Time series based dynamic air combat decision evaluation method

Abstract

The application discloses a kind of dynamic air combat decision evaluation methods based on time series, belong to decision evaluation field, first, the evaluation index system of dynamic air combat evaluation system is built, according to the air combat information of last round and flight dynamics, the air combat information corresponding to each maneuvering strategy in current round is obtained, and the evaluation index value in each strategy to be evaluated in current round is obtained by evaluation index function.Second, the subjective weight value of evaluation index is determined using improved analytic hierarchy process, the objective weight value is determined based on time series multi-attribute decision, and the comprehensive weight value is obtained by game theory.Then, the evaluation score of the decision to be evaluated is obtained according to the comprehensive weight value to activate the cloud model.One side selects the maneuvering strategy with the minimum weighted deviation degree as the optimal maneuvering strategy to execute, completes the decision effectiveness evaluation of current round, and repeats the above steps until the number of dynamic air combat rounds equals the set total number of air combat rounds.Therefore, each index value is accurately calculated, and the error caused by uncontrollable factors is avoided.

Description

Technical Field

[0001] This invention relates to the field of decision evaluation technology, and in particular to a dynamic air combat decision evaluation method based on time series. Background Technology

[0002] Decision-making technology is playing an increasingly important role in modern air combat. Because decision evaluation can provide a reference for subsequent decisions, it has gradually become a focus of attention for many scholars. The results of decision evaluation can be used to judge the quality of decisions and effectively avoid losses caused by decision-making errors. However, current research on decision evaluation mainly applies to single-round combat scenarios. Due to the multi-round nature of air combat, it is imperative to explore a new dynamic decision effectiveness evaluation model to solve the multi-decision evaluation problem in multi-round air combat.

[0003] In recent years, commonly used evaluation methods include multi-attribute decision-making, Bayesian networks, neural networks, genetic algorithms, grey relational analysis, rough sets, and combinations of multiple methods. These methods effectively solve the problems of optimizing evaluation models and allocating evaluation index weights, but they are only applicable to single-round adversarial decision-making problems, known as static decision performance evaluation. In actual air combat, there are often multiple rounds of engagement, with different situations for both sides in each round. This makes the decision-making and evaluation index data dynamic, and the evaluation of decision performance is also dynamic. Because the UAVs of both sides in air combat have complex nonlinear maneuvering characteristics, the situation and decisions of the current round are influenced by the situation and decisions of the previous round, which brings great challenges and difficulties to dynamic decision evaluation. Therefore, research results in this area are still scarce.

[0004] Multi-attribute decision making (MADM) refers to the decision problem of selecting the optimal solution or ranking solutions while considering multiple attributes. This method is only applicable to obtaining the objective weight values ​​of multiple decisions in a single round of air combat and does not meet the requirements of actual air combat. Therefore, it is necessary to study a new dynamic air combat decision evaluation method to handle the multi-decision evaluation problem in multi-round air combat. The difficulty of this multi-decision evaluation problem in multi-round air combat lies in the fact that multiple decisions must be evaluated simultaneously in the current round, and the evaluation in the current round is also affected by the situation and decisions of the previous round throughout the dynamic air combat. Summary of the Invention

[0005] This invention provides a dynamic air combat decision evaluation method based on time series to solve the problems of inaccurate calculation of evaluation indicators and the influence of uncontrollable factors on the calculation results.

[0006] This invention provides a time-series-based dynamic air combat decision evaluation method, comprising the following steps: constructing an evaluation index system for a dynamic air combat evaluation system containing multiple evaluation indicators; determining the evaluation index function values ​​in each evaluation strategy through multiple evaluation index functions; determining the subjective weight values ​​of multiple maneuver strategies to be evaluated in dynamic air combat using an improved analytic hierarchy process (AHP); determining the objective weight values ​​of multiple maneuver strategies to be evaluated in dynamic air combat using time-series-based multi-attribute decision-making; and adjusting the subjective weight values ​​and the objective weight values ​​using game theory to obtain a comprehensive weight value; activating a cloud generator using the weighted deviation of the evaluation decision to complete the air combat decision effectiveness evaluation for the current round, obtaining the dynamic air combat decision result for the current round; one side in the air combat selects the maneuver strategy with the smallest weighted deviation as the optimal maneuver strategy for execution, calculates the air combat information of both sides after executing the strategy, and determines whether the current air combat round has reached the total number of air combat rounds. If not, the dynamic air combat decision result for the next round is calculated; if so, the dynamic air combat ends.

[0007] In one embodiment of the present invention, the plurality of evaluation indicators include situational indicators and performance indicators, wherein the situational indicators include distance indicators, speed indicators, altitude indicators, and angle indicators, and the performance indicators include air combat performance indicators. The evaluation indicator system for constructing a dynamic air combat evaluation system comprising multiple evaluation indicators includes:

[0008] The dynamic air combat assessment system includes Red Force UAVs and Blue Team drones The time sequence of the dynamic air combat rounds is k = 1, 2, ..., K, where K is the total number of air combat rounds. Both the red and blue drones have a variety of maneuvers.

[0009] In one embodiment of the present invention, the evaluation index function values ​​in each evaluation strategy are determined through multiple evaluation index functions, including:

[0010] Information on both the red and blue teams' drones has been obtained in round k. In round k+1, the red team's drones... The values ​​of the UAV situational index function are expressed by equations (1)-(4):

[0011]

[0012]

[0013]

[0014]

[0015] Among them, R max R is the missile's maximum launch range. min It is the minimum launch distance of the missile, ri k+1 It is the i-th (i = 1, 2, ..., m) maneuver strategy in the (k+1)-th round. and The distance, R0 = (R max +R min ) / 2, σ=10(R max -R min ), In the (k+1)th round, the i-th maneuver strategy The distance index function value, In the (k+1)th round, the i-th maneuver strategy speed, In the (k+1)th round, the i-th maneuver strategy speed, In the (k+1)th round, the i-th maneuver strategy The speed index function value, In the (k+1)th round, the i-th maneuver strategy and The height difference In the (k+1)th round, the i-th maneuver strategy The height index function value, and These are the i-th maneuver strategies in round k+1. The target azimuth and target approach angle. In the (k+1)th round, the i-th maneuver strategy Angle index function value;

[0016] The air combat performance index function for round k+1 is expressed by equation (5):

[0017]

[0018] in, In the (k+1)th round, the i-th maneuver strategy involves the red team's drone. Performance index function value, and In the (k+1)th round, the i-th maneuver strategy and The performance index value is defined as:

[0019]

[0020] in, and In the (k+1)th round, the i-th maneuver strategy and The weapon's lethality parameters, and In the (k+1)th round, the i-th maneuver strategy and Detection capability parameters and In the (k+1)th round, the i-th maneuver strategy and Operable parameters, and In the (k+1)th round, the i-th maneuver strategy Survivability coefficient and In the (k+1)th round, the i-th maneuver strategy and The range coefficient, and In the (k+1)th round, the i-th maneuver strategy and The electronic countermeasures coefficient;

[0021] The calculation process of the red team's drone yields the blue team's drone. Evaluation index function for round k+1 and

[0022] In one embodiment of the present invention, the subjective weights of each maneuver strategy in round k+1 are set to be equal, and the subjective weight values ​​of multiple maneuver strategies to be evaluated in dynamic air combat are determined using an improved analytic hierarchy process, including:

[0023] A judgment matrix is ​​obtained by comparing the relative importance of the situation indicators and the performance indicators.

[0024] Take the judgment matrix antisymmetric matrix And construct its optimal transfer matrix.

[0025]

[0026] Construct the judgment matrix The quasi-consistent matrix

[0027]

[0028] right Normalize and add rows to obtain the matrix sum vector

[0029]

[0030] The sum vector Normalization is performed to obtain the subjective weight vector.

[0031]

[0032] The subjective weights of the situation indicators and the performance indicators are calculated separately. and Finally, the subjective weight value of the i-th maneuver strategy in round k+1 is obtained:

[0033]

[0034] In one embodiment of the present invention, determining the objective weight values ​​of multiple maneuver strategies to be evaluated in dynamic air combat using time-series-based multi-attribute decision-making includes:

[0035] According to equations (1)-(5), the red team's drone in round k+1 is obtained. and Blue Team drones The evaluation index function value is obtained by subtraction. Profit matrix

[0036]

[0037] Where m represents the number of maneuver strategies, and n represents the number of evaluation indicators in the dynamic air combat evaluation system. It is the state value of the j-th indicator in the i-th maneuver strategy of round k+1;

[0038] The evaluation indicators in the dynamic air combat evaluation system are all effectiveness indicators. The maximum and minimum values ​​of the j-th indicator (j = 1, 2, ..., n) among all maneuver strategies in round k+1 are selected as the positive ideal solution of that indicator in round k+1. and negative ideal solution

[0039]

[0040]

[0041] in, yes and The distance between them yes and The distance between them;

[0042] Indicator closeness Represented as:

[0043]

[0044] like remember

[0045] The objective weight value for the j-th indicator in the i-th maneuver strategy of round k+1:

[0046]

[0047] The actual objective weight value of the j-th indicator in the i-th maneuver strategy in round k+1:

[0048]

[0049] in, It is the actual objective weight value of the j-th indicator in the i-th maneuver strategy of the k-th round. This is the objective weight value of the j-th indicator in the i-th maneuver strategy of round k+1, where α1 and α2 are both greater than 0 and α1+α2=1. When k=1,

[0050] The actual objective weight vector of the i-th maneuver strategy in round k+1:

[0051]

[0052] In one embodiment of the present invention, a comprehensive weight value is obtained by adjusting the subjective weight value and the objective weight value through game theory, including:

[0053] Based on the calculated subjective and objective weight values, the comprehensive weight vector for round k+1 is obtained using game theory:

[0054]

[0055] in, and All are greater than 0 and and It is a row vector. It is a column vector;

[0056] The constraints are derived based on game theory:

[0057]

[0058] The Lagrange function is defined as follows: (This is based on the Lagrange method for finding the minimum value under constraints.)

[0059]

[0060] Let the Lagrange function right and The partial derivatives are zero:

[0061]

[0062] Solve the calculation:

[0063]

[0064]

[0065] The overall weight vector of the i-th maneuver strategy in round k+1:

[0066]

[0067] In one embodiment of the present invention, a cloud generator is activated using the weighted deviation of the evaluation decision to complete the evaluation of the air combat decision effectiveness for the current round, and to obtain the dynamic air combat decision result for the current round, including:

[0068] The decision matrix obtained from equation (15) Determine the cloud judgment matrix

[0069]

[0070] in, It is the closeness of the j-th indicator in the i-th maneuver strategy of round k+1;

[0071] The position vector of the centroid of the i-th maneuver strategy cloud is a i k+1 and height vector b i k+1 Represented as:

[0072]

[0073] Among them, a' (k+1) It is the cloud position vector under ideal conditions, and the j-th index in each maneuver strategy in round k+1. They are equal;

[0074] As the various maneuver strategies in the dynamic air combat assessment system change, the state of the dynamic air combat assessment system continuously changes, and the cloud center of gravity also changes accordingly. via cloud location vector and cloud height vector express:

[0075]

[0076] Multiplying the processed cloud centroid vector by the cloud height vector yields the weighted deviation of the i-th maneuver strategy:

[0077]

[0078]

[0079] The weighted deviation represents the difference between the weighted comprehensive cloud centroid of the i-th maneuver strategy in round k+1 and the weighted comprehensive cloud centroid under ideal conditions. Based on the system state after each strategy is executed in round k+1, the actual objective weight values ​​corresponding to each strategy are obtained and analyzed. The weighted deviation corresponding to each maneuver strategy is obtained using the cloud model. The cloud generator is activated based on the evaluation score.

[0080] In one embodiment of the present invention, one side in the air combat selects the maneuver strategy with the smallest weighted deviation as the optimal maneuver strategy and executes it, and calculates the air combat information of both sides after executing the strategy, including:

[0081] Air combat side The strategy with the minimum weighted deviation is selected as the optimal strategy. Combining the UAV's motion equations and dynamic equations, the air combat information of the UAV in the next round is obtained using equation (31):

[0082]

[0083] Where ψ, γ, and φ represent yaw angle, pitch angle, and roll angle, respectively. and Let v be the components of the UAV's velocity v along the three coordinate axes of the ground coordinate system, g be the acceleration due to gravity, and n be the acceleration due to gravity. x and n z For tangential overload and normal overload.

[0084] This invention proposes a method for evaluating the dynamic decision-making effectiveness of unmanned aerial vehicles (UAVs) considering the time-sequence of multi-round air combat and the time-varying weight coefficients of different evaluation indicators. By combining UAV flight dynamics, the method can more accurately obtain the values ​​of each indicator in the evaluation indicator system for each round. The theoretical objective weight of the evaluation indicator in the (k+1)th round is combined with the actual objective weight in the kth round to obtain the actual objective weight in the (k+1)th round. Based on game theory, a reasonable comprehensive weight is obtained, avoiding errors caused by uncontrollable factors.

[0085] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0086] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein:

[0087] Figure 1 A flowchart illustrating a time-series-based dynamic air combat decision evaluation method according to an embodiment of the present invention;

[0088] Figure 2 This is a schematic diagram of the evaluation index system of the dynamic air combat evaluation system provided in an embodiment of the present invention;

[0089] Figure 3 This is a flowchart of the dynamic air combat evaluation process from round k to round k+1 provided according to an embodiment of the present invention;

[0090] Figure 4 This is a flowchart for evaluating the decision-making effectiveness in the (k+1)th round according to an embodiment of the present invention;

[0091] Figure 5 The weighted deviation of the strategy in the 10th round provided by the embodiment of the present invention;

[0092] Figure 6 An evaluation cloud generator provided according to an embodiment of the present invention;

[0093] Figure 7 Provided according to embodiments of the present invention and The three-dimensional flight path. Detailed Implementation

[0094] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0095] This invention focuses on a time-series-based dynamic air combat decision evaluation method. It improves the MADM method by introducing a time series k = (2, 3, ..., K) (where K is the total number of dynamic air combat rounds) to provide objective weights for evaluating multiple decisions in multi-round air combat. First, a dynamic air combat decision effectiveness evaluation index system is established, and the values ​​of each evaluation index are obtained based on the information from each round of air combat. Second, an improved analytic hierarchy process (IAHP) is used to determine subjective weights, and a time-series-based MADM is proposed to determine the objective weights of the evaluation indicators for each round of dynamic air combat. Game theory is used to rationally adjust the subjective and objective weights to obtain a comprehensive weight. Then, the centroid of the cloud model is determined, and the cloud generator is activated based on the decision evaluation score to complete the air combat decision effectiveness evaluation for this round. Next, it is determined whether k is less than K. If so, the above steps are repeated; otherwise, the air combat ends. Finally, simulation examples verify the rationality of the proposed method.

[0096] Figure 1 A flowchart illustrating a time-series-based dynamic air combat decision evaluation method according to an embodiment of the present invention.

[0097] like Figure 1 As shown, this time-series-based dynamic air combat decision evaluation method includes the following steps:

[0098] In step S1, an evaluation index system for a dynamic air combat evaluation system containing multiple evaluation indicators is constructed.

[0099] Considering the involvement of Red Team drones (denoted as...) in air combat ) and the blue team's drone (denoted as The dynamic air combat rounds (time series) are k = (1, 2, ..., K) (K is the total number of air combat rounds). and Each has seven maneuvers. Establish a decision-making effectiveness evaluation index system based on situational and performance indicators, such as... Figure 2 As shown.

[0100] In step S2, the evaluation index function values ​​in each evaluation strategy are determined through multiple evaluation index functions.

[0101] In round k, information on both the red and blue sides' drones has been obtained. In round k+1... The UAV situational index function is expressed by equations (1)-(4):

[0102]

[0103]

[0104]

[0105]

[0106] In the formula, R max R is the missile's maximum launch range. min It is the minimum launch distance for missiles. It is the i-th (i = 1, 2, ..., m) maneuver strategy in the (k+1)-th round. and The distance, R0 = (R max +R min ) / 2, σ=10(R max -R min ), In the (k+1)th round, the i-th maneuver strategy The distance index function value. In the (k+1)th round, the i-th maneuver strategy speed,

[0107] In the (k+1)th round, the i-th maneuver strategy speed, In the (k+1)th round, the i-th maneuver strategy The speed index function value. In the (k+1)th round, the i-th maneuver strategy and The height difference In the (k+1)th round, the i-th maneuver strategy The height index function value. and These are the i-th maneuver strategies in round k+1. The target azimuth and target approach angle. In the (k+1)th round, the i-th maneuver strategy The angle index function value.

[0108] The performance index function of the UAV in round k+1 is expressed by equation (5):

[0109]

[0110] In the formula, represents the i-th maneuver strategy in the (k+1)-th round. The performance index function value. and In the (k+1)th round, the i-th maneuver strategy and The performance index value can be defined as:

[0111]

[0112] In the formula, and In the (k+1)th round, the i-th maneuver strategy and Weapon lethality parameters and In the (k+1)th round, the i-th maneuver strategy and Detection capability parameters and In the (k+1)th round, the i-th maneuver strategy and Operable parameters, and In the (k+1)th round, the i-th maneuver strategy and Survivability coefficient and In the (k+1)th round, the i-th maneuver strategy and The range coefficient, and In the (k+1)th round, the i-th maneuver strategy and The electronic countermeasure coefficient.

[0113] The calculation process of the red team's drone yields the blue team's drone. Evaluation index function for round k+1 and

[0114] In step S3, the subjective weight values ​​of multiple maneuver strategies to be evaluated in dynamic air combat are determined by the improved analytic hierarchy process, the objective weight values ​​of multiple maneuver strategies to be evaluated in dynamic air combat are determined by the time series-based multi-attribute decision-making, and the comprehensive weight value is obtained by adjusting the subjective weight values ​​and the objective weight values ​​through game theory.

[0115] First, determine the subjective weight vector for easier analysis, assuming that the subjective weights of each maneuver strategy are equal in round k+1.

[0116] Construct a judgment matrix. Experts, based on the 1-9 scale theory and their work experience, compared the relative importance of indicators among the situational and performance indicators to obtain the judgment matrix. (Situation indicators: n=4; Performance indicators: n=1).

[0117] Take the judgment matrix antisymmetric matrix And construct its optimal transfer matrix.

[0118]

[0119] Construct the judgment matrix The quasi-consistent matrix

[0120]

[0121] right Normalize and add rows to obtain the matrix sum vector

[0122]

[0123] The sum vector Normalization is performed to obtain the subjective weight vector.

[0124]

[0125] The subjective weights of the situation indicators and performance indicators are obtained by performing the above calculations separately. and Finally, the subjective weights of the i-th maneuver strategy in round k+1 are obtained:

[0126]

[0127] To introduce a time series k = (1,2,…,K) (IMADM) into the TOPSIS-based MADM method, the objective weights of multiple maneuver strategies to be evaluated are determined simultaneously.

[0128] According to equations (1)-(5), the (k+1)th round is obtained. and The evaluation index function value is obtained by subtraction. Profit matrix

[0129]

[0130] In the formula, m is the number of maneuver strategies ( and Each has 7 types of maneuvers (m=49). It refers to the number of evaluation indicators in the evaluation system. It is the state value of the j-th index in the i-th maneuver strategy of round k+1.

[0131] The evaluation indicators in the evaluation system are all effectiveness indicators. The maximum and minimum values ​​of the j-th indicator (j = 1, 2, ..., n) among all maneuver strategies in round k+1 are selected as the positive ideal solution for that indicator in round k+1. and negative ideal solution

[0132]

[0133]

[0134] In the formula, yes and The distance between them yes and The distance between them.

[0135] Indicator closeness It can be represented as:

[0136]

[0137] like remember

[0138] The objective weight value for the j-th indicator in the i-th maneuver strategy of round k+1:

[0139]

[0140] The actual objective weight value of the j-th indicator in the i-th maneuver strategy of round k+1:

[0141]

[0142] In the formula, It is the actual objective weight value of the j-th indicator in the i-th maneuver strategy of the k-th round. This is the objective weight value for the j-th indicator in the i-th maneuver strategy of round k+1. Both α1 and α2 are greater than 0, and α1 + α2 = 1. When k = 1,

[0143] The actual objective weight vector of the i-th maneuver strategy in round k+1:

[0144]

[0145] Based on the calculated subjective and objective weight vectors, the comprehensive weight vector for round k+1 is obtained using game theory:

[0146]

[0147] In the formula, and All are greater than 0 and and It is a row vector. It is a column vector.

[0148] The constraints are derived based on game theory:

[0149]

[0150] The Lagrange function is defined as follows: (This is based on the Lagrange method for finding the minimum value under constraints.)

[0151]

[0152] Let the Lagrange function right and The partial derivatives are zero:

[0153]

[0154] Solve the calculation:

[0155]

[0156]

[0157] The overall weight vector of the i-th maneuver strategy in round k+1:

[0158]

[0159] In step S4, the cloud generator is activated by using the weighted deviation of the evaluation decision to complete the evaluation of the air combat decision effectiveness of the current round, and obtain the dynamic air combat decision result of the current round. One side in the air combat selects the maneuver strategy with the smallest weighted deviation as the optimal maneuver strategy to execute, and calculates the air combat information of both sides after the execution of the strategy. It is also determined whether the current air combat round has reached the total number of air combat rounds. If it has not reached the total number of air combat rounds, the dynamic air combat decision result of the next round is calculated. If it has reached the total number of air combat rounds, the dynamic air combat ends.

[0160] The decision matrix obtained from equation (15) Determine the cloud judgment matrix

[0161]

[0162] In the formula, It is the closeness of the j-th indicator in the i-th maneuver strategy of the (k+1)-th round.

[0163] The position vector of the centroid of the i-th maneuver strategy cloud and height vector It can be represented as:

[0164]

[0165] In the formula, a' (k+1) It is the cloud position vector under ideal conditions, and the j-th index in each maneuver strategy in round k+1. They are equal.

[0166] As the various maneuver strategies in the evaluation system change, the system state continuously changes, and the cloud center of gravity shifts accordingly. Cloud location vectors and cloud height vector express:

[0167]

[0168] Multiplying the processed cloud centroid vector by the cloud height vector yields the weighted deviation of the i-th maneuver strategy:

[0169]

[0170]

[0171] The weighted deviation represents the difference between the weighted composite cloud center of gravity of the i-th maneuver strategy in round k+1 and the weighted composite cloud center of gravity under ideal conditions. Based on the system state after each strategy is executed in round k+1, the actual objective weight values ​​corresponding to each strategy are obtained and analyzed. The weighted deviation corresponding to each maneuver strategy is obtained using the cloud model. The cloud generator is activated based on the evaluation score, and it is determined whether the current air combat round number k is less than K.

[0172] The strategy with the minimum weighted deviation is selected as the optimal strategy. Combining the UAV's motion equations and dynamic equations, the air combat information of the UAV in the next round is obtained using equation (31):

[0173]

[0174] In the formula, ψ, γ, and φ represent the yaw angle, pitch angle, and roll angle, respectively. and Let v be the components of the UAV's velocity v along the three coordinate axes of the ground coordinate system; g is the acceleration due to gravity, which is taken as 9.8 m / s² in this paper. 2 ;n x and n z For tangential overload and normal overload.

[0175] like Figure 3 and Figure 4 As shown, the dynamic air combat evaluation process from round k to round k+1 and the decision effectiveness evaluation process for round k+1 are illustrated.

[0176] The following specific embodiment illustrates the time-series-based dynamic air combat decision evaluation method of the present invention.

[0177] and The maneuver time is denoted as Δt = 2s, the total number of air combat rounds is K = 10, and the maximum radar detection range is R. r =80km, the maximum range of the missile is R max =60km, the minimum range of the missiles is R min = 1km. and The initial information on the drone air combat between the two sides is shown in Table 1.

[0178] Table 1 and Initial information on the drone air combat between the two sides

[0179]

[0180] In multi-round dynamic air combat, the initial distance between the two sides is far beyond the maximum attack range of the missiles. and First, accelerate straight flight is selected. After a period of time, a maneuver is chosen based on the dynamic decision-making effectiveness evaluation results obtained from the current air combat information. Taking round 10 as an example, the evaluation index function is used to determine the maneuvering strategy under each maneuvering strategy. and The corresponding evaluation index function values ​​for both parties are shown in Tables 2 and 3.

[0181] Table 2 Evaluation index function value

[0182]

[0183]

[0184]

[0185] Table 3 Evaluation index function value

[0186]

[0187]

[0188] In the table, Indicates the first maneuver strategy: and All selected maneuver number 1.

[0189] Secondly, based on the idea of ​​the IMADM method, α1 = 0.2 and α2 = 0.8 are selected to obtain the actual objective weight vectors of all maneuver strategies in this round at once. The subjective weight vectors of each maneuver strategy in this round are then obtained according to IAHP. (Considering the complexity of dynamic air combat, it is assumed that the subjective weight vectors of each maneuver strategy are the same).

[0190] Then, using the ideas of game theory, the subjective and objective weight coefficients are reasonably adjusted to obtain the comprehensive weight vector for this round, as shown in Table 4.

[0191] Table 4 Comprehensive Weight Vector

[0192]

[0193]

[0194] According to cloud model theory, the language values ​​of each indicator are represented by (E). x E n H eThe evaluation set (none, very poor, poor, poor, average, good, good, very good, excellent) is quantified as (0, 0.15, 0.3, 0.4, 0.5, 0.6, 0.7, 0.85, 1), and the qualitative language of the ideal integrated cloud model is "excellent".

[0195] The weighted deviation of each maneuver strategy in round 10 was obtained through simulation experiments, such as... Figure 5 As shown. Further calculations yield the corresponding decision performance evaluation score, which is then input into the cloud generator to generate a comment set, resulting in the decision performance evaluation. Taking the maneuver strategy corresponding to the minimum and maximum weighted deviations in round 10 as an example, the generated cloud generator is as follows: Figure 6 As shown.

[0196] Figure 5 In the middle, strategy 16 is... Execute Maneuver No. 2 The weighted deviation corresponding to maneuver #3 is the smallest, i.e., Strategy 10. Execute Maneuver No. 3 The weighted deviation is the largest when performing maneuver number 2.

[0197] According to the principle of maximum membership, by Figure 6 It can be seen that when strategy 16 was executed in round 10, "Very Good" and "Excellent" were activated, indicating that the evaluation result of strategy 16 was between "Very Good" and "Excellent", which enabled... It holds a significant advantage. Executing strategy 10 activated "good" and "fair," indicating that the evaluation result of strategy 10 falls between "good" and "fair," enabling... They hold a significant advantage.

[0198] comprehensive and The changes in the situation after implementing Strategy 16 and Strategy 10 suggest that... Select strategy 16 for execution. Using the same method as above, the entire dynamic air combat process is obtained. and Selection of maneuvers and 3D trajectory diagrams, as shown Figure 7 As shown.

[0199] The flowchart of the dynamic air combat decision evaluation method based on time series proposed in this embodiment of the invention describes a method for constructing an evaluation index system for a dynamic air combat evaluation system. Based on the air combat information from the previous round and flight dynamics, air combat information corresponding to each maneuver strategy in the current round is obtained. Evaluation index values ​​for each strategy to be evaluated in the current round are obtained through an evaluation index function. Next, the subjective weight values ​​of the evaluation indicators are determined using an improved analytic hierarchy process (AHP), and objective weight values ​​are determined based on multi-attribute decision-making in the time series, with a comprehensive weight value obtained through game theory. Then, the evaluation score of the decision to be evaluated is obtained based on the comprehensive weight value, activating the cloud model. One side selects the maneuver strategy with the smallest weighted deviation as the optimal maneuver strategy and executes it, completing the decision effectiveness evaluation for the current round. The above steps are repeated until the number of dynamic air combat rounds equals the set total number of air combat rounds. This accurately calculates the values ​​of each index, thereby avoiding errors caused by uncontrollable factors.

[0200] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0201] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0202] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or N executable instructions for implementing custom logic functions or processes, and the scope of preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of the invention pertain.

Claims

1. A dynamic air combat decision evaluation method based on time series, characterized in that, Includes the following steps: An evaluation index system for a dynamic air combat evaluation system is constructed, comprising multiple evaluation indicators. These multiple evaluation indicators include situational indicators and performance indicators. The situational indicators include distance, speed, altitude, and angle indicators, and the performance indicators include air combat performance indicators. The evaluation index system for a dynamic air combat evaluation system comprises: The dynamic air combat assessment system includes Red Force UAVs and Blue Team drones The time sequence of the dynamic air combat rounds is k=1, 2, …, K, where K is the total number of air combat rounds. Both the red and blue drones have a variety of maneuvers. The values ​​of the evaluation index functions in each evaluation strategy are determined by using multiple evaluation index functions. The subjective weight values ​​of multiple maneuver strategies to be evaluated in dynamic air combat are determined by using the improved analytic hierarchy process, the objective weight values ​​of multiple maneuver strategies to be evaluated in dynamic air combat are determined by using multi-attribute decision-making based on time series, and the comprehensive weight value is obtained by adjusting the subjective weight values ​​and the objective weight values ​​through game theory. The cloud generator is activated by the weighted deviation of the evaluation decision to complete the evaluation of the air combat decision effectiveness in the current round, and obtain the dynamic air combat decision result in the current round. The air combat side selects the maneuver strategy with the smallest weighted deviation as the optimal maneuver strategy to execute, and calculates the air combat information of both sides after the execution of the strategy. It is also determined whether the current air combat round has reached the total number of air combat rounds. If it has not reached the total number of air combat rounds, the dynamic air combat decision result for the next round is calculated. If it has reached the total number of air combat rounds, the dynamic air combat ends. The evaluation index function values ​​for each evaluation strategy are determined through multiple evaluation index functions, including: Information on both the red and blue teams' drones has been obtained in round k. In round k+1, the red team's drones... The state indices of the unmanned aerial vehicle (UAV) are expressed by equations (1)-(4): 　　　（1） 　　　（2） 　　　（3） 　　　（4） in, This is the missile's maximum launch range. It is the minimum launch distance for missiles. In the (k+1)th round, the i-th maneuver strategy and The distance between them, where i = 1, 2, ..., m; , , In the (k+1)th round, the i-th maneuver strategy The distance index function value, In the (k+1)th round, the i-th maneuver strategy speed, In the (k+1)th round, the i-th maneuver strategy speed, In the (k+1)th round, the i-th maneuver strategy The speed index function value, In the (k+1)th round, the i-th maneuver strategy and The height difference In the (k+1)th round, the i-th maneuver strategy The height index function value, and These are the i-th maneuver strategies in round k+1. The target azimuth and target approach angle. In the (k+1)th round, the i-th maneuver strategy Angle index function value; The air combat performance index function for round k+1 is expressed by equation (5): 　　　（5） in, In the (k+1)th round, the i-th maneuver strategy involves the red team's drone. Performance index function value, and In the (k+1)th round, the i-th maneuver strategy and The performance index value is defined as: 　　　（6） in, and In the (k+1)th round, the i-th maneuver strategy and The weapon's lethality parameters, and In the (k+1)th round, the i-th maneuver strategy and Detection capability parameters and In the (k+1)th round, the i-th maneuver strategy and Operable parameters, and In the (k+1)th round, the i-th maneuver strategy and Survivability coefficient and In the (k+1)th round, the i-th maneuver strategy and The range coefficient, and In the (k+1)th round, the i-th maneuver strategy and The electronic countermeasures coefficient; The calculation process of the red team's drone yields the blue team's drone. Evaluation index function for round k+1 , , , and ; Using time-series-based multi-attribute decision-making, objective weight values ​​for multiple maneuver strategies to be evaluated in dynamic air combat are determined, including: According to equations (1)-(5), the red team's drone in round k+1 is obtained. and Blue Team drones The evaluation index function value is obtained by subtraction. Profit matrix : 　　　（12） in, It is the number of maneuver strategies. It refers to the number of evaluation indicators in the dynamic air combat evaluation system. It is the state value of the j-th indicator in the i-th maneuver strategy of round k+1; The evaluation indicators in the dynamic air combat evaluation system are all effectiveness indicators. The maximum and minimum values ​​of the j-th indicator among all maneuver strategies in round k+1 are selected as the positive ideal solution of that indicator in round k+1. and negative ideal solution Where j = 1, 2, ..., n: 　　　（13） 　　　（14） in, yes and The distance between them yes and The distance between them; Indicator closeness Represented as: 　　　（15） like ,remember ; The objective weight value for the j-th indicator in the i-th maneuver strategy of round k+1: 　　　（16） The actual objective weight value of the j-th indicator in the i-th maneuver strategy in round k+1: 　　　（17） in, It is the actual objective weight value of the j-th indicator in the i-th maneuver strategy of the k-th round. It is the objective weight value of the j-th indicator in the i-th maneuver strategy of round k+1. and All are greater than 0 and When k=1, ; The actual objective weight vector of the i-th maneuver strategy in round (k+1): （18）； The cloud generator is activated by utilizing the weighted deviation of the evaluation decision to complete the evaluation of the air combat decision effectiveness for the current round, obtaining the dynamic air combat decision results for the current round, including: The decision matrix obtained from equation (15) Determine the cloud judgment matrix : 　　　（26） in, It is the closeness of the j-th indicator in the i-th maneuver strategy of round k+1; The position vector of the centroid of the i-th maneuver strategy cloud and height vector Represented as: 　　　（27） in, It is the cloud position vector under ideal conditions, and the j-th index in each maneuver strategy in round k+1. They are equal; As the various maneuver strategies in the dynamic air combat assessment system change, the state of the dynamic air combat assessment system continuously changes, and the cloud center of gravity also changes accordingly. via cloud location vector and cloud height vector express: 　　　（28） Multiplying the processed cloud centroid vector by the cloud height vector yields the weighted deviation of the i-th maneuver strategy: 　　　（29） 　　　（30） The weighted deviation represents the difference between the weighted comprehensive cloud centroid of the i-th maneuver strategy in round k+1 and the weighted comprehensive cloud centroid under ideal conditions. Based on the system state after each strategy is executed in round k+1, the actual objective weight values ​​corresponding to each strategy are obtained and analyzed. The weighted deviation corresponding to each maneuver strategy is obtained using the cloud model. The cloud generator is activated based on the evaluation score.

2. The method according to claim 1, characterized in that, Assuming that the subjective weights of all maneuver strategies are equal in round k+1, the subjective weight values ​​of multiple maneuver strategies to be evaluated in dynamic air combat are determined using an improved analytic hierarchy process (AHP), including: A judgment matrix is ​​obtained by comparing the relative importance of the situation indicators and the performance indicators. ; Take the judgment matrix antisymmetric matrix And construct its optimal transfer matrix. : 　　　（7） Construct the judgment matrix The quasi-consistent matrix : 　　　（8） right Normalize and add rows to obtain the matrix , and vector : 　　　（9） The sum vector Normalization is performed to obtain the subjective weight vector. : 　　　（10） The subjective weights of the situation indicators and the performance indicators are calculated separately. and Finally, the subjective weight value of the i-th maneuver strategy in round k+1 is obtained: （11）。 3. The method according to claim 2, characterized in that, A comprehensive weight value is obtained by adjusting the subjective weight value and the objective weight value using game theory, including: Based on the calculated subjective and objective weight values, the comprehensive weight vector for round k+1 is obtained using game theory: 　　　（19） in, and All are greater than 0 and , and It is a row vector. It is a column vector; The constraints are derived based on game theory: 　　　（20） The Lagrange function is defined as follows: (This is based on the Lagrange method for finding the minimum value under constraints.) 　　　（21） Let the Lagrange function right , and The partial derivatives are zero: 　　　（22） Solve the calculation: 　　　（23） 　　　（24） The overall weight vector of the i-th maneuver strategy in round k+1: （25）。 4. The method according to claim 1, characterized in that, One side in the air combat selects the maneuver strategy with the smallest weighted deviation as the optimal maneuver strategy and calculates the air combat information of both sides after the strategy is executed, including: Air combat side The strategy with the minimum weighted deviation is selected as the optimal strategy. Combining the UAV's motion equations and dynamic equations, the air combat information of the UAV in the next round is obtained using equation (31): 　　　（31） in, , and Expressed as yaw angle, pitch angle, and roll angle. , and For drone speed The components on the three coordinate axes of the ground coordinate system, It is the acceleration due to gravity. and For tangential overload and normal overload.

Images