Simulation-deduction-oriented combat plan evaluation method under complex confrontation condition
By constructing positive and negative ideal solutions and using Wasserstein distance measurements, the problem that the existing technology cannot effectively evaluate combat solutions under complex confrontation conditions is solved, and higher evaluation credibility and efficiency are achieved.
Patent Information
- Application Number
- CN202411381029.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-30
- Publication Date
- 2025-02-14
AI Technical Summary
When existing evaluation technologies deal with large-sample simulation deduction data under complex confrontation conditions, they cannot fully utilize the information in the multimodal distribution, resulting in the inability to effectively evaluate the advantages and disadvantages of the combat plan.
By setting the solution performance distribution density function, a positive and negative ideal solution is constructed, and the distance between all alternatives and positive and negative ideal solutions is measured using the Wasserstein distance, thereby performing the preferred sorting of combat solutions.
This method can make more comprehensive use of complex adversarial information in simulation deduction data, improve the credibility of combat scheme evaluation, and avoid complex operations related to continuous distribution.
Smart Images
Figure CN119442589A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of military operations technology, and in particular relates to a combat plan evaluation method under complex confrontation conditions oriented to simulation deduction. Background Art
[0002] An operational plan refers to a specific operational plan and action plan formulated according to strategic intent and operational tasks. It usually includes enemy situation assessment, task allocation, force deployment, action coordination and various support measures, and is the basic basis for combat preparation and implementation of combat operations. For specific combat tasks, the staff department will generally formulate multiple sets of alternative operational plans based on the correct understanding of the superior's intentions and the actual situation on the battlefield. It will use sand table simulations, live-fire exercises or combat simulations as appropriate to comprehensively evaluate the alternative plans in terms of damage, survival, cost-effectiveness, etc., and give a ranking of advantages and disadvantages for commanders' decision-making reference. In addition, the evaluation of operational plans is also of great significance to issues such as operational design, force construction, and equipment development, and all of this must be based on the scientific and credible evaluation conclusions.
[0003] Generally speaking, the credibility of an assessment is highly dependent on low-cost, high-quality data sources, which is precisely the advantage of combat simulation. As today's war style has long entered the information and even intelligent era, the combat tempo is fast and the intensity is high. New technologies, new equipment, and new tactics are constantly emerging. All kinds of true and false information, real and fake actions, special situations and dangerous situations emerge in an endless stream, resulting in a highly dynamic and extremely complex confrontation environment. Combat simulation can restore complex battlefield environments with high fidelity and efficiently collect large sample deduction data, providing a solid foundation for the credible evaluation of combat plans.
[0004] However, existing evaluation technologies, such as AHP and TOPSIS, are no longer suitable for current combat needs and the development of simulation technology, because they all process large sample simulation data by simply reducing them to a single statistic, such as the mean, median, and interval number. However, given the diversity of confrontation strategies and the unpredictability of the threat environment, the simulation samples generally present a multimodal distribution. Therefore, existing evaluation technologies will inevitably ignore the rich information contained in complex confrontation conditions and fail to give full play to the advantages of combat simulation. Summary of the invention
[0005] In order to solve the above problems, the present invention provides a combat plan evaluation method under complex confrontation conditions for simulation and deduction. In view of the limitations of existing evaluation technologies in processing large sample simulation and deduction data under complex confrontation conditions, by setting the scheme effectiveness distribution density function, constructing positive (negative) ideal schemes that are theoretically better (worse) than all alternative combat plans, and using Wasserstein distance to measure the distance between all alternative plans and positive and negative ideal schemes in terms of probability distribution, the combat plans are optimally ranked accordingly.
[0006] The first aspect of the present invention provides a method for evaluating combat plans under complex confrontation conditions for simulation and deduction, the method comprising:
[0007] S1. Generate a sample matrix for simulation and deduction of combat plan effectiveness;
[0008] S2. Construct the effectiveness distribution function of the combat plan based on the effectiveness simulation and deduction sample matrix;
[0009] S3, constructing a positive ideal solution and calculating the effectiveness distribution function of the positive ideal solution;
[0010] S4, constructing a negative ideal solution and calculating the effectiveness distribution function of the negative ideal solution;
[0011] S5. Measure the distance between the operational plan and the positive ideal plan and the negative ideal plan respectively;
[0012] S6. Based on the distance between the operational plan and the positive ideal plan and the negative ideal plan, the evaluation value of the operational plan is calculated, and the operational plans are prioritized.
[0013] In S1:
[0014] Generate N sets of alternative combat plans S1, S2, ..., S according to combat missions and combat objectives. N ;
[0015] Select a suitable combat simulation system according to the usage requirements, set combat scenarios and complex confrontation strategies in it, and independently deduce each combat plan M times;
[0016] For all n∈[1,2,…,N] and m∈[1,2,…,M], select solution S n The mth deduction data is used to calculate the single sample performance evaluation value of the scheme using the evaluation method n∈[1,2,…,N], m∈[1,2,…,M], and the evaluation method includes weighted sum, weighted product, and AHP method;
[0017] Form an N×M-dimensional combat plan effectiveness simulation and deduction sample matrix:
[0018]
[0019] In S2:
[0020] For any solution S n , with random variable F n Represents its effectiveness, and the corresponding effectiveness distribution function is set as:
[0021]
[0022] in, # is the deduplication operator, <x * > n For x * In X n The number of occurrences in , δ(·) is the Dirac delta function.
[0023] In S3:
[0024] For the positive ideal solution S + , random variable F + To express its effectiveness, then:
[0025] F + =max(F 1 ,F 2 ,…,F N )
[0026] F + The efficiency distribution function is:
[0027]
[0028] in, α x* is the coefficient;
[0029] For any F n ,have:
[0030]
[0031] Among them, the function Ι(·,·) is defined as:
[0032]
[0033] In S4:
[0034] For the negative ideal solution S - , random variable F - To express its effectiveness, then:
[0035] F - =min(F 1 ,F 2 ,…,F N )
[0036] F - The efficiency distribution function is:
[0037]
[0038] in, is the coefficient.
[0039] In S5:
[0040] Measuring any solution Sn With the positive ideal solution S + The distance between the random variables F n With F + Distance in terms of probability distribution:
[0041]
[0042]
[0043]
[0044]
[0045] Use linear programming algorithm to find the minimum value D n , as Scheme S n With the positive ideal solution S + distance;
[0046] Use the same method to measure any solution S n With negative ideal solution S - Distance:
[0047]
[0048]
[0049]
[0050]
[0051] Use linear programming algorithm to find the minimum value C n , as Scheme S n With negative ideal solution S - distance.
[0052] In S6:
[0053] For all solutions S n Calculate the evaluation value V n :
[0054]
[0055] The higher the evaluation value, the better the solution S n The better the effectiveness is, the alternative combat plans will be arranged in descending order according to the evaluation value, which is the result of the optimal sorting of the plans.
[0056] A second aspect of the present invention discloses a combat plan evaluation system under complex confrontation conditions for simulation and deduction, the system comprising a processing unit, and the processing unit is configured to perform the following steps:
[0057] S1. Generate a sample matrix for simulation and deduction of combat plan effectiveness;
[0058] S2. Construct the effectiveness distribution function of the combat plan based on the effectiveness simulation and deduction sample matrix;
[0059] S3, constructing a positive ideal solution and calculating the effectiveness distribution function of the positive ideal solution;
[0060] S4, constructing a negative ideal solution and calculating the effectiveness distribution function of the negative ideal solution;
[0061] S5. Measure the distance between the operational plan and the positive ideal plan and the negative ideal plan respectively;
[0062] S6. Based on the distance between the operational plan and the positive ideal plan and the negative ideal plan, the evaluation value of the operational plan is calculated, and the operational plans are prioritized.
[0063] The third aspect of the present invention discloses an electronic device. The electronic device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the method for evaluating combat plans under complex confrontation conditions for simulation and deduction described in the first aspect of the present disclosure are implemented.
[0064] The fourth aspect of the present invention discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for evaluating combat plans under complex confrontation conditions for simulation and deduction described in the first aspect of the present disclosure are implemented.
[0065] In summary, the beneficial effects of the present invention are: first, the present invention can make full use of the complex adversarial information contained in the multi-modal distribution of the deduction samples, avoid the limitation of using a single statistic, and help improve the credibility of the program evaluation; second, the effectiveness distribution of the alternative combat plans and the positive and negative ideal plans in the present invention is essentially a discrete distribution, which can avoid the complex calculations related to the continuous distribution. BRIEF DESCRIPTION OF THE DRAWINGS
[0066] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0067] Figure 1 It is a schematic diagram of the implementation steps of the method of the present invention.
[0068] Figure 2Schematic diagram of the effectiveness distribution of alternative combat plans in an embodiment of the present invention.
[0069] Figure 3 Schematic diagram of the performance distribution of an ideal solution in an embodiment of the present invention.
[0070] Figure 4 FIG. 4 is a schematic diagram of the performance distribution of the negative ideal solution in an embodiment of the present invention. DETAILED DESCRIPTION
[0071] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0072] The first aspect of the present invention provides a method for evaluating combat plans under complex confrontation conditions for simulation and deduction, the method comprising:
[0073] S1. Generate a sample matrix for simulation and deduction of combat plan effectiveness;
[0074] S2. Construct the effectiveness distribution function of the combat plan based on the effectiveness simulation and deduction sample matrix;
[0075] S3, constructing a positive ideal solution and calculating the effectiveness distribution function of the positive ideal solution;
[0076] S4, constructing a negative ideal solution and calculating the effectiveness distribution function of the negative ideal solution;
[0077] S5. Measure the distance between the operational plan and the positive ideal plan and the negative ideal plan respectively;
[0078] S6. Based on the distance between the operational plan and the positive ideal plan and the negative ideal plan, the evaluation value of the operational plan is calculated, and the operational plans are prioritized.
[0079] The following will be combined Figure 1 The combat plan evaluation method under complex confrontation conditions for simulation and deduction proposed in the present invention is described in detail.
[0080] (1) Samples for performance evaluation of the collection scheme
[0081] First, according to the specific combat mission and combat purpose, N sets of alternative combat plans are formulated: S1, S2, ..., S N. According to the usage requirements, select a suitable combat simulation system, set up highly realistic combat scenarios and complex confrontation strategies in it, and independently deduce each combat plan M times, and it is recommended to take M ≥ 50. Then, for all n∈[1,2,…,N] and m∈[1,2,…,M], select plan S n The mth deduction data is used to calculate the single sample performance evaluation value of the scheme using existing classical evaluation methods (including but not limited to weighted sum, weighted product, AHP, etc.) Finally, an N×M-dimensional combat plan effectiveness simulation and deduction sample matrix is formed:
[0082]
[0083] In the embodiment, with a typical joint combat scenario as the background, the Red side command organization, by understanding the intention of the superiors, set combat tasks and combat determination, and formulated three sets of alternative combat plans: S1, S2, and S3. At the same time, the Blue side also formulated three response plans H1, H2, and H3 by analyzing the possible combat actions of the Red side. Afterwards, based on a certain type of combat simulation system, the highly realistic combat scene construction and the deployment of combat forces of both sides were completed, and the number of scenario simulations M=100 was set. In order to simulate complex confrontation conditions, the Blue side adopts a mixed strategy, that is, for any plan S n In each independent deduction, the blue team randomly selects the response plans H1, H2, and H3 with probabilities of 0.2, 0.5, and 0.3. Based on the single red-blue confrontation simulation deduction data and the evaluation indicators in Table 1, the weighted sum method is used to calculate the single sample effectiveness evaluation value, and the evaluation value is normalized in the interval [0, 100]. Finally, a 3×100-dimensional solution effectiveness simulation deduction sample matrix is formed.
[0084] Table 1
[0085]
[0086] (2) Constructing the solution effectiveness distribution function
[0087] For any solution S n , let the random variable F n Represents its effectiveness, and the corresponding effectiveness distribution function is set as:
[0088]
[0089] in, # is the deduplication operator, which is used to remove duplicate values in the data, such as: #{1,2,2,3,3,3,4}={1,2,3,4}. <x * > n For x * In X nThe number of occurrences in , δ(·) is the Dirac delta function, defined as follows:
[0090]
[0091] And satisfy:
[0092]
[0093] According to the above properties of the Dirac delta function, f n (x) is a continuous distribution in form, but it is a discrete distribution in essence and can only take #X n Discrete values in .
[0094] In the embodiment, based on the simulation deduction sample matrix, the efficiency distribution function of three sets of alternative solutions is obtained using formula (1). The distribution graph is shown in Figure 2 Obviously, the distribution graphs of all solutions are multimodal, which is caused by the mixed strategy adopted by the blue side. It reveals the inconsistency of the effectiveness of any red side's solution when randomly responding to different blue side solutions, so it is not suitable to be represented by a single statistic.
[0095] (3) Constructing a positive ideal solution
[0096] Assume that there is a way to integrate the advantages of all the alternative combat plans to form a theoretically optimal plan, that is, the positive ideal plan S + , then let the random variable F + Indicates its effectiveness, according to S + The optimality of is defined as:
[0097] F + =max(F 1 ,F 2 ,…,F N )
[0098] Based on this, we can deduce that F + The efficiency distribution function is:
[0099]
[0100] in, coefficient It can be calculated by combining formula (1). At the same time, for any F n ,have:
[0101]
[0102] Among them, the function Ι(·,·) is defined as:
[0103]
[0104] In addition, f + Although (x) is a continuous distribution in form, it is a discrete distribution in essence and can only take discrete values in X.
[0105] In the embodiment, the image of the positive ideal solution performance distribution function is shown in Figure 3 , the distribution obviously tends to the high energy value area, showing optimality in the theoretical sense.
[0106] (4) Constructing a negative ideal solution
[0107] Assume that there is a way to integrate the disadvantageous elements of all alternative combat plans to form a theoretically worst plan, that is, the negative ideal plan S - , then let the random variable F - Indicates its effectiveness, according to S - The worst property of is defined as:
[0108] F - =min(F 1 ,F 2 ,…,F N )
[0109] Based on this, we can deduce that F - The efficiency distribution function is:
[0110]
[0111] Among them, the coefficient It can be calculated by combining formula (1). - Although (x) is a continuous distribution in form, it is a discrete distribution in essence and can only take discrete values in X.
[0112] In the embodiment, the image of the negative ideal solution performance distribution function is shown in Figure 4 , the distribution is obviously biased towards the low efficiency value area, showing the worst performance in the theoretical sense.
[0113] (5) Measuring the distance between alternative solutions and positive and negative ideal solutions
[0114] First, measure any solution S n With the positive ideal solution S + The distance between the random variables F n With F + Distance in the sense of probability distribution. Here, Wasserstein distance is used, which is equivalent to solving the following minimization problem:
[0115]
[0116]
[0117]
[0118]
[0119] Using the linear programming algorithm (existing technology), the minimum value is D n , which is Scheme S n With the positive ideal solution S + distance.
[0120] Secondly, using the same method, measure any solution S n With negative ideal solution S - The distance is to solve the following minimization problem:
[0121]
[0122]
[0123]
[0124]
[0125] Using the linear programming algorithm, the minimum value is C n , which is Scheme S n With negative ideal solution S - distance.
[0126] In the embodiment, the built-in linprog function in the scipy library of the python language is used to solve the linear programming, and the solver adopts the HiGHS simplex method.
[0127] (6) Scheme optimization ranking
[0128] For all solutions S n , calculate the evaluation value V n :
[0129]
[0130] The higher the evaluation value, the better the effectiveness of the plan. Therefore, the alternative combat plans are arranged in descending order according to the evaluation value, which is the result of the optimal sorting of the plans.
[0131] In the embodiment, the evaluation values of the three alternative solutions are calculated as follows:
[0132] V1=0.541V2=0.716V3=0.254
[0133] Therefore, the schemes are ranked as follows:
[0134] S2>S1>S3
[0135] The symbol > indicates better than.
[0136] A second aspect of the present invention discloses a combat plan evaluation system under complex confrontation conditions for simulation and deduction, the system comprising a processing unit, and the processing unit is configured to perform the following steps:
[0137] S1. Generate a sample matrix for simulation and deduction of combat plan effectiveness;
[0138] S2. Construct the effectiveness distribution function of the combat plan based on the effectiveness simulation and deduction sample matrix;
[0139] S3, constructing a positive ideal solution and calculating the effectiveness distribution function of the positive ideal solution;
[0140] S4, constructing a negative ideal solution and calculating the effectiveness distribution function of the negative ideal solution;
[0141] S5. Measure the distance between the operational plan and the positive ideal plan and the negative ideal plan respectively;
[0142] S6. Based on the distance between the operational plan and the positive ideal plan and the negative ideal plan, the evaluation value of the operational plan is calculated, and the operational plans are prioritized.
[0143] The third aspect of the present invention discloses an electronic device. The electronic device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the method for evaluating combat plans under complex confrontation conditions for simulation and deduction described in the first aspect of the present disclosure are implemented.
[0144] The fourth aspect of the present invention discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for evaluating combat plans under complex confrontation conditions for simulation and deduction described in the first aspect of the present disclosure are implemented.
[0145] In summary, the beneficial effects of the present invention are: first, the present invention can make full use of the complex adversarial information contained in the multi-modal distribution of the deduction samples, avoid the limitation of using a single statistic, and help improve the credibility of the program evaluation; second, the effectiveness distribution of the alternative combat plans and the positive and negative ideal plans in the present invention is essentially a discrete distribution, which can avoid the complex calculations related to the continuous distribution.
[0146] Please note that the technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification. The above-mentioned embodiments only express several implementation methods of the present application, and their descriptions are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that for ordinary technicians in this field, without departing from the concept of the present application, several variations and improvements can be made, which all belong to the scope of protection of the present application. Therefore, the scope of protection of the patent in this application shall be based on the attached claims.
Claims
1. A combat plan evaluation method under complex confrontation conditions for simulation and deduction, characterized in that: The method comprises: S1. Generate a sample matrix for simulation and deduction of combat plan effectiveness; S2. Construct the effectiveness distribution function of the combat plan based on the effectiveness simulation and deduction sample matrix; S3, constructing a positive ideal solution and calculating the effectiveness distribution function of the positive ideal solution; S4, constructing a negative ideal solution and calculating the effectiveness distribution function of the negative ideal solution; S5. Measure the distance between the operational plan and the positive ideal plan and the negative ideal plan respectively; S6. Based on the distance between the operational plan and the positive ideal plan and the negative ideal plan, the evaluation value of the operational plan is calculated, and the operational plans are prioritized.
2. According to claim 1, a combat plan evaluation method under complex confrontation conditions for simulation and deduction is characterized in that: In S1: Generate N sets of alternative combat plans S1, S2, ..., S according to combat missions and combat objectives. N ; Select a suitable combat simulation system according to the usage requirements, set combat scenarios and complex confrontation strategies in it, and independently deduce each combat plan M times; For all n∈[1,2,…,N] and m∈[1,2,…,M], select solution S n The mth deduction data is used to calculate the single sample performance evaluation value of the scheme using the evaluation method n∈[1,2,…,N], m∈[1,2,…,M], and the evaluation method includes weighted sum, weighted product, and AHP method; Form an N×M-dimensional combat plan effectiveness simulation and deduction sample matrix:
3. The method for evaluating combat plans under complex confrontation conditions for simulation and deduction according to claim 2 is characterized in that: In S2: For any solution S n , with random variable F n Represents its effectiveness, and the corresponding effectiveness distribution function is set as: in, # is the deduplication operator, <x * > n For x * In X n The number of occurrences in , δ(·) is the Dirac delta function.
4. The method for evaluating combat plans under complex confrontation conditions for simulation and deduction according to claim 3 is characterized in that: In S3: For the positive ideal solution S + , random variable F + To express its effectiveness, then: F + =max(F 1 ,F 2 ,…,F N ) F + The efficiency distribution function is: in, is the coefficient; For any F n ,have: The function I(·,·) is defined as:
5. The method for evaluating combat plans under complex confrontation conditions for simulation and deduction according to claim 4 is characterized in that: In S4: For the negative ideal solution S - , random variable F - To express its effectiveness, then: F - =min(F 1 ,F 2 ,…,F N ) F - The efficiency distribution function is: in, is the coefficient.
6. The method for evaluating combat plans under complex confrontation conditions for simulation and deduction according to claim 5 is characterized in that: In S5: Measuring any solution S n With the positive ideal solution S + The distance between the random variables F n With F + Distance in terms of probability distribution: Use linear programming algorithm to find the minimum value D n , as Scheme S n With the positive ideal solution S + distance; Use the same method to measure any solution S n With negative ideal solution S - Distance: Use linear programming algorithm to find the minimum value C n , as Scheme S n With negative ideal solution S - distance.
7. The method for evaluating combat plans under complex confrontation conditions for simulation and deduction according to claim 6 is characterized in that: In S6: For all solutions S n Calculate the evaluation value V n : The higher the evaluation value, the better the solution S n The better the effectiveness is, the alternative combat plans will be arranged in descending order according to the evaluation value, which is the result of the optimal sorting of the plans.
8. A combat plan evaluation system for complex confrontation conditions oriented to simulation and deduction, characterized in that: The system comprises a processing unit configured to perform the following steps: S1. Generate a sample matrix for simulation and deduction of combat plan effectiveness; S2. Construct the effectiveness distribution function of the combat plan based on the effectiveness simulation and deduction sample matrix; S3, constructing a positive ideal solution and calculating the effectiveness distribution function of the positive ideal solution; S4, constructing a negative ideal solution and calculating the effectiveness distribution function of the negative ideal solution; S5. Measure the distance between the operational plan and the positive ideal plan and the negative ideal plan respectively; S6. Based on the distance between the operational plan and the positive ideal plan and the negative ideal plan, the evaluation value of the operational plan is calculated, and the operational plans are prioritized.
9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the method for evaluating combat plans under complex confrontation conditions for simulation and deduction as described in any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the method for evaluating combat plans under complex confrontation conditions for simulation and deduction as described in any one of claims 1 to 7 are implemented.