An equipment system construction method and system

Abstract

This invention belongs to the field of equipment system simulation modeling, and particularly relates to an equipment system construction method and system. The method first obtains an updated index set reflecting the dynamic changes in the type and state of the target object, and initializes a preset target-resource bidirectional update graph. With the dual optimization objectives of minimizing deployment cost and maximizing response efficiency range, resource allocation strategies are intelligently matched from the strategy library, and the strategies are mapped onto the bidirectional update graph in the task simulation platform for real-time deduction and closed-loop iteration. When the simulation results meet the preset effective performance and the deployment cost is minimized, the corresponding optimized deployment scheme is output and encapsulated into the final new equipment system. This invention achieves accurate perception of dynamic threats and adaptive construction with quantitative trade-offs.

Description

Technical Field

[0001] This invention belongs to the field of equipment system simulation modeling, and in particular relates to an equipment system construction method and system. Background Technology

[0002] In the construction of existing equipment systems, resource allocation and deployment plans are typically planned based on empirical judgments or offline analysis of target objects and their static relationships. For sets of target objects characterized by diversity, dynamism, and strong correlation, the continuous changes in their types, states, and relationships make it difficult for static or empirically based plans to adapt in real time, leading to significant fluctuations in system effectiveness. To maintain effectiveness, existing methods often tend to frequently initiate high-cost global resource adjustments or simply add more resources, lacking a mechanism for synergistic optimization between the effectiveness of dynamic system adjustments and the costs of configuration updates. Furthermore, existing methods generally lack a formal description of the inherent quantitative relationship between the dynamic evolution of target objects and resource allocation adjustments, and also lack a closed loop that embeds real-time simulation verification into the decision-making process. This results in systems prone to problems such as slow response, inefficient resource allocation, and insufficient reliability of adjustment decisions.

[0003] In summary, in scenarios where the target object is dynamically evolving and its combined impact may amplify rapidly, the technical problem that needs to be solved is how to reduce the overall response performance decline caused by insufficient dynamic adaptability of the system through quantitative modeling and closed-loop verification without exceeding the constraints of acceptable configuration costs. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention proposes a method and system for constructing an equipment system. The method first acquires a set of updated indicators reflecting the dynamic changes in the type and state of the target object, and initializes a preset target-resource bidirectional update graph. With the dual optimization objectives of minimizing deployment cost and maximizing response efficiency, resource allocation strategies are intelligently matched from a strategy library, and these strategies are mapped onto the bidirectional update graph within a task simulation platform for real-time simulation and closed-loop iteration. When the simulation results meet the preset effective performance and minimize deployment cost, the corresponding optimized deployment scheme is output and encapsulated into the final equipment system. This invention achieves accurate perception, quantitative trade-offs, and cost-effective adaptive construction of dynamic threats.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for constructing an equipment system, comprising:

[0007] Obtain the target object update indicator set, including the increase or decrease of the target object type and the target object status update;

[0008] Initialize a preset target-resource bidirectional update graph. On the target-resource bidirectional update graph, in response to the target object update index set, with the goal of minimizing the configuration update index in the resource configuration association graph and maximizing the response performance range on the target object association graph, match the corresponding resource allocation strategy in the preset resource allocation strategy library.

[0009] The target-resource bidirectional update graph includes a resource configuration association graph, a target object association graph, and a mapping relationship between resource configuration update indicators and the influence intensity and influence range of target objects; the response effectiveness range is characterized by the change in the influence intensity and influence range of the target objects; the configuration update indicator is characterized by the cost corresponding to each resource configuration update.

[0010] The matched resource allocation strategy is mapped to the target-resource bidirectional update graph within the configured task simulation platform for real-time simulation adjustment. When the response performance range meets the preset response performance range index and the configuration update index is minimized, the resource configuration association graph under the corresponding conditions is obtained and encapsulated into an equipment construction system.

[0011] Specifically, the process of constructing the target object association graph includes:

[0012] Obtain the initial target object information and corresponding deployment association information; the target object information includes deployment location information, tag information, and single influence intensity value;

[0013] Based on the single impact intensity value of the target object and its label and corresponding deployment association information, combined with the association analysis algorithm, a sequence of target object association relationships and a sequence of association impact indicators are obtained. The association impact indicators are used to measure the joint impact intensity value of the target to be protected after at least two related target objects are combined. The sequence of association impact indicators includes: the joint impact intensity value for the target to be protected, and the contribution weight of each target node that makes up the joint impact intensity value.

[0014] Using the location information of each target object as the target node and the sequence of relationships between target objects as the connection relationships, combined with a graph algorithm, an initial target object graph is obtained.

[0015] Specifically, the process of constructing the target object association graph also includes:

[0016] The sequence of related influence indicators is mapped to the initial target object graph by connecting the contribution weight of the single influence intensity value of the target object label and each target node to the joint influence intensity value under the corresponding target node, thus obtaining the initial target object association graph.

[0017] Obtain the historical target object status update information of each target node in the initial target object graph, as well as the adjustment values ​​of the target object association relationship and the joint influence intensity value under the corresponding timestamp;

[0018] Based on the historical target object state update information of each target node and combined with the support vector machine, a target state prediction function is obtained, and the target state prediction function is attached to the corresponding target node with the target node label.

[0019] Specifically, the process of constructing the target object association graph also includes:

[0020] Based on the historical target object status update information of each target node and the target object association adjustment value under the corresponding timestamp, the association influence adjustment function between two target nodes with an association relationship and the association influence strength adjustment function between the target node and the joint influence strength value are constructed through the radial kernel function.

[0021] The association influence adjustment function is embedded into the association connection relationship between corresponding pairs of target nodes according to the label of the target node; at the same time, the association influence strength adjustment function is mapped to the contribution weight of each target node to the joint influence strength value, so as to update the association contribution connection in real time.

[0022] Specifically, the process of constructing the target object association graph also includes:

[0023] The initial target object association graph is collected in real time to obtain the real-time state graph and the target object state prediction graph at each time point; the target object state prediction graph is constructed by the target state prediction function attached to each target node in the initial target object association graph and the predicted state change of the target object within a preset time length.

[0024] Based on the real-time status diagram of the target object at adjacent time points, the target object type is increased or decreased to determine the increase or decrease target node set and the corresponding increase or decrease target node set status information.

[0025] Specifically, the process of constructing the target object association graph also includes:

[0026] The set of target nodes to be added or removed, the set of target nodes to be retained, and the corresponding real-time status information are fed back to the initial target object association graph. The target object association sequence and the association influence index sequence are adjusted and updated in real time to obtain the first updated real-time target object association graph.

[0027] Simultaneously, the target object state prediction map is fed back to the initial target object association map. Taking the timestamp corresponding to the first updated real-time target object association map as the starting point and the preset time length as the time axis, a hierarchically updated target object association map is constructed.

[0028] Specifically, the process of constructing the target-resource bidirectional update graph includes:

[0029] The location of each piece of equipment deployed in the initial target object graph is taken as the initial resource node in the resource configuration association graph. The collaborative response association information between the initial resource nodes is combined with the collaborative response association degree obtained by the association analysis algorithm as the connection to construct the initial resource configuration association graph and obtain the initial configuration update index corresponding to the initial resource configuration association graph.

[0030] After the initial resource configuration association graph is deployed, monitor and obtain the joint influence intensity change value and single influence intensity change value and the corresponding change direction in the initial target object graph;

[0031] Based on a comprehensive evaluation algorithm that combines the change value of joint influence intensity with the change value of single influence intensity and the corresponding change direction, the response effectiveness range and the corresponding effectiveness suppression evaluation value are obtained; the effectiveness suppression evaluation value is constructed by the change amount of the largest target node that meets the joint influence intensity threshold before and after equipment deployment.

[0032] Specifically, the process of constructing the target-resource bidirectional update graph also includes:

[0033] The system collects the performance change data of each resource node in the resource configuration association graph at continuous time points, as well as the change in the response performance range of each resource node at consecutive timestamps and the change in the corresponding performance suppression evaluation value and direction. The collected data is then filtered using a distributed Kalman filter algorithm.

[0034] Based on the preprocessed performance change data, the change in response performance range, and the corresponding change in performance suppression evaluation value, combined with the convolutional radial kernel function, a performance-performance suppression mapping function is obtained, and mapped to the corresponding resource node and target node according to the labels of resource node and target node.

[0035] By combining the configuration update index corresponding to each equipment update in the historical target-resource bidirectional update graph, the change in the response performance range and the change in the corresponding performance suppression evaluation value of the target object association graph after the deployment update, and the change direction of the change, a linear algorithm is used to obtain the configuration update index-performance suppression mapping function, and then the resource node and target node are mapped to the corresponding resource node and target node according to the labels of the resource node and target node.

[0036] Define a goal-resource priority constraint game model, including a resource allocation optimization party and a goal object dynamic evolution party. The decision variable of the resource allocation optimization party is the resource allocation strategy of the resource allocation association graph, and the state variable of the goal object dynamic evolution party is the state of the goal node at each time point in the hierarchically updated goal object association graph.

[0037] Set single-impact safety thresholds and joint-impact safety thresholds, and use the set single-impact safety thresholds and joint-impact safety thresholds, response effectiveness range, effectiveness suppression evaluation value, and adjustable total cost value to construct a game constraint space, and map it to the target-resource priority constraint game model;

[0038] Specifically, the process of constructing the target-resource bidirectional update graph also includes:

[0039] Based on the performance-efficiency inhibition mapping function and the configuration update index-efficiency inhibition mapping function mapped in the target-resource bidirectional update graph, a game payoff index function is constructed and mapped to the target-resource priority constraint game model.

[0040] Using the strategies in the resource allocation strategy library as the solution space, and aiming to maximize the game payoff index function, the target-resource priority constraint game model is solved by combining a heuristic search algorithm to obtain the optimal resource allocation strategy sequence that maximizes the cumulative payoff.

[0041] The first resource allocation strategy in the optimal resource allocation strategy sequence obtained by the solution is mapped to the target-resource bidirectional update graph in the configured task simulation platform for execution and verification.

[0042] When the simulation confirms the execution of the resource allocation strategy, and the dynamic evolution of the target object satisfies the game constraint space, and the configuration update index of the resource allocation strategy is the smallest among all alternative strategies that satisfy the game constraint space, the resource configuration association graph at the current moment is obtained and encapsulated and output as an equipment construction system.

[0043] An equipment system construction system, comprising:

[0044] The data acquisition module is used to obtain the target object update indicator set, including the increase or decrease of the target object type and the target object status update.

[0045] The matching module is used to initialize a preset target-resource bidirectional update graph. On the target-resource bidirectional update graph, in response to the target object update index set, with the goal of minimizing the configuration update index in the resource configuration association graph and maximizing the response performance range on the target object association graph, the module matches the corresponding resource allocation strategy in the preset resource allocation strategy library.

[0046] The target-resource bidirectional update graph includes a resource configuration association graph, a target object association graph, and a mapping relationship between resource configuration update indicators and the influence intensity and influence range of target objects; the response effectiveness range is characterized by the change in the influence intensity and influence range of the target objects; the configuration update indicator is characterized by the cost corresponding to each resource configuration update.

[0047] The simulation module is used to map the matched resource allocation strategy onto the target-resource bidirectional update graph within the configured task simulation platform for real-time simulation adjustment. When the response performance range meets the preset response performance range index and the configuration update index is minimized, the resource configuration association graph under the corresponding conditions is obtained and encapsulated into an equipment construction system.

[0048] Compared with the prior art, the beneficial effects of the present invention are:

[0049] This invention addresses the shortcomings of existing technologies by constructing a target-resource bidirectional update graph and a target-resource priority constraint game model, achieving a paradigm shift from static planning to dynamic adaptive construction of equipment systems. In particular, through real-time updated threat correlation prediction graphs and bidirectional update graphs, the system achieves accurate perception and quantitative characterization of dynamic threats, solving the problems of delayed response and static analysis inherent in traditional methods. Secondly, by introducing a game model with security as the absolute priority and cost as the optimization objective, strategy optimization is performed in a simulation environment, ensuring that decisions achieve economic optimality while meeting the highest security constraints, overcoming the dilemma of imbalance between response and efficiency. Finally, this application significantly improves the system's targeting, adaptability, and overall combat effectiveness by transforming system construction from an experience-dependent offline activity into a data- and model-driven adaptive process. Attached Figure Description

[0050] Figure 1 Here is a flowchart of an equipment system construction method provided in Embodiment 1 of the present invention;

[0051] Figure 2 This is a system module diagram of an equipment system construction provided in Embodiment 2 of the present invention. Detailed Implementation

[0052] Example 1

[0053] To facilitate understanding of the background and applicable scenarios of this application, the specific implementation methods are described below in conjunction with the task of resource allocation in modern dynamic systems.

[0054] In the task of building and adaptively adjusting systems that are multi-domain heterogeneous and dynamically coupled, the target objects and system resources exhibit complex characteristics that are multi-dimensional, highly dynamic, and strongly correlated:

[0055] Within the same task area, there are both routine targets and emerging targets that are dynamically introduced. Resource allocation needs to strike a balance between cost control and response effectiveness, requiring both timely responses to current targets and forward-looking adaptation to their evolutionary trends. It is also constrained by multiple factors, including resource performance degradation, deployment timing constraints, and overall allocation limitations. In practice, decision-making units typically coordinate various heterogeneous resources to form a collaborative system based on pre-set plans and real-time information. The target situation is usually in a dynamic state, with new targets emerging, existing targets relocating, and collaboration patterns changing between targets. Core indicators such as resource allocation costs, response efficiency, and action timeliness fluctuate accordingly. This complexity is further amplified by a typical phenomenon: the same target node may be matched with multiple resource allocation strategies under different analytical dimensions or time scales, leading to cross-resource and cross-time-series strategy conflicts and efficiency redundancy.

[0056] Existing methods generally follow two paths: First, they prioritize static target analysis, constructing a fixed resource allocation relationship diagram first, and then adapting to target changes through subsequent local adjustments; second, they focus on experience-driven strategy matching, directly selecting schemes similar to target characteristics from a pre-set strategy library. The former often results in high adjustment costs when facing changes in target types or sudden state shifts, commonly exhibiting problems such as resource-target mismatch, insufficient response range, and wasted configuration costs. While the latter possesses rapid response potential, its strategy adaptation accuracy and performance stability are insufficient to meet high-reliability configuration requirements when faced with complex constraints such as changes in target relationships, dynamic degradation of resource performance, and conflicts in multi-target optimization. Furthermore, it struggles to achieve dynamic optimization when introducing hierarchical target prediction information. The common limitation of both approaches is that once the dynamic association and quantified mapping of targets and resources are postponed to the strategy verification stage, the resource layout formed based on early, fixed judgments is already determined, making it difficult to fundamentally avoid resource waste and performance loss in subsequent adjustments.

[0057] To resolve the above issues, please refer to Figure 1 The present invention provides the following embodiment: a method for constructing an equipment system, comprising the following steps:

[0058] S1. Obtain the target object update indicator set, including the increase or decrease of the target object type and the target object status update;

[0059] S2. Initialize a preset target-resource bidirectional update graph. On the target-resource bidirectional update graph, in response to the target object update index set, with the goal of minimizing the configuration update index in the resource configuration association graph and maximizing the response performance range on the target object association graph, match the corresponding resource allocation strategy in the preset resource allocation strategy library.

[0060] It should be further explained that the target-resource bidirectional update graph in this embodiment includes a resource configuration association graph, a target object association graph, and a mapping relationship between resource configuration update indicators and the influence intensity and influence range of target objects. The response effectiveness range is characterized by the change in the influence intensity and influence range of the target object. The configuration update indicator is characterized by the cost corresponding to each resource configuration update. The target-resource bidirectional update graph is the core data model for constructing a dynamic adaptive resource system in this application. It couples the "target object association graph" representing the target environment with the resource configuration association graph representing one's own resources, and embeds a series of quantitative mapping functions and game optimization mechanisms to form a closed-loop perception-decision-evaluation framework. Specifically, the target object association graph uses target objects as nodes and their collaborative relationships as edges, and attaches dynamic models such as target state prediction functions and association influence adjustment functions to quantify the evolution of their single influence intensity, association intensity, and joint influence intensity. The resource configuration association graph uses resource units as nodes and collaborative response relationships as edges, and associates performance-effectiveness suppression mapping functions and configuration update indicator-effectiveness suppression mapping functions to characterize resource performance, collaborative effectiveness, and adjustment costs. Two subgraphs are connected to a pre-defined target-resource priority constraint game model via a mapping function. This allows the system to dynamically solve and verify the optimal resource allocation strategy based on real-time perception and hierarchical prediction of the target situation, while satisfying the triple constraints of security, efficiency, and cost. Ultimately, it outputs an executable resource configuration association graph as the system construction scheme, thereby achieving continuous, accurate, and efficient resource adaptation to dynamic targets. In this embodiment, the cost refers to the total consumption of various resources that must be comprehensively considered when dynamically adjusting the resource allocation system. It directly reflects economic expenditures such as equipment procurement, transportation, deployment, and maintenance, and also includes the time window required for deployment reorganization, the potential loss of benefits due to abandoning other schemes by choosing the current strategy, and the inherent consumption such as equipment performance degradation and fluctuations in collaborative efficiency caused by frequent adjustments. This comprehensive consumption is quantified as a "configuration update index" and used as a core constraint condition alongside the security threshold and efficiency target. This index is input into the target-resource priority constraint game model to ensure that the optimization algorithm solves an equilibrium strategy with both high adversarial effectiveness and high feasibility within strict resource boundaries.

[0061] It should be further noted that the preset resource allocation strategy library in this embodiment was constructed by the technicians of this embodiment based on historical resource allocation strategies and a graph database;

[0062] S3. Map the matched resource allocation strategy onto the target-resource bidirectional update graph within the configured task simulation platform for real-time simulation adjustment. When the response performance range meets the preset response performance range index and the configuration update index is minimized, obtain the resource configuration association graph under the corresponding conditions and encapsulate it into an equipment construction system.

[0063] In the equipment construction system constructed in this embodiment, "response effectiveness range," "effectiveness suppression assessment value," and "configuration update index" are three core metrics used for quantitative evaluation and optimization decisions. "Response effectiveness range" is a spatial effectiveness index, measured by calculating the geographical coverage area formed by all target nodes affected by the resource allocation scheme and suppressed below a preset safety threshold after the scheme takes effect. "Effectiveness suppression assessment value" is a collaborative disintegration effectiveness index, obtained by comparing the change in the size of the largest collaborative node group in the target object association graph that can form a joint influence intensity exceeding the safety threshold before and after system adjustment. Its integer value intuitively quantifies the degree of damage the resource allocation scheme inflicts on the target party's high-end system collaborative combat capabilities. "Configuration update index," as an economic constraint index, is used to comprehensively quantify the total cost incurred in each system adjustment, covering economic costs, time windows, and opportunity costs. These three indicators together constitute the quantitative basis for evaluating system effectiveness and driving the target-resource game model to perform multi-objective optimization solutions.

[0064] In this embodiment, the target state prediction function, the association influence adjustment function, and the association influence strength adjustment function embedded in the target object association graph are responsible for forward-lookingly predicting the state transition of a single target node, quantifying in real time the changes in the strength of the collaborative relationship between target nodes, and accurately calculating the dynamic effect of the above changes on the overall joint influence strength of the target group. Meanwhile, the performance-efficiency suppression mapping function and the configuration update index-efficiency suppression mapping function embedded in the bidirectional update graph establish quantitative prediction relationships between the performance degradation of a resource unit itself and its counter-efficiency output, and between the comprehensive cost of system adjustment and the expected efficiency gain. Based on this, a target-resource priority constraint game model is defined, formalizing the dynamically evolving target environment and the proactively optimized resource allocation formula into two game players. Through a mathematical framework aimed at maximizing long-term cumulative efficiency gains, within a constraint space composed of a safety baseline, efficiency target, and resource ceiling, the model automatically optimizes and verifies strategies, thereby driving the entire resource system to achieve continuous and accurate adaptation and optimal efficiency in a dynamic environment.

[0065] It should be further explained that the construction process of the target object association graph in this embodiment includes:

[0066] S201. Obtain the initial target object information and corresponding deployment association information; the target object information includes deployment location information, tag information, and single influence strength value; in this embodiment, the deployment location information of various target objects is collected in the initial stage, the corresponding category tags and their respective single influence strengths are clarified, and the deployment association between these target objects is obtained simultaneously, laying the foundation for the subsequent construction of the target object association graph, and realizing the systematic sorting and integration of information; it should be further noted that the single influence strength value in this embodiment is evaluated and labeled by those skilled in the art based on historical experience;

[0067] S202. An expert evaluation system correlation analysis algorithm is configured based on the single influence intensity value of the target object, its label, and the corresponding deployment association information to obtain a target object association sequence and an association influence index sequence. The association influence index is used to measure the joint influence intensity value of the target to be protected after at least two related target objects are combined. The association influence index sequence includes: the joint influence intensity value for the target to be protected, and the contribution weight of each target node that makes up the joint influence intensity value.

[0068] It should be further explained that the specific meaning of the correlation impact index in this embodiment is to combine the category attributes, individual impact strength, and mutual deployment correlation characteristics of multiple target objects with deployment correlation relationships. Through a correlation analysis algorithm incorporating expert evaluation experience, it comprehensively quantifies the overall threat level to the predetermined protected target when these target objects cooperate or conduct joint operations. At the same time, it clarifies the proportion of the threat strength of each target object in the overall threat level. For example, in a typical cooperative scenario including a mobile penetration unit and an electromagnetic suppression unit, the two have a significant functional cooperation relationship. The mobile penetration unit mainly generates directional physical suppression impact, while the electromagnetic suppression unit mainly generates regional signal interference impact. When the two cooperate, the electromagnetic suppression unit creates favorable conditions for the mobile penetration unit's advance by interfering with key systems, thereby generating a synergistic enhancement effect of "1+1>2". The correlation analysis algorithm in this process quantifies the joint impact strength of the protected party under this cooperative mode and analyzes the contribution ratio of the physical suppression capability of the mobile penetration unit and the signal interference capability of the electromagnetic suppression unit in the joint strength, thereby clearly defining the primary and secondary roles and synergistic value of different types of target objects in system-wide operations.

[0069] S203. Using the location information of each target object as a target node and the sequence of target object associations as the association connections, a graph algorithm is used to obtain an initial target object graph. For example, in a typical multi-domain collaborative mission area, the forward deployment position of the mobile penetration unit, the preset working position of the electronic countermeasures unit, and the normalized working airspace of the reconnaissance and detection unit are each taken as independent target nodes. The functional collaboration relationships between these three constitute a target object association sequence. This sequence clearly defines the association logic whereby the electronic countermeasures unit provides electromagnetic environment control support for the penetration unit's actions, and the reconnaissance and detection unit provides real-time situational awareness and information support for the former two. Using this association sequence as the connection relationship between nodes, a graph algorithm is used to structurally model the spatial distribution and association logic of each target node, ultimately forming an initial target object graph that clearly presents the location distribution and collaborative association situation of various target objects.

[0070] S204. The sequence of associated impact indicators is mapped onto the initial target object graph, using the contribution weight of the single impact strength value of each target node to the joint impact strength value under the corresponding target node as the association contribution connection, to obtain the initial target object association graph. It should be further explained that the motivation for doing this in this embodiment is to compensate for the deficiency of the initial target object graph, which only presents the location and association relationship of target nodes and lacks quantitative threat information, thus providing data support for subsequent accurate assessment of joint impact strength and matching of resource allocation strategies. The principle is to establish a correspondence between associated impact indicators and target nodes based on the target object labels, and to connect the contribution weight of the single impact strength of each target node to the joint threat. The contribution weight and the joint impact strength value under different numbers of nodes are mapped as quantitative attributes to the nodes and associated connections of the initial target object graph, so that the graph structure has both spatial association and quantitative threat characteristics. For example, in the initial target object graph of the operation, the labels of ground armored group "firepower assault", vehicle-mounted electronic jamming unit "electromagnetic suppression" and UAV reconnaissance unit "target guidance" are used as the association basis. The impact strength values ​​of each pair and the three in combination, as well as the contribution weight of each node's single threat to different joint scenarios, are mapped to the corresponding target nodes and the collaborative connections between nodes, forming an initial target object association graph that allows for intuitive viewing of the quantitative threat level and collaborative contribution ratio of each target node.

[0071] S205. Obtain the historical target object status update information of each target node in the initial target object map and the adjustment values ​​of the target object association relationship and the joint influence intensity value under the corresponding timestamp; for example, in the historical data records corresponding to the initial target object map, collect the status update information of each target node, such as the deployment area migration of the ground armored group, the working power switching of the vehicle-mounted electronic jamming unit, and the patrol airspace adjustment of the UAV reconnaissance unit, and simultaneously extract the timestamp corresponding to each status update, as well as the adjustment of the strength of the tactical coordination association between the three types of target nodes under the timestamp, and the change of the overall threat level under different joint scenarios, to form a complete historical update dataset, providing data support for the subsequent construction of prediction and adjustment functions.

[0072] S206. Based on the historical target object state update information of each target node and combined with a support vector machine, a target state prediction function is obtained, and the target state prediction function is attached to the corresponding target node with a target node label. It should be further explained that the motivation for constructing the target state prediction function in this embodiment is to capture the future state change trend of each target node in advance, get rid of the static limitations of traditional threat analysis, provide data support for the forward-looking adjustment of resource allocation strategies, and ensure that the equipment system can adapt to threat evolution in advance. The principle is to use a support vector machine to learn and train the historical state update information of each target node, explore the inherent laws of state change, generate a function model that can accurately predict the future state, and then establish the association between the function and the corresponding node through the target node label and attach it. Example: For historical deployment area migration data of ground armored groups, historical power switching data of vehicle-mounted electronic jamming units, and historical airspace adjustment data of UAV reconnaissance detachments, input them into support vector machines for model training to obtain prediction functions for future state changes of each node. Then, attach the deployment trend prediction function of ground armored groups, the power change prediction function of vehicle-mounted electronic jamming units, and the airspace adjustment prediction function of UAV reconnaissance detachments to each target node in the initial target object association graph with the corresponding node labels.

[0073] S207. Based on the historical target object status update information of each target node and the target object association adjustment value under the corresponding timestamp, a radial kernel function is used to construct the association influence adjustment function between two target nodes with an association relationship and the association influence strength adjustment function between the target node and the joint influence strength value. It should be further explained that the association influence adjustment function in this embodiment is used to quantify the dynamic influence degree of one node on the other node when the state of one of the two target nodes with an association relationship changes, providing a quantitative basis for the real-time adjustment of the association relationship between target nodes; the association influence adjustment function is used to quantify the dynamic effect of the state change of one or more target nodes on the overall joint influence strength value, supporting the accurate update of the joint influence strength. The motivation for constructing these two systems is to address the problem that traditional threat analysis cannot capture the dynamic influence between nodes and the nonlinear correlation between nodes and joint threats, breaking through the limitations of static analysis and providing core algorithmic support for the dynamic updating and hierarchical prediction of target object correlation graphs. The principle is to utilize the powerful nonlinear fitting capability of the radial kernel function to mine the potential patterns between target node state updates, correlation adjustment values, and joint influence intensity adjustment values ​​in historical data, and establish a quantitative mapping model among the three, so that the adjustment function can accurately reflect the correlation influence of various dynamic changes, providing reliable algorithmic support for the real-time updating and prediction of the subsequent threat situation.

[0074] Furthermore, in the combat scenario corresponding to this embodiment, when a new hypersonic weapon platform is added as a target node, the correlation impact adjustment function will immediately quantify the collaborative combat impact between the new node and the existing ground armored cluster and vehicle-mounted electronic jamming unit, dynamically update the tactical correlation strength among the three, and form a new target object correlation sequence. The correlation impact adjustment function will simultaneously calculate the overall impact strength change of the joint operation of the three types of target nodes after the addition of the new node, as well as the adjustment result of the contribution weight of each node's single threat to the joint threat, and update the correlation impact index sequence in real time. If the original UAV reconnaissance unit reduces the number of target nodes due to mission withdrawal, the two functions will quickly recalculate the correlation impact degree between the remaining ground armored cluster and vehicle-mounted electronic jamming unit, adjust the correlation sequence between the two, and recalculate the joint impact strength of the two nodes and their respective contribution weights, completing the real-time update of the correlation impact index sequence.

[0075] S208. The association influence adjustment function is embedded into the association connection relationship between corresponding pairs of target nodes according to the label of the target node; at the same time, the association influence strength adjustment function is mapped to the contribution weight of each target node to the joint influence strength value, for real-time updating of association contribution connections; the motivation of this process in this embodiment is to enable the target object association graph to have dynamic quantitative adjustment capability, so that the association influence between pairs of target nodes and the association effect between target nodes and the joint influence strength can be directly calculated in real time through the function in the graph structure, without rebuilding the graph to respond to changes in node state, improve the timeliness and accuracy of threat assessment, and establish a direct association between the function and the graph structure. To ensure the traceability of adjustment logic, for example, in the initial target object association graph of this embodiment, the association influence adjustment functions of ground armored groups and vehicle-mounted electronic jamming units, ground armored groups and UAV reconnaissance units, and vehicle-mounted electronic jamming units and UAV reconnaissance units are embedded into the tactical coordination connection relationship between pairs of nodes according to the labels of the corresponding nodes. At the same time, the association influence adjustment functions corresponding to the joint influence strength values ​​of each target node and the pairs and the three-way joint influence are mapped to the quantitative association connection between each node and the corresponding joint influence strength value, so that the graph structure can directly calculate the association relationship and joint influence strength adjustment result after the node state change in real time through the embedded functions.

[0076] S209. Real-time acquisition of the initial target object association graph to obtain the real-time status graph and target object status prediction graph at each time point; the target object status prediction graph is constructed by the target status prediction function attached to each target node in the initial target object association graph, predicting the state change of the target object within a preset time length; for example, in the combat scenario of this embodiment, real-time data acquisition is performed on the initial target object association graph, and the real-time status graph and target object status prediction graph are obtained synchronously at each time point: the real-time status graph accurately presents the real-time status of each target node, such as the current deployment position of the ground armored cluster, the working power of the vehicle-mounted electronic jamming unit, and the actual patrol airspace of the UAV reconnaissance unit at that time point; the target object status prediction graph is constructed by integrating these prediction results, based on the target status prediction function attached to each target node in the initial target object association graph, predicting the state change of the ground armored cluster's deployment area migration trend, the power adjustment range of the vehicle-mounted electronic jamming unit, and the patrol airspace change range of the UAV reconnaissance unit within a preset time length.

[0077] S210. Based on the real-time status diagram of the target object at adjacent time points, determine whether the target object type has increased or decreased, and obtain the set of target nodes that have increased or decreased and the corresponding status information of the set of target nodes that have increased or decreased. It should be further noted that the status information in this embodiment includes at least the tag information of the target object that has increased or decreased to clarify its type attribute, the deployment location information to define its operational space distribution, the single impact intensity value to characterize its basic threat intensity, and the potential association relationship characteristics with existing target nodes, so as to provide core basic data for subsequently integrating the set of target nodes that have increased or decreased into the target object association diagram and adjusting and updating the association relationship sequence and the association impact indicator sequence.

[0078] S211. The added or removed target node set and the retained target node set, along with their corresponding real-time status information, are fed back to the initial target object association graph. The target object association sequence and association influence index sequence are adjusted and updated in real time to obtain the first updated real-time target object association graph. It should be further explained that the dynamic update process of the target object association graph in this embodiment is a multi-step iterative reconstruction process driven by a quantization function and expert rules. Specifically, the system first receives information including the added or removed target nodes and their complete real-time status information. This information covers node labels, deployment coordinates, current single influence strength values, and predefined association feature vectors. Second, the system calls the pre-embedded association influence adjustment function in the graph to accurately calculate the quantified change value of the association strength between the newly added node and the existing nodes in the graph, or between the remaining nodes after deleting a node, due to factors such as spatial relationships and capability complementarity. Based on this, the system further calls the association influence adjustment function, combining the updated association strength with the real-time threat value of each node, to recalculate the new joint influence strength value generated by target node groups of different sizes and combinations, and dynamically allocates the contribution weight of each node in the joint threat. The above calculation process is all verified and logically optimized in real time through an association analysis algorithm configured with an expert evaluation system, thereby generating an updated and logically consistent sequence of target object association relationships and a sequence of association influence indicators. Finally, based on these two sets of optimized core sequences, the system automatically completes the real-time reconstruction of the graph topology and node edge attribute mapping, and outputs the first updated real-time target object association graph.

[0079] This specific example assumes that an electronic jamming vehicle node A and an anti-radiation missile node B already exist in the initial target object association graph. These two nodes have established a strong association due to typical effective soft and hard synergy. When the system detects a newly added high-altitude unmanned reconnaissance aircraft node C, the dynamic update process is initiated. The system first extracts complete state information of node C, including its tag, real-time coordinates, reconnaissance capability feature vector, and initial threat value. Next, the system calls a pre-defined association impact adjustment function in the graph to quantitatively analyze the potential interaction patterns between node C and node A, and between node C and node B. The analysis identifies that node C can provide precise target guidance for node A and provide strike effect assessment for node B. Based on this synergy logic, a moderate association value is calculated between node C and nodes A and B, respectively. Subsequently, the system calls the association impact adjustment function to incorporate node C into the threat combination assessment system. Based on the updated association strength, this function performs joint threat calculation on the combination of nodes A, B, and C. The calculation results show that the joint impact strength of the cooperative link formed by the three nodes—node C responsible for reconnaissance and location, node A implementing electronic jamming, and node B performing hard destruction—significantly exceeds that of the original combination consisting only of nodes A and B. At the same time, the function dynamically determines the specific contribution weight of node C in this new joint threat. Finally, based on the new association relationships and threat indicators, the system automatically completes the reconstruction of the graph structure: adding a new node C entity, establishing directed connection edges between node C and nodes A and B, and refreshing the association attributes and threat indicator sequences of all affected nodes in the graph, thereby completing a complete real-time update of the association graph.

[0080] S212. Simultaneously, the target object state prediction map is fed back to the initial target object association map. Taking the timestamp corresponding to the first updated real-time target object association map as the starting point and the preset time length as the time axis, a hierarchically updated target object association map is constructed.

[0081] It should be further explained that, in this embodiment, after feeding back the target object state prediction map to the initial target object association map, the system uses the latest timestamp of the first updated real-time target object association map as the current time reference and a preset duration as the total prediction span to initiate the construction of a hierarchically updated target object association map. The specific construction process includes: first, discretizing the total prediction span according to multiple preset continuous and progressive time slices; second, for each future time slice, calling the target state prediction function attached to each target node, and predicting the position coordinates and influence intensity value of each node at the corresponding future time based on the historical and current state data of the target node; simultaneously, calling the node... The system employs an inter-node embedded correlation influence adjustment function to extrapolate and quantify the adjustment value of the correlation strength between nodes at a future time based on the predicted changes in node states. Then, based on the predicted influence strength values ​​of each node and the updated correlation strength, the system recalculates the joint influence strength values ​​and contribution weight distribution of different target node combinations at future time points using the correlation influence adjustment function. Finally, the system integrates all target node states, inter-node correlations, and combined threat indicators calculated for each time slice into an independent prediction layer. Multiple prediction layers are then overlaid and correlated in chronological order to construct a hierarchical prediction map with a clear time dimension, reflecting the dynamic evolution path of the threat situation. This prediction map provides a quantitative and forward-looking threat situation blueprint for the advanced configuration and adaptive adjustment of equipment systems.

[0082] To illustrate the steps of this embodiment in more detail, the following example is given: In the first updated real-time target object association graph, there are two key target nodes: electronic jamming vehicle node E and reconnaissance drone node F. Node E's current state information includes its location at coordinates (X1, Y1), its movement northeast at speed V1, and the parameters of its built-in "target state prediction function" (e.g., a kinematic trajectory prediction model) have been trained based on its historical movement data. Node F's current state information includes its location at coordinates (X2, Y2), its remaining fuel level of F2, and its threat value partially dependent on its endurance. The "target state prediction function" mounted on this node is a performance degradation prediction function based on a fuel consumption model. Nodes E and F have an association edge due to their information support relationship. The "association influence adjustment function" embedded on this edge quantifies the relationship between their collaborative effectiveness and factors such as relative distance and line-of-sight.

[0083] The system takes this moment as the starting point T0, sets the total prediction duration to 30 minutes, and divides it into three prediction layers of equal length, corresponding to the future times T1 (T0+10 minutes), T2 (T0+20 minutes), and T3 (T0+30 minutes).

[0084] For the construction of the prediction layer at time T1: The system first calls the target state prediction function of node E, inputs its current position, speed, and direction, predicts that it will reach the coordinates (X1', Y1') at time T1, and based on the geographical environment of the new position, calculates that the effective interference range will change, and its single influence intensity value is updated to Te1. At the same time, the system calls the target state prediction function of node F, and based on its current fuel F2 and consumption rate, predicts that its fuel will drop to F2' at time T1, and accordingly its threat value is updated to Tf1, and Tf1 < Tf0 (the current threat value); then, the system calls the associated influence adjustment function between nodes E and F, inputs their predicted coordinates (X1', Y1') and (X2', Y2') (the predicted coordinates of node F change little due to endurance problems), and calculates that due to the increase in distance and possible changes in line-of-sight conditions, its associated strength will weaken from the initial ω0 to ω1. Finally, based on the predicted Te1, Tf1, and ω1, the associated influence adjustment function is called to recalculate the combined influence intensity value J1 of the combination of E and F at time T1. The system judges that J1 has slightly decreased compared to the current combined threat value J0.

[0085] For the construction of the prediction layer at times T2 and T3: The system performs recursive deduction. At the T2 layer, the target state prediction function of node E may further predict that it will conduct the next stage of maneuver or enter the standby state, and the single influence intensity value is adjusted to Te2. The fuel prediction function of node F determines that its fuel may be on the verge of exhaustion F2'' ≈ 0, and its threat value drops sharply to Tf2 (close to the failure threshold). The associated strength ω2 calculated by the associated influence adjustment function further weakens. At this time, the calculated combined threat value J2 decreases significantly. At the T3 layer, the prediction shows that node F may withdraw from the battlefield due to fuel exhaustion, that is, it is marked as the "failed" or "removed" state in the figure. Node E may exist alone. At this time, the system needs to evaluate the single influence intensity of only node E remaining and whether new associations are generated with other potential nodes. Finally, the prediction result at the T3 layer may show that the combined threat in this local area has basically collapsed. Here, F2'' is the predicted value of the remaining fuel of the target node F at the second prediction time (T2);

[0086] The system packages the complete predicted states at times T1, T2, and T3 into three structured prediction layers. These layers include the predicted attributes of each node, the predicted correlations between nodes, and the combined influence intensity values ​​at each level. Organized chronologically, these three layers together form a complete dynamic evolution view of the local target combination from "manifestation" to "decline" and finally "dissipation" over the next half hour. The hierarchical prediction map in this embodiment is output to the decision-making system, and its evolution trend visually indicates that the urgency of addressing this group of targets is decreasing over time. This forward-looking assessment provides the decision-making unit with clear quantitative evidence, enabling it to prioritize the reallocation of limited response resources to other target directions where the influence intensity is increasing or remains high during the prediction period.

[0087] It should be further explained that the construction process of the target-resource bidirectional update graph in this embodiment includes:

[0088] Using the location of each new type of equipment deployed corresponding to the initial target object map as the initial resource node in the resource configuration association graph, and combining the corresponding collaborative response association information between the initial resource nodes with the collaborative response association degree obtained by the association analysis algorithm configured with the expert evaluation system as the connection, an initial resource configuration association graph is constructed, and the initial configuration update index corresponding to the initial resource configuration association graph is obtained; in this embodiment, the initial configuration update index is characterized by the total cost of constructing the initial resource configuration association graph. It should be further explained that, in this embodiment, the coordinates of three high-value target objects (e.g., missile launchers located in different directions) identified in the initial target object map are used as the benchmark, and a long-range air defense system, a mobile radar station, and an electronic warfare jamming vehicle are deployed as initial resource nodes in their corresponding defense areas. The collaborative response association information of the three is analyzed by the expert evaluation system: the system evaluates that the radar station provides target indication for the air defense system, and the electronic warfare vehicle provides electromagnetic cover for the former two. Based on this, the collaborative association degree between the radar station and the air defense system is quantified as 0.8, and the collaborative association degree between the electronic warfare vehicle and the air defense system and the radar station is 0.7, and these quantified association degrees are used as the weights of the connection edges. The initial resource configuration association graph constructed in this way contains three nodes and three weighted connection edges. The corresponding initial configuration update index is the total cost of building this complete system, which specifically includes the transportation and deployment costs of transporting the three sets of equipment to the designated position, the integration costs of information system integration and debugging, and the special construction costs of establishing and maintaining a highly reliable data link and collaborative network between the three.

[0089] After monitoring and acquiring the initial resource configuration association map deployment, the system monitors and acquires the changes in the joint influence intensity and individual influence intensity of the target object map, as well as the corresponding directions of change. For example, in this embodiment, after completing the initial resource configuration association map deployment, the system initiates continuous monitoring of the initial target object map to quantitatively evaluate deployment effectiveness. The specific implementation process includes: real-time collection of dynamic activity data of the target object through an integrated network of deployed radar, electronic reconnaissance, and signals intelligence sensors; the system calls a built-in threat assessment calculation model to perform calculation and analysis within a preset assessment time window after equipment deployment takes effect. Taking a typical coordinated attack combination consisting of a target high-altitude reconnaissance UAV and an escorting fighter jet as an example, the system monitors and acquires the specific changes in threat indicators before and after deployment. Before deployment, the quantified individual influence intensity value of the reconnaissance UAV is 0.8, and that of the fighter jet is 0.9; the joint influence intensity value generated by their coordinated cooperation is 1.2. After deployment is completed and a 30-minute effectiveness period has elapsed, the system re-evaluates and shows that the individual influence intensity value of the reconnaissance UAV has decreased to 0.3, the individual influence intensity value of the fighter jet has decreased to 0.5, and the joint influence intensity value of the two has significantly decreased to 0.6. All changes showed a clear downward trend. By calculating the difference between the assessment values ​​before and after deployment, the system obtained a quantified change in the joint impact intensity of -0.6, and individual impact intensity changes of -0.5 and -0.4 for the two independent targets, respectively. This complete monitoring data directly verifies the immediate countermeasure effectiveness of the initial equipment deployment and provides precise quantitative input for calculating the response effectiveness range and effectiveness suppression assessment values ​​in subsequent steps.

[0090] Based on a comprehensive evaluation algorithm that combines the change value of joint influence intensity with the change value of single influence intensity and the corresponding change direction, the response effectiveness range and the corresponding effectiveness suppression evaluation value are obtained; the effectiveness suppression evaluation value is constructed from the change amount of the largest target node that meets the joint influence intensity threshold before and after the deployment of the new equipment.

[0091] It should be further explained that this embodiment obtains the response effectiveness range and the corresponding effectiveness suppression assessment value based on the combined impact intensity change value and the single impact intensity change value and the corresponding change direction, combined with a comprehensive evaluation algorithm. Specifically, the implementation is as follows: First, a single threat suppression threshold and a combined threat suppression threshold are set; then, target nodes whose single impact intensity value drops below the single threat suppression threshold after equipment deployment, or whose combined impact intensity value of any threat combination to which they belong drops below the combined threat suppression threshold, are marked as nodes that have been effectively suppressed; next, based on the geographical coordinates of all effectively suppressed target nodes, their minimum bounding polygon is calculated using the convex hull algorithm, and the area covered by this polygon is defined as the response effectiveness range; this embodiment... The specific implementation method for constructing the effectiveness suppression assessment value includes: First, a clear joint impact security threshold is preset; second, in the target object association graph before equipment deployment, a traversal algorithm is used to search and determine the combination of all target node combinations whose joint impact strength value exceeds the security threshold and contains the most nodes, and the number of nodes is recorded as the maximum threat scale value before deployment; then, after the equipment is deployed and takes effect, this search process is repeated in the updated target object association graph to determine the largest node combination that meets the condition of exceeding the security threshold, and the number of nodes is recorded as the maximum threat scale value after deployment; finally, the maximum threat scale value before deployment is subtracted from the maximum threat scale value after deployment, and the difference is the effectiveness suppression assessment value. This value quantitatively reflects the degree of weakening achieved by the equipment system in dealing with the target's largest-scale coordinated offensive capability.

[0092] A specific example is as follows: The joint impact safety threshold is set to 1.0. In the target object association graph, initially there exists a collaborative combination consisting of land-based, sea-based, and air-based platform nodes. Its joint impact strength value is calculated to be 1.5, exceeding the safety threshold. According to the rules defined in this application, the system traverses all possible node combinations in the graph and confirms that there is no combination containing four or more nodes with a joint impact strength exceeding 1.0. Therefore, the current maximum collaborative scale value is recorded as 3. After applying this method to optimize resource allocation and deploy the integrated response system, the system re-evaluates the same target object association graph. At this point, the joint impact strength value of the original three-node combination has decreased to 0.8, below the safety threshold. Furthermore, after system verification, the joint impact strength value of all node combinations with two or more nodes in the graph does not exceed 1.0. Therefore, the maximum collaborative scale value after deployment is updated to 0. According to the quantitative indicators defined in this application, the effectiveness suppression evaluation value is the difference between the maximum collaborative scale value before and after deployment, i.e., 3-0=3. From the perspective of dynamic game theory and effectiveness quantification, the results show that the resource allocation scheme optimized by this method can suppress the maximum potential scale of effective over-limit synergy in the target object system from 3 nodes to 0, which intuitively and quantitatively verifies the effectiveness of the scheme in suppressing the target synergy effect and optimizing the system response efficiency.

[0093] The system collects the performance change data of each resource node in the resource configuration association graph at continuous time points, as well as the change in the response performance range of each resource node at consecutive timestamps and the change in the corresponding performance suppression evaluation value and direction. The collected data is then filtered using a distributed Kalman filter algorithm.

[0094] Based on the preprocessed performance change data, the change in response performance range, and the corresponding change in performance suppression evaluation value, combined with the convolutional radial kernel function, a performance-performance suppression mapping function is obtained, and mapped to the corresponding resource node and target node according to the labels of resource node and target node.

[0095] It should be further explained that the performance-efficiency suppression mapping function in this embodiment aims to construct a quantitative prediction relationship between the performance state changes of resource nodes and their efficiency suppression evaluation value output. The construction steps include: First, synchronously collecting time-series data of the performance parameters of each resource node in the resource configuration association graph at continuous time points, as well as the change amount and direction of the response efficiency range and efficiency suppression evaluation value fed back by the monitoring and evaluation system at the same timestamp; Second, using a distributed Kalman filter algorithm to fuse and filter the above multi-source time-series data to extract a consistent and smooth system state estimate; Subsequently, using the filtered performance change sequence as input and the corresponding efficiency change sequence as output, using a convolutional radial basis function for nonlinear regression fitting, training to obtain a mathematical model that characterizes the mapping relationship of "performance evolution-efficiency output"; Finally, based on the unique identifier of the resource node and its associated target node type label, the trained function is used as a computable attribute and dynamically attached to the association edge between the corresponding resource node and the target node in the target-resource bidirectional update graph, so that the system can perform a forward-looking quantitative evaluation of the efficiency suppression effect of the resource node on a specific target node based on the real-time or predicted performance state of the resource node.

[0096] It should be further explained that the performance-effectiveness suppression mapping function constructed in this embodiment is not limited to the static evaluation of the current effectiveness of resource nodes, but extends to support forward-looking lifecycle management based on effectiveness thresholds. Specifically, the system first sets a safety warning threshold for the effectiveness suppression evaluation value, which defines the minimum effectiveness standard for maintaining an effective response. By inversely solving the performance-effectiveness suppression mapping function, the system can derive the critical value of the resource performance parameter corresponding to ensuring that the effectiveness evaluation value is not lower than this warning threshold. To achieve prediction, the system continuously monitors the performance data of resource nodes and uses historical data to fit the decay function of their performance parameters over time. By solving the decay function and the aforementioned critical value of the performance parameter simultaneously, the future moment when the performance of the resource node is expected to drop to the critical point can be calculated, i.e., the predicted failure time point. For example, in the resource configuration association graph, the key performance parameter of an electronic countermeasures vehicle node is its effective jamming power. Through its unique performance-effectiveness suppression mapping function analysis, it can be seen that when the jamming power drops to 150 kilowatts, its suppression effectiveness (i.e., effectiveness suppression evaluation value) against a specific radar guidance unit node in the graph will reach the safety warning threshold. The system detected that the node's current power is 200 kilowatts, and its historical degradation trend conforms to a linear model of 2 kilowatts per day. Based on this model, the node is expected to reach its performance criticality point in 25 days. The early warning information generated by the system provides a clear decision-making window for initiating maintenance or allocating alternative resources before failure, thus transforming resource assurance from a passive response to proactive and precise planning based on an efficiency model. This mechanism significantly enhances the continuous adaptability and operational reliability of the entire resource system supported by the target-resource bidirectional update graph.

[0097] By combining the configuration update index corresponding to each new equipment update in the historical target-resource bidirectional update graph, the change in the response performance range and the change in the corresponding performance suppression evaluation value of the target object association graph after the deployment update, and the change direction of the change with the change direction, a linear algorithm is used to obtain the configuration update index-performance suppression mapping function, and then the resource node and target node are mapped to the corresponding resource node and target node according to the labels of the resource node and target node.

[0098] It should be further explained that the configuration update index-effectiveness suppression mapping function in this embodiment is used to quantify the proportional relationship between the economic cost of equipment system adjustment and the resulting improvement in combat effectiveness, providing a quantitative cost-effectiveness analysis basis for core decision-making. Its construction steps include: First, the system collects each complete equipment update record from the historical update log of the target-resource bidirectional update map. The key data items extracted include the total cost consumed in this update action, i.e., the configuration update index value; the change in the response effectiveness range measured in the target object association map within a preset evaluation period after the update; and the change in the effectiveness suppression evaluation value. The direction of the effectiveness change is recorded as either improvement or reduction. Second, the collected historical dataset is cleaned and standardized to ensure data consistency. Next, a multiple linear regression algorithm is used to fit the dataset, with the configuration update index value as the independent variable and the change in the response effectiveness range and the change in the effectiveness suppression evaluation value as the dependent variables. The regression coefficients are solved using the least squares method, thereby training a linear mapping function model. This model can predict the range of comprehensive effectiveness output that may be obtained after investing specific costs in system adjustment. Finally, based on the resource node identifiers and target node identifiers of their main adversaries involved in the historical update record, the system dynamically associates and binds the trained configuration update metric-performance suppression mapping function as a key attribute to the adversarial relationship edges between the corresponding resource nodes and target nodes in the target-resource bidirectional update graph. This mapping enables the system to quickly assess the expected performance gains under different cost budgets for specific threat response needs in subsequent planning, or to reverse-engineer the approximate resource input required under a given performance objective, thus achieving scientific decision-making under resource constraints.

[0099] A target-resource priority constraint game model is defined, including a resource allocation optimization party and a target object dynamic evolution party. The decision variables of the resource allocation optimization party are the resource allocation strategies in the resource configuration association graph, and the state variables of the target object dynamic evolution party are the target node states at each time point in the hierarchically updated target object association graph. It should be further noted that this target-resource priority constraint game model is a formal framework describing the adversarial interaction between an equipment system and a dynamic threat environment. This target-resource priority constraint game model explicitly defines two game participants: the resource allocation optimization party, as the decision-making subject, whose decision variables are specific strategy schemes selected from a preset resource allocation strategy library for adjusting or reconstructing the resource configuration association graph in the target-resource bidirectional update graph. These strategies specifically stipulate the addition and deletion of resource nodes, attribute adjustments, and changes in the collaborative connections between nodes; and the target object dynamic evolution party, as the environmental response subject, whose state variables are fully described by the hierarchically updated target object association graph, covering the location, influence intensity, and other attributes of each target node from the current moment to multiple predicted future time points, as well as the association strength between nodes and the resulting joint influence intensity values ​​at various levels. The core rule of this model is that every time the resource allocation optimization party makes a strategic decision and implements it, the state of the target object dynamic evolution party will change accordingly based on its built-in prediction function (such as the target state prediction function and the correlation influence adjustment function), thus forming a sequential game process of "decision-state transition". Its ultimate optimization goal is to find the optimal resource allocation strategy sequence that maximizes the long-term cumulative benefits of the equipment party under the premise of strictly satisfying a series of security and resource constraints.

[0100] A single-impact safety threshold and a joint-impact safety threshold are set, and a game constraint space is constructed using these thresholds, along with the response effectiveness range, effectiveness suppression evaluation value, and total deployable cost value. This space is then mapped to the target-resource priority constraint game model. It should be further explained that the core purpose of setting the single-impact safety threshold and the joint-impact safety threshold in this embodiment is to establish a minimum standard that must be guaranteed for the continuous and stable operation of the resource allocation system, ensuring that any optimization decision is made under the premise of meeting basic safety requirements. Introducing the response effectiveness range and effectiveness suppression evaluation value as constraints aims to transform abstract effectiveness requirements into concrete and quantifiable performance targets, thereby ensuring that the optimization process focuses on achieving verifiable and substantial effects. Setting the constraint of total deployable cost value is to incorporate the fundamental reality of limited resources into the model, forcing the search for the optimal solution within a given cost budget, thus ensuring that the generated resource allocation strategy is both highly efficient and feasible. These three types of constraints together construct a game constraint space that simultaneously covers the safety baseline, efficiency target, and cost ceiling, and fully maps it to the goal-resource priority constraint game model. This systematically guides the optimization algorithm to find an equilibrium strategy that truly meets the requirements of complex tasks within the solution space composed of the three key dimensions of safety, efficiency, and cost.

[0101] Based on the performance-efficiency inhibition mapping function and the configuration update index-efficiency inhibition mapping function mapped in the target-resource bidirectional update graph, a game payoff index function is constructed and mapped to the target-resource priority constraint game model.

[0102] Based on the performance-efficiency suppression mapping function and the configuration update index-efficiency suppression mapping function mapped in the target-resource bidirectional update graph, a game payoff index function is constructed and mapped to the target-resource priority-constrained game model. The fundamental motivation is to provide a calculable and comparable quantitative standard for the merits of each resource allocation strategy selection during the game process. This standard must uniformly measure the combat effectiveness gain brought by the strategy and the system adjustment cost incurred, thereby driving the optimization algorithm to automatically select the strategy sequence with the highest cost-effectiveness ratio under the premise of satisfying security constraints. The specific construction process includes: first, defining a payoff calculation unit for a single round of the game, which takes the target-resource bidirectional update graph state at a certain moment in the game process and the selected specific resource allocation strategy as input; its... Next, the performance-efficiency suppression mapping function associated with the current resource node and the adversarial target node is invoked. Based on the predicted performance state of the resource node after strategy execution, the expected response efficiency range and efficiency suppression evaluation value that the strategy can generate are calculated, and the two are weighted and combined into the expected efficiency gain. At the same time, the associated configuration update index-efficiency suppression mapping function is invoked, or the configuration update index value marked by the strategy itself is directly obtained and quantified as the strategy execution cost. Finally, the benefit index function is constructed as the linear weighted difference between the expected efficiency gain and the strategy execution cost, where the weight coefficient of the efficiency gain term is much larger than the weight coefficient of the cost term. This mathematically ensures that the optimization process prioritizes efficiency satisfaction and automatically seeks cost minimization under this premise, thus forming a complete quantitative decision basis.

[0103] Using the strategies in the resource allocation strategy library as the solution space, and aiming to maximize the game payoff index function, the target-resource priority constraint game model is solved by combining a heuristic search algorithm to obtain the optimal resource allocation strategy sequence that maximizes the cumulative payoff.

[0104] It should be further explained that this embodiment uses the sequence space formed by all feasible strategy combinations in the resource allocation strategy library as the solution space, and takes maximizing the cumulative value of the game payoff index function within the preset total game duration as the optimization objective. A heuristic search algorithm integrating depth-first branch and bound and Monte Carlo simulation evaluation is used to iteratively solve the objective-resource priority constrained game model. The specific implementation process includes:

[0105] First, the search tree is initialized, with its root node representing the current state of the target-resource bidirectional update graph and the hierarchically updated target object association graph. Starting from the root node, the algorithm generates all feasible candidate strategies based on the resource allocation strategy library as branch actions for the current node. For each candidate strategy, the algorithm performs action expansion: calculating its cost using the mapping configuration update metric-performance suppression mapping function, and predicting its performance benefit using the performance-performance suppression mapping function, forming a preliminary node benefit estimate. Simultaneously, the algorithm calls the target object dynamic evolution model to predict the threat situation at the next moment after executing the strategy, generating child node states.

[0106] Secondly, the algorithm employs a strategy combining depth-first search and upper bound confidence interval (UCT) for tree search and node selection. During the exploration process, the number of visits and cumulative simulated payoffs are maintained for each node. When selecting child nodes, a tradeoff is made based on the immediate estimate of the game payoff indicator function, the long-term average payoff of the node, and the exploration factor to balance in-depth development of the current optimal branch with extensive exploration of underexplored branches.

[0107] Third, Monte Carlo stochastic simulations are performed on the selected node paths until the preset game time depth is reached. In the simulation, for subsequent nodes that are not fully expanded, strategies are randomly selected from the resource allocation strategy library to simulate actions, and feasibility is strictly verified according to the game constraint space (including single / joint influence safety thresholds and total allocateable cost). Any path that violates the constraints will be pruned immediately, and its payoff will be recorded as negative infinity.

[0108] Fourth, the cumulative payoff obtained from the simulation path is backpropagated to update the statistical information of all nodes on the path. Through multiple iterations, the algorithm gradually converges, and finally selects the strategy sequence with the highest average cumulative simulated payoff from the expanded branches of the root node as the output optimal resource allocation strategy sequence that maximizes the cumulative payoff. This process ensures that a near-globally optimal game equilibrium solution that satisfies multiple complex constraints is efficiently found in the huge strategy sequence solution space.

[0109] The first resource allocation strategy in the optimal resource allocation strategy sequence obtained by the solution is mapped to the target-resource bidirectional update graph in the configured task simulation platform for execution and verification.

[0110] It should be further explained that the steps in this embodiment of mapping the first resource allocation strategy in the obtained optimal resource allocation strategy sequence to the target-resource bidirectional update graph within the configured task simulation platform for execution and verification include: First, obtaining the complete description data of the first resource allocation strategy in the optimal resource allocation strategy sequence and parsing it into a specific set of executable operation instructions and parameters, including the identifier of the resource node to be adjusted, attribute settings, and instructions for establishing or dissolving collaborative relationships between nodes. Second, obtaining the complete state snapshot data of the target-resource bidirectional update graph at the current moment, including the deployment attributes of all resource nodes and the real-time status and hierarchical predicted status of all target nodes, and synchronously injecting it into the virtual environment of the task simulation platform. Third, within the simulation platform, based on the obtained operation instructions and parameters, driving the simulation engine to accurately execute the equipment deployment adjustment actions in the virtual spacetime, and synchronously obtaining the continuous threat situation change data within a future evaluation window deduced by the target object dynamic evolution model based on the strategy effect. Fourth, during the simulation process, the calculated values ​​of the single impact intensity of each target node and the combined impact intensity of all related combinations are acquired in real time and compared with the preset single impact safety threshold and combined impact safety threshold one by one. At the same time, the data of various virtual resources consumed during the strategy execution process are accumulated, the total cost is calculated and checked against the adjustable total cost constraint. Fifth, the performance output data in the evaluation window after the strategy is executed is acquired, including the achieved response performance range and performance suppression evaluation value, and compared with the revenue indicators predicted by the game model. If the actual performance meets the preset response performance range indicator, the actual cost does not exceed the budget, and all safety constraints are met throughout the process, the verification is passed and the strategy is confirmed as the final execution plan; otherwise, the complete verification result data package is fed back to the target-resource priority constraint game model to trigger a new round of optimization and solution iteration. When the simulation confirms the execution of the resource allocation strategy, and the dynamic evolution of the target object satisfies the game constraint space, and the configuration update index of the resource allocation strategy is the smallest among all alternative strategies that satisfy the game constraint space, the resource configuration association graph at the current moment is obtained and encapsulated and output as a new equipment construction system.

[0111] It should be further explained that, in a specific defense scenario, this embodiment first obtains a threat update indicator, showing that the target has added a hypersonic weapon platform and adjusted the deployment status of its electronic warfare units. The system then calls the pre-constructed target object association graph, which, based on historical data and prediction functions, already includes nodes, associations, and quantified threat values ​​for existing threats such as ground armored clusters and UAV reconnaissance units. Based on this graph, the system initializes a target-resource bidirectional update graph. The current resource configuration association graph in the graph consists of nodes of long-range air defense systems, mobile radar stations, and electronic warfare jamming vehicles, and their collaborative connections. To cope with the new hypersonic threat, the system aims to minimize adjustment costs and maximize the range of response effectiveness against the enemy. It matches the alternative strategy of "adding a new anti-missile interception system and upgrading the radar station data link" from the strategy library. Next, the system uses the "performance-effectiveness suppression mapping function" mapped in the graph to predict the effectiveness of the strategy, uses the "configuration update indicator-effectiveness suppression mapping function" to calculate its cost, and inputs these data into the target-resource priority constraint game model. Under strict constraints of safety thresholds and total cost, the model employs a heuristic search algorithm for multi-round game simulations to ultimately find an optimal sequence encompassing the chosen strategy. Subsequently, the first strategy in this sequence—deploying an anti-missile system and upgrading radar stations—is mapped onto a mission simulation platform for verification. Simulation results show that executing this strategy successfully suppresses the threat posed by the hypersonic weapon platform, with the combined threat value of all target nodes falling below the safety threshold, and the actual cost remaining within budget and lower than other feasible solutions. After successful verification, the system obtains the updated resource configuration relationship graph, which includes a newly added anti-missile system node, updated radar station node attributes, and enhanced collaborative connections. Finally, this graph is encapsulated and output as a novel equipment construction system that is effective, cost-efficient, and tailored to the latest threat landscape.

[0112] This embodiment effectively addresses the core pain points of traditional methods, such as the lag in static analysis, equipment-threat adaptation discrepancies, and cost-efficiency imbalances, by constructing a dynamic target object association graph and a hierarchical prediction graph, and integrating a quantitative mapping function with a priority-constrained game model. This significantly improves the scientific rigor and practical adaptability of new equipment system construction. In particular, by collecting multi-dimensional information such as the single influence strength of target object location tags, and combining it with expert evaluation algorithms to construct an initial target object association graph, it embeds a target state prediction function based on support vector machines and a correlation influence and threat adjustment function constructed using a radial kernel function. This enables real-time quantitative updates of association relationships and threat indicators when target nodes are added, deleted, or their states change. This breaks through the limitations of traditional static analysis, accurately captures threat synergy and dynamic evolution trends, and provides forward-looking situational support for equipment deployment. Secondly, it constructs a target-resource bidirectional... The updated graph, through distributed Kalman filtering preprocessing of data, trains two types of mapping functions: performance-effective defense and configuration update index-effective defense. This establishes a quantitative correlation between equipment performance, adjustment costs, and combat effectiveness, overcoming the difficulty of coordinating effectiveness and cost in traditional methods and providing precise quantitative basis for strategy optimization. Furthermore, a target-resource priority constraint game model is defined, clarifying the game relationship between the resource allocation optimization party and the dynamically evolving target object. A constraint space is constructed by combining security thresholds, effectiveness requirements, and cost limits. A heuristic search algorithm is used to solve for the optimal resource allocation strategy sequence, achieving a dynamic balance of security, effectiveness, and cost objectives, avoiding resource misallocation and effectiveness redundancy. Finally, the feasibility of the strategy is verified through a mission simulation platform, ensuring that the output equipment system meets both the response effectiveness range and defense assessment requirements while minimizing deployment costs. Overall, this application, through a complete technical system of "dynamic threat perception - quantitative correlation modeling - game optimization solution - simulation verification and implementation," significantly improves the response speed, adaptation accuracy, and resource utilization efficiency of equipment systems to dynamic and complex threats, effectively undermines the target's collaborative combat capabilities, provides efficient and reliable technical support for the construction of new equipment systems, significantly reduces the later adjustment costs of traditional methods, avoids effectiveness losses, and adapts to the practical needs of modern all-domain operations with intertwined multi-domain threats.

[0113] Example 2

[0114] Please see Figure 2 Another embodiment of the present invention provides: an equipment system construction system, comprising:

[0115] The data acquisition module is used to obtain the target object update indicator set, including the increase or decrease of the target object type and the target object status update.

[0116] The matching module is used to initialize a preset target-resource bidirectional update graph. On the target-resource bidirectional update graph, in response to the target object update index set, with the goal of minimizing the configuration update index in the resource configuration association graph and maximizing the response performance range on the target object association graph, the module matches the corresponding resource allocation strategy in the preset resource allocation strategy library.

[0117] The target-resource bidirectional update graph includes a resource configuration association graph, a target object association graph, and a mapping relationship between resource configuration update indicators and the influence intensity and influence range of target objects; the response effectiveness range is characterized by the change in the influence intensity and influence range of the target objects; the configuration update indicator is characterized by the cost corresponding to each resource configuration update.

[0118] The simulation module is used to map the matched resource allocation strategy onto the target-resource bidirectional update graph within the configured task simulation platform for real-time simulation adjustment. When the response performance range meets the preset response performance range index and the configuration update index is minimized, the resource configuration association graph under the corresponding conditions is obtained and encapsulated into an equipment construction system.

[0119] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments under the guidance of the present invention without departing from the spirit and scope of the claims. All of these variations are within the protection scope of the present invention.

Claims

1. A method for constructing an equipment system, characterized in that, include: Obtain the target object update indicator set, including the increase or decrease of the target object type and the target object status update; Initialize a preset target-resource bidirectional update graph. On the target-resource bidirectional update graph, in response to the target object update index set, with the goal of minimizing the configuration update index in the resource configuration association graph and maximizing the response performance range on the target object association graph, match the corresponding resource allocation strategy in the preset resource allocation strategy library. The target-resource bidirectional update graph includes a resource configuration association graph, a target object association graph, and a mapping relationship between resource configuration update indicators and the influence intensity and influence range of target objects; the response effectiveness range is characterized by the change in the influence intensity and influence range of the target objects; the configuration update indicator is characterized by the cost corresponding to each resource configuration update. The target object association graph uses target objects as nodes and their collaborative relationships as edges, and is equipped with target state prediction functions, association influence adjustment functions, and association influence intensity adjustment functions to quantify the evolution of their individual influence intensity, association intensity, and joint influence intensity. Specifically, the target state prediction function, association influence adjustment function, and association influence intensity adjustment function embedded in the target object association graph are responsible for respectively deducing the state transition of a single target node, quantifying in real-time changes in the strength of collaborative relationships between target nodes, and calculating the dynamic effect of the state transition of a single target node or the real-time quantification of changes in the strength of collaborative relationships between target nodes on the overall joint influence intensity of the target group. The response effectiveness range is a spatial effectiveness index, measured by calculating the geographical coverage area formed by all target nodes affected by the resource allocation scheme and suppressed below a preset safety threshold after the scheme takes effect. The configuration update index is used to comprehensively quantify the total cost incurred in each equipment system adjustment, covering economic costs, time windows, and opportunity costs. The matched resource allocation strategy is mapped to the target-resource bidirectional update graph within the configured task simulation platform for real-time simulation adjustment. When the response performance range meets the preset response performance range index and the configuration update index is minimized, the resource configuration association graph under the corresponding conditions is obtained and encapsulated into an equipment construction system.

2. The equipment system construction method as described in claim 1, characterized in that, The process of constructing the target object association graph includes: Obtain the initial target object information and the corresponding deployment association information; the target object information includes deployment location information, tag information, and single influence intensity value; Based on the single impact intensity value of the target object and its label and corresponding deployment association information, combined with the association analysis algorithm, a sequence of target object association relationships and a sequence of association impact indicators are obtained. The association impact indicators are used to measure the joint impact intensity value of the target to be protected after at least two related target objects are combined. The sequence of association impact indicators includes: the joint impact intensity value for the target to be protected, and the contribution weight of each target node that makes up the joint impact intensity value. Using the location information of each target object as the target node and the sequence of relationships between target objects as the connection relationships, combined with a graph algorithm, an initial target object graph is obtained.

3. The equipment system construction method as described in claim 2, characterized in that, The process of constructing the target object association graph also includes: The sequence of related influence indicators is mapped to the initial target object graph by connecting the contribution weight of the single influence intensity value of the target object label and each target node to the joint influence intensity value under the corresponding target node, thus obtaining the initial target object association graph. Obtain the historical target object status update information of each target node in the initial target object graph, as well as the adjustment values ​​of the target object association relationship and the joint influence intensity value under the corresponding timestamp; Based on the historical target object state update information of each target node and combined with the support vector machine, a target state prediction function is obtained, and the target state prediction function is attached to the corresponding target node with the target node label.

4. The equipment system construction method as described in claim 3, characterized in that, The process of constructing the target object association graph also includes: Based on the historical target object status update information of each target node and the target object association adjustment value under the corresponding timestamp, the association influence adjustment function between two target nodes with an association relationship and the association influence strength adjustment function between the target node and the joint influence strength value are constructed through the radial kernel function. The association influence adjustment function is embedded into the association connection relationship between corresponding pairs of target nodes according to the label of the target node; at the same time, the association influence strength adjustment function is mapped to the contribution weight of each target node to the joint influence strength value, which is used to update the association contribution connection in real time.

5. The equipment system construction method as described in claim 4, characterized in that, The process of constructing the target object association graph also includes: The initial target object association graph is collected in real time to obtain the real-time state graph and the target object state prediction graph at each time point; the target object state prediction graph is constructed by the target state prediction function attached to each target node in the initial target object association graph, predicting the state change of the target object within a preset time length. Based on the real-time state diagram of the target object at adjacent time points, the increase or decrease of the target object type is determined, and the increase or decrease of the target node set and the corresponding increase or decrease of the target node set state information are obtained.

6. The equipment system construction method as described in claim 5, characterized in that, The process of constructing the target object association graph also includes: The set of target nodes to be added or removed, the set of target nodes to be retained, and the corresponding real-time status information are fed back to the initial target object association graph. The target object association sequence and the association influence index sequence are adjusted and updated in real time to obtain the first updated real-time target object association graph. Simultaneously, the target object state prediction map is fed back to the initial target object association map. Taking the timestamp corresponding to the first updated real-time target object association map as the starting point and the preset time length as the time axis, a hierarchically updated target object association map is constructed.

7. The equipment system construction method as described in claim 6, characterized in that, The process of constructing the target-resource bidirectional update graph includes: The location of each piece of equipment deployed in the initial target object graph is taken as the initial resource node in the resource configuration association graph. The collaborative response association information between the initial resource nodes is combined with the collaborative response association degree obtained by the association analysis algorithm as the connection to construct the initial resource configuration association graph and obtain the initial configuration update index corresponding to the initial resource configuration association graph. After the initial resource configuration association graph is deployed, monitor and obtain the joint influence intensity change value and single influence intensity change value and the corresponding change direction in the initial target object graph; Based on a comprehensive evaluation algorithm that combines the change value of joint influence intensity with the change value of single influence intensity and the corresponding change direction, the response effectiveness range and the corresponding effectiveness suppression evaluation value are obtained; the effectiveness suppression evaluation value is constructed by the change amount of the largest target node that meets the joint influence intensity threshold before and after equipment deployment.

8. The equipment system construction method as described in claim 7, characterized in that, The process of constructing the target-resource bidirectional update graph also includes: The system collects the performance change data of each resource node in the resource configuration association graph at continuous time points, as well as the change in the response performance range of each resource node at consecutive timestamps and the change in the corresponding performance suppression evaluation value and direction. The collected data is then filtered using a distributed Kalman filter algorithm. Based on the preprocessed performance change data, the change in response performance range, and the corresponding change in performance suppression evaluation value, combined with the convolutional radial kernel function, a performance-performance suppression mapping function is obtained, and mapped to the corresponding resource node and target node according to the labels of resource node and target node. By combining the configuration update index corresponding to each equipment update in the historical target-resource bidirectional update graph, the change in the response performance range and the change in the corresponding performance suppression evaluation value of the target object association graph after the deployment update, and the change direction of the change, a linear algorithm is used to obtain the configuration update index-performance suppression mapping function, and then the resource node and target node are mapped to the corresponding resource node and target node according to the labels of the resource node and target node. Define a goal-resource priority constraint game model, including a resource allocation optimization party and a goal object dynamic evolution party. The decision variable of the resource allocation optimization party is the resource allocation strategy of the resource allocation association graph, and the state variable of the goal object dynamic evolution party is the state of the goal node at each time point in the hierarchically updated goal object association graph. Set single-impact safety thresholds and joint-impact safety thresholds, and use the set single-impact safety thresholds and joint-impact safety thresholds, response effectiveness range, effectiveness suppression evaluation value, and adjustable total cost value to construct a game constraint space, and map it to the target-resource priority constraint game model.

9. The equipment system construction method as described in claim 8, characterized in that, The process of constructing the target-resource bidirectional update graph also includes: Based on the performance-efficiency inhibition mapping function and the configuration update index-efficiency inhibition mapping function mapped in the target-resource bidirectional update graph, a game payoff index function is constructed and mapped to the target-resource priority constraint game model. Using the strategies in the resource allocation strategy library as the solution space, and aiming to maximize the game payoff index function, the target-resource priority constraint game model is solved by combining a heuristic search algorithm to obtain the optimal resource allocation strategy sequence that maximizes the cumulative payoff. The first resource allocation strategy in the optimal resource allocation strategy sequence obtained by the solution is mapped to the target-resource bidirectional update graph in the configured task simulation platform for execution and verification. When the simulation confirms the execution of the resource allocation strategy, and the dynamic evolution of the target object satisfies the game constraint space, and the configuration update index of the resource allocation strategy is the smallest among all alternative strategies that satisfy the game constraint space, the resource configuration association graph at the current moment is obtained and encapsulated and output as an equipment construction system.

10. An equipment system construction system, used to implement the equipment system construction method according to any one of claims 1-9, characterized in that, include: The data acquisition module is used to obtain the target object update indicator set, including the increase or decrease of the target object type and the target object status update. The matching module is used to initialize a preset target-resource bidirectional update graph. On the target-resource bidirectional update graph, in response to the target object update index set, with the goal of minimizing the configuration update index in the resource configuration association graph and maximizing the response performance range on the target object association graph, the module matches the corresponding resource allocation strategy in the preset resource allocation strategy library. The target-resource bidirectional update graph includes a resource configuration association graph, a target object association graph, and a mapping relationship between resource configuration update indicators and the influence intensity and influence range of target objects; the response effectiveness range is characterized by the change in the influence intensity and influence range of the target objects; the configuration update indicator is characterized by the cost corresponding to each resource configuration update. The simulation module is used to map the matched resource allocation strategy onto the target-resource bidirectional update graph within the configured task simulation platform for real-time simulation adjustment. When the response performance range meets the preset response performance range index and the configuration update index is minimized, the resource configuration association graph under the corresponding conditions is obtained and encapsulated into an equipment construction system.

Images