Iterative optimization method for collaborative interception combat system architecture

Abstract

The invention relates to a collaborative interception combat system architecture iterative optimization method, and belongs to the field of aerospace and system engineering. The invention provides a method for optimizing a collaborative interception combat system architecture by using a multi-objective optimization algorithm in order to solve the problems of low resource allocation efficiency and insufficient dynamic adaptability of an existing anti-guide interception combat system. According to the method, a top view based on a DoDAF framework is combined with an MATLAB interception simulation model, and an NSGA-3 multi-objective optimization algorithm is used for carrying out iterative optimization on a collaborative interception combat system architecture. According to the method, the process is simple, the optimal architecture suitable for different threat scenes is generated through dynamic resource allocation and collaborative interception decision, and rapid iterative optimization of the collaborative interception combat system architecture is achieved.

Description

Technical Field

[0001] This invention relates to the fields of aerospace and systems engineering, and more specifically, to an optimization method for a cooperative interception combat system architecture. Background Technology

[0002] With the ever-changing global strategic security landscape, the threat of ballistic missiles is intensifying, particularly their significantly enhanced precision strike capabilities against national security, critical military facilities, and civilian targets. Modern missile defense systems typically consist of various types of interceptor platforms, radars, satellite sensors, and command and control centers, each playing a crucial role in different phases of combat. However, with the increasing variety and quantity of equipment, the interdependencies and constraints between systems become increasingly complex. Therefore, the design of missile defense systems needs to optimize the coordination mechanisms, resource allocation, and decision-making processes among various components from an architecture perspective. This not only enhances the overall combat effectiveness of the system but also enables flexible responses to attacks of varying scales, making full use of existing resources.

[0003] When dealing with multiple threats, especially ballistic missile swarm attacks, traditional interception systems typically rely on pre-set, fixed interception strategies. These strategies often fail to account for changes in the attack context and the dynamic changes in system resources. Interceptor resources may not be optimally utilized due to inaccurate assessment of the number and priority of targets, and some important targets may even be overlooked. Therefore, these fixed strategies may not be able to respond flexibly to complex and high-density threats, significantly reducing overall defense effectiveness. To address these issues, cooperative interception systems need to introduce advanced multi-target optimization algorithms. By optimizing the anti-missile system and the dynamic allocation of interceptor resources, and achieving multi-platform, multi-level cooperative operations, the cooperative interception combat system can improve its ability to respond to swarm attacks, enhance defense effectiveness, maximize the system's resource advantages, and thus improve overall cooperative interception combat capabilities. Summary of the Invention

[0004] To address the aforementioned shortcomings, the purpose of this invention is to provide an iterative optimization method for a collaborative interception combat system architecture. For a typical collaborative interception combat scenario involving a single ballistic missile attack, based on the established DoDAF2.0 framework top-level model and a MATLAB platform interception simulation model, the applicability and effectiveness of the established collaborative interception combat system iterative optimization model are analyzed and verified through digital simulation programs. The decision variables and objective functions required for the iterative optimization of the system architecture are designed. Based on the joint simulation results, a multi-objective optimization algorithm is introduced to optimize the decision variables, thereby achieving rapid iterative optimization of the system architecture and finding the optimal system architecture under the current combat scenario, providing support for the design of a collaborative interception combat system. The specific steps are as follows:

[0005] Step 1: Establish the top-level view and bottom-level interception simulation model of the collaborative interception combat system architecture.

[0006] Step 2: Design the decision variables and objective function of the collaborative interception combat system architecture.

[0007] Step 3 runs the system architecture to obtain the initial scheme, and continuously generates new decision variables based on the NSGA-3 multi-objective optimization algorithm, iteratively running and optimizing the collaborative interception combat system architecture.

[0008] Preferably, in step (1), the collaborative interception combat system architecture includes equipment elements such as space-based satellites, ground radars, ground command and control centers, ground intelligence centers, interceptor missile launchers, and mid-to-terminal interceptors.

[0009] Preferably, in step (1), the architecture is designed based on the DoDAF2.0 architecture framework, describing the structural composition of the combat system, system functions, combat mission activities, and data interaction relationships between various combat equipment elements.

[0010] Preferably, in step (1), for the specific operational process designed for the collaborative interception combat system, the simulation model of the corresponding activity at each stage is integrated into the system architecture so that the system architecture can be iteratively optimized in the future.

[0011] Preferably, in step (2), the design of decision variables and objective functions in the collaborative interception system architecture is related to the combat process and combat activities in step (1).

[0012] Preferably, in step (2), when iteratively optimizing the collaborative interception system architecture using a multi-objective optimization algorithm, it is necessary to design decision variables for the multi-objective optimization algorithm. The decision variables include the interception system model, the number of interceptor missiles launched, and the interception target allocation scheme. The interception system model refers to the different mid- and terminal phase interceptor models of the interception system. The number of interceptor missiles launched refers to the number of interceptor missiles used by each interception system to intercept each incoming ballistic missile. The interception target allocation scheme is to allocate mid- and terminal phase interception systems for each incoming ballistic missile based on the defense zone priority, the distance of the missile from the defense zone, and the missile threat level.

[0013] Preferably, in step (2), when iteratively optimizing the cooperative interception system architecture using a multi-objective optimization algorithm, it is necessary to design the objective function of the multi-objective optimization algorithm. The objective function is driven by the effectiveness of the cooperative interception system, including the effectiveness of a single interception system, the interception success rate, and the interception cost. The interception success rate is determined by the interception probability of a single interceptor missile and the number of interceptor missiles. The interception cost depends on the equipment mobilized and consumed. The effectiveness of a single interception combat system is calculated using the ADC method. The total effectiveness of the cooperative interception system is calculated by integrating the effectiveness of a single interception system, the interception success rate, and the interception cost.

[0014] Preferably, in step (3), the NSGA-3 multi-objective optimization algorithm is used to design a multi-objective optimization simulation model in the system architecture and integrate it into the system architecture. By iteratively simulating the system architecture, the system performance, interception success rate and interception cost of each architecture scheme are obtained, thereby obtaining the solution set of the architecture scheme that satisfies Pareto optimality.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention is based on the forward-driven design of system analysis and evaluation from the top-level view model of the cooperative interception combat system to the bottom-level interception simulation model. It provides data input for the optimization algorithm of the system architecture, designs a set of decision variables and objective functions applicable to the cooperative interception combat system, and proposes an iterative optimization method to realize the iterative optimization of the system architecture. This enables the rapid and effective selection of the equipment system architecture design scheme that meets the performance requirements from a variety of equipment system architecture design schemes, which is more in line with the development of existing technologies and the background of anti-missile operations. Attached Figure Description

[0016] Figure 1 This is a flowchart of the iterative optimization method for the collaborative interception combat system architecture of the present invention.

[0017] Figure 2 This is a view and relationship diagram of the collaborative interception system architecture based on DODAF 2.0 of the present invention.

[0018] Figure 3 This is a flowchart of the collaborative interception operation of the present invention.

[0019] Figure 4 This is a diagram illustrating the effectiveness evaluation indicators of the collaborative interception combat system of the present invention.

[0020] Figure 5 The flowchart of the NSGA-3 multi-objective optimization algorithm of this invention is shown below.

[0021] Figure 6 This is a schematic diagram illustrating the multi-target algorithm optimization approach for the collaborative interception combat system architecture of this invention. Detailed Implementation

[0022] To make the content of this invention clearer and easier to understand, the content of this invention will be clearly and completely described below in conjunction with the embodiments and accompanying drawings.

[0023] The iterative optimization method for cooperative interception combat system architecture provided by this invention is as follows: Figure 1 As shown. The specific process is as follows:

[0024] Step S100: Establish a collaborative interception combat system architecture model.

[0025] Specifically, based on the operational process, the DoDAF2.0 architecture framework is used to describe, design, analyze, and construct the collaborative interception combat system architecture model from multiple perspectives, such as operational, system, service, and capability perspectives. Simultaneously, the architecture integrates a low-level interception simulation model, enabling the construction and optimization of the collaborative interception combat system to achieve continuous iterative improvement and performance enhancement based on actual operational needs and effectiveness indicators. This approach allows for more effective evaluation and enhancement of the overall combat capability of the system, ensuring its high efficiency and adaptability in a changing battlefield environment. Its main views and relationships are as follows: Figure 2 As shown.

[0026] Anti-missile interception combat simulation process as follows Figure 3 The specific process is as follows:

[0027] In step S101, when facing an incoming ballistic missile, our satellites and ground radar detect the position and velocity vector of the enemy ballistic missile. Since the trajectory of the incoming missile will vary depending on factors such as booster specific impulse, missile aerodynamic parameters, launch angle, and payload weight, and due to the large number of variable parameters, accurately predicting the trajectory requires obtaining many actual parameters of the enemy missile before the battle, which is quite difficult. Considering that the boost phase and reentry phase account for a very small proportion of the total time and range during the entire flight, the entire trajectory of the incoming missile can be approximated as an elliptical trajectory determined by the initial ground velocity and launch angle, with the equation of motion as follows:

[0028]

[0029] Among them, h M v represents the target ground height at any given time. M Let θ be the target velocity; γ be the angle between the target velocity and the local horizontal plane; θ1 be the target range angle. Assume the missile acquires an initial velocity v at the launch point. M0 The launch angle γ0 and the maximum range angle, according to the two-body theory, satisfy the following relationship:

[0030]

[0031] It is easy to see from the above formula that, for a given range, the condition for v0 to reach its minimum value is:

[0032]

[0033] In ballistic prediction, the enemy missile's maximum range is a relatively easy parameter to obtain. At this point, it flies along the minimum energy trajectory, and the missile's equivalent ground initial velocity can be obtained. Then, based on the actual distance between the protected target and the enemy's launch point, the corresponding launch angle is obtained. Finally, using the ground initial velocity and launch angle as initial conditions, the missile's motion parameters at any time t are calculated by extrapolation based on the system of equations.

[0034] Step S102 involves determining the predicted trajectory of an enemy ballistic missile. The ground intelligence center collects and processes data for threat assessment. Based on the target missile's flight trajectory and expected impact point, a suitable interceptor platform (land-based midcourse or terminal phase anti-missile interceptor system) is selected, available interceptors are deployed, and multi-layered interceptor systems are coordinated to ensure each interceptor has a clearly defined target assignment, avoiding resource waste. The ground command center then issues a launch command to the interceptor platform.

[0035] In step S103, the interceptor missile's flight process in the middle and terminal phases is divided into three stages: the initial stage, the active stage, and the passive stage. In the initial stage, in order to obtain a certain initial speed and altitude, the interceptor missile flies according to a predetermined program based on external indication information during the active stage. When the boost ends, it obtains the required speed and direction, and the warhead moves to the vicinity of the predetermined interception point under the action of gravity. Then, the kinetic kill warhead's infrared seeker is activated, guiding the warhead to perform limited maneuvers until the missile and the target meet.

[0036] To simplify the model, the trajectory of the mid-course interceptor is approximated as an elliptical trajectory with the Earth's center as the focus, thus transforming the complex trajectory planning problem into the classic Lambert problem.

[0037] The Lambert problem for interceptor trajectory planning is described as follows: In a geocentric coordinate system, given the position vector r1 = (r1 - r2) of the interceptor launch point Q1... 1x ,r 1y ,r 1z ), the position vector of the encounter point Q2 is r2=(r 2x ,r 2y ,r 2z ) and the flight time t of the interceptor missile f Plan an elliptical orbit with semi-major axis a, eccentricity m, and semi-major diameter p to achieve the interceptor missile being launched from point Q1 at time t1 and encountering the target at point Q2 at time t2.

[0038] The ballistic equations for the free flight phase of a ballistic missile target are as follows:

[0039]

[0040] In the formula, f is the true near-surface angle, and its formula for calculating the partial near-surface angle E is:

[0041]

[0042] The position vector and velocity vector after the interceptor missile is launched are obtained as follows:

[0043]

[0044] In the formula, u is the Earth's gravitational constant.

[0045] According to the equation of an ellipse, for a given two points, there are infinitely many elliptical trajectories passing through those two points. For mid-course interception, the constraint of flight time needs to be considered. Therefore, given the flight time of the interceptor missile, a unique trajectory passing through the interception point and the encounter point can be determined.

[0046] Step S104, terminal interception, employs proportional guidance for kinematic modeling of the interceptor missile. The principle is that when the interceptor missile intercepts the target, its velocity vector rotation angular velocity is always proportional to the target's linear rotation angular velocity. This method is widely used due to its advantages such as insensitivity to kinematic models and ease of engineering application. The specific principle is as follows:

[0047] The relative kinematic equations for proportional guidance are as follows:

[0048]

[0049] q=σ+η

[0050] q = σ T +η T

[0051]

[0052] Where r is the relative distance between the missile and the target, r = 0 when the missile hits the target, and q is the line-of-sight angle. T These are the angles between the missile's velocity vector and the baseline, respectively, and are referred to as the missile's and target's heading angles. η, η T These are the angles between the missile's and the target's velocity vectors and the line of sight, respectively, and are called the missile's and target's velocity vector lead angles. The scaling factor K must satisfy:

[0053]

[0054] First, the lower limit of the K value should satisfy the above formula, but at the same time, due to the limitation of missile normal overload, the K value is selected in the range of 3-5.

[0055] The above steps calculate and achieve terminal trajectory tracking of enemy ballistic missiles. When the enemy ballistic missile is in its terminal phase, interceptor missiles are launched at regular intervals to intercept it. The operation ends when all interceptor missiles have been launched or the enemy missile has been destroyed.

[0056] After the interception operation in step S105 is completed, the ground command and control center will conduct an effectiveness assessment of the coordinated interception combat system and save the results to facilitate subsequent iterative optimization of the system architecture.

[0057] Step S200: Design the decision variables and objective function of the collaborative interception combat system.

[0058] Specifically, the decision variables include three aspects: the type of interceptor system, the number of interceptor missiles launched, and the target allocation scheme. The objective function is driven by system effectiveness and includes three aspects: the effectiveness of a single interceptor system, the interception success rate, and the interception cost. The overall combat effectiveness calculation considers the interception success rate, the interception cost, and the interception effectiveness of the mid- / terminal interceptor system.

[0059] Regarding the selection of interception system models, assume that different interception systems have different levels of coordination. There are m sets of mid-course interception systems (M) and n sets of terminal interception systems (N). Mid-course interception must be executed; therefore, the appropriate system should be selected. set Terminal interception selection Sets, combination methods include kind.

[0060] Each midcourse intercept system fires the following number of missiles against a single target: Each terminal intercept system launches a certain number of missiles against a single target. The number of interceptor missiles affects the outcome of the strike; combinations include... kind.

[0061] The formula for calculating the target allocation scheme is as follows:

[0062] F z =I i G e L ie

[0063] Among them, I i Prioritizing the defense zone of incoming missile targets, G e As for the threat level of the incoming missile, L ie This refers to the distance between the incoming missile and the defense zone.

[0064] The target allocation scheme is determined using a roulette wheel method. Based on the above... One scheme is assigned to l enemy ballistic missiles, with a total of Allocation scheme.

[0065] The interception cost C of the collaborative interception system c The calculation formula is related to the type and quantity of interceptor missiles consumed:

[0066]

[0067] In the formula, C c (*) represents the cost of a single interceptor missile in the mid-to-terminal intercept system.

[0068] The interception success rate P of the collaborative interception system sv The calculation formula is:

[0069]

[0070] In the formula, P ht (*) represents the kill probability of a single interceptor missile in the mid- and terminal phase interception system.

[0071] Interception effectiveness E of a single mid-to-terminal interception system s The ADC algorithm is used for calculation. The ADC method defines equipment combat effectiveness as: a measure of the degree to which a system is expected to meet the requirements of a specific combat mission under given constraints; it is a function of availability (A), reliability (D), and capability (C). Its model is as follows:

[0072]

[0073] To assess the operational effectiveness of our interceptors, based on the fighter jet's combat process, we have compiled and summarized a system of indicators for evaluating our interception operational effectiveness, such as... Figure 4 As shown.

[0074] This patent divides the interception combat system into two subsystems: the interceptor platform and the mission payload (including information acquisition payload and firepower attack payload). Each subsystem may be in a normal or faulty state before executing the mission (in this article, faulty state refers to non-critical faults, where the interceptor weapon still has the ability to perform the mission, but its combat effectiveness will be weakened). Therefore, the availability vector is...

[0075] A = [a1, a2, a3, a4]

[0076] Where a1 represents the probability that both the interceptor platform and the mission payload subsystem are in normal condition when the mission begins; a2 represents the probability that the interceptor platform subsystem is normal and the mission payload subsystem is faulty when the mission begins; a3 represents the probability that the interceptor platform subsystem is faulty and the mission payload subsystem is normal when the mission begins; and a4 represents the probability that both the interceptor platform and the mission payload subsystem are in a faulty state when the mission begins. Once the system structure is determined, the probability of each operating state depends on the reliability and maintainability of the system and its components. Therefore, the probability formula for the system being in a normal or faulty state can be expressed as:

[0077]

[0078] In the formula, P zc and P gc T represents the probability that the system is in a normal and faulty state, respectively; MTBF The mean time between failures (MTBF) is the average operating time of a repairable system between two consecutive failures; T MTTRThe mean time to repair a system failure is the average time required for the system to recover from a failure.

[0079] The reliability D calculation of an interception combat system is a measure of the system's state at one or more instants during the mission, given the system's initial state at the start of the mission. Reliability matrix.

[0080]

[0081] In the formula, d ij (i,j = 1, 2, 3, 4) represents the state of the interceptor weapon during its mission, from state a. i Transform into state a j The probability of.

[0082] This patent, through analysis of the composition and operational process of an interception combat system, concludes that the system's capabilities primarily depend on three aspects: pre-launch preparation capability (C1), ground command and control system capability (C2), radar target designation capability (C3), and interceptor missile target attack capability (C4). The calculation methods for each capability indicator are as follows:

[0083] ①Pre-launch preparation capability

[0084] Pre-launch readiness capability C1 is determined by information acquisition capability, launch preparation time, and technical support capability, and is calculated using the following formula:

[0085]

[0086] In the formula, k 11 k 12 k 13 These are the weighting coefficients, and their sum is 1. C 11 C 12 C 13 These represent the expected values ​​for information acquisition capabilities, launch preparation time, and technical support capabilities, respectively.

[0087] ② Ground command and control system capabilities

[0088] Ground command and control system capability C2 consists of target threat assessment capability C 21 Information processing ability C 22 Reaction time C 23 The decision is made using the following formula:

[0089]

[0090] In the formula, k 21 k 22 k 23 These are the weighting coefficients, and their sum is 1. C′ 21 C′ 22 C′23 These are the expected values ​​for target threat assessment capability, information processing capability, and reaction time, respectively.

[0091] ③ Radar target designation capability

[0092] The radar target indication capability C3 is determined by the radar range C. 31 Detection accuracy C 32 Target recognition capability C 33 The decision is made using the following formula:

[0093]

[0094] In the formula, k 31 k 32 k 33 These are the weighting coefficients, and their sum is 1. C′ 31 C′ 32 C′ 33 These represent the expected values ​​for radar range, detection accuracy, and target recognition capability, respectively.

[0095] ④ Interceptor missile attack target capability

[0096] The interceptor missile's ability to attack targets is determined by the flight control capability of C4. 41 Anti-interference capability C 42 Guidance accuracy C 43 And hit damage ability C 44 The decision is made using the following formula:

[0097]

[0098] In the formula, k 41 k 42 k 43 k 44 These are the weighting coefficients, and their sum is 1. C 41 C′ 42 C′ 43 C′ 44 These represent the expected values ​​for flight control capability, anti-jamming capability, guidance accuracy, and hit-and-damage capability, respectively.

[0099] Based on the various performance indicators of the interception combat system, the weight coefficients k of the four types of capability indicators are determined using the Analytic Hierarchy Process (AHP). w The capability index value can be calculated as follows:

[0100] C = k w ·[C1C2C3C4]

[0101] The overall effectiveness of a coordinated interception system is determined by the effectiveness of individual interception systems, the interception success rate, and the interception cost. The calculation formula is:

[0102]

[0103] In the formula, E represents the total effectiveness of the coordinated interception combat system, and k is the proportionality coefficient. C represents the average performance of all interception systems. all The cost of all our interceptor missiles (including those not yet launched).

[0104] Step S300 uses the NSGA-3 multi-objective optimization algorithm to optimize the system architecture.

[0105] Specifically, in a collaborative interception combat system, the performance, number, and operational plans of interceptors significantly impact the overall capability and operational effectiveness of the system. Traditional operational effectiveness assessments may primarily focus on overall operational performance, but are insufficient to cover other important design requirements, such as cost-effectiveness and system sustainability. Therefore, a multi-objective optimization method based on system effectiveness and machine learning is employed to comprehensively evaluate and optimize the three key objectives of the collaborative interception combat system: interception system effectiveness, interception cost, and interception success rate.

[0106] The NSGA-3 multi-objective optimization algorithm flow is as follows: Figure 5 As shown, when performing population selection operations in multi-objective optimization problems, using crowding distance or clustering operators to ensure population diversity is computationally inefficient, resulting in high time and space complexity. The NSGA-III algorithm uses predefined reference points to guide the search direction. By pre-constructing uniformly distributed reference points to guide the algorithm's search direction, the algorithm tends to search for solutions closer to the reference points. This ensures that the generated Pareto front is also as uniformly distributed as possible, solving the problem that the Pareto front (the hyperplane formed by all reference points) is difficult to visualize in multi-objective optimization problems.

[0107] The schematic diagram illustrates the implementation process of iterative optimization of the cooperative interception combat system architecture using the NSGA-3 multi-objective optimization algorithm. Figure 6As shown, the specifics are as follows: In the first computational architecture, the initial architecture scheme is obtained through initial calculations. The objective function values ​​for measuring the architecture are calculated, corresponding to the interception system effectiveness, interception success rate, and interception cost, respectively, and added to the historical architecture schemes. Each scheme in the historical architecture schemes consists of three types of decision variables: interception system type, number of interceptor missiles launched, and interception target allocation scheme, as well as the corresponding operational effectiveness, interception success rate, and interception cost. Subsequently, the decision variables are optimized in the Pareto optimal solution set of the historical architecture schemes based on a multi-objective optimization algorithm. These optimized parameters will affect which architecture models are actually executed in the architecture view. A new architecture scheme is obtained through these new decision variables. Simulation is run to obtain the mid- and terminal interception system effectiveness, interception success rate, and interception cost corresponding to the new architecture scheme. The new architecture scheme of the collaborative interception system may not satisfy the Pareto optimal solution set, and multiple iterations are required to obtain a Pareto optimal architecture scheme solution set that satisfies multiple objective functions.

Claims

1. This invention proposes an iterative optimization method for a collaborative interception combat system architecture, the specific steps of which are as follows: (1) Establish a top-level view and a bottom-level interception simulation model of the collaborative interception combat system architecture. (2) Design the decision variables and objective function of the collaborative interception combat system architecture. (3) The initial scheme of the operating system architecture is obtained, and new decision variables are continuously generated based on the NSGA-3 multi-objective optimization algorithm. The collaborative interception combat system architecture is iteratively optimized.

2. The iterative optimization method for a collaborative interception combat system architecture as described in claim 1, characterized in that, In step (1), a collaborative interception combat system architecture model based on the DoDAF2.0 framework is involved, which integrates the top-level view with the bottom-level interception simulation model: the collaborative interception combat system architecture model based on the DoDAF2.0 framework includes a panoramic view, capability view, combat view, system view and data information view. At the same time, the mid- and terminal interception simulation code is integrated into the collaborative interception combat system architecture model based on the collaborative interception combat process.

3. The iterative optimization method for a collaborative interception combat system architecture as described in claim 1, characterized in that, In step (2), for the interception combat system architecture model to be optimized, decision variables including the type of interception system, the number of interceptor missiles launched, and the interception target allocation scheme are proposed.

4. The iterative optimization method for a collaborative interception combat system architecture as described in claim 1, characterized in that, In step (2), for the interception combat system architecture model to be optimized, an objective function is proposed, driven by the effectiveness of the collaborative interception system, including the effectiveness of individual mid- and terminal interception systems, interception success rate, and interception cost. The objective function aims to improve the effectiveness of the collaborative interception system while maximizing the interception success rate and minimizing resource consumption, thereby achieving a cost-benefit balance.

5. The iterative optimization method for a collaborative interception combat system architecture as described in claim 1, characterized in that, In step (3), new decision variables are generated based on the NSGA-3 multi-objective optimization algorithm, and the collaborative interception combat system architecture is iteratively optimized. The optimization process is driven by the effectiveness of the collaborative interception system, and the decision variables are updated multiple times to achieve the optimal collaborative interception combat system architecture scheme.

6. The iterative optimization method for a collaborative interception combat system architecture as described in claim 1, characterized in that, In step (3), the method further includes dynamically correcting and adjusting the simulation model based on real-time battlefield feedback data during the optimization process, so as to achieve iterative optimization of the collaborative interception combat system architecture based on real-time data.

Images