Complex spacecraft system credibility evaluation and decision-making method based on fuzzy information entropy
Through the trustworthy evaluation and decision-making method of complex spacecraft systems based on fuzzy information entropy, the problem that system engineering work in the development of complex spacecraft is difficult to ensure engineering mission goals, and the improvement of system engineering execution efficiency and credibility are achieved.
Patent Information
- Application Number
- CN202411642036.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-18
- Publication Date
- 2025-03-11
AI Technical Summary
During the development of complex spacecraft, it is difficult for system engineering to ensure the realization of engineering mission goals, and face the problems of technical management contradictions, related work being cut off, problems being exposed too late, repeated iterations in multiple stages, and inconsistent work goals with engineering goals.
The trustworthy evaluation and decision-making method of complex spacecraft systems based on fuzzy information entropy is adopted, and various trustworthy dimensions are quantified through triangular fuzzy numbers, the system engineering work evaluation data matrix is constructed, and priority is used to fuzzy sort to determine the credibility of various system engineering work in the system. The trustworthy improvement strategy library is established through expert methods, the impact of different strategies on the trustworthy state of the system is analyzed, and the trustworthy improvement rate and benefits are calculated.
It effectively improves the efficiency of system engineering execution, supports trustworthy improvement strategy decisions, ensures the realization of project task goals, establishes a feedback loop between the engineering organization and the technical system, and continuously improves credibility.
Smart Images
Figure CN119599457A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a complex spacecraft system trustworthy evaluation and decision-making method based on fuzzy information entropy, and belongs to the field of complex spacecraft development system engineering. Background Art
[0002] The development of spacecraft relies on the system engineering method, which decomposes engineering tasks to form system indicators at all levels to carry out system development at all levels, and then obtains the overall spacecraft system through system integration. Spacecraft represented by Chang'e 5 and Chang'e 6 have the characteristics of large technical span, complex system composition, strong coupling of on-orbit mission links, many irreversible links and high process uncertainty. It is difficult to decompose tasks and integrate systems according to deterministic system boundaries, clearly decomposed mission actions and mature technical status indicators. It is also difficult to ensure that the developed engineering system can achieve the engineering mission goals 100% from the overall system level.
[0003] Faced with the complex spacecraft development tasks with high requirements for subsequent key tasks, many heterogeneous participants, and strong technological innovation, the system engineering process will inevitably face problems such as "superposition of technical management and other contradictions", "artificial severance of related work", "frequent rework due to late exposure of problems", "multi-stage repeated iterations", and "inconsistency between work goals and engineering goals". It is necessary to start from the perspective of system engineering executors, focus on the credible concept of whether system engineering work can achieve engineering task goals, combine credible static characteristics and dynamic evolution characteristics, and construct a credible assessment method and credible improvement decision-making method for complex spacecraft systems to ensure that system engineering work can achieve engineering task goals. Summary of the invention
[0004] The technical problem solved by the present invention is: to overcome the shortcomings of the prior art, and to provide a complex spacecraft system trustworthiness assessment and decision-making method based on fuzzy information entropy, which can measure the credibility of system engineering work from the perspective of achieving engineering mission goals, assist organizations at all levels to better implement system engineering, improve the efficiency of system engineering execution, support trustworthiness enhancement strategy decision-making, and ensure the achievement of engineering mission goals.
[0005] The technical solution of the present invention is: a complex spacecraft system trustworthiness evaluation and decision-making method based on fuzzy information entropy, comprising:
[0006] According to the actual situation of various system engineering work, triangular fuzzy numbers are used to quantify various credible dimensions and obtain the quantified value X of each credible dimension. i ;
[0007] The results of quantifying various trust dimensions using triangular fuzzy numbers are used to construct a system engineering work evaluation data matrix. The system engineering work evaluation data is fuzzily sorted using priority to determine the trustworthiness of each system engineering work in the system.
[0008] Establish a system credibility improvement strategy library through expert methods, analyze the relationship between different credibility improvement strategies and the credibility of various system engineering tasks in the system; use the fuzzy ranking value of relative weights and apply the transcendental logarithmic multi-factor production function to calculate the impact of different strategies on the system credibility state and calculate the credibility improvement rate of different strategies;
[0009] Compute the benefits of various credibility enhancement strategies.
[0010] Preferably, triangular fuzzy numbers can quantify various credible dimensions and transform credible subjective judgments into quantitative data. Specifically:
[0011] The triangular fuzzy number α is expressed as α=(α l ,α m ,α u ), using this value to represent the work task goal of a certain dimension; where (α l ,α m ,α u ) are all real numbers, and α l <α m <α u , α m is called the principal value of the triangular fuzzy number α, l and α u They are called the upper and lower bounds of α respectively. (α m -α l ) and (α u -α m ) are called the lower and upper limits of α respectively;
[0012] When α l =α m =α u , α becomes a real number in the ordinary sense; when (α u -α l ) is larger, the triangular fuzzy number α=(α l ,α m ,a u ) is more blurred;
[0013] Triangular fuzzy numbers can represent multiple fuzzy evaluations of a system engineering work in a certain credible dimension.
[0014] Preferably, the systems engineering work evaluation data matrix is:
[0015] Construct a system engineering work evaluation data matrix A in a certain system = (A ij ) n×m , where the matrix A ijThe value represents the triangular fuzzy number of a system engineering work i in a certain credible dimension j, the subscript n represents the total category of system engineering work, and the subscript m represents the total category of credible dimensions.
[0016] Preferably, the system engineering work evaluation data is fuzzily sorted by priority to determine the degree of trustworthiness of each system engineering work in the system for the trust dimension j, specifically:
[0017] For system engineering task i, a priority function is constructed in the corresponding closed interval, where the closed interval is A i (α)=[A iL (α),A iR (α)], i = 1, 2, ..., n, triangular fuzzy number α∈[0,1], left priority function Right priority function They are:
[0018]
[0019] Then construct the consistency positive and negative judgment matrix:
[0020]
[0021] Among them, λ represents the risk attitude towards credible judgment;
[0022] According to the eigenvector corresponding to the maximum eigenroot of U(α,λ):
[0023] ω(α,λ)=(ω1(α,λ),ω2(α,λ),…,ω i (α,λ),…,ω n (α,λ) T
[0024] The fuzzy number A is obtained by integration i The corresponding sorted vector:
[0025] W i (λ)=∫0 1 ω i (α,λ)dα,i=1,2,…,n
[0026] in:
[0027]
[0028] Let W(λ)=(W1(λ),W2(λ),…,W n (λ) T Normalization is performed, and the larger the ranking value, the stronger the credibility of the system engineering work. The calculated normalized fuzzy ranking is set as the relative weight of each credibility dimension to calculate the overall credibility value of the system.
[0029] Preferably, the impact of different strategies on the system trust state is recorded as the total trust value TAV:
[0030]
[0031] X i is the quantitative value of each credible dimension; W i For X i The corresponding weight is the normalized fuzzy ranking value.
[0032] Preferably, the trust improvement efficiency of different strategies is calculated, specifically:
[0033] Use strategies to increase a credibility dimension x i When the state is , the calculation method of the reliability improvement rate IROT is:
[0034] IROT=(TAV after -TAV before ) / TAV before
[0035] Among them, TAV before That is, the total credibility value before using the strategy, TAV after That is, the total credibility value after using this strategy.
[0036] Preferably, the method for calculating the benefits of various trust enhancement strategies is:
[0037] BENEFIT=IROT / PM nominal
[0038] Among them, IROT is the credibility improvement rate; PM nominal The cost expenditure of this strategy is expressed using a structured cost model.
[0039] Preferably, a system trust improvement strategy library is established through an expert method, which includes the trust dimension quantified value data X of each system engineering work and the sorting information W for different trust dimensions. i .
[0040] Compared with the prior art, the present invention has the following advantages:
[0041] (1) The present invention can effectively conduct multi-dimensional trustworthy evaluation of the system engineering work of complex spacecraft development at the executor level, and can measure the trustworthiness of the system engineering work from the perspective of achieving engineering mission objectives, thereby assisting organizations at all levels to better implement system engineering and improve the efficiency of system engineering execution;
[0042] (2) The present invention can assist decision makers in establishing a trust improvement strategy library, analyze the nonlinear impact of various strategies on trust improvement, and perform fuzzy sorting of various strategies through quantitative methods while considering economic costs, support trust improvement strategy decision-making, and ensure the achievement of engineering task objectives;
[0043] (3) The present invention establishes a feedback loop between the engineering organization system and the engineering technology system through credible evaluation and decision-making. It can continuously improve the organization system's understanding of the credibility of the technology system through positive feedback between the two systems, while ensuring that the technology system ultimately achieves the engineering task objectives. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] Figure 1 It is a trustworthy evaluation and decision framework diagram of the present invention;
[0045] Figure 2 It is the trustworthy evaluation matrix of the present invention. DETAILED DESCRIPTION
[0046] The development of complex spacecraft is a process in which the development organization adopts a system engineering approach to carry out system engineering work. It is very important for the development organization to evaluate whether the system engineering work can achieve the engineering mission objectives, which helps to optimize the system engineering work and ensure the achievement of the engineering mission objectives.
[0047] Therefore, each system engineering work can be trusted based on the trusted characteristic attributes of the engineering task objectives, thereby quantitatively measuring the overall trust status of the engineering system and optimizing the choice of trust improvement strategies based on the impact of different strategies on the trustworthiness of system engineering work.
[0048] To achieve the above purpose, Figure 1 As shown, the technical solutions adopted by the present invention are as follows:
[0049] According to the actual situation of various system engineering work, triangular fuzzy numbers are used to quantify various credible dimensions and obtain the quantified value X of each credible dimension. i ;
[0050] The results of quantifying various trust dimensions using triangular fuzzy numbers are used to construct a system engineering work evaluation data matrix. The system engineering work evaluation data is fuzzily sorted using priority to determine the trustworthiness of each system engineering work in the system.
[0051] Establish a system credibility improvement strategy library through expert methods, analyze the relationship between different credibility improvement strategies and the credibility of various system engineering tasks in the system; use the fuzzy ranking value of relative weights and apply the transcendental logarithmic multi-factor production function to calculate the impact of different strategies on the system credibility state and calculate the credibility improvement rate of different strategies;
[0052] Compute the benefits of various credibility enhancement strategies.
[0053] The above method is described in detail:
[0054] 1. Standard triangular fuzzy numbers can quantify various credible dimensions and transform credible subjective judgments into quantitative data. Specifically:
[0055] The general triangular fuzzy number α is expressed as α = (α l ,α m ,α u ), using this value to represent the work task goal of a certain dimension; where (α l ,α m ,α u ) are all real numbers, and α l <α m <α u , α m is called the principal value of the triangular fuzzy number α, l and α u They are called the upper and lower bounds of α respectively. (α m -α l ) and (α u -α m ) are called the lower and upper limits of α respectively;
[0056] When α l =α m =α u , α becomes a real number in the ordinary sense; when (α u -α l ) is larger, the triangular fuzzy number α=(α l ,α m ,α u ) is more blurred;
[0057] Standard triangular fuzzy numbers can represent multiple fuzzy evaluations of a system engineering work in a certain credible dimension.
[0058] 2. The system engineering work evaluation data matrix is:
[0059] Construct a system engineering work evaluation data matrix A in a certain system = (A ij ) n×m , where the matrix A ij The value represents the triangular fuzzy number of a system engineering work i in a certain credible dimension j, the subscript n represents the total category of system engineering work, and the subscript m represents the total category of credible dimensions.
[0060] Credibility matrix Figure 2 shown.
[0061]
[0062] 3. Use the priority to fuzzily sort the system engineering work evaluation data to determine the degree of trustworthiness of each system engineering work in the system for the trust dimension j, specifically:
[0063] For system engineering task i, a priority function is constructed in the corresponding closed interval, where the closed interval is A i (α)=[A iL (α),A iR (α)], i = 1, 2, ..., n, triangular fuzzy number α∈[0,1], left priority function Right priority function They are:
[0064]
[0065] Then construct the consistency positive and negative judgment matrix:
[0066]
[0067] Among them, λ represents the risk attitude towards credible judgment;
[0068] According to the eigenvector corresponding to the maximum eigenroot of U(α,λ):
[0069] ω(α,λ)=(ω1(α,λ),ω2(α,λ),…,ω i (α,λ),…,ω n (α,λ) T
[0070] in,
[0071]
[0072] The fuzzy number A is obtained by integration i The corresponding sorted vector:
[0073] w i (λ)=∫0 1 ω i (α,λ)dα,i=1,2,…,n
[0074] Let W(λ)=(W1(λ),W2(λ),…,W n (λ) T Normalization is performed, and the larger the ranking value, the stronger the credibility of the system engineering work. The calculated normalized fuzzy ranking is set as the relative weight of each credibility dimension to calculate the overall credibility value of the system.
[0075] 4. The impact of different strategies on the system's trustworthy state is recorded as the total trustworthiness value TAV:
[0076]
[0077] X i is the quantitative value of each credible dimension; W i For X i The corresponding weight is the normalized fuzzy ranking value.
[0078] 5. Calculate the trust improvement efficiency of different strategies, specifically:
[0079] Use strategies to improve a credibility dimension X i When the state is , the calculation method of the reliability improvement rate IROT is:
[0080] IROT=(TAV after -TAV befote ) / TAV before
[0081] Among them, TAV before That is, the total credibility value before using the strategy, TAV after That is, the total credibility value after using this strategy.
[0082] 6. The method for calculating the benefits of various trust enhancement strategies is:
[0083] BENEFIT=IROT / PM nominal
[0084] Among them, IROT is the credibility improvement rate; PM nominal The cost expenditure of this strategy is expressed using a structured cost model.
[0085] Example:
[0086] According to the complex spacecraft system engineering method, the core system engineering work is determined around the three main tasks of design, verification and control, the trust dimension is selected according to the mission objectives, and possible trust improvement strategies are proposed.
[0087] A. Fuzzy measurement of credible values in system engineering work
[0088] Determine the credible evaluation dimensions including indicator goals and demand goals based on the characteristics of the engineering task objectives.
[0089] According to the actual situation of various system engineering work, standard triangular fuzzy numbers are used to quantify various credible dimensions, converting credible subjective judgments into quantitative data, and obtaining the quantitative value X of each credible dimension. i (This value is a quantitative value of the system engineering work i. This value is a constant value and can be replaced by a traditional reliability index.) Specifically:
[0090] A general triangular fuzzy number α can be expressed as α = (a l , α m , α u ), and use this value to represent the work task goal of a certain dimension. l , α m , α u ) are all real numbers, and a l <α m <α u , α m is called the principal value of the triangular fuzzy number α, l and α u They are called the upper and lower bounds of α respectively. (α m -α l ) and (α u -α m ) are called the lower limit and upper limit of α respectively. l =α m =α u , α becomes a real number in the ordinary sense. l -α i ) is larger, the triangular fuzzy number α=(α l , α m , αu ) The more blurred.
[0091] Standard triangular fuzzy numbers can be used to represent five fuzzy evaluations of a system engineering work in a certain credibility dimension (very untrustworthy, untrustworthy, general, trustworthy, and very trustworthy).
[0092] B. Multi-dimensional evaluation of system credibility
[0093] Determine the core system engineering work of each system around design, verification and control, establish a system trustworthiness assessment matrix based on various trustworthiness dimensions, and use information entropy methods to carry out multi-dimensional measurement of the trustworthiness of system engineering work.
[0094] 1) Construct a system-level system engineering work evaluation data matrix, where the behavior of system engineering work is listed as a credible feature. The values in the matrix represent the credible fuzzy evaluation value of a certain work in a certain dimension. Specifically, the system engineering work evaluation data matrix in a certain system is:
[0095]
[0096] Each value in the matrix represents the triangular fuzzy number of a system engineering work in a certain credible dimension. The subscript n represents the total category of system engineering work, and the subscript m represents the total category of credible dimensions.
[0097] 2) Information entropy is used to comprehensively measure the contribution of the credible features in each system. The higher the entropy value, the higher the credible contribution of the credible feature to the system. Specifically:
[0098] For any fuzzy number α on the interval [a, b] l , α m , α u ), information entropy is used to measure the credibility of multidimensional work evaluation data.
[0099]
[0100] Where E(α) represents information entropy, 0≤E(a)≤1, E(a) j , the larger the E(a) is, the greater the trustworthiness of the j-th dimension's contribution to the trustworthiness of the entire system.
[0101] 3) Determine the relative weight of each credible feature based on the normalized fuzzy ranking, and determine the credibility of the system based on the credibility values of each dimension.
[0102] The evaluation data is fuzzily sorted using priority to determine the credibility of each system engineering task in the system.
[0103] In the corresponding closed interval (α∈[0,1]) construct left and right priority functions Then construct the consistency positive and negative judgment matrix U:
[0104]
[0105] According to the eigenvector ω(α, λ) corresponding to the maximum eigenroot of U(a, λ) obtained, ω(α, λ) = (ω1(α, λ), ω2(α, λ), …, ω n (α, λ) T , the integral synthesis obtains the fuzzy number The corresponding sorting vector W i (λ)=∫0 1 ω i (α, λ)dα, i=1, 2,...,n, let W(λ)=(W1(λ), W2(λ),...,W n (λ) T Normalization is performed. The larger the ranking value, the stronger the credibility of the system engineering work. In addition, the calculated normalized fuzzy ranking can be set as the relative weight of each credibility dimension to calculate the overall credibility value of the system.
[0106] C. Dynamic improvement strategy for system trustworthiness
[0107] 1) Establish a system trust improvement strategy library
[0108] A system credibility enhancement strategy library is established through expert methods, and the nonlinear relationship between strategies and different credibility indicator dimensions is analyzed.
[0109] Analyze the possible impact of different trust improvement strategies on the trust status of different dimensions of the system, and classify them according to the positive and negative effects.
[0110] The strategy library contains part A of the credible dimension quantitative value data X of each system engineering work i , and the ranking information W for different trust dimensions in part B i .
[0111] 2) Impact of the trusted state of the strategy
[0112] In view of the nonlinear influence characteristics, the fuzzy ranking value of relative weight is adopted and the transcendental logarithmic multi-factor production function is applied to calculate the impact of different strategies on the system's trustworthy state.
[0113] Considering the multidimensional characteristics of trust and the nonlinear impact of strategies on trust values of different dimensions, the transcendental logarithmic multi-factor production function is applied to calculate the impact of different strategies on the trust state of the system.
[0114]
[0115] Among them, TAV is the total credibility value, X i is the quantitative value of each credible dimension; W i For X i The corresponding weight is the normalized fuzzy ranking value.
[0116] 3) Evaluation of the credibility of the strategy
[0117] The credibility improvement effect of different strategies is calculated based on the ratio value.
[0118] Use strategies to improve a credibility dimension X i When the reliability is in a state, the reliability improvement rate can be calculated according to the following formula.
[0119] IROT=(TAV after -TAV before ) / TAV before
[0120] Among them, IROT is the reliability improvement rate, TAV before That is, the total credibility value before using the strategy, TAV after That is, the total credibility value after using this strategy.
[0121] D. Trust Enhancement Strategy Decision
[0122] Since all trust strategies require certain cost expenditures, the benefits of trust enhancement strategies can be determined based on the ratio of efficiency to cost, thereby making strategic decisions on trust enhancement.
[0123] Apply the following formula to calculate the benefits of various credibility enhancement strategies.
[0124] BENEFIT=IROT / PM nominal
[0125] Among them, BENEFIT is the credibility benefit, IROT is the credibility improvement rate, PM nominal The cost expenditure of this strategy can be expressed using a structured cost model.
[0126] The contents not described in detail in the specification of the present invention belong to the prior art known to the professional and technical personnel in this field.
Claims
1. A method for generating test data for a visual intelligence algorithm, characterized in that include; According to the actual situation of various system engineering work, triangular fuzzy numbers are used to quantify various credible dimensions and obtain the quantified value X of each credible dimension. i ; The results of quantifying various trust dimensions using triangular fuzzy numbers are used to construct a system engineering work evaluation data matrix. The system engineering work evaluation data is fuzzily sorted using priority to determine the trustworthiness of each system engineering work in the system. Establish a system credibility improvement strategy library through expert methods, analyze the relationship between different credibility improvement strategies and the credibility of various system engineering tasks in the system; use the fuzzy ranking value of relative weights and apply the transcendental logarithmic multi-factor production function to calculate the impact of different strategies on the system credibility state and calculate the credibility improvement rate of different strategies; Compute the benefits of various credibility enhancement strategies.
2. The complex spacecraft system trustworthiness assessment and decision-making method based on fuzzy information entropy according to claim 1, characterized in that: Triangular fuzzy numbers can quantify various credible dimensions and transform credible subjective judgments into quantitative data. Specifically: The triangular fuzzy number α is expressed as α=(α l , α m , α u ), using this value to represent the work task goal of a certain dimension; where (α l , α m , α u ) are all real numbers, and α l <α m <α u , α m is called the principal value of the triangular fuzzy number α, l and α u They are called the upper and lower bounds of α respectively. (α m -α l ) and (α u -α m ) are called the lower and upper limits of α respectively; When α l =α m =α u , α becomes a real number in the ordinary sense; when (α u -α l ) is larger, the triangular fuzzy number α=(α l , α m , α u ) is more blurred; Triangular fuzzy numbers can represent multiple fuzzy evaluations of a system engineering work in a certain credible dimension.
3. The complex spacecraft system trustworthiness assessment and decision-making method based on fuzzy information entropy according to claim 1, characterized in that: The system engineering work evaluation data matrix is: Construct a system engineering work evaluation data matrix A in a certain system = (A ij ) n×m , where the matrix A ij The value represents the triangular fuzzy number of a system engineering work i in a certain credible dimension j, the subscript n represents the total category of system engineering work, and the subscript m represents the total category of credible dimensions.
4. The complex spacecraft system trustworthiness assessment and decision-making method based on fuzzy information entropy according to claim 1 is characterized by: The system engineering work evaluation data is fuzzily sorted using priority to determine the degree of trustworthiness of each system engineering work in the system for the trust dimension j, specifically: For system engineering task i, a priority function is constructed in the corresponding closed interval, where the closed interval is A i (α)=[A iL (α), A iR (α)], i = 1, 2, ..., n, triangular fuzzy number α∈[0, 1], left priority function Right priority function They are: i=1,2,...,n,j=1,2,...,m Then construct the consistency positive and negative judgment matrix: Among them, λ represents the risk attitude towards credible judgment; According to the eigenvector corresponding to the maximum eigenroot of U(α,λ): ω(α,λ)=(ω1(α,λ),ω2(α,λ),...,ω i (a, l),...,ω n (a, l)) T The fuzzy number A is obtained by integration i The corresponding sorted vector: in: Let W(λ)=(W1(λ), W2(λ),...,W n (λ) T Normalization is performed, and the larger the ranking value, the stronger the credibility of the system engineering work. The calculated normalized fuzzy ranking is set as the relative weight of each credibility dimension to calculate the overall credibility value of the system.
5. The complex spacecraft system trustworthiness assessment and decision-making method based on fuzzy information entropy according to claim 1 is characterized by: The impact of using different strategies on the system's trust state is recorded as the total trust value TAV: X i is the quantitative value of each credible dimension; W i For X i The corresponding weight is the normalized fuzzy ranking value.
6. The complex spacecraft system trustworthiness assessment and decision-making method based on fuzzy information entropy according to claim 1, characterized in that: Calculate the trust improvement efficiency of different strategies, specifically: Use strategies to improve a credibility dimension X i When the state is , the calculation method of the reliability improvement rate IROT is: IROT=(TAV after -TAV before ) / TAV before Among them, TAV before That is, the total credibility value before using the strategy, TAV after That is, the total credibility value after using this strategy.
7. The complex spacecraft system trustworthiness assessment and decision-making method based on fuzzy information entropy according to claim 1 is characterized by: The method for calculating the benefits of various credible improvement strategies is: BENEFIT=IROT / PM nominal Among them, IROT is the credibility improvement rate; PM nominal The cost expenditure of this strategy is expressed using a structured cost model.
8. The complex spacecraft system trustworthiness assessment and decision-making method based on fuzzy information entropy according to claim 1 is characterized by: Establish a system trustworthiness improvement strategy library through expert methods, including the trustworthiness dimension quantitative value data of each system engineering work X i , and the ranking information W for different credible dimensions i .