute of the structural failure mode. For any failure mode x∈FM, a∈A, we have f(x,a)∈R. a f(x,a) is a mapping function from failure mode x to attribute a; this binary relation means that a failure mode has a specific value or data corresponding to a single attribute.

[0007] Step 2: Based on all known risk control measures in SFMIS, obtain expert scores for the effectiveness of each risk control measure in reducing the probability and severity of risk occurrence, and construct a risk control effectiveness assessment information system;

[0008] RCUAIS=(Me,A (k) ,R (k) ,F (k) )

[0009] In the formula, Me represents all design improvement measures R in SFMIS. Med and the use of compensation measures R MeU A collection of objects;

[0010] A (k) It is a risk control utility attribute, A (k) ={ΔS (k) ,ΔP (k)}, where ΔS (k) The effectiveness of risk control measures proposed by k experts in the failure mode risk severity level S is ΔP. (k) This refers to the utility of risk control measures proposed by k experts at the probability level P, i.e., the utility of severity control S and the utility generated by the probability level P, which are abbreviated as the severity control utility ΔS. (k) and the probability of occurrence control utility ΔP (k) ;

[0011] R (k) It is the severity control effect ΔS (k) and the probability of occurrence control utility ΔP (k) The range of values;

[0012] F (k) It is a binary relation between the set of control measures Me and the attributes, that is, for any risk control measure y∈Me, the severity control utility given by the l-th expert is: The probability control utility is y represents risk control measures. This represents the range of values ​​for the severity control utility given by the l-th expert. This represents the range of values ​​for the probability control utility given by the Lth expert.

[0013] Step 3: Map the utility ratings given by experts to prospect theory, minimizing the influence of experts' subjective preferences, and calculate the pairwise opinion distance between experts to obtain the expert evaluation opinion distance matrix D = [d q,t (y)] k×k d q,t (y) represents the q-th expert e q And the tth expert e t Distance of the given utility evaluation opinion;

[0014] In the risk control effectiveness assessment information system RCUAIS=(Me,A (k) ,R (k) ,F (k) In the given information, for any risk control measure y∈Me, the distance matrix of the pairwise expert opinions on utility evaluation is D=[d q,t (y)] k×k ;

[0015] The qth expert e q And the tth expert e t The distance d of the given utility evaluation opinion q,t (y) is:

[0016]

[0017] In the formula, The severity control effect f of the risk control measure y given by the qth expert. q (y,ΔS (q) The mapping value in the prospect theory function, the probability control utility is: f represents the severity and control effectiveness of the risk control measure y given by the t-th expert. t (y,ΔS (t) The mapping value in the prospect theory function, the probability control utility is:

[0018] d q,t (y) satisfies:

[0019] d q,t (y)=d t,q (y)

[0020] d q,q (y)=0

[0021] Mapping the utility ratings given by experts to prospect theory, minimizing the influence of experts' subjective preferences, and calculating the pairwise opinion distance between experts, we obtain the expert opinion distance matrix D = [d q,t (y)] k×k ;

[0022] Step 4: Considering the conflict between expert opinions and group consensus, calculate the degree of conflict δ between each expert and the expert group based on the obtained opinion distance matrix;

[0023] In the risk control effectiveness assessment information system RCUAIS=(Me,A (k) ,R (k) ,F (k)For any risk control measure y∈Me, define the degree of conflict δ between the l-th expert and the group opinion. l (y) is:

[0024]

[0025] In the formula, δ l The larger (y) is, the greater the conflict between the expert's opinion and the group's opinion, and thus the greater the impact on reaching a group consensus;

[0026] Step 5: Calculate the expert weights based on the conflict degree of each expert;

[0027] In the risk control effectiveness assessment information system RCUAIS=(Me,A (k) ,R (k) ,F (k) For any risk control measure y∈Me, the weight w of the l-th expert is... l (y) is:

[0028]

[0029] In the formula, δ l (y) represents the degree of conflict between the l-th expert's opinion and the group's opinion; It takes the sequence number from 1 to k {δ l The maximum value of (y)};

[0030] Step 6: Based on the risk control utility score obtained in Step 2 and the expert weight obtained in Step 5, the total utility of each risk control measure is obtained by weighted summation and arithmetic square.

[0031] In the Structural Failure Mode Information System (SFMIS) = (FM, A, R, F), for any failure mode x ∈ FM, there is a design improvement measure f(x, MeD) = {MeD}. i |i=1,2,...,m}、Use compensation measures f(x,MeU={MeU} j |j=1,2,...,n},MeD i For the i-th design improvement measure, MeU j For the j-th use of compensatory measures; m is the number of design improvement measures, n is the number of compensatory measures used; the risk control utility assessment information system for failure mode x is RCUAIS = (Me,A (k) ,R (k) ,F (k) ), where Me = {MeD} i ∪MeU j |i=1,2,...,m;j=1,2,...,n}, the total risk control utility U of all design improvement measures for failure mode x ctr(MeD) is:

[0032]

[0033] Failure mode x Total utility of risk control using compensatory improvement measures U ctr (MeU) is:

[0034]

[0035] In the formula, and These are the weights that experts assign to the severity control utility and the probability control utility when evaluating the severity control utility. and These are the mapping values ​​of the severity control utility and the probability of occurrence control utility given by expert l in the prospect theory function. Based on the total utility of known failure mode risk control measures, the residual risk of structural failure modes is calculated.

[0036]

[0037]

[0038] In the formula, MeD i For the i-th design improvement measure, MeU i For the i-th use of compensation measures, It assesses the degree of conflict between the l-th expert and group opinions when taking the i-th design improvement measure to evaluate the utility of controlling the probability of occurrence. This is the degree of conflict between the l-th expert and group opinions when assessing the effectiveness of severity control. It takes δ from index 1 to k. l The maximum value, It assesses the degree of conflict between the l-th expert and the group opinion when the i-th compensatory measure is taken to control the probability of occurrence. It assesses the degree of conflict between the l-th expert and group opinions when taking the i-th compensatory measure to evaluate the severity control utility. It takes δ from index 1 to k. l The maximum value;

[0039]

[0040] In the formula, It is the probability control utility given by the l-th expert when the i-th design improvement measure is taken. This is the severity control effectiveness given by the l-th expert when the i-th design improvement measure is implemented. It is the probability control utility given by the l-th expert when taking the i-th compensatory measure. It is the severity control utility given by the l-th expert when taking the i-th compensatory measure, and A is the prospect coefficient;

[0041] Step 7: Calculate the residual risk for each failure mode based on the total utility of each risk control measure;

[0042] In the Structural Failure Mode Information System (SFMIS) = (FM, A, R, F), for any failure mode x ∈ FM, there is a design improvement measure f(x, MeD) = {MeD}. i |i=1,2,...,m}、Use compensation measures f(x,MeU={MeU} j Given |j=1,2,...,n}, the two total utility values ​​of risk control obtained after evaluation by k experts are U and U respectively. ctr (MeD) and U ctr (MeU), the residual risk RR of failure mode x is:

[0043] RR x =RL x -U ctr (MeD)-U ctr (MeU)

[0044] In the formula, RL x U represents the inherent risk level of failure mode x. ctr (MeD) represents the total utility of risk control for design improvement measures, U ctr (MeU) represents the total utility of risk control using compensatory measures;

[0045] Step 8: Based on the residual risk obtained in Step 7 for each failure mode, calculate the total utility and incremental ratio of risk control measures; the total utility of failure mode x risk control is:

[0046] U ctrx =U ctrx (MeU)+U ctrx (MeD)

[0047] In the formula: U ctrx (MeD) represents the total utility of structural failure mode x design improvement measures and risk control, U ctrx (MeU) represents the total utility of risk control using compensatory measures for structural failure mode x.

[0048] The risk level increase ratio of structural failure mode x is:

[0049] I x =U ctrx / RL x

[0050] In the formula: U ctrx For failure mode x total utility of risk control, RLx The inherent risk level of structural failure mode x.

[0051] In one specific embodiment of the present invention, the foreground coefficient A is set to 0.89.

[0052] This invention provides a method for quantitatively assessing residual risks in equipment structures. Based on the results of the FMEA (Factors-Driven Engineering Assessment) during the design phase, this invention defines a relevant information system for assessing residual risks in equipment structures, and determines an expert weight calculation method based on experience conflicts, as well as the calculation method and process for residual risk assessment. Compared to previous risk assessment methods, this invention simultaneously considers the difficulty in quantifying residual risks of equipment structural failure modes and the existence of experience conflicts among experts. Based on defining the concept of equipment residual risk and analyzing its characteristics, it proposes a basic approach to residual risk assessment. Using the results of the FMEA during the design phase as a premise, it defines a structural failure mode information system and a risk control effectiveness assessment information system for equipment residual risk assessment, and determines an expert weight calculation method based on experience conflicts, as well as the calculation method and process for residual risk assessment. An example calculation of equipment residual risk assessment is conducted, and the residual risks of equipment structural failure modes are obtained according to the proposed method and process. Attached Figure Description

[0053] Figure 1 A flowchart illustrating the specific implementation of FMEA for the design phase;

[0054] Figure 2 Diagram of equipment residual risk assessment method;

[0055] Figure 3 Flowchart for quantitative assessment of equipment residual risk. Detailed Implementation

[0056] To clarify the residual structural risks remaining after FMEA (Failure Mode and Effects Analysis) in the equipment design phase, it is necessary to quantitatively measure the impact of two types of risk control measures in FMEA on failure mode risk. If we consider this issue from the perspective of the structural failure mode risk generation mechanism, structural failure is influenced by multiple factors, including the structural material properties, the specific form of external loads, and the complex environmental conditions. Identifying these influencing factors is already a complex problem; further studying how each risk control measure acts on these influencing factors exacerbates the complexity, making a complete solution virtually impossible. Quantitatively measuring the effectiveness of risk control measures from the perspective of structural failure mechanism lacks feasibility. The workflow for conducting FMEA in the design phase is as follows: Figure 1 As shown, the expert group provides a quantitative probability level and severity level for each failure mode based on their experience and knowledge.

[0057] like Figure 2As shown, the inventive concept of this invention is to comprehensively evaluate the effectiveness of each risk control measure in controlling failure mode risk based on the experience and knowledge of different types of expert groups, and then calculate the residual risk of the failure mode. This quantitative evaluation of the effectiveness of risk control measures avoids the feasibility problem of measuring the effectiveness of risk control from the perspective of the structural failure mode risk generation mechanism. The steps of this invention are as follows: Figure 3 As shown, based on defining the concept of equipment residual risk and analyzing its characteristics, and taking the FMEA results from the design phase as a premise, this paper defines a structural failure mode information system and a risk control effectiveness evaluation information system for equipment residual risk assessment. It also determines an expert weight calculation method based on experience conflict and a calculation method and process for residual risk assessment. A case study of equipment residual risk assessment was conducted, and the residual risk of equipment structural failure modes was obtained according to the proposed methodology and process.

[0058] According to the above principles and requirements of the present invention, the specific steps of the method of the present invention are as follows:

[0059] Step 1: Based on the existing equipment structural design FMEA results, construct a structural failure mode information system SFMIS = (FM, A, R, F), where FM is the set of objects composed of all structural failure modes in FMEA.

[0060] A is a set of structural failure mode attributes, where A = {RL, MeD, MeU}. RL is the inherent risk level, which is a comprehensive measure of severity S and probability of occurrence P. MeD is the design improvement measure, and MeU is the compensation measure.

[0061] R is the set of attribute value ranges, that is, the set of all values ​​that all failure modes can take under each attribute.

[0062] F = {f a :FM→R a |a∈A} is a binary relation between failure modes and attributes, where f a R represents a binary relation mapping function. a Let f(x,a) represent the set of values ​​for attribute a, where a is an attribute of the structural failure mode. For any failure mode x∈FM, a∈A, we have f(x,a)∈R. a f(x,a) is a mapping function from failure mode x to attribute a. Specifically, this binary relation means that a failure mode has a specific value or data corresponding to a single attribute.

[0063] Step 2: Based on all known risk control measures in SFMIS, obtain expert scores for the effectiveness of each risk control measure in reducing the probability and severity of risk occurrence, and construct a risk control effectiveness assessment information system.

[0064] RCUAIS=(Me,A(k) ,R (k) ,F (k) )

[0065] In the formula, Me represents all design improvement measures R in SFMIS. Med and the use of compensation measures R MeU A collection of objects.

[0066] A (k) It is a risk control utility attribute, A (k) ={ΔS (k) ,ΔP (k)}, where ΔS (k) ΔP (k) These are the effects of the risk control measures proposed by k experts on the severity level S and the probability level P of the failure mode risk, respectively; that is, the effects of severity control utility S and the probability level P, which are referred to as severity control utility ΔS. (k) and the probability of occurrence control utility ΔP (k) .

[0067] R (k) It is the severity control effect ΔS (k) and the probability of occurrence control utility ΔP (k) The range of values.

[0068] F (k) It is a binary relation between the set of control measures Me and the attributes, that is, for any risk control measure y∈Me, the severity control utility given by the l-th expert is: The probability control utility is y represents risk control measures. This represents the range of values ​​for the severity control utility given by the l-th expert. This represents the range of values ​​for the probability control utility given by the Lth expert.

[0069] Step 3: Map the utility ratings given by experts to prospect theory, minimizing the influence of experts' subjective preferences, and calculate the pairwise opinion distance between experts to obtain the expert evaluation opinion distance matrix D = [d q,t (y)] k×k d q,t (y) represents the q-th expert e q And the tth expert e t The distance between the given utility evaluation opinions.

[0070] In the risk control effectiveness assessment information system RCUAIS=(Me,A (k) ,R (k) ,F (k)In the given information, for any risk control measure y∈Me, the distance matrix of the pairwise expert opinions on utility evaluation is D=[d q,t (y)] k×k ;

[0071] The qth expert e q And the tth expert e t The distance d of the given utility evaluation opinion q,t (y) is:

[0072]

[0073] In the formula, The severity control effect f of the risk control measure y given by the qth expert. q (y,ΔS (q) The mapping value in the prospect theory function, and the probability control utility, are similarly expressed as follows: f represents the severity and control effectiveness of the risk control measure y given by the t-th expert. t (y,ΔS (t) The mapping value in the prospect theory function, according to prospect theory, the probability control utility is similarly...

[0074] d q,t (y) satisfies:

[0075] d q,t (y)=d t,q (y)

[0076] d q,q (y)=0

[0077] Mapping the utility ratings given by experts using prospect theory minimizes the influence of expert subjective preferences (mapping using prospect theory aims to minimize subjective preferences). The pairwise opinion distances between experts are then calculated, yielding the expert opinion distance matrix D = [d...]. q,t (y)] k×k ;

[0078] This step maps expert evaluations based on prospect theory, calculates the expert opinion distance, and obtains the expert evaluation opinion distance matrix.

[0079] Step 4: Considering the situation where expert opinions conflict with group consensus, calculate the degree of conflict δ between each expert and the expert group based on the obtained opinion distance matrix.

[0080] In the risk control effectiveness assessment information system RCUAIS=(Me,A (k) ,R (k) ,F (k)For any risk control measure y∈Me, define the degree of conflict δ between the l-th expert and the group opinion. l (y) is:

[0081]

[0082] In the formula, δ l The larger (y) is, the greater the conflict between the expert's opinion and the group's opinion, and thus the greater the impact on achieving group consensus.

[0083] This step considers the situation where there is a conflict between expert opinions and group consensus. Based on the obtained opinion distance matrix, the degree of conflict δ between each expert and the expert group is calculated.

[0084] Step 5: Calculate the expert weights based on the conflict degree of each expert.

[0085] In the risk control effectiveness assessment information system RCUAIS=(Me,A (k) ,R (k) ,F (k) For any risk control measure y∈Me, the weight w of the l-th expert is... l (y) is:

[0086]

[0087] In the formula, δ l (y) represents the degree of conflict between the l-th expert's opinion and the group's opinion. It takes the sequence number from 1 to k {δ l The maximum value of (y)}.

[0088] Step 6: Based on the risk control utility score obtained in Step 2 and the expert weight obtained in Step 5, the total utility of each risk control measure is obtained by weighted summation and arithmetic square.

[0089] In the Structural Failure Mode Information System (SFMIS) = (FM, A, R, F), for any failure mode x ∈ FM, there is a design improvement measure f(x, MeD) = {MeD}. i |i=1,2,...,m}、Use compensation measures f(x,MeU={MeU} j |j=1,2,...,n},MeD i For the i-th design improvement measure, MeU j This is the j-th compensation measure used. m represents the number of design improvement measures, and n represents the number of compensation measures used. The risk control utility assessment information system for failure mode x is RCUAIS = (Me,A (k) ,R (k) ,F (k) ), where Me = {MeD} i ∪MeUj |i=1,2,...,m;j=1,2,...,n}, the total risk control utility U of all design improvement measures for failure mode x ctr (MeD) is:

[0090]

[0091] Similarly, the total utility U of risk control using compensatory improvement measures for failure mode x is... ctr (MeU) is:

[0092]

[0093] In the formula, and These are the weights that experts assign to the severity control utility and the probability control utility when evaluating the severity control utility. and These are the mapping values ​​of severity control utility and probability control utility given by expert l in the prospect theory function. Based on the total utility of known failure mode risk control measures, the residual risk of structural failure modes is calculated.

[0094]

[0095] In the formula, MeD i For the i-th design improvement measure, MeU i For the i-th use of compensation measures, It assesses the degree of conflict between the l-th expert and group opinions when taking the i-th design improvement measure to evaluate the utility of controlling the probability of occurrence. This is the degree of conflict between the l-th expert and group opinions when assessing the effectiveness of severity control. It takes δ from index 1 to k. l The maximum value, It assesses the degree of conflict between the l-th expert and the group opinion when the i-th compensatory measure is taken to control the probability of occurrence. It assesses the degree of conflict between the l-th expert and group opinions when taking the i-th compensatory measure to evaluate the severity control utility. It takes δ from index 1 to k. l The maximum value;

[0096]

[0097] In the formula, It is the probability control utility given by the l-th expert when the i-th design improvement measure is taken. This is the severity control effectiveness given by the l-th expert when the i-th design improvement measure is implemented. It is the probability control utility given by the l-th expert when taking the i-th compensatory measure. It is the severity control utility given by the l-th expert when taking the i-th compensation measure, and A is the prospect coefficient, which is 0.89 in this invention.

[0098] Based on the risk control utility score and expert weights obtained in this step, the total utility of each risk control measure is obtained through weighted summation and arithmetic squaring.

[0099] Step 7: Calculate the residual risk for each failure mode based on the total utility of each risk control measure.

[0100] In the Structural Failure Mode Information System (SFMIS) = (FM, A, R, F), for any failure mode x ∈ FM, there is a design improvement measure f(x, MeD) = {MeD}. i |i=1,2,...,m}、Use compensation measures f(x,MeU={MeU} j Given |j=1,2,...,n}, the two total utility values ​​of risk control obtained after evaluation by k experts are U and U respectively. ctr (MeD) and U ctr (MeU), the residual risk RR of failure mode x is:

[0101] RR x =RL x -U ctr (MeD)-U ctr (MeU)

[0102] In the formula, RL x U represents the inherent risk level of failure mode x. ctr (MeD) represents the total utility of risk control for design improvement measures, U ctr (MeU) represents the total utility of risk control using compensation measures.

[0103] This step calculates the residual risk for each failure mode based on the total utility of each risk control measure.

[0104] Step 8: Based on the residual risk obtained in Step 7 for each failure mode, calculate the total utility and increment of risk control measures. Failure mode x Total utility (increment) of risk control is:

[0105] U ctrx =U ctrx (MeU)+U ctrx (MeD)

[0106] In the formula: U ctrx (MeD) represents the total utility of structural failure mode x design improvement measures and risk control, U ctrx(MeU) represents the total utility of risk control using compensatory measures for structural failure mode x.

[0107] The risk level increase ratio of structural failure mode x is:

[0108] I x =U ctrx / RL x

[0109] In the formula: U ctrx For failure mode x total utility of risk control, RL x The inherent risk level of structural failure mode x.

[0110] In one embodiment of the present invention, a certain type of equipment is selected as the evaluation object. For different failure modes of the same structural component of the equipment, the same control measures are combined to establish an equipment structural failure mode information system (SFMIS = (FM, A, R, F)). For example, the following table shows:

[0111]

[0112] Based on existing FMEA results, five experts with different professional knowledge, experience, and skills were selected from both the equipment design unit and the equipment user unit to form an evaluation team. The effectiveness of risk control measures was evaluated in terms of controlling the probability of failure and controlling the severity of failure. Specifically, the five experts from the design unit evaluated the effectiveness of design assurance measures, and the five experts from the user unit evaluated the effectiveness of compensation measures. This yielded the expert-evaluated effectiveness of all risk control measures for each structural failure mode. The experts evaluated the effectiveness of individual measures in controlling both the probability of failure and the severity of failure modes. The control effectiveness comments are shown in the table below:

[0113]

[0114] Based on the assessment results, establish a risk control effectiveness assessment information system RCUAIS = (Me,A) (k) ,R (k) ,F (k) The number of experts is k = 5. The expert evaluation utility results for structural failure mode A1 are shown in the table below:

[0115]

[0116] This invention uses the calculation of the residual risk of this failure mode as an example to illustrate the assessment process and results of the residual risk of equipment structure.

[0117] After completing the risk control effectiveness assessment, based on the assessment results, and following steps 3 to 5, the expert opinion distance, individual expert conflict degree, and expert weight are calculated sequentially. In the A1 failure mode, the expert weights are shown in the table below:

[0118]

[0119] Based on the expert weights, the total utility of risk control measures and the residual risk for each structural failure mode are calculated sequentially according to steps 6 and 7. The utility of each risk control measure under failure mode A1 is shown in the table below:

[0120]

[0121] Considering that the risk level given in the FMEA results during the design phase is negatively correlated with its numerical value (i.e., a smaller value indicates a higher risk level), the residual risk level should be adjusted to the inherent risk level plus the risk control utility when calculating the residual risk level, i.e., RR. x =RL x +U ctr (MeD)+U ctr (MeU). The residual risk assessment results for all structural failure modes of the equipment are shown in the table below:

[0122]

[0123] Based on the residual risks of each failure mode of the equipment given in the table above, and combined with the inherent risks of the failure modes, the residual risk calculation results are shown in the table below, according to the formula in step 8:

[0124]

Claims

1. A method for residual risk analysis of complex equipment structures based on quantitative assessment, characterized in that, Specifically, the following steps are included: Step 1: Based on the existing equipment structural design FMEA results, construct a structural failure mode information system SFMIS = (FM, A, R, F), where FM is the set of objects composed of all structural failure modes in FMEA, A is the set of structural failure mode attributes, and A = {RL, MeD, MeU}, where RL is the inherent risk level, which is a comprehensive measure of severity S and probability of occurrence P, MeD is the design improvement measure, and MeU is the use compensation measure; R is the set of attribute value ranges, that is, the set of all values ​​that all failure modes can take under each attribute; F = {f a :FM→R a |a∈A} is a binary relation between failure modes and attributes, where f a R represents a binary relation mapping function. a Let f(x,a) represent the set of values ​​for attribute a, where a is an attribute of the structural failure mode. For any failure mode x∈FM, a∈A, we have f(x,a)∈R. a f(x,a) is a mapping function from failure mode x to attribute a; this binary relation means that a failure mode has a specific value or data corresponding to a single attribute. Step 2: Based on all known risk control measures in SFMIS, obtain expert scores for the effectiveness of each risk control measure in reducing the probability and severity of risk occurrence, and construct a risk control effectiveness assessment information system; RCUAIS=(Me,A (k) ,R (k) ,F (k) ) In the formula, Me represents all design improvement measures R in SFMIS. Med and the use of compensation measures R MeU A collection of objects; A (k) It is a risk control utility attribute, A (k) ={ΔS (k) ,ΔP (k) }, where ΔS (k) The effectiveness of risk control measures proposed by k experts in the failure mode risk severity level S is ΔP. (k) This refers to the utility of risk control measures proposed by k experts at the probability level P, i.e., the utility of severity control S and the utility generated by the probability level P, which are abbreviated as the severity control utility ΔS. (k) and the probability of occurrence control utility ΔP (k) ; R (k) It is the severity control effect ΔS (k) and the probability of occurrence control utility ΔP (k) The range of values; F (k) It is a binary relation between the set of control measures Me and the attributes, that is, for any risk control measure y∈Me, the severity control utility given by the l-th expert is: The probability control utility is y represents risk control measures. This represents the range of values ​​for the severity control utility given by the l-th expert. This represents the range of values ​​for the probability control utility given by the Lth expert. Step 3: Map the utility ratings given by experts to prospect theory, minimizing the influence of experts' subjective preferences, and calculate the pairwise opinion distance between experts to obtain the expert evaluation opinion distance matrix D = [d q,t (y)] k×k d q,t (y) represents the q-th expert e q And the tth expert e t Distance of the given utility evaluation opinion; In the risk control effectiveness assessment information system RCUAIS=(Me,A (k) ,R (k) ,F (k) In the given information, for any risk control measure y∈Me, the distance matrix of the pairwise expert opinions on utility evaluation is D=[d q,t (y)] k×k ; The qth expert e q And the tth expert e t The distance d of the given utility evaluation opinion q,t (y) is: In the formula, The severity control effect f of the risk control measure y given by the qth expert. q (y,ΔS (q) The mapping value in the prospect theory function, the probability control utility is: f represents the severity and control effectiveness of the risk control measure y given by the t-th expert. t (y,ΔS (t) The mapping value in the prospect theory function, the probability control utility is: d q,t (y) satisfies: d q,t (y)=d t,q (y) d q,q (y)=0 Mapping the utility ratings given by experts to prospect theory, minimizing the influence of experts' subjective preferences, and calculating the pairwise opinion distance between experts, we obtain the expert opinion distance matrix D = [d q,t (y)] k×k ; Step 4: Considering the conflict between expert opinions and group consensus, calculate the degree of conflict δ between each expert and the expert group based on the obtained opinion distance matrix; In the risk control effectiveness assessment information system RCUAIS=(Me,A (k) ,R (k) ,F (k) For any risk control measure y∈Me, define the degree of conflict δ between the l-th expert and the group opinion. l (y) is: In the formula, δ l The larger (y) is, the greater the conflict between the expert's opinion and the group's opinion, and thus the greater the impact on reaching a group consensus; Step 5: Calculate the expert weights based on the conflict degree of each expert; In the risk control effectiveness assessment information system RCUAIS=(Me,A (k) ,R (k) ,F (k) For any risk control measure y∈Me, the weight w of the l-th expert is... l (y) is: In the formula, δ l (y) represents the degree of conflict between the l-th expert's opinion and the group's opinion; It takes the sequence number from 1 to k {δ l The maximum value of (y)}; Step 6: Based on the risk control utility score obtained in Step 2 and the expert weight obtained in Step 5, the total utility of each risk control measure is obtained by weighted summation and arithmetic square. In the Structural Failure Mode Information System (SFMIS) = (FM, A, R, F), for any failure mode x ∈ FM, there is a design improvement measure f(x, MeD) = {MeD}. i |i=1,2,...,m}、Use compensation measures f(x,MeU={MeU} j |j=1,2,...,n},MeD i For the i-th design improvement measure, MeU j For the j-th use of compensatory measures; m is the number of design improvement measures, n is the number of compensatory measures used; the risk control utility assessment information system for failure mode x is RCUAIS = (Me,A (k) ,R (k) ,F (k) ), where Me = {MeD} i ∪MeU j |i=1,2,...,m;j=1,2,...,n}, the total risk control utility U of all design improvement measures for failure mode x ctr (MeD) is: Failure mode x Total utility of risk control using compensatory improvement measures U ctr (MeU) is: In the formula, and These are the weights that experts assign to the severity control utility and the probability control utility when evaluating the severity control utility. and These are the mapping values ​​of the severity control utility and the probability of occurrence control utility given by expert l in the prospect theory function. Based on the total utility of known failure mode risk control measures, the residual risk of structural failure modes is calculated. In the formula, MeD i For the i-th design improvement measure, MeU i For the i-th use of compensation measures, It assesses the degree of conflict between the l-th expert and group opinions when taking the i-th design improvement measure to evaluate the utility of controlling the probability of occurrence. This is the degree of conflict between the l-th expert and group opinions when assessing the effectiveness of severity control. It takes δ from index 1 to k. l The maximum value, It assesses the degree of conflict between the l-th expert and the group opinion when the i-th compensatory measure is taken to control the probability of occurrence. It assesses the degree of conflict between the l-th expert and group opinions when taking the i-th compensatory measure to evaluate the severity control utility. It takes δ from index 1 to k. l The maximum value; In the formula, It is the probability control utility given by the l-th expert when the i-th design improvement measure is taken. This is the severity control effectiveness given by the l-th expert when the i-th design improvement measure is implemented. It is the probability control utility given by the l-th expert when taking the i-th compensatory measure. It is the severity control utility given by the l-th expert when taking the i-th compensatory measure, and A is the prospect coefficient; Step 7: Calculate the residual risk for each failure mode based on the total utility of each risk control measure; In the Structural Failure Mode Information System (SFMIS) = (FM, A, R, F), for any failure mode x ∈ FM, there is a design improvement measure f(x, MeD) = {MeD}. i |i=1,2,...,m}、Use compensation measures f(x,MeU={MeU} j Given |j=1,2,...,n}, the two total utility values ​​of risk control obtained after evaluation by k experts are U and U respectively. ctr (MeD) and U ctr (MeU), the residual risk RR of failure mode x is: RR x =RL x -HE ctr (MeD)-U ctr (MeU) In the formula, RL x U represents the inherent risk level of failure mode x. ctr (MeD) represents the total utility of risk control for design improvement measures, U ctr (MeU) represents the total utility of risk control using compensatory measures; Step 8: Based on the residual risk obtained in Step 7 for each failure mode, calculate the total utility and incremental ratio of risk control measures; the total utility of failure mode x risk control is: U ctrx =U ctrx (MeU)+U ctrx (MeD) In the formula: U ctrx (MeD) represents the total utility of structural failure mode x design improvement measures and risk control, U ctrx (MeU) represents the total utility of risk control using compensatory measures for structural failure mode x. The risk level increase ratio of structural failure mode x is: I x =U ctrx / RL x In the formula: U ctrx For failure mode x total utility of risk control, RL x The inherent risk level of structural failure mode x.

2. The method for residual risk analysis of complex equipment structures based on quantitative assessment as described in claim 1, characterized in that, The prospect coefficient A is set to 0.89.

Images