Full-life-cycle comprehensive fuzzy evaluation method based on time sequence function and related product

Abstract

The invention relates to the technical field of engineering suitability evaluation, in particular to a full-life-cycle comprehensive fuzzy evaluation method based on a time sequence function and a related product, and the method comprises the steps: dividing evaluation time periods of a full life cycle, and constructing an influence factor coefficient matrix and a mutual feedback effect matrix of each time period; performing function fitting on the time sequences of the influence factor matrix and the mutual feedback effect matrix by adopting a least square method to obtain a time sequence function; an evaluation matrix in each time period is constructed, a stationary point value of a time sequence function is solved, evaluation values of all schemes in different evaluation time periods are calculated, a comprehensive evaluation function is used for comparing the multiple schemes, and finally the optimal scheme is determined. According to the method, a time sequence function is adopted for modeling influence factors and mutual feedback effects of all time periods, and the dynamic change trend of all schemes in different stages can be accurately reflected; through calculation of an evaluation matrix and a correlation coefficient, advantages and disadvantages of each scheme can be systematically integrated.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering suitability evaluation, and specifically relates to a comprehensive fuzzy evaluation method based on a time series function for the entire life cycle and related products. Background Art

[0002] Currently, in the analysis and evaluation of complex systems in multiple fields, there is a lack of a systematic and standardized multi-index comprehensive evaluation method for the comprehensive suitability evaluation of multiple solutions or scenarios. Existing technologies usually evaluate single indicators, such as the stability analysis of geological conditions, the assessment of engineering construction risks, or the single calculation of resource consumption. These single-index evaluation methods are difficult to comprehensively reflect the coupling relationship and dynamic changes of multiple factors in a complex system in scenarios that require the integration of multiple indicators and multi-stage characteristics.

[0003] For the comprehensive evaluation of multiple indicators, existing methods generally fall into the following categories:

[0004] Single-index evaluation methods based on monitoring data or mathematical models, which are applicable to clear index scenarios. However, these methods have limited applicability in dealing with complex scenarios with no or little monitoring data, and lack the ability to analyze the mutual relationship between multiple indicators.

[0005] In the case of no data support, qualitative analysis is relied on expert experience. This method is relatively convenient in terms of operation, but the results are easily affected by subjective factors, resulting in insufficient objectivity and scientificity of the evaluation.

[0006] In recent years, fuzzy mathematics has gradually been applied to the evaluation of various types of indicators, and the uncertainty is characterized by introducing fuzzy membership degrees. However, most existing methods only target static scenarios and ignore the dynamic characteristics of multiple indicators within the entire life cycle. Summary of the Invention

[0007] The technical problem to be solved by the present invention is that the existing multi-solution decision-making evaluation method cannot effectively handle the dynamics of influencing factors changing over time, and fails to fully consider the time series characteristics of influencing factors and the interaction relationship between solutions. The purpose is to provide a comprehensive fuzzy evaluation method based on a time series function for the entire life cycle and related products, which realizes the evaluation of different solutions at different evaluation time periods, and compares multiple solutions through a comprehensive evaluation function, so as to provide a more comprehensive and reasonable decision-making basis for users and ensure that the finally selected solution can remain optimal throughout the life cycle.

[0008] The present invention is achieved through the following technical solutions:

[0009] A comprehensive fuzzy evaluation method based on a time series function for the entire life cycle, comprising:

[0010] Evaluate multiple solutions over the entire life cycle, divide the evaluation periods, and determine the influence factor coefficient matrix and the mutual feedback matrix for each period;

[0011] Use the least squares method to fit the time series of the influence factor matrix and the mutual feedback matrix with functions, and construct the time series functions corresponding to the influence factor matrix and the mutual feedback matrix;

[0012] Construct the evaluation matrix for each period, and calculate the evaluation values of each solution in different evaluation periods by solving the stationary point values of the time series function, and construct the evaluation matrix for the entire life cycle;

[0013] Obtain the correlation coefficients of different solutions in each period, and compare the solutions for the entire life cycle through the comprehensive evaluation function to obtain the optimal solution.

[0014] Specifically, the method for dividing the evaluation periods includes:

[0015] Denote the set of multiple solutions to be evaluated as {A i | i = 1, 2,..., m}, where m is the total number of solutions;

[0016] Record the critical moments corresponding to each solution to form the preliminary evaluation moment set of solution A i ;

[0017] Organize the evaluation moment sets of all solutions and align the evaluation moments. If there are unaligned moments, insert new moments according to the state changes in adjacent periods to obtain the supplementary evaluation moment set of solution A i ;

[0018] Combine the preliminary evaluation moment set and the supplementary evaluation moment set to obtain the complete evaluation moment set of solution A i :

[0019] Unify the division of the complete evaluation moment sets T t (A i ) of all solutions to generate the unified evaluation period set {T t | t = 1, 2,..., q} for the entire life cycle. Each evaluation period T t is the time unit for fuzzy evaluation, and q is the total number of unified evaluation periods for the entire life cycle.

[0020] Specifically, within each evaluation period T t , assign trapezoidal fuzzy numbers to each influence factor and perform defuzzification to obtain each solution A i in period Tt Evaluation value matrix of influencing factors;

[0021] Normalize the evaluation values of influencing factors within each time period to construct the influencing factor coefficient matrix G t , where Where, represents the influencing factor coefficient of Plan A t within the time period T i ;

[0022] Calculate the mutual feedback effect between each plan within the same time period to obtain the mutual feedback effect matrix P t , where Where, represents the degree of mutual influence of Plan A i with other plans within the time period T t ;

[0023] Specifically, the timing function includes constant function, piecewise function, oscillating function, smoothing function and step function; the oscillating function includes a descending oscillating function approaching the asymptote and an ascending oscillating function approaching the asymptote; the smoothing function includes a descending smoothing function and an ascending smoothing function; the step function includes an ascending step function approaching the asymptote and a descending step function approaching the asymptote.

[0024] Specifically, the method for constructing the evaluation matrix for the entire life cycle includes:

[0025] Construct the evaluation matrix N t (t) of each plan within each time period T i (t), N i (t) = G t (A i ) · P t (A i ) · M t (A i ), where G t (A i ) represents the influencing factor coefficient matrix of Plan A t within the time period T i , P t (A i ) represents the mutual feedback effect matrix of Plan A i within the time period T t , M t (A i ) is the importance degree matrix of the influencing factors of Plan A i ;

[0026] Solve the stationary point values of the timing function of the influencing factor matrix and the timing function of the mutual feedback effect matrix to obtain the evaluation values of the influencing factors of each plan at different evaluation time periods;

[0027] Obtain Solution A i Evaluation matrix for the entire life cycle i ∈ m, where m is the total number of solutions, q is the total number of unified evaluation time periods in the entire life cycle, and N i (tq) is the evaluation value of Solution A i at the q-th moment.

[0028] Specifically, the method for obtaining the correlation coefficient includes:

[0029] Define (τ i ) T = N i (t t ), where τ i = max{N i (t t )}, extract the solution value of each Solution A i at its optimal time period and form a new vector where, i ∈ m, J, K,..., L ∈ T represents the set of typical moments within the evaluation time period T t ;

[0030] Calculate the difference sequence for each time period t c , By calculating the difference sequences for all time periods, obtain the mean value and the standard deviation

[0031] Calculate the mean value of its change vector and the standard deviation where,

[0032] Calculate the correlation coefficient where, the sign function has the following value-taking rules: takes the value of 0 when takes a positive value when takes a negative value when

[0033] Specifically, the comprehensive evaluation function where, N i (t t ) is the evaluation value of Solution A within the time period t t , and obtain the optimal solution i

[0034] Optionally, screen out the evaluation values with a correlation coefficient not greater than 0, and determine the optimal solution according to the comprehensive evaluation function.​

[0035] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-described method is implemented.

[0036] A computer program product includes a computer program / instructions, and when the computer program / instructions are executed by a processor, the above-described method is implemented.

[0037] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0038] By dividing the evaluation time periods of the entire life cycle and constructing the influence factor coefficient matrix and the mutual feedback action matrix for each time period, the least squares method is used to fit the time series of the influence factor matrix and the mutual feedback action matrix to obtain a time series function; by constructing an evaluation matrix within each time period and solving the stationary point value of the time series function, the evaluation values of each scheme at different evaluation time periods are calculated, and a comprehensive evaluation function is used to compare multiple schemes, and finally the optimal scheme is determined.

[0039] The present invention uses a time series function to model the influence factors and mutual feedback actions of each time period, which can accurately reflect the dynamic change trends of each scheme at different stages; through the calculation of the evaluation matrix and the correlation coefficient, the advantages and disadvantages of each scheme can be systematically integrated, avoiding the limitations of the traditional method that does not fully consider the time change factors; by comparing multiple schemes through a comprehensive evaluation function, the optimal scheme can be accurately identified among different schemes, thereby improving the scientificity and rationality of decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings illustrate exemplary embodiments of the present invention and, together with the description thereof, are used to explain the principles of the present invention. These drawings are included to provide a further understanding of the present invention, and the drawings are included in this specification and form a part of this specification, and do not constitute a limitation on the embodiments of the present invention.

[0041] Figure 1 is a schematic flowchart of the comprehensive fuzzy evaluation method for the entire life cycle based on a time series function according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0042] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the relevant content and do not limit the present invention.

[0043] In addition, it should be noted that for the sake of convenience of description, only parts related to the present invention are shown in the drawings.

[0044] Without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the drawings and in combination with the embodiments.

[0045] Embodiment 1

[0046] As Figure 1 shown, a comprehensive fuzzy evaluation method for the whole life cycle based on a time series function is provided, including:

[0047] S1. Evaluate multiple solutions during the whole life cycle, divide the evaluation time periods, and determine the influence factor coefficient matrix and the mutual feedback action matrix for each time period.

[0048] According to the key state change nodes of multiple solutions during the whole life cycle (such as the initial state, the major adjustment moment, the end state, etc.), the entire life cycle is divided into multiple time units (evaluation time periods) to ensure that the time axes of all solutions are aligned. The influence factor coefficient matrix is that for each time period, the influence factors of each solution (such as cost, efficiency, risk, etc.) are converted into a numerical coefficient matrix through fuzzy evaluation, reflecting the contribution degree of each factor to the solution. The mutual feedback action matrix is that within the same time period, the mutual influence between different solutions (such as resource competition, synergy effect, etc.) is quantified to form a matrix to describe the correlation strength of each solution.

[0049] Assign trapezoidal fuzzy numbers to the influence factors of each time period (such as mapping "high, medium, low" to a numerical interval), obtain a definite value through defuzzification (such as the centroid method), and generate an influence factor coefficient matrix after normalization. Quantify the mutual feedback action between solutions through expert scoring or data-driven methods to generate a mutual feedback action matrix.

[0050] S2. Use the least squares method to fit the time series of the influence factor matrix and the mutual feedback action matrix with functions, and construct time series functions corresponding to the influence factor matrix and the mutual feedback action matrix.

[0051] The time series functions include constant functions, piecewise functions, oscillating functions, smooth functions, and step functions; the oscillating functions include a descending oscillating function approaching the asymptote and an ascending oscillating function approaching the asymptote; the smooth functions include a descending smooth function and an ascending smooth function; the step functions include an ascending step function approaching the asymptote and a descending step function approaching the asymptote.

[0052] S3. Construct the evaluation matrix for each time period, and calculate the evaluation values of each solution in different evaluation time periods by solving the stationary point values of the time series functions, and construct the evaluation matrix for the whole life cycle; the evaluation matrix is a matrix that integrates the static evaluation values of each time period to form a dynamic evaluation in the time dimension, reflecting the comprehensive performance of the solution during the whole life cycle.

[0053] The calculation of the stationary point value is to determine the critical evaluation moments (such as the peak performance and the risk trough) by solving the stationary points (extreme points) of the derivative of the time series function, and generate the evaluation values at the typical moments.

[0054] S4. Obtain the correlation coefficients of different solutions in each time period, and compare the solutions of the whole life cycle through the comprehensive evaluation function to obtain the optimal solution.

[0055] The correlation coefficient is used to measure the degree of association between the evaluation value of the solution in each time period and the optimal state. It is dynamically weighted by combining the trend consistency and the deviation degree. The total score of the solution is obtained by weighted summation (correlation coefficient × evaluation value), and the solution with the highest total score is selected as the optimal solution.

[0056] Embodiment 2

[0057] This embodiment provides a method for dividing the whole life cycle time periods, forming multiple evaluation time periods. Through the recording, alignment and supplementary processing of the critical moments of the solutions, a time period set suitable for comprehensive fuzzy evaluation is generated to ensure the scientificity and consistency of the evaluation results.

[0058] The method includes:

[0059] Denote the set of multiple solutions to be evaluated as {A i | i = 1, 2,..., m}, where m is the total number of solutions, and each solution A i corresponds to different evaluation objects or technical solutions.

[0060] Record the critical moments corresponding to each solution to form the preliminary evaluation moment set of solution A i In this embodiment, the critical moment is generally the moment when each solution undergoes a significant change compared with the initial state under the action of the influencing factors.

[0061] Sort out the evaluation moment sets of all solutions and align the evaluation moments. The alignment operation ensures the consistency of the evaluation time periods of all solutions at the time nodes, which is conducive to unified comparative analysis. If there are unaligned moments, new moments are inserted according to the state changes in adjacent time periods to obtain the supplementary evaluation moment set of solution A i

[0062] Integrate the preliminary evaluation moment set and the supplementary evaluation moment set to obtain the complete evaluation moment set of solution A i : The complete evaluation moment set is the set of all critical time points in the whole life cycle of the solution, including the initially recorded moments and the supplementary inserted moments. ​​

[0063] Unify the set \(T\) of complete evaluation moments for all scenarios t (A i ) is uniformly partitioned to generate a unified set \(\{T t |t = 1, 2, \ldots, q\}\) of evaluation periods for the entire life cycle. Each evaluation period \(T t is a time unit for performing fuzzy evaluation.

[0064] Within each evaluation period \(T t , trapezoidal fuzzy numbers are assigned to each influencing factor and defuzzification processing is performed to obtain the matrix of evaluation values of influencing factors for each scenario \(A i in the period \(T t ;

[0065] Assigning trapezoidal fuzzy numbers maps the verbal evaluation to trapezoidal fuzzy numbers to quantify the attributes of influencing factors with uncertainty. For example, the fuzzy sets are described by excellent or large or good feasibility (\(\delta\)), good or relatively large or better feasibility (\(\gamma\)), medium or average or passable (\(\beta\)), poor or small or poor feasibility (\(\alpha\)), and the site suitability evaluation just meets the characteristics of the trapezoidal function.

[0066] The basic definitions are as follows: The membership function of the fuzzy number on the universe of discourse \(R\) satisfies the following relational expression. The fuzzy number

[0067]

[0068] where: \(\alpha\leq\beta\leq\gamma\leq\delta\in\psi\),

[0069] When \(\alpha = \beta = \gamma = \delta\), represents a non-fuzzy number, and the exact numerical value can also be represented by .

[0070] The membership degree \(\mu ij of \(G G \) in the normalized evaluation scheme coefficient matrix \(G\): \(R\rightarrow[0, 1]\). For the same influencing factor \(X i , obviously, the weight coefficients of the scenario set \(A i =\{(X1, G1), (X2, G2), \ldots, (X i , G i )\) satisfy the trapezoidal function model and can be scaled by (\(\alpha\), \(\beta\), \(\gamma\), \(\delta\)) for the grades of poor, medium, good, and excellent.

[0071] Define a method for defuzzifying numbers:

[0072] Let: \(S\) be a bounded convex fuzzy number set λ (0 < λ ≤ 1), whose level cut set is a real closed interval. When λ j = j / n (j = 1, 2... n), denote If R is the set of real numbers, the function D: S → R, for all have: And Then D is called the defuzzification function on S. From the definition of trapezoidal fuzzy numbers in the previous text,

[0073] The generated value matrix where each row corresponds to an evaluation period and each column is an influencing factor. Normalize the above matrix to eliminate the dimension difference, map the evaluation values of each factor to a unified scale, and integrate the normalized evaluation values of all solutions to form the coefficient matrix G within the period t .

[0074] Normalize the evaluation values of the influencing factors within each period to construct the influencing factor coefficient matrix G t , where Among them, represents the influencing factor coefficient of solution A t within the period T i ;

[0075] Through domain experts directly assigning values according to experience, determine the value of the mutual feedback effect, and then calculate the mutual feedback effect between each solution within the same period to obtain the mutual feedback effect matrix P t , where Among them, represents the degree of mutual influence of solution A i with other solutions within the period T t .

[0076] Example Three

[0077] The timing function includes constant function, piecewise function, oscillating function, smoothing function and step function; the oscillating function includes a descending oscillating function approaching the asymptote and an ascending oscillating function approaching the asymptote; the smoothing function includes a descending smoothing function and an ascending smoothing function; the step function includes an ascending step function approaching the asymptote and a descending step function approaching the asymptote.

[0078] The common reference function relations are as follows, and the parameters are all set according to needs.

[0079] The type of constant function is: y = m, x ≥ 0.

[0080] The type of piecewise function is:

[0081] The types of descending oscillating functions approaching asymptotes are as follows: x ≥ 0, λ > 0, a > 0, b > 0, c > 0, d > 0.

[0082] The types of ascending oscillating functions approaching asymptotes are as follows: x ≥ 0, λ > 0, a > 0, b > 0, c > 0, d > 0, e > 0.

[0083] The types of descending smooth functions are as follows: x ≥ 0, λ > 0, a > 0, b > 0.

[0084] The types of ascending smooth functions are as follows: x ≥ 0, λ > 0, a > 0, b > 0, c > 0.

[0085] The types of step functions with ascending asymptotes (the value of p is determined according to requirements) are: y = n1arctan(k1x) + n2arctan[k2(x - m)] + … + n p arctan[k p (x - n)], x ≥ 0, n1, n2, …, n p > 0, k1, k2, …, k p > 0.

[0086] The types of step functions with descending asymptotes (the value of p is determined according to requirements) are:

[0087] y = n1arctan(k1x) - n2arctan[k2(x - m)] - ··· - n p arctan[k p (x - n)], x ≥ 0, n1, n2, …, n p >0, k1, k2, …, k p > 0.

[0088] For specific influencing factor indicators, their functional relationships should be fitted according to the actual situation on-site, combined with the engineering geological analogy method and literature data method.

[0089] Example 4

[0090] The method for constructing an evaluation matrix for the entire life cycle includes:

[0091] Construct the evaluation matrix N t of each plan within each time period T i (t), N i (t) = G t (A i ) · P t (A i ) · M t (A i), where G t (A i ) represents that within the time period T t , the influence factor coefficient matrix of Scheme A i , indicating the influence degree of each influence factor on the scheme within this time period. P t (A i ) represents the mutual feedback action matrix of Scheme A i within the time period T t , indicating the mutual action and influence among the schemes within this time period. M t (A i ) is the importance degree matrix of the influence factors of Scheme A i , indicating the relative importance of each influence factor to the scheme. Multiplying the three matrices can obtain the evaluation matrix of Scheme A i within the time period T t , reflecting the evaluation result of the comprehensive influence of this scheme within a specific time period.

[0092] Solve the stationary point values of the time series function of the influence factor matrix and the time series function of the mutual feedback action matrix. The stationary point value is a specific value of the time series function, usually representing the equilibrium state or the optimal state of the system. In this embodiment, the extreme point (the first derivative is zero) or the non-differentiable point of the time series function is adopted. The stationary point value represents the optimal evaluation results of the influence factors and the mutual feedback actions of Scheme A i under different evaluation time periods. By solving the stationary point values, the evaluation values of the influence factors of each scheme at different evaluation time periods are obtained.

[0093] Combining the evaluation matrices within all evaluation time periods T t , construct the evaluation matrix N i of the full life cycle of each scheme A i , and obtain the evaluation matrix of the full life cycle of Scheme A i i ∈ m, m is the total number of schemes, q is the total number of unified evaluation time periods of the full life cycle, N i (tq) is the evaluation value of Scheme A i at the qth moment. The matrix will comprehensively summarize the evaluation values of all schemes within each time period, forming a complete full life cycle evaluation system to quantify the performance of each scheme at different time periods.

[0094] Example Five

[0095] The method for obtaining the correlation coefficient includes:

[0096] Define (τ i ) T = N i (t t ), where τ i = max{N​i (t t )}, that is, the value is equal to the evaluation value of Scheme A i within its optimal period. Extract the optimal period of each Scheme A i and the scheme value within its optimal period and form a new vector where i ∈ m, J, K,..., L ∈ T represents the set of typical moments within the evaluation period T t ;

[0097] Calculate the difference sequence d c for each period t i (t c ), By calculating the difference sequences of all periods, obtain the mean value and the standard deviation

[0098] Calculate the mean value of its change vector and the standard deviation where,

[0099] Calculate the correlation coefficient where the sign function has the following value-taking rules: When, it takes the value of 0, and there is no change in at least one party. When , it takes a positive value, indicating a same-direction change. When , it takes a negative value, indicating an opposite-direction change.

[0100] Weighted sum the correlation coefficients of each period with the evaluation value to obtain the comprehensive evaluation function where N i (t t ) is the evaluation value of Scheme A t within the period t i , and obtain the optimal scheme The scheme with the highest total score is the optimal choice within the whole life cycle.

[0101] In this embodiment, the optimal period and correlation coefficient of each scheme are extracted, and the performance changes of the scheme in different evaluation periods are measured by calculating the statistical values of the difference and change vector. Finally, by synthesizing the evaluation results of each period and comparing the schemes through the comprehensive evaluation function, the optimal scheme is selected.

[0102] In addition, based on the optimal scheme, a better scheme can be extracted.

[0103] First, perform validity screening. Only retain the time period data with a correlation coefficient greater than 0, and eliminate the negatively correlated or uncorrelated time periods, that is, screen out the evaluation values with a correlation coefficient not greater than 0, so as to remove the time periods that have no positive impact on the scheme evaluation and ensure that only those time periods with positive contributions are considered.

[0104] Among the time periods retained after screening, the larger the correlation coefficient, the stronger the positive correlation between the performance of the scheme in this time period and the external conditions, and it should be given priority as the decision-making basis.

[0105] Determine the optimal scheme according to the comprehensive evaluation function, that is, only use the time periods with a correlation coefficient greater than 0 to calculate the total score of the scheme, ensure that the decision result is dominated by the positive synergy time periods, and the finally selected scheme not only has the best overall performance in the whole life cycle, but also can be better than other schemes at the critical moment (that is, the moment with the largest evaluation value).

[0106] Example Six

[0107] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned method is implemented.

[0108] Without loss of generality, computer-readable media can include computer storage media and communication media. Computer storage media includes volatile and non-volatile, removable and non-removable media implemented by any method or technology for storing information such as computer-readable instructions, data structures, program modules, or other data. Computer storage media includes RAM, ROM, EPROM, EEPROM, flash memory or other solid-state storage technologies, CD-ROM, DVD or other optical storage, magnetic tape cartridges, tapes, magnetic disk storage or other magnetic storage devices. Of course, those skilled in the art know that computer storage media are not limited to the above several. The above-mentioned system memory and mass storage devices can be collectively referred to as memory.

[0109] A computer program product includes a computer program / instructions, and when the computer program / instructions are executed by a processor, the above-mentioned method is implemented.

[0110] A computer program product includes a computer program or instruction set for performing specific tasks or implementing specific functions. These programs or instructions are designed to be executable by a processor to achieve a series of predefined steps or operations. The program product may be stored in various forms of computer storage media, such as memory, hard disk, solid-state drive, optical disc or other forms of digital storage devices. It may exist in the form of compiled binary code or in the form of scripts or bytecodes executable by an interpreter. Through carefully designed algorithms and logical instructions, the program product enables the processor to process data in a specific order and manner to complete various functions such as data analysis, user interaction, and device control.

[0111] In the description of this specification, the descriptions with reference to the terms "one embodiment / way", "some embodiments / ways", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment / way or example are included in at least one embodiment / way or example of this application. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment / way or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments / ways or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments / ways or examples described in this specification and the features of different embodiments / ways or examples.

[0112] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, "a plurality of" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0113] Those skilled in the art should understand that the above embodiments are only for clearly explaining the present invention and are not intended to limit the scope of the present invention. For those skilled in the art, other changes or modifications can be made based on the above invention, and these changes or modifications are still within the scope of the present invention.

Claims

1. A comprehensive fuzzy evaluation method for the whole life cycle based on a time series function, characterized in that, Including: Evaluating multiple solutions over the entire life cycle, dividing the evaluation time periods and determining the influence factor coefficient matrix and the mutual feedback action matrix for each time period; Using the least squares method to perform function fitting on the time series of the influence factor matrix and the mutual feedback action matrix, and constructing the corresponding time series functions for the influence factor matrix and the mutual feedback action matrix; Constructing the evaluation matrix for each time period, and calculating the evaluation values of each solution at different evaluation time periods by solving the stationary point values of the time series functions, and constructing the evaluation matrix for the entire life cycle; Obtaining the correlation coefficients of different solutions within each time period, and comparing the solutions for the entire life cycle through the comprehensive evaluation function to obtain the optimal solution.

2. The comprehensive fuzzy evaluation method for the whole life cycle based on the time series function according to claim 1, wherein The method for dividing the evaluation time periods includes: Denote the set of multiple solutions to be evaluated as {A i | i = 1, 2,..., m}, where m is the total number of solutions; Record the critical moments corresponding to each plan to form Plan A i The initial evaluation moment set Sort out the evaluation moment sets of all plans And align the evaluation moments. If there are unaligned moments, insert new moments according to the state changes in adjacent time periods to obtain the supplementary evaluation moment set of Plan A i The supplementary evaluation moment set Set of comprehensive preliminary evaluation times and set of supplementary evaluation times to obtain i the complete set of evaluation times for Plan A: The complete set of evaluation moments \(T\) for all scenarios t (A i ) is uniformly divided to generate a unified set of evaluation time periods \(\{T t |t = 1, 2, \ldots, q\}\) for the entire life cycle. Each evaluation time period \(T t is a time unit for fuzzy evaluation, and \(q\) is the total number of unified evaluation time periods for the entire life cycle.

3. The full-life-cycle comprehensive fuzzy evaluation method based on a timing function according to claim 2, characterized in that, During each evaluation period T t assign trapezoidal fuzzy numbers to each influencing factor and perform defuzzification processing to obtain the evaluation value matrix of the influencing factors of each solution A i During period T t ; Normalize the evaluation values of influencing factors in each time period to construct the influencing factor coefficient matrix G t , where Among them, represents the influencing factor coefficient of Scheme A t within the time period T i ; Calculate the mutual feedback effect between each plan within the same time period to obtain the mutual feedback effect matrix P t , where Among them represents Plan A i within time period T t The degree of mutual influence with other plans 4. A comprehensive fuzzy evaluation method for the whole life cycle based on a timing function according to claim 1, characterized in that The time series functions include constant functions, piecewise functions, oscillating functions, smoothing functions, and step functions; the oscillating functions include a descending oscillating function approaching the asymptote and an ascending oscillating function approaching the asymptote; the smoothing functions include a descending smoothing function and an ascending smoothing function; the step functions include an ascending step function approaching the asymptote and a descending step function approaching the asymptote.

5. A comprehensive fuzzy evaluation method for the full life cycle based on a timing function according to claim 1, characterized in that, The method for constructing the evaluation matrix for the entire life cycle includes: Construct each time period T t The evaluation matrix N of each solution within i (t), N i (t)=G t (A i )·P t (A i )·M t (A i ), where G t (A i ) represents the influence factor coefficient matrix of solution A within time period T t , P i (A t ) represents the mutual feedback action matrix of solution A within time period T i , and M i (A t ) is the importance degree matrix of the influence factors of solution A t (A i ) is the importance degree matrix of the influence factors of solution A i ; Solving the stationary point values of the time series functions of the influence factor matrix and the time series functions of the mutual feedback action matrix to obtain the influence factor evaluation values of each solution at different evaluation time periods; Obtain Solution A i Evaluation matrix for the entire life cycle i ∈ m, where m is the total number of solutions, q is the total number of unified evaluation periods in the entire life cycle, N i (t q ) is the evaluation value of Solution A i at the q-th moment.

6. The comprehensive fuzzy evaluation method for the whole life cycle based on the timing function according to claim 5, characterized in that, The method for obtaining the correlation coefficients includes: Define (τ i ) T = N i (t t ), where τ i = max{N i (t t )}, extract the solution value of each solution A i at its optimal time period and form a new vector where, i ∈ m, J, K,..., L ∈ T represents the set of typical moments in the evaluation period T t within; Calculate for each period t c of the difference sequence, By calculating the difference sequences for all periods, the mean value and the standard deviation Calculate the mean of its change vector and standard deviation wherein Calculate the correlation coefficient Among them, the sign function The value-taking rule is as follows: When it takes the value of 0, when When it takes a positive value, when When it takes a negative value.

7. A comprehensive fuzzy evaluation method for the whole life cycle based on a timing function according to claim 6, characterized in that Comprehensive evaluation function Among them, N i (t t ) is the evaluation value of Scheme A within the time period t t , and the optimal scheme is obtained i ​ 8. A comprehensive fuzzy evaluation method for the whole life cycle based on a timing function according to claim 7, characterized in that, Screening out the evaluation values with correlation coefficients not greater than 0, and determining the optimal solution according to the comprehensive evaluation function.

9. A computer-readable storage medium storing a computer program, characterized in that, The computer program, when executed by a processor, implements the method according to any one of claims 1-8.

10. A computer program product, comprising a computer program / instructions, characterized in that, The computer program / instructions, when executed by a processor, implement the method according to any one of claims 1-8.

Images