A system damage net reconstruction method for time-sensitive target disposal

Abstract

The application discloses a kind of system killing net reconstruction methods for time-sensitive target disposal, design a kind of reconstruction mechanism based on formal matrix modeling and multi-objective evolutionary algorithm to solve the contradiction that existing system is difficult to consider both established task and sudden time-sensitive target.Firstly, based on F2T2EA function chain, the function set of combat equipment, communication topology matrix and task allocation matrix are constructed, and the formal modeling of system killing net is completed;Then when monitoring the time-sensitive target, a double-objective optimization model containing the maximization of strike effectiveness and the minimization of equipment use cost is established, and a triple-restraint logic of equipment capability, link communication and killing chain closure is constructed;Finally, the NSGA-III algorithm based on reference point selection mechanism is used for high-dimensional search, and the Pareto optimal solution set is obtained and the execution scheme is output.The application realizes the rapid response to the new time-sensitive target and the optimal allocation of resources under the premise of ensuring the logic closure of killing chain, and significantly improves the robustness and resource utilization efficiency of battlefield operational system.

Description

Technical Field

[0001] This invention belongs to the field of military command and control and combat system optimization technology, and in particular relates to a system kill network reconstruction method for dealing with time-sensitive targets. Background Technology

[0002] In the context of contemporary networked and system-of-systems warfare, the F2T2EA kill chain, built around the stages of Find, Fix, Track, Target, Engage, and Assess, has become a crucial organizational framework for handling time-sensitive targets. Time-sensitive targets are typically characterized by sudden appearance, short exposure windows, and high threat levels; delays in handling them can easily lead to lost operational opportunities or damage to critical friendly platforms. In system-of-systems warfare, the rapid construction of the F2T2EA kill chain is core to dealing with time-sensitive targets (sudden appearance, short exposure windows). However, existing technologies face three major technical challenges when dealing with on-the-spot reconfiguration:

[0003] (1) Existing kill networks typically form task links in advance around predetermined targets. When new time-sensitive targets appear unexpectedly on the battlefield, the original task allocation and coordination links often cannot be directly adapted, resulting in a decrease in the effectiveness of the system strike. If the static configuration method is still used, it is difficult to take into account both new targets and existing combat tasks in a short period of time; if the single-target optimization method is used, it is easy to be biased towards a single effectiveness index, making it difficult to take into account both strike effect and resource consumption.

[0004] (2) Existing reconstruction methods mainly rely on rule deduction, low-dimensional optimization or traditional multi-objective optimization algorithms. The former is not adaptable to complex battlefields, while the latter is prone to problems such as uneven distribution of solution sets and insufficient convergence speed in high-dimensional, multi-constraint kill net reconstruction problems, making it difficult to meet the real-time and scheme diversity requirements of time-sensitive target disposal.

[0005] Therefore, there is an urgent need for a system kill reconstruction method that can perform rapid global optimization in a high-dimensional constrained space and ensure the closed loop of the kill chain logic. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a system kill network reconstruction method for handling time-sensitive targets. By constructing a reconstruction optimization model containing logical constraints and combining it with the reference point-guided NSGA-Ⅲ evolution mechanism, a paradigm shift from "rule-based reasoning" to "intelligent evolution" is achieved.

[0007] To achieve the objectives of this invention, a system kill network reconstruction method for time-sensitive target handling is proposed, comprising the following steps:

[0008] S1: Matrix-based state modeling of the kill network in the combat system

[0009] The Red Force's combat equipment is mapped to six functional sets based on the F2T2EA functional chain, and equipment attribute sets for each functional link are established. Based on the activity logic permissions between equipment, a communication relationship matrix describing the communication topology is constructed. Based on the task allocation logic, an allocation matrix describing the mapping relationship between equipment and targets is constructed, thus completing the formal representation of the system kill network state.

[0010] S2: Dynamic Extended Representation of Time-Sensitive Target Events

[0011] When a high-threat time-sensitive target is observed, the newly added target is merged with the original target set to form an extended target set to be dealt with.

[0012] S3: Establish and reconstruct decision variables

[0013] The decision variable matrix is ​​reconstructed, and the allocation state of equipment at each stage to different targets is mapped by integer encoding, thus transforming the reconstruction problem into a search problem in a high-dimensional decision space.

[0014] S4: Construction of a Multi-Objective Reconstruction Optimization Model

[0015] Construct a mathematical optimization model with the objective of maximizing strike effectiveness as objective 1 and minimizing equipment usage cost as objective 2;

[0016] S5: Construction of Logical Constraints for Multidimensional Conditions

[0017] The decision space is logically limited by constructing constraints including equipment capability constraints, link communication constraints, and kill chain closure constraints.

[0018] S6: Multi-objective cooperative optimization based on NSGA-III

[0019] Based on the integer encoding vector, an initial population is generated, and the NSGA-III algorithm is used to find the Pareto optimal solution set in the multi-objective search space through non-dominated sorting and reference point adaptive mechanism.

[0020] S7: Optimal Decision Output

[0021] Based on preset performance and cost preference weights, the system selects the execution plan from the Pareto optimal plan, updates the configuration of the original kill network, and enables rapid inclusion and coordinated strikes against time-sensitive targets.

[0022] Preferably, in step S1, the six types of functions are: discovery, location, tracking, aiming, engagement, and evaluation.

[0023] Preferably, in step S4, the objective function of the kill net multi-target reconstruction optimization model includes:

[0024] S41 maximizes strike effectiveness

[0025]

[0026] in, Indicates the first The value coefficient of each objective Indicates surrounding the first The combined strike effectiveness of the kill chain formed by the targets; the combined strike effectiveness is jointly determined by the coordinated hit probability of equipment in each stage of detection, location, tracking, aiming, engagement, and assessment;

[0027] S42 Minimize Equipment Usage Costs

[0028]

[0029] in, , , , , and Used to filter links that have changed in the kill network after reconstruction; , , , , and These represent the cost vectors corresponding to the detection equipment, positioning equipment, tracking equipment, aiming equipment, engagement equipment, and evaluation equipment, respectively; the equipment usage cost includes the platform usage cost and the ammunition consumption cost of the strike equipment.

[0030] Preferably, in step S5, the constraints include the following:

[0031] S51 Equipment Capability Constraints

[0032] This is used to limit the participation of a single piece of equipment in the coordinated handling of multiple targets, provided that it does not exceed its multi-target processing capability;

[0033]

[0034] in, , , , , and They represent the first The first discovered equipment, the first The first positioning device, the first The first tracking device, the first The first aiming device, the first The combat equipment and the first The assessment equipment's multi-target combat capability;

[0035] S52 link communication constraints

[0036] This is used to ensure that there is an effective communication link between adjacent functional equipment so that they can jointly form a kill chain.

[0037]

[0038] For example, consider the communication relationship constraints between the detection equipment and the positioning equipment: If , indicating the first The first discovered equipment and the first If there is no communication link between the positioning devices, then there is ,Right now and At least one of them must be 0; this means that neither can be simultaneously positive for the first. Attacks are carried out on individual targets, thus failing to form an effective kill chain; if This indicates that there is a communication relationship between the two, then we have ,Right now and They can all take 1 at the same time, thus jointly participating in the process of determining the first... The strike on a single target;

[0039] S53 Kill Chain Closure Constraint

[0040] This is used to ensure that a complete "detection-location-tracking-aiming-engagement-assessment" closed loop is formed for each target, and to ensure that the target damage effect meets the preset threshold requirements;

[0041]

[0042] in, , , , , and Representing the target The joint detection accuracy, joint positioning accuracy, joint tracking accuracy, joint aiming accuracy, joint engagement accuracy, and joint assessment accuracy; Indicates the first The damage threshold of each target is set to ensure that every link in the kill chain is physically connected and logically sound.

[0043] Preferably, in step S6, the NSGA-Ⅲ multi-objective optimization algorithm specifically includes:

[0044] S61 Based on the integer encoding method described in step S3, randomly generate an initial population and conduct a feasibility assessment;

[0045] S62 uses tournament selection, simulated binary crossover, and polynomial mutation operations to generate offspring populations.

[0046] S63 performs a fast non-dominated sort after merging the parent and offspring populations.

[0047] S64 normalizes the objective function and, using a pre-defined set of reference points, calculates the proximity score and diversity score of individuals to execute a survival scoring mechanism.

[0048] S65 Repeat steps S62 to S64 until the termination condition is met, and output the Pareto optimal solution set.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] 1. This invention models newly added threat targets as time-sensitive target events and drives the reconstruction process with an expanded target set, which can better reflect the needs of handling ad-hoc targets in real battlefields.

[0051] 2. This invention introduces a dual-objective collaborative optimization model that combines strike effectiveness and equipment usage cost, overcoming the limitations of previous single-indicator approaches. By utilizing the Pareto front solution set, it provides commanders with a "solution library" that balances effectiveness and cost, ensuring that high-threat targets are dealt with quickly while maximizing the optimization of battlefield resource allocation.

[0052] 3. By constructing a triple constraint model of "equipment capability-link communication-kill chain closure", this invention ensures that any output scheme satisfies the F2T2EA functional logic, fundamentally solving the problem of "logic chain break" in the operational application of the reconfiguration scheme and improving the engineering feasibility of the scheme.

[0053] 4. This invention employs the NSGA-Ⅲ algorithm based on reference points, which, compared to traditional genetic algorithms, exhibits stronger convergence efficiency and solution set diversity when handling high-dimensional, multi-constraint reconstruction problems. This method can significantly shorten reconstruction time, meeting the timeliness requirements for handling time-sensitive targets.

[0054] 5. The reconfiguration mechanism of this invention can achieve incremental optimization through local link adjustment and resource reconfiguration without affecting the established mission, thus ensuring the strong robustness and adaptability of the combat system in response to sudden threats. Attached Figure Description

[0055] Figure 1 This is a functional framework diagram of the F2T2EA kill chain according to an embodiment of this application;

[0056] Figure 2 This is a schematic diagram of kill net reconstruction for time-sensitive targets according to an embodiment of this application;

[0057] Figure 3 This is a flowchart illustrating the NSGA-Ⅲ algorithm solution process in an embodiment of this application.

[0058] Figure 4 This is a schematic diagram illustrating a maritime and air defense and anti-missile demonstration scenario in an embodiment of this application.

[0059] Figure 5 This is a comparison diagram of the convergence of NSGA-Ⅲ and NSGA-Ⅱ in the objective function according to embodiments of this application;

[0060] Figure 6 This is a comparison chart of the convergence of NSGA-Ⅲ and NSGA-Ⅱ in the objective function of the embodiments of this application;

[0061] Figure 7 This is a comparison diagram of NSGA-Ⅲ and NSGA-Ⅱ on the Pareto front in the embodiments of this application;

[0062] Figure 8 A comparison chart of evaluation metrics for different algorithms;

[0063] Figure 9 The kill network structure diagram is reconstructed before the appearance of time-sensitive targets;

[0064] Figure 10 This is a diagram of the kill network structure after the time-sensitive target appears and is reconstructed. Detailed Implementation

[0065] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0066] Example 1

[0067] This application provides a system kill network reconstruction method for handling time-sensitive targets, specifically including the following steps:

[0068] S1: Matrix-based state modeling of the kill network in the combat system

[0069] like Figure 1 As shown, the Red Force's combat equipment is divided into four categories according to the F2T2EA functional chain: detection, location, tracking, aiming, engagement, and evaluation equipment, as detailed below:

[0070] (1)

[0071] (2)

[0072] in, , , , , and They represent the first Class discovery unit, the first Class positioning unit, first Class tracking unit, the first Type of aiming unit, the first Type of combat unit and the first Evaluation unit.

[0073] Simultaneously, a communication relationship matrix between adjacent functional equipment and an allocation matrix between each functional equipment and enemy targets are constructed.

[0074] The communication matrix can be represented as:

[0075] (3)

[0076] in, Indicates the first The first Type of equipment and the first The first There are communication links between the equipment of this type.

[0077] The allocation matrix can be represented as:

[0078] (4)

[0079] in, Indicates the first The first This type of equipment was assigned to strike the first One enemy target.

[0080] S2: Dynamic Extended Representation of Time-Sensitive Target Events

[0081] like Figure 2 As shown, when a new high-threat target appears on the battlefield, it is defined as a time-sensitive target and merged with the original target set to form an extended target set, which is then handled during the reconstruction phase.

[0082] Blue Team Time-Sensitive Target Set Include The targets to be attacked are listed below:

[0083] (5)

[0084] When added When a blue team target is discovered, the new target set is denoted as . Then the expanded target set is:

[0085] (6)

[0086] S3: Establish and reconstruct decision variables

[0087] For newly added time-sensitive targets, reconstructed decision matrices are constructed for the discovery, location, tracking, targeting, engagement, and evaluation stages.

[0088] (7)

[0089]

[0090] Among them, when , , , , , When the value is 0, it indicates that the first... The original allocation of equipment for detection, location, tracking, targeting, engagement, and assessment remains unchanged; when these variables take values... When, it indicates the first The equipment was reassigned from the original target to the new target. .

[0091] S4: Construction of a Multi-Objective Reconstruction Optimization Model

[0092] The multi-objective reconstruction optimization model includes the following objective functions:

[0093] S41 Objective 1: Maximize strike effectiveness, represented by the product of the target value coefficient and the joint strike effectiveness;

[0094] (8)

[0095] in, Indicates the first The value coefficient of each objective Indicates surrounding the first The joint strike effectiveness of the kill chain formed by the targets; wherein, the joint strike effectiveness is determined by the coordinated hit probability of the equipment in each link of detection, location, tracking, aiming, engagement and assessment.

[0096] (9)

[0097]

[0098] in, , , , , and They represent the first The first discovered equipment, the first The first positioning device, the first The first tracking device, the first The first aiming device, the first The combat equipment and the first The evaluation equipment for the first The accuracy of each target; , , , , and Representing the target The joint detection accuracy, joint positioning accuracy, joint tracking accuracy, joint aiming accuracy, joint engagement accuracy, and joint assessment accuracy; , , , , and For design variables.

[0099] S42 Objective 2: Minimize equipment usage costs, taking into account both platform usage costs and ammunition consumption costs.

[0100] (10)

[0101] in, , , , , and Used to filter targets in the reconstructed kill net. A link where the target changes.

[0102] The cost of equipment use includes the cost of platform use and the cost of ammunition consumption for combat equipment.

[0103] (11)

[0104] in, , , , , and These represent the cost vectors corresponding to the detection equipment, positioning equipment, tracking equipment, aiming equipment, engagement equipment, and evaluation equipment, respectively.

[0105] S5: Construction of Logical Constraints for Multidimensional Conditions

[0106] The constraints include equipment capability constraints, link communication constraints, and kill chain closure constraints, as detailed below:

[0107] S51 Equipment Capability Constraints: These constraints limit the participation of a single piece of equipment in the coordinated handling of multiple targets, provided that it does not exceed its multi-target processing capability.

[0108] (12)

[0109] in, , , , , and They represent the first The first discovered equipment, the first The first positioning device, the first The first tracking device, the first The first aiming device, the first The combat equipment and the first The assessment evaluates the multi-target combat capability of the equipment; where multi-target combat capability refers to the upper limit of tasks that the equipment can handle concurrently within the same combat cycle or the maximum load it can bear. This constraint ensures that the processing load of a single piece of equipment per unit time does not exceed its rated maximum combat capability, thereby avoiding functional failure caused by equipment overload.

[0110] S52 Link Communication Constraint: This is used to ensure that adjacent functional equipment can form a kill chain only when there is an effective communication link between them.

[0111] (13)

[0112] For example, consider the communication relationship constraints between the detection equipment and the positioning equipment: If , indicating the first The first discovered equipment and the first If there is no communication link between the positioning devices, then there is ,Right now and At least one of them must be 0. This means that neither can simultaneously affect the first. Attacks are carried out on individual targets, thus preventing the formation of an effective kill chain. If... This indicates that there is a communication relationship between the two, then we have ,Right now and They can all take 1 at the same time, thus jointly participating in the process of determining the first... The strike on the target.

[0113] S53 Kill Chain Closure Constraint: Used to ensure that a complete "detection-location-tracking-aiming-engagement-assessment" closed loop is formed for each target, and to ensure that the damage effect of the target meets the preset threshold requirements.

[0114] (14)

[0115] in, Indicates the first Damage threshold for each target.

[0116] By constraining the link validity of the communication relationship matrix and the logical integrity of the kill chain closure, it is ensured that the six types of functional equipment assigned to the same target form a complete kill chain that satisfies the communication link topology, thus realizing the logical closed loop of the combat mission from detection to assessment.

[0117] S6: Multi-objective cooperative optimization based on NSGA-III

[0118] like Figure 3 As shown, the NSGA-III multi-objective optimization algorithm is used to solve the reconstruction optimization model to obtain the Pareto optimal reconstruction scheme set, specifically including the following steps:

[0119] S61 uses integer encoding to encode the reconstructed decision variables for each stage of discovery, location, tracking, aiming, engagement, and assessment, and randomly generates an initial population;

[0120] S62 performs tournament selection, simulated binary crossover, and polynomial mutation on parent individuals to generate offspring populations.

[0121] Two individuals are selected from the parent population through a tournament selection process. and Subsequently, two offspring are generated using simulated binary crossover operations. and The details are as follows:

[0122] (15)

[0123] (16)

[0124] in, From random numbers Determine; Cross-distribution index Used to adjust the degree of similarity between offspring and parents.

[0125] After the crossover operation is completed, each child Further random polynomial mutation operations are performed to generate new offspring individuals:

[0126] (17)

[0127] in, It is the distribution index of variation; and Representing variables respectively The upper and lower bounds.

[0128] S63 performs a fast non-dominated sort after merging the parent and offspring populations.

[0129] S64 normalizes the objective function and performs individual screening based on reference points and survival scoring mechanisms.

[0130] Normalization techniques are used to standardize the non-dominated frontier, scaling the objective values ​​of all solutions on the frontier to an interval. Inside.

[0131] (18)

[0132] in, Solution In the The function value on each target This indicates that all solutions in this frontier layer are at the th... The minimum value on each objective is achieved by normalizing the objective functions with different dimensions, which solves the optimization bias caused by the difference in magnitude between multiple objective functions and ensures the effectiveness of the reference point guidance mechanism.

[0133] Calculating the curvature of the approximate Pareto front using the center point ,based on The norm is used to calculate the proximity score by determining the distances between individuals. Then, the minimum distance between individuals is calculated to obtain the diversity score. Finally, a survival score is calculated, and the next generation of the population is selected based on this score.

[0134] S65 Repeat steps S62 to S64 until the termination condition is met, and output the Pareto optimal solution set. Under this model framework, no matter how the battlefield environment changes, as long as the above matrix logic is satisfied, the system can output the theoretically optimal cooperative configuration.

[0135] S7: Based on performance and cost preferences, select the option to be executed from the Pareto optimal options, update the configuration of the original kill network, and thus achieve rapid inclusion and coordinated strike against time-sensitive targets.

[0136] Based on typical operational scenarios of the "maritime air defense and anti-missile combat system", a system kill network reconstruction method for dealing with time-sensitive targets is verified.

[0137] S71 Experimental Scenario Setup

[0138] like Figure 4 As shown, air defense and missile defense operations refer to the operational process in modern warfare that comprehensively utilizes multi-functional systems such as reconnaissance, command, strike, and communication to coordinate the detection, interception, and strike of aerial threats.

[0139] As shown in Table 1, the Blue Force deployed 5 Elf UAVs, 5 Coyote UAVs, and 5 anti-ship missiles to conduct probing strikes against the Red Force ships. Simultaneously, high-threat cruise missiles were introduced during the operation to further attack identified Red Force targets. As shown in Table 2, the Red Force's system consisted of 5 UAVs, 5 unmanned surface vessels, 1 aircraft carrier, and various types of missiles. From the Red Force's operational perspective, it was essential to utilize existing operational resources to rapidly reconstruct an effective kill network to intercept incoming threats.

[0140] Table 1. Blue Team Force Composition Information

[0141]

[0142] Table 2. Composition of Red Forces

[0143]

[0144] Comparison of evaluation metrics for different algorithms in S72

[0145] like Figures 5-7 As shown, a comparative analysis of NSGA-III and NSGA-II was conducted through computational experiments, focusing on the convergence characteristics of the objective function and the Pareto front distribution. The comprehensive evaluation results show that the two algorithms exhibit significant performance differences across multiple dimensions, as detailed below:

[0146] (1) Convergence analysis of objective function 1.

[0147] like Figure 5 As shown, experimental results indicate that NSGA-III exhibits higher convergence efficiency, with a final convergence value of approximately -8590, slightly better than NSGA-II's approximately -8560. The convergence trajectory reveals that NSGA-III established an advantage early in the optimization process and maintained stable progress throughout the evolution. Particularly noteworthy is the more stable convergence process exhibited by NSGA-III compared to the occasional stagnation observed in NSGA-II, demonstrating a stronger ability to escape local optima.

[0148] (2) Convergence analysis of objective function 2.

[0149] like Figure 6 As shown, NSGA-III also demonstrates superior optimization performance on the second objective, achieving a final cost of 28,000, 12% lower than NSGA-II. The convergence curves show that both algorithms improve rapidly in the initial stage (generations 0–40), but NSGA-III declines faster during this critical phase. More significant differences emerge in the mid-to-late stages (generations 60–100), where NSGA-III maintains a stable optimization trend and outperforms NSGA-II in terms of average improvement per generation.

[0150] (3) Pareto frontier characteristics analysis.

[0151] like Figure 7 As shown, the analysis of the distribution of non-dominated solutions reveals significant differences between the two algorithms in terms of solution quality and distribution pattern. NSGA-III obtained 23 high-quality non-dominated solutions with a more compact distribution, closer to the true Pareto front; while NSGA-II's solution set distribution is more dispersed, and its convergence performance is poorer. The reference point selection mechanism in NSGA-III effectively promotes convergence towards the Pareto front while maintaining population diversity.

[0152] (4) Performance advantage analysis.

[0153] Experimental results clearly demonstrate that NSGA-III outperforms NSGA-II in both convergence speed and multi-objective balancing capability. This performance improvement is primarily attributed to its innovative reference-point-based selection mechanism, specifically reflected in:

[0154] First, it can more effectively maintain diversity through systematic adaptation of reference points;

[0155] Second, it can rely on elite retention operations to create stronger convergence pressure;

[0156] Third, it can better balance global exploration and local development by leveraging niche conservation strategies.

[0157] These improvements enable NSGA-III to overcome the limitations of NSGA-II in multi-objective optimization problems, particularly in approaching the true Pareto front faster while maintaining population diversity. Experimental results demonstrate that NSGA-III exhibits superior overall performance for complex engineering optimization problems requiring balance among multiple conflicting objectives.

[0158] To further quantitatively evaluate the performance differences between algorithms, this paper uses generational distance (GD), spacing index (SP), inverse generational distance (IGD), and hypervolume (HV) as evaluation metrics. Figure 8 As shown, Figure 6A histogram comparison of the two algorithms on these metrics is presented. The results clearly show that NSGA-III outperforms NSGA-II in all four metrics. Specifically, NSGA-III has a lower GD value, indicating that its solution set is closer to the true Pareto front; a lower IGD value, indicating that it achieves a better balance between convergence and diversity; a lower SP value, indicating that its solution set is more uniformly distributed; and a higher HV value, further indicating that it has a wider coverage in the objective space and higher overall solution set quality. These results are consistent with theoretical expectations because NSGA-III introduces a reference point-guided selection mechanism and a niche preservation strategy on the basis of NSGA-II, enabling it to achieve better diversity and convergence simultaneously in multi-objective optimization problems.

[0159] S73 Changes in the kill web before and after the appearance of time-sensitive targets

[0160] like Figure 9 and Figure 10 As shown, Figure 9 This is a diagram showing the pre-reconstruction kill network structure before the appearance of time-sensitive targets. Figure 10 This is a diagram of the kill network structure after reconstruction following the appearance of a time-sensitive target. Figure 9 and Figure 10 The diagram illustrates the structural changes in the kill net before and after the appearance of time-sensitive targets (cruise missiles 1 to 5), visually demonstrating the adaptive adjustment capability of the proposed reconstruction method in dynamic combat environments. The following reconstruction features are clearly visible in the diagram:

[0161] First, each newly emerging time-sensitive target was rapidly assigned corresponding interception resources, as indicated by the red line in the diagram. This demonstrates that the reconstructed kill network has brought these high-threat targets within its fire coverage and can respond effectively in the shortest possible time.

[0162] Secondly, as some equipment resources were reallocated to address new time-sensitive targets, the original attack plans were also partially adjusted to optimize overall combat effectiveness and maintain resource balance. For example, the Elf UAV_3 originally allocated to Missile-1_11 was reassigned to Missile-3_3 after reconstruction. This adjustment not only reflects the flexibility of resource reallocation but also demonstrates the algorithm's collaborative optimization capabilities under multiple constraints such as strike effectiveness, resource cost, and time requirements.

[0163] Overall, while maintaining the basic stability of the original system architecture, the reconstructed kill network significantly improved its response capability to sudden time-sensitive targets and its overall combat effectiveness through local link adjustments and resource reallocation.

Claims

1. A system kill network reconstruction method for time-sensitive target handling, characterized in that, Includes the following steps: S1: Matrix-based state modeling of the kill network in the combat system The Red Force's combat equipment is mapped to six functional sets based on the F2T2EA functional chain, and equipment attribute sets for each functional link are established; based on the activity logic permissions between equipment, a communication relationship matrix describing the communication topology is constructed. Based on the task allocation logic, an allocation matrix describing the mapping relationship between equipment and targets is constructed to complete the formal representation of the system kill network state. S2: Dynamic Extended Representation of Time-Sensitive Target Events When a high-threat time-sensitive target is observed, the newly added target is merged with the original target set to form an extended target set to be dealt with. S3: Establish and reconstruct decision variables The decision variable matrix is ​​reconstructed, and the allocation state of equipment at each stage to different targets is mapped by integer encoding, thus transforming the reconstruction problem into a search problem in a high-dimensional decision space. S4: Construction of a Multi-Objective Reconstruction Optimization Model Construct a mathematical optimization model with the objective of maximizing strike effectiveness as objective 1 and minimizing equipment usage cost as objective 2; S5: Construction of Logical Constraints for Multidimensional Conditions The decision space is logically limited by constructing constraints including equipment capability constraints, link communication constraints, and kill chain closure constraints. S6: Multi-objective cooperative optimization based on NSGA-III Based on the integer encoding vector, an initial population is generated, and the NSGA-III algorithm is used to find the Pareto optimal solution set in the multi-objective search space through non-dominated sorting and reference point adaptive mechanism. S7: Optimal Decision Output Based on preset performance and cost preference weights, the system selects the execution plan from the Pareto optimal plan, updates the configuration of the original kill network, and enables rapid inclusion and coordinated strikes against time-sensitive targets.

2. The system kill network reconstruction method according to claim 1, characterized in that, In step S1, the six types of functions are: discovery, location, tracking, aiming, engagement, and evaluation.

3. The system kill network reconstruction method according to claim 1, characterized in that, In step S4, the objective function of the kill net multi-target reconstruction optimization model includes: S41 maximizes strike effectiveness in, Indicates the first The value coefficient of each objective Indicates surrounding the first The combined strike effectiveness of the kill chain formed by the targets; the combined strike effectiveness is jointly determined by the cooperative hit probability of the equipment in each stage of detection, location, tracking, aiming, engagement and assessment. S42 Minimize Equipment Usage Costs in, , , , , and Used to filter links that have changed in the kill network after reconstruction; , , , , and These represent the cost vectors corresponding to the detection equipment, positioning equipment, tracking equipment, aiming equipment, engagement equipment, and evaluation equipment, respectively; the equipment usage cost includes the platform usage cost and the ammunition consumption cost of the strike equipment.

4. The system kill network reconstruction method according to claim 1, characterized in that, In step S5, the constraints include the following: S51 Equipment Capability Constraints This is used to limit the participation of a single piece of equipment in the coordinated handling of multiple targets, provided that it does not exceed its multi-target processing capability; in, , , , , and They represent the first The first discovered equipment, the first The first positioning device, the first The first tracking device, the first The first aiming device, the first The combat equipment and the first The assessment equipment's multi-target combat capability; S52 link communication constraints This is used to ensure that there is an effective communication link between adjacent functional equipment so that they can jointly form a kill chain. For example, consider the communication relationship constraints between the detection equipment and the positioning equipment: If , indicating the first The first discovered equipment and the first If there is no communication link between the positioning devices, then there is ,Right now and At least one of them must be 0; this means that neither can be simultaneously positive for the first. Attacks are carried out on individual targets, thus failing to form an effective kill chain; if This indicates that there is a communication relationship between the two, then we have ,Right now and They can all take 1 at the same time, thus jointly participating in the process of determining the first... The strike on a single target; S53 Kill Chain Closure Constraint This is used to ensure that a complete "detection-location-tracking-aiming-engagement-assessment" closed loop is formed for each target, and to ensure that the target damage effect meets the preset threshold requirements; in, , , , , and Representing the target The joint detection accuracy, joint positioning accuracy, joint tracking accuracy, joint aiming accuracy, joint engagement accuracy, and joint assessment accuracy; Indicates the first The damage threshold of each target is set to ensure that every link in the kill chain is physically connected and logically sound.

5. The system kill network reconstruction method according to claim 1, characterized in that, In step S6, the NSGA-Ⅲ multi-objective optimization algorithm specifically includes: S61 Based on the integer encoding method described in step S3, randomly generate an initial population and conduct a feasibility assessment; S62 uses tournament selection, simulated binary crossover, and polynomial mutation operations to generate offspring populations. S63 performs a fast non-dominated sort after merging the parent and offspring populations. S64 normalizes the objective function and, using a pre-defined set of reference points, calculates the proximity score and diversity score of individuals to execute a survival scoring mechanism. S65 Repeat steps S62 to S64 until the termination condition is met, and output the Pareto optimal solution set.

Images