A military assessment method, device, equipment and storage medium based on simulation situation data

Abstract

The present invention provides a military assessment method, device, equipment and storage medium based on simulation situation data, which obtains military assessment scenario data and calls a four-dimensional assessment model to select an algorithm configuration scheme corresponding to the military assessment scenario from an intelligent algorithm library based on an efficiency factor, a noise robustness factor, a result reliability factor, and a scenario adaptability factor. According to the algorithm configuration scheme, a heterogeneous graph convolutional network HGCN algorithm is used to construct a military knowledge graph, and a meta-learning framework is called to provide a dynamic assessment model to the military knowledge graph; real-time calculation results of the dynamic assessment model are obtained, and the real-time calculation results are processed through a distributed computing architecture to analyze military effectiveness trends and update assessment parameters, and the indicator weights are dynamically adjusted to match the algorithm strategy, thereby solving the problem that the assessment system cannot quickly adapt to new scenarios.

Description

Technical Field

[0001] The present invention relates to the field of military assessment, and in particular to a military assessment method, device, equipment and storage medium based on simulation situation data. Background Art

[0002] In the field of modern military operations and training, accurate assessment of the effectiveness of military operations, the feasibility of combat plans, and changes in battlefield situation has become a key technical requirement.

[0003] Traditional evaluation models rely on manual construction based on expert experience, with fixed indicator systems and a lack of a refined indicator library tailored to different combat scenarios. Modifying evaluation indicators requires recoding and deployment, which takes an average of over 72 hours. This is particularly true in dynamic battlefield environments, where the evaluation system cannot quickly adapt to new scenarios as combat missions change.

[0004] Although the development of military simulation technology can generate rich battlefield situation data, it is currently impossible to effectively convert this data into evaluation conclusions with decision-making guidance, which limits the in-depth application of military simulation technology in combat research, military training, and combat plan optimization.

[0005] In view of this, this application is filed. Summary of the Invention

[0006] The present invention discloses a military assessment method, apparatus, device and storage medium based on simulation situation data, aiming to solve the problem that the assessment system cannot quickly adapt to new scenarios.

[0007] A first embodiment of the present invention provides a military assessment method based on simulation situation data, comprising:

[0008] Obtaining military assessment scenario data, and calling a four-dimensional assessment model to select an algorithm configuration scheme corresponding to the military assessment scenario from an intelligent algorithm library based on an efficiency factor, a noise robustness factor, a result reliability factor, and a scenario adaptability factor, wherein the military assessment scenario data includes a combat mission description, environmental parameters, force size data, and assessment requirement characteristics;

[0009] According to the algorithm configuration scheme, a heterogeneous graph convolutional network (HGCN) algorithm is used to construct a military knowledge graph, and a meta-learning framework is called to provide a dynamic evaluation model to the military knowledge graph, wherein the dynamic evaluation model is generated based on a scene semantic parser, a meta-model matching engine, and an adaptive optimization module;

[0010] Obtain real-time calculation results of the dynamic evaluation model, process the real-time calculation results through a distributed computing architecture to analyze military effectiveness trends and update evaluation parameters, and dynamically adjust the strategy of matching indicator weights with algorithms.

[0011] Preferably, military assessment scenario data is obtained, and a four-dimensional assessment model is called to select an algorithm configuration scheme corresponding to the military assessment scenario from an intelligent algorithm library based on the efficiency factor, noise robustness factor, result reliability factor, and scenario adaptability factor, specifically:

[0012] The military evaluation scenario data is processed by a scenario semantic parser to generate a scenario feature vector. The candidate algorithms are evaluated by a four-dimensional evaluation model and classified into different applicable categories. For each algorithm, the comprehensive evaluation function value of the algorithm is calculated according to the following expression:

[0013]

[0014] in, is the calculation efficiency factor, is the noise robustness factor, is the result reliability factor, is the scenario adaptability factor. The four-dimensional evaluation model is configured to assign the algorithm with the largest comprehensive evaluation function value to the corresponding technical link. After all technical links are assigned corresponding algorithms, an algorithm configuration scheme is formed, which is expressed as follows:

[0015] Algorithm_Config= {argmax(Qgraph), argmax(Qeval), argmax(Qopt),argmax(Qtime)}

[0016] Among them, Qgraph is the evaluation value of the graph modeling algorithm, Qeval is the evaluation value of the main evaluation algorithm, Qopt is the evaluation value of the optimization algorithm, and Qtime is the evaluation value of the timing algorithm.

[0017] Preferably, the algorithm configuration scheme adopts the heterogeneous graph convolutional network HGCN algorithm to construct the military knowledge graph, specifically:

[0018] Military entities and evaluation indicators are extracted from the military assessment scenario as graph nodes. The correlation coefficient and influence coefficient between the indicators are calculated to determine the association strength. When the association strength is greater than the set threshold, a connection edge is established between the corresponding nodes. For each pair of indicators, the association strength is calculated according to the following expression:

[0019]

[0020] in, For indicators and The correlation coefficient of is the influence coefficient, To integrate the spatiotemporal attenuation factor, the HGCN algorithm is configured to classify indicator relationships into three categories: causal chain relationships, hierarchical dependency relationships, and spatiotemporal constraint relationships according to the strength of association. After establishing the heterogeneous graph structure, the indicator relationship matrix is ​​calculated, which is expressed as follows:

[0021]

[0022] Among them, R is the n×n indicator relationship matrix, is the correlation strength between the i-th indicator and the j-th indicator, and the propagation weight is calculated by the influence propagation algorithm:

[0023]

[0024] in, is a nonlinear activation function, is the Sigmoid gating function, is the gradient relationship between indicators.

[0025] Preferably, the real-time calculation results of the dynamic evaluation model are obtained, and the real-time calculation results are processed by a distributed computing architecture to analyze military effectiveness trends and update evaluation parameters, specifically:

[0026] Continuously acquire simulation situation data, capture the time-varying characteristics of military effectiveness through a stream-batch integrated distributed computing architecture, and update the evaluation parameters through a state update function that includes time decay. The state update function expression is:

[0027]

[0028] in, is the evaluation state at time t, is the state at the previous moment, is the input data at time t, Decay_factor is the comprehensive spatiotemporal decay factor, To affect the propagation coefficient, is the indicator relationship weight, is a nonlinear activation function, is the indicator change.

[0029] Preferably, before capturing the time-varying characteristics of military effectiveness through the integrated stream-batch distributed computing architecture, the method further includes partitioning the simulation situation data, wherein the partitioning strategy adopts a hybrid hash partitioning algorithm, which is expressed as:

[0030] partition = (hash(unitID) + spatial_index) % N

[0031] Among them, unitID is the combat unit identifier, spatial_index is the spatial index, and N is the number of partitions.

[0032] Preferably, the comprehensive spatiotemporal attenuation factor is calculated according to the following rules:

[0033] When the evaluation involves time constraints: ,in is the time attenuation coefficient, is the time interval;

[0034] When evaluating spatial constraints: ,in is the spatial attenuation coefficient, is the spatial distance;

[0035] When the evaluation involves both temporal and spatial constraints: , realizing the joint attenuation effect of time and space.

[0036] A second embodiment of the present invention provides a military assessment device based on simulated situation data, comprising:

[0037] a configuration scheme selection unit, configured to obtain military assessment scenario data and invoke a four-dimensional assessment model to select an algorithm configuration scheme corresponding to the military assessment scenario from an intelligent algorithm library based on an efficiency factor, a noise robustness factor, a result reliability factor, and a scenario adaptability factor, wherein the military assessment scenario data includes a combat mission description, environmental parameters, force size data, and assessment requirement characteristics;

[0038] a model generation unit, configured to construct a military knowledge graph using a heterogeneous graph convolutional network (HGCN) algorithm according to the algorithm configuration scheme, and to call a meta-learning framework to provide a dynamic evaluation model to the military knowledge graph, wherein the dynamic evaluation model is generated based on a scenario semantic parser, a meta-model matching engine, and an adaptive optimization module;

[0039] A feedback adjustment unit is used to obtain the real-time calculation results of the dynamic evaluation model, process the real-time calculation results through a distributed computing architecture to analyze military effectiveness trends and update evaluation parameters, and dynamically adjust the strategy of matching indicator weights with algorithms.

[0040] A third embodiment of the present invention provides a military assessment device based on simulated situation data, characterized in that it includes a memory and a processor, wherein a computer program is stored in the memory, and the computer program can be executed by the processor to implement a military assessment method based on simulated situation data as described in any one of the above items.

[0041] The fourth embodiment of the present invention provides a computer-readable storage medium, characterized in that it stores a computer program, and the computer program can be executed by a processor of the device where the computer-readable storage medium is located to implement a military assessment method based on simulation situation data as described in any one of the above items.

[0042] The present invention discloses a military assessment method, device, equipment and storage medium based on simulation situation data. The method first obtains military assessment scenario data and calls a four-dimensional assessment model to select an algorithm configuration scheme corresponding to the military assessment scenario from an intelligent algorithm library based on an efficiency factor, a noise robustness factor, a result reliability factor, and a scenario adaptability factor. Then, a heterogeneous graph convolutional network HGCN algorithm is used to construct a military knowledge graph according to the algorithm configuration scheme, and a meta-learning framework is called to provide a dynamic assessment model to the military knowledge graph, wherein the dynamic assessment model is generated based on a scenario semantic parser, a meta-model matching engine and an adaptive optimization module; finally, the real-time calculation results of the dynamic assessment model are obtained, and the real-time calculation results are processed through a distributed computing architecture to analyze the military effectiveness trend and update the assessment parameters, and the indicator weight and algorithm matching strategy are dynamically adjusted, thereby solving the problem that the assessment system cannot quickly adapt to new scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 1 is a flow chart of a military assessment method based on simulation situation data provided by the first embodiment of the present invention;

[0044] Figure 2 1 is a module diagram of a military assessment device based on simulation situation data provided by the second embodiment of the present invention. DETAILED DESCRIPTION

[0045] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0046] In order to better understand the technical solution of the present invention, the embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0047] It should be understood that the embodiments described are only a portion of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative work are within the scope of protection of the present invention.

[0048] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The singular forms "a", "an", "the" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise.

[0049] It should be understood that the term "and / or" as used herein is merely a description of the relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document generally indicates that the associated objects are in an "or" relationship.

[0050] The word "if," as used herein, may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to the determination" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)," depending on the context.

[0051] The "first" and "second" mentioned in the embodiments are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It is understood that the specific order or precedence of "first" and "second" can be interchanged where appropriate. It should be understood that the objects distinguished by "first" and "second" can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein.

[0052] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0053] The present invention discloses a military assessment method, apparatus, device and storage medium based on simulation situation data, aiming to solve the problem that the assessment system cannot quickly adapt to new scenarios.

[0054] See also Figure 1 A first embodiment of the present invention provides a military assessment method based on simulated situation data, which can be performed by a military assessment device based on simulated situation data (hereinafter referred to as the assessment device), and in particular, is performed by one or more processors within the assessment device to implement at least the following steps:

[0055] S101, obtaining military assessment scenario data, and invoking a four-dimensional assessment model to select an algorithm configuration scheme corresponding to the military assessment scenario from an intelligent algorithm library based on an efficiency factor, a noise robustness factor, a result reliability factor, and a scenario adaptability factor, wherein the military assessment scenario data includes a combat mission description, environmental parameters, force size data, and assessment requirement characteristics;

[0056] In this embodiment, the evaluation device can be a terminal with data processing capabilities such as a server, desktop computer, or laptop computer. The evaluation device can be installed with a corresponding operating system and application software, and the functions required by this embodiment can be achieved through the combination of the operating system and application software.

[0057] Taking an air defense combat simulation evaluation scenario as an example, the evaluation device first receives user-entered military evaluation scenario data: keyword descriptions such as "air defense combat," "electronic countermeasures," and "rapid response," along with evaluation requirements for 12 core metrics, including radar detection accuracy and interception success rate. The scenario semantic parser uses the BERT-BiLSTM-CRF model to process the input data, achieving a recognition accuracy of 92.5%. It extracts combat elements and generates a scenario feature vector, which includes quantitative features such as data complexity of 0.8, time urgency of 0.9, accuracy requirement of 0.85, and scenario diversity of 0.6.

[0058] The four-dimensional evaluation model then evaluates the candidate algorithms in the intelligent algorithm library, where the efficiency factor is calculated. The reciprocal of the average execution time of the algorithm is calculated, and the AHP algorithm is 0.15, the fuzzy evaluation algorithm is 0.6, and the improved neural network algorithm is 0.3; the noise robustness factor Based on the performance evaluation of the algorithm under 10% noise interference, the results are 0.6, 0.4, and 0.8 respectively; the reliability factor 0.21, 0.20, 0.23 calculated by correlation with historical validation data; scenario adaptability factor The scores were determined to be 0.015, 0.08, and 0.12, respectively, based on the proportion of the number of scenarios in which the algorithms were applicable. The system calculated the comprehensive evaluation function value Q for each algorithm, with the AHP algorithm scoring 0.975, the fuzzy evaluation algorithm scoring 1.28, and the improved neural network algorithm scoring 1.45.

[0059] Based on the evaluation results, the system assigned the optimal algorithm to each technical component: the heterogeneous graph convolutional network (HGCN) (Qgraph=3.2) was selected for graph modeling; the AHP-entropy combination algorithm (Qeval=2.8) was selected for the main evaluation phase; the NSGA-II genetic algorithm (Qopt=3.1, crossover probability 0.85, mutation probability 0.01) was selected for optimization; and the LSTM model (Qtime=2.9) was selected for time series analysis. The resulting algorithm configuration (Algorithm_Config={HGCN, AHP-entropy combination, NSGA-II, LSTM}) reduced model construction time for air defense combat assessment from hours to minutes, increasing response speed 20 times compared to traditional methods. In actual combat exercises, the system increased the air defense success rate from 63% to 89%.

[0060] S102: constructing a military knowledge graph using a heterogeneous graph convolutional network (HGCN) algorithm according to the algorithm configuration scheme, and invoking a meta-learning framework to provide a dynamic evaluation model to the military knowledge graph, wherein the dynamic evaluation model is generated based on a scenario semantic parser, a meta-model matching engine, and an adaptive optimization module;

[0061] In the air defense combat assessment, the system constructs a military knowledge graph based on the HGCN algorithm selected in the algorithm configuration. It first extracts 12 core indicators from the "air defense combat" scenario, including radar detection accuracy, interception success rate, electronic countermeasures effectiveness, fire coordination delay, missile reload speed, radar power-up timing, interference source localization capability, anti-interference index, detection range, reaction speed, survivability, and electronic protection capability, as graph nodes. By analyzing historical combat data, the system calculates the strength of associations between indicators. For example, the correlation coefficient ρ between radar power-up timing and detectability is 0.94, the impact coefficient is 0.8, and the spatiotemporal decay factor is e^(-0.1×Δt), resulting in an association strength of 0.75. When the association strength exceeds the threshold of 0.5, the system establishes an edge between the corresponding nodes, forming a causal chain relationship of "radar power-up timing → detectability → survivability → firepower sustainability" with a weight of 0.8 → 0.9 → 0.7.

[0062] The HGCN algorithm categorizes indicator relationships into three categories: causal chains, such as radar power-on timing directly affecting intercept success rate; hierarchical dependencies, such as air defense effectiveness consisting of detection capability (weight 2.1) and interception capability (weight 2.8); and spatiotemporal constraints, such as the decay of interference source localization capability over time and distance. The system generates a 12×12 indicator relationship matrix R, where the correlation strength between radar accuracy and detection range is 0.8, and the correlation strength between detection range and interception accuracy is 0.6. The influence propagation algorithm calculates that when the gradient change in radar detection accuracy exceeds 0.15 / min, it marks the decision window. The propagation weights are processed and output using a nonlinear activation function and sigmoid gating.

[0063] The meta-learning framework provides a dynamic evaluation model based on the constructed knowledge graph. After the scenario semantic parser identifies the "rapid response" requirement, the meta-model matching engine uses an improved cosine similarity algorithm to match the air defense combat template, achieving a 92% match. The adaptive optimization module uses the NSGA-II genetic algorithm to optimize the objective function F = [accuracy, real-time performance], with a population size of 200 and 50 iterations, ultimately generating a dynamic evaluation model with a three-layer evaluation architecture. This model solves the problem of isolated indicator analysis and enables a globally coordinated assessment of battlefield elements. It achieves a 20-fold increase in response speed compared to traditional methods, and achieves a 93.7% accuracy rate in identifying key factors.

[0064] S103, obtaining the real-time calculation results of the dynamic evaluation model, processing the real-time calculation results through a distributed computing architecture to analyze military effectiveness trends and update evaluation parameters, and dynamically adjusting the strategy of matching indicator weights with algorithms.

[0065] During the real-time evaluation phase of air defense operations, the system deploys a stream-batch integrated distributed computing pipeline to continuously acquire simulation situation data. The data input layer uses Apache Pulsar to process 500,000 radar data per second, which enters the computing layer after time offset correction by the DTW spatiotemporal alignment module. The system implements a hash partitioning strategy based on the combat unit ID to ensure the locality of data in the same unit, and uses a 15-second sliding window and a 5-second step size for window aggregation processing. When the blue side's anti-landing firepower caused the red side's beachhead breakthrough speed to drop from 85 meters / minute to 42 meters / minute at 06:35, the state update function A sharp drop in efficiency was captured, where the spatiotemporal attenuation factor of 0.8 reflects the influence weight of the historical state, the current input weight of 0.2 reflects the real-time situation change, and the impact propagation item calculates the linkage effect through the weight of 0.85 between the radar detection accuracy and the interception success rate.

[0066] The distributed computing architecture reduces recalculation by 80% through incremental state management, and a fault-tolerance mechanism based on Chandy-Lamport snapshots ensures fault recovery time of less than 500ms. When a gradient change in intercept success rate exceeding 0.15 / minute is detected, the system automatically marks it as a decision window and triggers dynamic weighting adjustments, increasing the weight of electronic countermeasure effectiveness from 0.15 to 0.20 to adapt to the changing battlefield situation. The real-time mode, based on Flink, achieves millisecond-level response, pushing evaluation results via WebSocket with latency under 200ms (P95 indicator). An innovative battlefield data compression algorithm reduces network load by 65%. The system continuously monitors the Shapley values ​​of various indicators. When the radar power-on timing influence exceeds 0.3 for a sustained period, it automatically recommends a 30-second early power-on strategy, which is expected to increase the intercept success rate by 25%. This enables closed-loop optimization of evaluation parameters and dynamic strategy adjustment.

[0067] In one possible implementation of the present invention, in air defense combat simulations, the system first intelligently partitions the massive amount of simulation situation data before processing it in a stream-batch integrated distributed computing architecture. For example, the system uses a hybrid hash partitioning algorithm partition = (hash(unitID) + spatial_index) % N to pre-process the 500,000 radar data per second generated by 25 combat units on both sides of the red and blue sides. The hash value of combat unit T001 is added to its spatial index 128.7 and then modulo the number of partitions 16 to ensure that the radar detection, communication status, location information and other data of the unit are assigned to the same partition node. The spatial index is divided based on the combat area. The navigation area, beach area, and depth area correspond to different spatial_index values, so that the data of combat units with similar geographical locations can be distributed and processed nearby, reducing cross-partition communication overhead.

[0068] This partitioning strategy ensures data locality. When the Blue Force's anti-landing firepower impacted the Red Force's assault group at 06:35, data on the damage rate, breakthrough speed changes, and fire suppression intensity of each amphibious armored vehicle within the same landing formation were all processed within the same partition, avoiding data transmission delays associated with distributed computing. The system uses unitID hashing to ensure uniform data distribution and prevent hotspot partitions. The spatial_index ensures efficient spatial correlation analysis. After partitioning, each partition node performs parallel processing using a 15-second sliding window and a 5-second step size. This integrated streaming and batch architecture supports both real-time situation monitoring and offline analysis of historical data, reducing terabyte-level data processing time by 24 times.

[0069] In one possible implementation of the present invention, the comprehensive spatiotemporal attenuation factor is calculated according to the following rules:

[0070] When the evaluation involves time constraints: ,in is the time attenuation coefficient, is the time interval;

[0071] When evaluating spatial constraints: ,in is the spatial attenuation coefficient, is the spatial distance;

[0072] When the evaluation involves both temporal and spatial constraints: , realizing the joint attenuation effect of time and space.

[0073] In the military evaluation system, the comprehensive time-space attenuation factor is dynamically calculated according to the constraint type of the indicator relationship. When evaluating the time-constrained impact of electronic interference on communication quality, the system sets the time attenuation coefficient = 0.1, for a 10-minute time interval , the calculation results show that Decay_factor = e^(-0.1×10) = 0.37, indicating that the electronic jamming effect decays exponentially over time. When analyzing the spatial constraint relationship between friendly fire support and the combat effectiveness of the assault unit, the system uses the spatial decay coefficient =0.2, when the space distance When the distance is 5 kilometers, the calculated Decay_factor = e^(-0.2×5) =0.37, reflecting that the fire support effect decreases significantly with increasing distance.

[0074] In complex battlefield environments, certain military effectiveness indicators are subject to both temporal and spatial constraints. The system employs a joint attenuation model for comprehensive calculation. Taking radar detection accuracy as an example, for a 5-minute time interval and a 3-kilometer spatial distance, the system calculates the temporal attenuation e^(-0.1×5)=0.61 and the spatial attenuation e^(-0.15×3)=0.64, respectively, ultimately yielding a combined spatiotemporal attenuation factor of Decay_factor = 0.61×0.64 = 0.39. This joint attenuation mechanism ensures that the evaluation model accurately reflects the dynamic characteristics of military effectiveness in both temporal and spatial dimensions. It adaptively adjusts the attenuation coefficient based on different combat scenarios and indicator characteristics, achieving precise modeling of both the temporal and spatial correlations of battlefield situations.

[0075] In one possible implementation of the present invention, the influence of each factor is quantified using a military variant of the Shapley value in the process of analyzing the military effectiveness trend. The calculation formula is:

[0076]

[0077] Where φi(v) is the Shapley value of factor i, S is the subset that does not contain i, v is the alliance value function, β is the time decay coefficient, and t is the time variable.

[0078] Furthermore, the distribution data of the Shapley values ​​of each factor is obtained, the statistical characteristics of the distribution data are calculated, and the abnormal contribution range of the key influencing factors is identified through the anomaly detection algorithm; the evaluation strategy and weight configuration are dynamically adjusted based on the key influencing factor analysis and the abnormal contribution range.

[0079] See also Figure 2 A second embodiment of the present invention provides a military assessment device based on simulation situation data, comprising:

[0080] A configuration scheme selection unit 201 is used to obtain military assessment scenario data and call a four-dimensional assessment model to select an algorithm configuration scheme corresponding to the military assessment scenario from an intelligent algorithm library based on an efficiency factor, a noise robustness factor, a result reliability factor, and a scenario adaptability factor, wherein the military assessment scenario data includes combat mission description, environmental parameters, force size data, and assessment requirement characteristics;

[0081] A model generation unit 202 is configured to construct a military knowledge graph using a heterogeneous graph convolutional network (HGCN) algorithm according to the algorithm configuration scheme, and to call a meta-learning framework to provide a dynamic evaluation model to the military knowledge graph, wherein the dynamic evaluation model is generated based on a scenario semantic parser, a meta-model matching engine, and an adaptive optimization module;

[0082] The feedback adjustment unit 203 is used to obtain the real-time calculation results of the dynamic evaluation model, process the real-time calculation results through a distributed computing architecture to analyze military effectiveness trends and update evaluation parameters, and dynamically adjust the strategy of matching indicator weights with algorithms.

[0083] A third embodiment of the present invention provides a military assessment device based on simulated situation data, characterized in that it includes a memory and a processor, wherein a computer program is stored in the memory, and the computer program can be executed by the processor to implement a military assessment method based on simulated situation data as described in any one of the above items.

[0084] The fourth embodiment of the present invention provides a computer-readable storage medium, characterized in that it stores a computer program, and the computer program can be executed by a processor of the device where the computer-readable storage medium is located to implement a military assessment method based on simulation situation data as described in any one of the above items.

[0085] The present invention discloses a military assessment method, device, equipment and storage medium based on simulation situation data. The method first obtains military assessment scenario data and calls a four-dimensional assessment model to select an algorithm configuration scheme corresponding to the military assessment scenario from an intelligent algorithm library based on an efficiency factor, a noise robustness factor, a result reliability factor, and a scenario adaptability factor. Then, a heterogeneous graph convolutional network HGCN algorithm is used to construct a military knowledge graph according to the algorithm configuration scheme, and a meta-learning framework is called to provide a dynamic assessment model to the military knowledge graph, wherein the dynamic assessment model is generated based on a scenario semantic parser, a meta-model matching engine and an adaptive optimization module; finally, the real-time calculation results of the dynamic assessment model are obtained, and the real-time calculation results are processed through a distributed computing architecture to analyze the military effectiveness trend and update the assessment parameters, and the indicator weight and algorithm matching strategy are dynamically adjusted, thereby solving the problem that the assessment system cannot quickly adapt to new scenarios.

[0086] For example, the computer programs described in the third and fourth embodiments of the present invention can be divided into one or more modules, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules can be a series of computer program instruction segments capable of performing specific functions, which are used to describe the execution process of the computer program in the device for implementing a military assessment based on simulated situation data. For example, the device described in the second embodiment of the present invention.

[0087] The processor may be a central processing unit (CPU), or other general-purpose processors, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the military assessment method based on simulated situation data, and utilizes various interfaces and lines to connect the various parts of the military assessment method based on simulated situation data.

[0088] The memory can be used to store the computer programs and / or modules. The processor implements various functions of a military assessment method based on simulated situational data by running or executing the computer programs and / or modules stored in the memory and accessing data stored in the memory. The memory may primarily include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for a function (such as a sound playback function or a text conversion function); the data storage area may store data generated based on the use of the mobile phone (such as audio data and text message data). Furthermore, the memory may include high-speed random access memory (RAM) and non-volatile memory, such as a hard disk, internal memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0089] If the implemented module is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention can implement all or part of the process steps in the above-mentioned method embodiments by using a computer program to instruct the relevant hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium. It should be noted that the content of the computer-readable medium can be appropriately increased or decreased based on the requirements of legislation and patent practice in a jurisdiction. For example, in some jurisdictions, based on legislation and patent practice, computer-readable media does not include electric carrier signals and telecommunication signals.

[0090] It should be noted that the device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. In addition, in the drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art can understand and implement the present invention without inventive effort.

[0091] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A military assessment method based on simulation situation data, characterized in that: include: Obtaining military assessment scenario data, and calling a four-dimensional assessment model to select an algorithm configuration scheme corresponding to the military assessment scenario from an intelligent algorithm library based on an efficiency factor, a noise robustness factor, a result reliability factor, and a scenario adaptability factor, wherein the military assessment scenario data includes a combat mission description, environmental parameters, force size data, and assessment requirement characteristics; According to the algorithm configuration scheme, the heterogeneous graph convolutional network (HGCN) algorithm is used to construct a military knowledge graph. Specifically, military entities and evaluation indicators are extracted from the military evaluation scenario as graph nodes. The correlation coefficient and influence coefficient between the indicators are calculated to determine the association strength. When the association strength is greater than the set threshold, a connection edge is established between the corresponding nodes. For each pair of indicators, the association strength is calculated according to the following expression: in, For indicators and The correlation coefficient of is the influence coefficient, To integrate the spatiotemporal attenuation factor, the HGCN algorithm is configured to classify indicator relationships into three categories: causal chain relationships, hierarchical dependency relationships, and spatiotemporal constraint relationships according to the strength of association. After establishing the heterogeneous graph structure, the indicator relationship matrix is ​​calculated, which is expressed as follows: Among them, R is the n×n indicator relationship matrix, is the correlation strength between the i-th indicator and the j-th indicator, and the propagation weight is calculated by the influence propagation algorithm: in, is a nonlinear activation function, is the indicator relationship weight, is the Sigmoid gating function, For the gradient relationship between indicators, a meta-learning framework is called to provide a dynamic evaluation model to the military knowledge graph, wherein the dynamic evaluation model is generated based on a scenario semantic parser, a meta-model matching engine, and an adaptive optimization module; Obtain real-time calculation results of the dynamic evaluation model, process the real-time calculation results through a distributed computing architecture to analyze military effectiveness trends and update evaluation parameters, and dynamically adjust the strategy of matching indicator weights with algorithms.

2. A military assessment method based on simulation situation data according to claim 1, characterized in that: Obtain military assessment scenario data, and call the four-dimensional assessment model to select an algorithm configuration scheme corresponding to the military assessment scenario from the intelligent algorithm library based on the efficiency factor, noise robustness factor, result reliability factor, and scenario adaptability factor. Specifically, The military evaluation scenario data is processed by a scenario semantic parser to generate a scenario feature vector. The candidate algorithms are evaluated by a four-dimensional evaluation model and classified into different applicable categories. For each algorithm, the comprehensive evaluation function value of the algorithm is calculated according to the following expression: in, is the calculation efficiency factor, is the noise robustness factor, is the result reliability factor, is the scenario adaptability factor. The four-dimensional evaluation model is configured to assign the algorithm with the largest comprehensive evaluation function value to the corresponding technical link. After all technical links are assigned corresponding algorithms, an algorithm configuration scheme is formed, which is expressed as follows: Algorithm_Config= {argmax(Qgraph), argmax(Qeval), argmax(Qopt), argmax(Qtime)} Among them, Qgraph is the evaluation value of the graph modeling algorithm, Qeval is the evaluation value of the main evaluation algorithm, Qopt is the evaluation value of the optimization algorithm, and Qtime is the evaluation value of the timing algorithm.

3. A military assessment method based on simulation situation data according to claim 1, characterized in that: The real-time calculation results of the dynamic evaluation model are obtained, and the real-time calculation results are processed by a distributed computing architecture to analyze military effectiveness trends and update evaluation parameters, specifically: Continuously acquire simulation situation data, capture the time-varying characteristics of military effectiveness through a stream-batch integrated distributed computing architecture, and update the evaluation parameters through a state update function that includes time decay. The state update function expression is: in, is the evaluation state at time t, is the state at the previous moment, is the input data at time t, Decay_factor is the comprehensive spatiotemporal decay factor, To affect the propagation coefficient, is the indicator relationship weight, is a nonlinear activation function, is the indicator change.

4. A military assessment method based on simulation situation data according to claim 3, characterized in that: Before capturing the temporal variation characteristics of military effectiveness through the integrated stream-batch distributed computing architecture, the method further includes partitioning the simulation situation data, wherein the partitioning strategy adopts a hybrid hash partitioning algorithm, which is expressed as: partition = (hash(unitID) + spatial_index) % N Among them, unitID is the combat unit identifier, spatial_index is the spatial index, and N is the number of partitions.

5. A military assessment method based on simulation situation data according to claim 3, characterized in that: The comprehensive spatiotemporal attenuation factor is calculated according to the following rules: When the evaluation involves time constraints: ,in is the time attenuation coefficient, is the time interval; When evaluating spatial constraints: ,in is the spatial attenuation coefficient, is the spatial distance; When the evaluation involves both temporal and spatial constraints: , realizing the joint attenuation effect of time and space.

6. A military assessment device based on simulation situation data, characterized in that: include: a configuration scheme selection unit, configured to obtain military assessment scenario data and invoke a four-dimensional assessment model to select an algorithm configuration scheme corresponding to the military assessment scenario from an intelligent algorithm library based on an efficiency factor, a noise robustness factor, a result reliability factor, and a scenario adaptability factor, wherein the military assessment scenario data includes a combat mission description, environmental parameters, force size data, and assessment requirement characteristics; The model generation unit is configured to construct a military knowledge graph using the heterogeneous graph convolutional network (HGCN) algorithm according to the algorithm configuration scheme, specifically by extracting military entities and evaluation indicators from the military evaluation scenario as graph nodes, determining the association strength by calculating the correlation coefficient and influence coefficient between the indicators, establishing a connection edge between the corresponding nodes when the association strength is greater than a set threshold, and calculating the association strength for each pair of indicators according to the following expression: in, For indicators and The correlation coefficient of is the influence coefficient, To integrate the spatiotemporal attenuation factor, the HGCN algorithm is configured to classify indicator relationships into three categories: causal chain relationships, hierarchical dependency relationships, and spatiotemporal constraint relationships according to the strength of association. After establishing the heterogeneous graph structure, the indicator relationship matrix is ​​calculated, which is expressed as follows: Among them, R is the n×n indicator relationship matrix, is the correlation strength between the i-th indicator and the j-th indicator, and the propagation weight is calculated by the influence propagation algorithm: in, is a nonlinear activation function, is the indicator relationship weight, is the Sigmoid gating function, is the gradient relationship between indicators; Invoking a meta-learning framework to provide a dynamic evaluation model to the military knowledge graph, wherein the dynamic evaluation model is generated based on a scenario semantic parser, a meta-model matching engine, and an adaptive optimization module; A feedback adjustment unit is used to obtain the real-time calculation results of the dynamic evaluation model, process the real-time calculation results through a distributed computing architecture to analyze military effectiveness trends and update evaluation parameters, and dynamically adjust the strategy of matching indicator weights with algorithms.

7. A military assessment device based on simulation situation data, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program, and the computer program can be executed by the processor to implement a military assessment method based on simulation situation data as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that A computer program is stored, and the computer program can be executed by a processor of the device where the computer-readable storage medium is located to implement a military assessment method based on simulation situation data as described in any one of claims 1 to 5.

Images