System contribution rate evaluation method and system

Abstract

The present application relates to a kind of system contribution rate evaluation method and system, belong to equipment system technical field, solve the problem that the system contribution rate calculation is not accurate due to the lack of objectivity of index weight in existing;Including: constructing combat ring and netted index system in combat system;Each combat ring is refined into multiple task chains;Sample data is collected by combat simulation, the index value of each index in the end level in netted index system and the influence value of task chain are calculated, and initial index matrix is obtained;Using principal component analysis method and Lasso algorithm, the index weight of each index in the end level is obtained by iteration learning to initial index matrix;According to the combat simulation data to be evaluated and each index weight, the combat system capability comprehensive value is calculated;Remove the equipment to be evaluated from combat system, according to the change rate of combat system capability comprehensive value, the system contribution rate of the equipment to be evaluated is obtained. The objective evaluation of system contribution rate is realized.

Description

Technical Field

[0001] This invention relates to the field of equipment system technology, and in particular to a method and system for evaluating system contribution rate. Background Technology

[0002] The system contribution rate is used to measure the role of equipment in completing a mission. It focuses on measuring the contribution of a certain type or class of equipment system to completing a mission, achieving mission objectives, and meeting capability requirements. It aims to measure the magnitude of the role or influence of the components of an equipment system on the entire equipment system, thereby supporting equipment development and procurement decisions.

[0003] The assessment of the equipment system's contribution rate is based on an established indicator system. This involves collecting corresponding performance indicator data, calculating the weight of each indicator, and employing various evaluation models to assess the equipment system's contribution rate under mission-driven conditions. Current methods mostly utilize hierarchical tree-structured indicator systems, and various improved hierarchical tree-structured indicator systems. However, in complex real-world applications, these systems often involve simplification, and their importance relies heavily on expert experience, resulting in significant subjectivity.

[0004] In practical applications, various indicators of equipment may be coupled, exhibiting coupling relationships within the same layer or across layers, forming a network-like indicator system. Traditional subjective or objective indicator weighting methods, such as the analytic hierarchy process (AHP) and the chain ratio method, become inapplicable. Especially as the number of indicators increases, the complexity of the network-like indicator system changes exponentially, with increasingly complex correlations and complicated calculation methods for indicator weights. Relying solely on expert experience to assign weights to individual indicators in practical applications, even with simplified indicator coupling relationships, results in significant discrepancies with reality. For assessing the system contribution rate of equipment, inaccurate input will lead to biased assessment conclusions, hindering a clear understanding of the equipment's contribution to the mission. Summary of the Invention

[0005] Based on the above analysis, the embodiments of the present invention aim to provide a method and system for evaluating the system contribution rate, so as to solve the problem that the calculation of the system contribution rate is inaccurate due to the lack of objectivity of the indicator weights.

[0006] On one hand, embodiments of the present invention provide a method for evaluating the contribution rate of a system, comprising the following steps:

[0007] Based on the combat mission and various equipment, construct the combat loop and network indicator system in the combat system; based on the network indicator system, refine each combat loop into multiple task chains;

[0008] By collecting sample data through combat simulation, the values ​​of each indicator at the final level in the network indicator system are calculated based on each sample data. The impact value of the task chain is calculated based on each indicator value. Based on the impact value of the task chain and the corresponding indicator values ​​of each sample data, an initial indicator matrix is ​​obtained. The initial indicator matrix is ​​iteratively learned using principal component analysis and Lasso algorithm to obtain the weights of each indicator at the final level.

[0009] Based on the combat simulation data to be evaluated and the weights of various indicators, the comprehensive value of the combat system capability is calculated; the equipment to be evaluated is removed from the combat system, and the system contribution rate of the equipment to be evaluated is obtained based on the rate of change of the comprehensive value of the combat system capability.

[0010] Based on further improvements to the above methods, combat missions are decomposed into one or more combat objectives; each piece of equipment contains at least one equipment capability, and each equipment capability corresponds to at least one piece of equipment; the network indicator system includes: primary combat objectives, secondary equipment capabilities, and tertiary indicators.

[0011] Based on the above method, further improvements are made to the network indicator system, which refines each combat link into multiple task chains. This is to construct a task chain from the final level, where each indicator passes through the capabilities of each piece of equipment to complete each combat objective.

[0012] Based on further improvements to the above method, the impact value of the task chain is calculated according to various indicator values, including:

[0013] Based on the various indicator values ​​included in the capabilities of each piece of equipment in the task chain, the correlation coefficient between two adjacent nodes is calculated as the relationship value of the corresponding edge; the average value of the relationship values ​​of each edge in each task chain is taken as the influence value of the task chain.

[0014] Based on further improvements to the above method, an initial indicator matrix is ​​obtained according to the impact value of the task chain and the indicator values ​​corresponding to each sample data, including:

[0015] The importance value of each indicator is obtained by weighting the impact value of each indicator in the task chain to which it belongs with the weight of the combat target in the task chain.

[0016] Based on the importance values ​​of each indicator at the final level, the indicator values ​​corresponding to each sample data are weighted separately to obtain the vector of indicator values ​​corresponding to each sample.

[0017] The vectors of all the indicator values ​​in the sample data constitute the initial indicator matrix.

[0018] Based on further improvements to the above method, principal component analysis and the Lasso algorithm are used to iteratively learn the initial index matrix to obtain the weights of each index in the final stage, including:

[0019] Principal component analysis is used to obtain a dictionary index matrix after dimensionality reduction of the initial index matrix. Based on the initial index matrix and the dictionary index matrix, the Lasso algorithm is used to obtain the regression coefficient matrix, which is then normalized and used as the index weight matrix. The average weight of each index in the index weight matrix is ​​calculated. When the standard deviation of the average weight is less than 1, the initial index matrix is ​​updated according to the average weight. Principal component analysis and the Lasso algorithm are used again to obtain a new average weight. When the standard deviation of the average weight is greater than 1, the iteration stops. The final average weight is the index weight.

[0020] Based on a further improvement of the above method, the regression coefficients are obtained using the Lasso algorithm, based on the initial index matrix and the dictionary index matrix, by solving the following objective function:

[0021]

[0022] Where S represents the initial index matrix, Z represents the dictionary index matrix, β represents the regression coefficient matrix, λ represents a constant, taking a decimal number between [0,1], and ||·||1 and ||·||2 represent the 1-norm and 2-norm, respectively.

[0023] Based on further improvements to the above method, and according to the operational simulation data to be evaluated and the weights of various indicators, the comprehensive value of the operational system capability is calculated, including:

[0024] Based on the values ​​and weights of the various indicators included in each equipment capability, the capability value of each equipment is calculated using the indicator aggregation method.

[0025] For each established combat loop, the capability value of the combat loop is obtained by multiplying the capability values ​​of the reconnaissance, decision-making, and attack equipment together and then dividing by the capability value of the target equipment.

[0026] The capability values ​​of each operational link corresponding to the same operational objective are added together to obtain the influence value of the operational link on the operational objective; according to the weight of each operational objective, the corresponding influence values ​​are weighted and summed to obtain the comprehensive capability value of the operational system.

[0027] Based on further improvements to the above method, the equipment to be evaluated is removed from the combat system. The system contribution rate of the equipment to be evaluated is then obtained using the following formula, based on the rate of change of the overall combat system capability value:

[0028]

[0029] Where G represents the combat system network, G -v This indicates that removing the equipment v to be evaluated from G will form a new combat system network, h G This represents the overall capability value of the original combat system. This represents the overall capability value of the new combat system.

[0030] On the other hand, embodiments of the present invention provide a system for evaluating the contribution rate of a system, comprising:

[0031] The combat system construction module is used to construct the combat loop and network indicator system in the combat system according to the combat mission and various equipment; based on the network indicator system, each combat loop is refined into multiple task chains;

[0032] The indicator weight calculation module is used to collect sample data through combat simulation, calculate the values ​​of each indicator at the last level in the network indicator system based on each sample data, calculate the impact value of the task chain based on each indicator value, and obtain the initial indicator matrix based on the impact value of the task chain and the corresponding indicator values ​​of each sample data. Iterative learning is performed on the initial indicator matrix using principal component analysis and Lasso algorithm to obtain the weights of each indicator at the last level.

[0033] The system contribution rate calculation module is used to calculate the comprehensive value of the combat system capability based on the combat simulation data to be evaluated and the weights of various indicators; remove the equipment to be evaluated from the combat system, and obtain the system contribution rate of the equipment to be evaluated based on the rate of change of the comprehensive value of the combat system capability.

[0034] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0035] 1. A network-like indicator system was established. Unlike the traditional simplified linear processing of indicator systems, the network-like indicator system fully considers the correlation between indicators driven by actual tasks, which is more in line with the actual situation. It replaces the tree-like indicator system construction technology and provides a more realistic and objective input for the evaluation of the contribution rate of equipment system.

[0036] 2. Based on a network-like indicator system, task chain analysis is performed for each indicator. Weights are assigned from both structural and data perspectives, fully considering the relationships between multi-level indicators. Iterative principal component analysis and LASSO learning techniques are combined to analyze and process data with coupling relationships, resulting in objective indicator weights that are closer to real-world applications and provide technical support for understanding the contribution of equipment in missions.

[0037] In this invention, the above-described technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of this invention will be set forth in the following description, and some advantages may become apparent from the description or be learned by practicing the invention. The objects and other advantages of this invention can be realized and obtained from what is particularly pointed out in the description and drawings. Attached Figure Description

[0038] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.

[0039] Figure 1 This is a flowchart of a method for evaluating the contribution rate of a system in Embodiment 1 of the present invention;

[0040] Figure 2 This is a schematic diagram of the system structure for evaluating the contribution rate of a system in Embodiment 2 of the present invention. Detailed Implementation

[0041] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0042] Example 1

[0043] A specific embodiment of the present invention discloses a method for evaluating the contribution rate of a system, such as... Figure 1 As shown, it includes the following steps:

[0044] S11. Based on the combat mission and various equipment, construct the combat ring and network indicator system in the combat system; based on the network indicator system, refine each combat ring into multiple task chains.

[0045] It should be noted that the combat mission is broken down into one or more operational targets to be interfered with or destroyed, and various equipment is selected according to the operational targets to form a combat system. The types of armed equipment include four categories: reconnaissance (S), decision-making (D), attack (A), and target acquisition (T). There can be multiple pieces of each type of armed equipment in the combat system. Each piece of equipment must contain at least one capability, and each capability must correspond to at least one piece of equipment.

[0046] For example, the combat mission is decomposed into striking enemy targets, and the combat system consists of a decision-making command and control system, two reconnaissance-type situational awareness systems, a strike system, and a target-type strike system. The strike system performs precision strikes against the target-type strike system. The command and control system has command and control capabilities, the situational awareness system has intelligence reconnaissance and detection capabilities, the two types of strike systems have strike capabilities, and all four types of equipment have information transmission capabilities.

[0047] Preferably, when a combat mission is classified into multiple combat objectives, a weight is assigned to each combat objective.

[0048] Each type of equipment capability is further refined into three levels of indicators. For example, command and control capability includes control accuracy and control timeliness indicators; intelligence reconnaissance and detection capability includes target reconnaissance distance and target identification probability indicators; strike capability includes target hit probability and target damage probability indicators; and information transmission capability includes link transmission time and communication quality indicators.

[0049] Since various types of equipment have the same equipment capabilities, the three-level indicators do not correspond to a single type of equipment, but form a network of indicator systems. This system includes, from top to bottom: primary combat objectives, secondary equipment capabilities, and tertiary indicators.

[0050] It should be noted that each level in the network indicator system of this embodiment can be further subdivided according to the actual situation, without limiting the specific level.

[0051] A combat loop is a closed loop formed by reconnaissance, decision-making, and attack equipment entities within an equipment system and an enemy target entity to accomplish a specific combat mission. For example, a typical combat loop T→S→D→A→T represents the process where enemy equipment T is detected by friendly reconnaissance equipment S, which transmits the relevant information to decision-making equipment D. Decision-making equipment D processes the information and issues an attack command to attack equipment A, which then attacks enemy equipment T. Since there may be more than one piece of equipment of different types, and there are synergistic or substitutable relationships between the same type of equipment, multiple pieces of equipment of the same type may appear in a combat loop, demonstrating different combat capabilities. For example, a combat loop T1→S1→S2→D2→D1→A1→T1 represents a combat loop that includes information sharing and coordinated command.

[0052] In order to more objectively assign weights to the final-level indicators, this embodiment, after establishing a combat loop based on the combat objective, considers the equipment capabilities of each piece of equipment in the combat loop, as well as the final-level indicators contained in each equipment capability, and refines multiple task chains. Each task chain starts from each final-level indicator, passes through a certain equipment capability of a certain piece of equipment to which it belongs, and then determines the equipment capability of the next piece of equipment based on the role and transmission relationship of the equipment capability of that piece of equipment in the task chain, until any combat objective is reached.

[0053] Compared with existing technologies, this embodiment establishes a network-like indicator system. Unlike the traditional simplified linear processing of indicator systems, the network-like indicator system fully considers the correlation between indicators driven by actual tasks, which is more in line with the actual situation. It replaces the tree-like indicator system construction technology and analyzes the task chain for each indicator, which prepares for subsequent weighting of indicators from both structural and data levels, thereby improving the authenticity and objectivity of the equipment system contribution rate assessment.

[0054] S12. Collect sample data through combat simulation, calculate the values ​​of each indicator at the final level in the network indicator system based on each sample data, calculate the impact value of the task chain based on each indicator value, and obtain the initial indicator matrix based on the impact value of the task chain and the corresponding indicator values ​​of each sample data. Use principal component analysis and Lasso algorithm to iteratively learn the initial indicator matrix to obtain the weights of each indicator at the final level.

[0055] It should be noted that each indicator has a specific calculation formula based on the actual situation. For example, the probability of target destruction = the number of enemy equipment destroyed by our side ÷ the total number of enemy equipment. The system collects performance parameters of each piece of equipment, interactive data from the combat simulation system, and monitoring data during historical combat simulations. The raw data from each simulation is used as a sample data point, and the values ​​of each indicator at the final level in the network indicator system are calculated based on each sample data point.

[0056] Furthermore, the impact value of the task chain is calculated based on the values ​​of various indicators, including:

[0057] Based on the various indicator values ​​included in the capabilities of each piece of equipment in the task chain, the correlation coefficient between two adjacent nodes is calculated as the relationship value of the corresponding edge; the average value of the relationship values ​​of each edge in each task chain is taken as the influence value of the task chain.

[0058] It should be noted that calculating the correlation coefficient between two adjacent nodes in the task chain, as the relationship value of the corresponding edge, includes: using canonical correlation analysis (CCA) to analyze the index items contained in two adjacent nodes in the task chain, and using the obtained canonical correlation coefficient as the relationship value of the corresponding edge in the task chain.

[0059] It should be noted that the first two nodes are the final-level indicator and its corresponding equipment capability. Since equipment capability includes multiple indicators, the final-level indicator may have mutual influence or causal relationship with other indicators, which can also be obtained by calculating the Pearson correlation coefficient.

[0060] The average value of the relationship values ​​of each edge in each task chain is used as the influence value of the task chain.

[0061] Finally, based on the impact value of the task chain and the corresponding indicator values ​​for each sample data point, an initial indicator matrix is ​​obtained, including:

[0062] The importance value of each indicator is obtained by weighting the impact value of each indicator in the task chain to which it belongs with the weight of the combat target in the task chain.

[0063] Based on the importance values ​​of each indicator at the final level, the indicator values ​​corresponding to each sample data are weighted separately to obtain the vector of indicator values ​​corresponding to each sample.

[0064] The vectors of all the indicator values ​​in the sample data constitute the initial indicator matrix.

[0065] It should be noted that since the indicators at the final level can be related to multiple pieces of equipment, and the mission chains generated by different equipment capabilities are different, each indicator at the final level corresponds to one or more mission chains, and the combat objectives to be achieved are different. Therefore, based on the influence value of the mission chain to which each indicator belongs at the final level, the importance value of that indicator is calculated, including: taking a weighted average of the influence value of the mission chain to which each indicator belongs at the final level and the weight of the combat objectives in that mission chain to obtain the importance value of that indicator. The importance values ​​of each indicator at the final level are then weighted on the corresponding indicator values ​​calculated for each sample, and arranged in a preset order to form a vector of indicator values, which serves as the row vector of the initial indicator matrix. Thus, the vectors of indicator values ​​from all sample data are combined to form the initial indicator matrix.

[0066] For example, with m sample data and n indicators at the final level of the network indicator system, an initial indicator matrix of m×n dimensions is obtained.

[0067] Compared with existing technologies, this embodiment is based on a network-like index system. It analyzes the task chain for each index, assigns weights to each index from the network structure, and fully considers the relationship between the index and equipment capabilities, as well as the relationship between equipment capabilities. Next, from the data level, it uses principal component analysis and Lasso algorithm to iteratively learn the initial index matrix to obtain the weights of each index at the final level.

[0068] Specifically, principal component analysis is used to obtain a dictionary index matrix after dimensionality reduction of the initial index matrix. Based on the initial index matrix and the dictionary index matrix, the Lasso algorithm is used to obtain the regression coefficient matrix, which is then normalized and used as the index weight matrix. The average weight of each index in the index weight matrix is ​​calculated. When the standard deviation of the average weight is less than 1, the initial index matrix is ​​updated according to the average weight. Principal component analysis and the Lasso algorithm are used again to obtain a new average weight. When the standard deviation of the average weight is greater than 1, the iteration stops. The final average weight is the index weight.

[0069] First, using principal component analysis, we obtain the dictionary index matrix after dimensionality reduction of the initial index matrix, including:

[0070] Based on the preset contribution rate, the principal component matrix after dimensionality reduction by principal component analysis of the initial index matrix is ​​obtained; the dictionary index matrix is ​​obtained by multiplying the initial index matrix with the transpose of the principal component matrix.

[0071] For example, if the preset contribution rate is 90%, then based on the preset contribution rate, the initial index matrix of m×n dimensions is reduced to p dimensions to obtain a principal component matrix of p×n dimensions; the initial index matrix of m×n dimensions is multiplied by the transpose of the principal component matrix of n×p dimensions to obtain an m×p-dimensional dictionary index matrix.

[0072] Furthermore, based on the initial index matrix and the dictionary index matrix, the regression coefficients are obtained using the Lasso algorithm by solving the following objective function:

[0073]

[0074] Where S represents the initial index matrix, Z represents the dictionary index matrix, β represents the regression coefficient matrix, λ represents a constant, taking a decimal number between [0,1], and ||·||1 and ||·||2 represent the 1-norm and 2-norm, respectively.

[0075] Specifically, based on the initial index matrix of m×n dimensions and the dictionary index matrix of m×p dimensions, the regression coefficients of p×n dimensions are obtained. After normalization, they are used as the index weight matrix. The average value of each column is taken to obtain the average weight of n indicators.

[0076] Next, the standard deviation of the n average weights is calculated. If the standard deviation is greater than 1, the learning stops and the n average weights are the indicator weights of the n indicators. Otherwise, the corresponding indicator values ​​in the initial indicator matrix are weighted according to the average weights, the initial indicator matrix is ​​updated, and principal component analysis and Lasso algorithm are used again to obtain new average weights. If the standard deviation of the average weights is greater than 1, the iteration stops and the final average weights are the indicator weights.

[0077] S13. Calculate the comprehensive value of the combat system capability based on the combat simulation data to be evaluated and the weights of each indicator; remove the equipment to be evaluated from the combat system, and obtain the system contribution rate of the equipment to be evaluated based on the rate of change of the comprehensive value of the combat system capability.

[0078] It should be noted that the comprehensive value of the combat system capability is calculated based on the combat simulation data to be evaluated and the weights of various indicators, including:

[0079] ①Based on the values ​​and weights of the various indicators included in the capabilities of each piece of equipment, the capability values ​​of each piece of equipment are calculated using the indicator aggregation method.

[0080] Preferably, a minimum threshold for weights is set. The weights of each indicator at the final level are compared with the minimum threshold. If the weight is less than the minimum threshold, it means that the indicator has little effect on the overall equipment capability. In this case, the indicator is no longer selected, and only indicators with weights greater than the minimum threshold are retained.

[0081] This embodiment does not limit the method of index aggregation. Based on the actual index strength of each equipment capability, the weighted average method, weighted product method, etc. can be selected.

[0082] ② For each established combat loop, multiply the equipment capability values ​​of reconnaissance, decision-making, and attack equipment together and divide by the equipment capability value of target equipment to obtain the capability value of the combat loop.

[0083] It should be noted that there is a sequential relationship between reconnaissance, decision-making, and attack equipment, so the product method is used to represent their comprehensive capabilities. Target-oriented equipment has a negative impact on combat capabilities, so its equipment capability value is used as the denominator.

[0084] ③ Add up the capability values ​​of each operational link corresponding to the same operational objective to obtain the influence value of the operational link on the operational objective; according to the weight of each operational objective, sum up the corresponding influence values ​​to obtain the comprehensive capability value of the operational system.

[0085] An intuitive understanding of the contribution of equipment to system capability is that if the removal or destruction of a piece of equipment from the system significantly reduces the remaining system's combat capability against enemy targets, then the equipment contributes a high level to the combat system capability. Conversely, if the removal of the equipment does not fundamentally change the combat capability of the old and new systems against enemy targets, then the addition of the equipment has not played a positive role in increasing the system capability.

[0086] The contribution rate of equipment system capability can be considered as the contribution rate of a certain piece of equipment to the system capability, or it can be considered as the contribution rate of a certain type of equipment to the system capability.

[0087] Taking the removal of a piece of equipment as an example, the ratio of the difference between the overall combat system capability assessment value with and without a certain piece of equipment to the original overall combat system capability assessment value is used as the system capability contribution rate of that piece of equipment. That is, the system contribution rate of a certain piece of equipment is expressed by the rate of change of the overall combat system capability value; the formula is as follows:

[0088]

[0089] Where G represents the combat system network, G -v This indicates that removing the equipment v to be evaluated from G will form a new combat system network, h G This represents the overall capability value of the original combat system. This represents the comprehensive value of the new combat system capability, which is calculated by performing steps S11-S13 on the new combat system network.

[0090] Compared with existing technologies, this embodiment provides a system contribution rate evaluation method that establishes a network-like indicator system. Unlike traditional simplified linear processing of indicator systems, the network-like indicator system fully considers the correlations between indicators driven by actual tasks, making it more realistic and replacing tree-like indicator system construction techniques. This provides a more realistic and objective input for evaluating the equipment system contribution rate. Based on the network-like indicator system, task chain analysis is performed on each indicator, assigning weights from both structural and data perspectives. It fully considers the relationships between multi-level indicators and combines iterative principal component analysis and LASSO learning techniques to analyze and process data with coupling relationships, deriving objective indicator weights that are closer to real-world applications and provide technical support for understanding the degree of equipment contribution in missions.

[0091] Example 2

[0092] Another embodiment of the present invention discloses a system for evaluating the contribution rate of a system, thereby implementing the system contribution rate evaluation method in Embodiment 1. The specific implementation of each module is described in the corresponding description in Embodiment 1. Figure 2 As shown, the system includes:

[0093] The combat system construction module 101 is used to construct the combat rings and network indicator system in the combat system according to the combat mission and various equipment; based on the network indicator system, each combat ring is refined into multiple task chains;

[0094] The indicator weight calculation module 102 is used to collect sample data through combat simulation, calculate the values ​​of each indicator at the last level in the network indicator system based on each sample data, calculate the impact value of the task chain based on each indicator value, obtain the initial indicator matrix based on the impact value of the task chain and the corresponding indicator values ​​of each sample data, and use principal component analysis and Lasso algorithm to iteratively learn the initial indicator matrix to obtain the weights of each indicator at the last level.

[0095] The system contribution rate calculation module 103 is used to calculate the comprehensive value of the combat system capability based on the combat simulation data to be evaluated and the weights of various indicators; remove the equipment to be evaluated from the combat system, and obtain the system contribution rate of the equipment to be evaluated based on the rate of change of the comprehensive value of the combat system capability.

[0096] Since this embodiment and the aforementioned method for evaluating the contribution rate of a system can be mutually referenced, and this is a repetition, they will not be repeated here. Because this system embodiment shares the same principle as the aforementioned method embodiment, it also possesses the corresponding technical effects of the aforementioned method embodiment.

[0097] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0098] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for evaluating the contribution rate of a system, characterized in that, Includes the following steps: Based on the combat missions and various equipment, construct a combat ring and network indicator system within the combat system; Break down combat missions into one or more combat objectives; Each piece of equipment contains at least one equipment ability, and each equipment ability corresponds to at least one piece of equipment. The network indicator system includes: primary combat objectives, secondary equipment capabilities, and tertiary indicators. Based on the network indicator system, each combat link is refined into multiple task chains, which is to construct a task chain from the final level, where each indicator passes through each equipment capability to complete each combat objective. By collecting sample data through combat simulation, the values ​​of each indicator at the final level in the network indicator system are calculated based on each sample data. The impact value of the mission chain is then calculated based on each indicator value, including: calculating the correlation coefficient between two adjacent nodes based on the indicator values ​​included in the capabilities of each piece of equipment in the mission chain, which is used as the relationship value of the corresponding edge; and using the average value of the relationship values ​​of each edge in each mission chain as the impact value of the mission chain. Based on the impact value of the task chain and the corresponding indicator values ​​of each sample data, an initial indicator matrix is ​​obtained, including: taking the weighted average of the impact value of each indicator at the final level and the weight of the combat target in the task chain to obtain the importance value of the indicator; weighting the indicator values ​​of each sample data according to the importance values ​​of each indicator at the final level to obtain the indicator value vector of each sample; the indicator value vectors of all sample data constitute the initial indicator matrix. Principal component analysis (PCA) and the Lasso algorithm are used to iteratively learn the initial index matrix to obtain the final-level index weights. This process includes: using PCA to obtain a dimensionality-reduced dictionary index matrix from the initial index matrix; using the Lasso algorithm to obtain the regression coefficient matrix based on the initial index matrix and the dictionary index matrix, and then normalizing it to obtain the index weight matrix; calculating the average weight of each index in the index weight matrix; updating the initial index matrix based on the average weight when the standard deviation of the average weight is less than 1; and then using PCA and the Lasso algorithm again to obtain the new average weight. When the standard deviation of the average weight is greater than 1, the iteration stops, and the final average weight is the index weight. Based on the combat simulation data to be evaluated and the weights of various indicators, the comprehensive value of the combat system capability is calculated; the equipment to be evaluated is removed from the combat system, and the system contribution rate of the equipment to be evaluated is obtained based on the rate of change of the comprehensive value of the combat system capability.

2. The method for evaluating the system contribution rate according to claim 1, characterized in that, The regression coefficients, obtained using the Lasso algorithm based on the initial index matrix and the dictionary index matrix, are derived by solving the following objective function: in, S Represents the initial index matrix. Z Represents the dictionary index matrix, β Represents the regression coefficient matrix. λ Represents a constant, taking the decimal value between [0,1]. and These represent the 1-norm and the 2-norm, respectively.

3. The method for evaluating the system contribution rate according to claim 1, characterized in that, The comprehensive value of the combat system capability is calculated based on the combat simulation data to be evaluated and the weights of various indicators, including: Based on the values ​​and weights of the various indicators included in each equipment capability, the capability value of each equipment is calculated using the indicator aggregation method. For each established combat loop, the capability value of the combat loop is obtained by multiplying the capability values ​​of the reconnaissance, decision-making, and attack equipment together and then dividing by the capability value of the target equipment. The capability values ​​of each operational link corresponding to the same operational objective are added together to obtain the influence value of the operational link on the operational objective; according to the weight of each operational objective, the corresponding influence values ​​are weighted and summed to obtain the comprehensive capability value of the operational system.

4. The method for evaluating the system contribution rate according to claim 3, characterized in that, The removal of the equipment to be evaluated from the combat system is described by the following formula, which calculates the system contribution rate of the equipment to be evaluated based on the rate of change of the overall capability value of the combat system: in, G Indicates the combat system network. Indicates that the equipment to be evaluated v from G After being removed from the network, it forms a new combat system network. This represents the overall capability value of the original combat system. This represents the overall capability value of the new combat system.

5. A system for evaluating the contribution rate of a system, used to implement the method for evaluating the contribution rate of the system according to any one of claims 1-4, characterized in that, include: The combat system construction module is used to construct the combat loop and network indicator system in the combat system based on combat missions and various equipment. Based on the network indicator system, each combat link is broken down into multiple task chains; The indicator weight calculation module is used to collect sample data through combat simulation, calculate the values ​​of each indicator at the last level in the network indicator system based on each sample data, calculate the impact value of the task chain based on each indicator value, and obtain the initial indicator matrix based on the impact value of the task chain and the corresponding indicator values ​​of each sample data. Iterative learning is performed on the initial indicator matrix using principal component analysis and Lasso algorithm to obtain the weights of each indicator at the last level. The system contribution rate calculation module is used to calculate the comprehensive value of the combat system capability based on the combat simulation data to be evaluated and the weights of various indicators. Remove the equipment to be evaluated from the combat system, and obtain the system contribution rate of the equipment to be evaluated based on the rate of change of the overall capability value of the combat system.

Images