Battle decision auxiliary system and method based on big data analysis

Abstract

The invention relates to the technical field of informatization war, and particularly provides a big data analysis-based combat decision auxiliary system and method, and the system comprises a data collection module, a data preprocessing module, a multi-source intelligence fusion module, a situation prediction depth mining and analysis module, a military strength deployment module, and an auxiliary optimization module. The method comprises the following steps: S101, data acquisition, connection and preprocessing; s102, fusing multi-source intelligence to judge a target; s103, accurately analyzing and predicting the attempt of the enemy; s104, carrying out multi-dimensional association recommendation on the available strength; s105, assisting in optimizing a combat scheme; according to the scheme, how to apply a big data analysis technology is introduced in detail, a set of efficient and intelligent combat scheme generation system is constructed, and scientific and accurate tactical and strategic guidance is provided for military decision makers. According to the system, key technical links such as data acquisition, fusion, analysis, modeling and visualization are integrated so as to cope with a complex and dynamic battlefield environment, and maximization of combat effectiveness is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of information warfare, and in particular to a combat decision-making support system and method based on big data analysis. Background Art

[0002] With the rapid development of information technology, modern warfare is becoming increasingly complex and dynamic.

[0003] Traditional combat decision-making methods often rely on the commander's experience and intuition, as well as limited battlefield information, which is inadequate in the face of information explosion and rapid changes in the modern battlefield. Therefore, in the current combat plan formulation process, the commander alone faces problems such as scattered data sources, low processing efficiency, and lack of in-depth analysis support for decision-making, resulting in poor overall combat plan formulation results. Summary of the invention

[0004] In view of the shortcomings of the prior art mentioned above, the purpose of the present invention is to provide a combat decision-making assistance system and method based on big data analysis, which uses big data analysis and data mining technology to provide intelligent auxiliary decision support for military commanders, so as to solve the problem that the prior art only relies on commanders to manually organize large amounts of redundant data to formulate combat plans, which is ineffective.

[0005] To achieve the above-mentioned and other related purposes, the present invention provides a combat decision support system based on big data analysis, the system comprising:

[0006] Data acquisition module, used to access multi-source battlefield data information in real time;

[0007] Data preprocessing module, used for cleaning, denoising, format conversion, and spatiotemporal calibration preprocessing operations on multi-source battlefield data;

[0008] A multi-source intelligence fusion module, which uses a data fusion algorithm to integrate, remove redundancy, and resolve conflicts in fragmented intelligence information from different data sources; the data fusion algorithm includes a target recognition confidence calculation submodule, which is used to quantify the credibility of each data source for target recognition and adjust the weight of the target attribute in the fusion result accordingly;

[0009] A situation prediction deep mining and analysis module, which generates a prediction model through a machine learning algorithm and performs deep mining and analysis on enemy combat information data;

[0010] A force deployment module, which integrates regional data, force data, and force status tasks to build a multi-dimensional correlation model including strategic resources and combat tasks. By inputting combat task requirements, it can obtain recommendations on available force combinations, configuration plans, and action sequences suitable for the current battlefield situation;

[0011] The auxiliary optimization module uses operations research methods combined with artificial intelligence algorithms to calculate a variety of alternative combat plans that meet predetermined strategic goals, tactical principles and constraints, and provides commanders with intuitive plan comparisons, advantages and disadvantages analysis, and decision-making suggestions through combat simulation, risk assessment, and benefit analysis, assisting commanders to quickly and scientifically select the combat plan that best suits battlefield reality and has the greatest combat advantages.

[0012] In one embodiment of the present invention, the data preprocessing module also includes a data integrity verification mechanism for detecting and filling missing data, and a data consistency verification mechanism for resolving data conflicts between different data sources.

[0013] In one embodiment of the present invention, in the situation prediction deep mining and analysis module, the prediction model includes an adaptive moment estimation algorithm, which dynamically and adaptively updates the prediction model according to newly acquired battlefield parameters to maintain prediction accuracy.

[0014] In one embodiment of the present invention, the force deployment module includes a force effectiveness evaluation submodule, a force demand prediction submodule, a force optimization configuration submodule, a force dynamic adjustment submodule, and a resource consumption tracking and supply planning submodule;

[0015] The force effectiveness evaluation submodule is pre-set with a combat effectiveness evaluation model, in which the combat effectiveness evaluation model combines the data characteristics of various strategic resources to quantitatively evaluate the combat capability of a single combat unit or the entire combat system;

[0016] The force demand prediction submodule integrates time series analysis, regression analysis, and neural network prediction methods. Based on historical combat data, enemy and friendly situation analysis, mission planning information, and combat rules, it predicts the type, quantity, configuration, and support requirements of forces required in different combat scenarios in the future, providing a basis for force scheduling and configuration.

[0017] The force optimization configuration submodule adopts an optimization algorithm, which includes linear programming, integer programming, and genetic algorithm, to calculate the optimal force allocation plan according to the combat plan requirements and force demand prediction results, the existing strategic resource status, and battlefield environment constraints;

[0018] The dynamic force adjustment submodule adopts a feedback control mechanism, which triggers the dynamic adjustment of strategic resources by real-time monitoring of battlefield changes, combat progress, and force loss;

[0019] The resource consumption tracking and supply planning submodule is used to monitor the consumption of strategic resources during combat in real time, including ammunition, oil, and material losses, and to formulate supply strategies and emergency supply plans based on factors such as remaining resource reserves, supply capabilities, and supply route security, using supply chain optimization and inventory management methods to ensure the continuous supply of strategic resources during combat.

[0020] In one embodiment of the present invention, in the auxiliary optimization module, the risk assessment includes risk factor identification, risk probability assessment, risk impact assessment, risk level classification, risk response strategy formulation, risk monitoring and early warning, risk assessment result application, and risk assessment update.

[0021] A combat decision support method based on big data analysis, and a combat decision support system based on the big data analysis, comprising the following steps:

[0022] S101, data acquisition and preprocessing: collect battlefield data in real time through acquisition equipment, including satellite remote sensing data, radar detection data, drone reconnaissance, electronic intelligence and image data, and perform denoising and standardization on the image data;

[0023] S102. Multi-source intelligence fusion to identify targets: Use data fusion algorithms to fuse pre-processed data, integrate complementary and redundant information in multi-source data, resolve conflicting information, and use target recognition confidence calculation to quantify the credibility of target recognition from each data source, and adjust the weight of target attributes in the fusion result accordingly, so as to achieve accurate recognition, positioning, tracking and attribute labeling of both enemy and friendly targets, and generate a unified and real-time updated target situation map;

[0024] S103. Accurately analyze and predict enemy intentions: Collect enemy combat information data extensively and generate prediction models through machine learning algorithms. Combined with real-time combat data related to combat missions, conduct in-depth analysis to build a comprehensive battlefield situation map to intuitively reveal the advantages, disadvantages, opportunities and threats of the current combat missions.

[0025] S104. Multi-dimensional correlation recommendation of available forces: Use big data correlation analysis and optimization algorithms to integrate regional data, force data, and force status tasks to build a multi-dimensional correlation model that includes strategic resources and combat tasks, taking into account multi-dimensional factors including battlefield environment, mission requirements, force status, and enemy and our strength comparison. After inputting combat mission requirements, the multi-dimensional correlation model intelligently recommends force combinations, configuration plans, and action sequences;

[0026] S105. Assist in selecting optimal combat plans: Combine the above intelligence analysis results with the recommended force deployment, use operations research methods combined with artificial intelligence algorithms, and calculate and generate a variety of combat plans that meet predetermined strategic goals, follow tactical principles, and comply with battlefield constraints.

[0027] In one embodiment of the present invention, in step S101, the acquisition equipment includes a sensor, a radar, and a satellite.

[0028] In one embodiment of the present invention, in step S103, the enemy combat information data includes historical action records, troop deployment changes, communication network traffic, and social media information.

[0029] In one embodiment of the present invention, in step S104, the multi-dimensional correlation model covers the relationship between strategic resources and combat tasks, and also goes deep into the levels of force effectiveness evaluation, demand forecasting, optimal configuration, dynamic adjustment, resource consumption tracking and supply planning.

[0030] In one embodiment of the present invention, in step S105, the artificial intelligence algorithm includes one or more of a genetic algorithm, a particle swarm optimization, and a reinforcement learning algorithm.

[0031] As described above, the combat decision support system and method based on big data analysis of the present invention has the following beneficial effects:

[0032] The present invention collects, integrates and analyzes combat data, and uses big data technology and artificial intelligence algorithms to generate scientific, efficient and accurate combat plans, which can effectively improve the efficiency and accuracy of combat plan generation, reduce the subjectivity and errors of manual analysis and experience judgment, and provide strong support for military decision-making; enhance the flexibility and adaptability of combat plans, and can quickly adjust and optimize combat plans according to real-time battlefield changes, improve the adaptability of combat operations; improve the intelligence level of military decision-making, and discover potential patterns and trends through big data analysis and mining technology, so as to provide commanders with scientific and accurate decision-making basis. Therefore, the present invention significantly improves the scientificity, accuracy and response speed of military decision-making by efficiently integrating multi-source data, deeply analyzing battlefield situations, and intelligently generating and optimizing combat plans; this scheme aims to introduce in detail how to use big data analysis technology to build a set of efficient and intelligent combat plan generation system, and provide scientific and accurate tactical and strategic guidance for military decision makers; the system will integrate key technical links such as data collection, fusion, analysis, modeling and visualization to cope with complex and dynamic battlefield environments and maximize combat effectiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 Shown is a flow chart of the combat decision-making assistance method based on big data analysis disclosed in the present invention.

[0034] Figure 2 Shown is a block diagram of the combat decision support system based on big data analysis disclosed in the present invention.

[0035] Component number description

[0036] Data collection module 1; data preprocessing module 2; multi-source intelligence fusion module 3; situation prediction deep mining and analysis module 4; force deployment module 5; auxiliary optimization module 6. DETAILED DESCRIPTION

[0037] The following is a description of the implementation of the present invention by means of specific embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification.

[0038] See also Figure 1 to Figure 2 It should be noted that the structures, proportions, sizes, etc. illustrated in the drawings in this specification are only used to match the contents disclosed in the specification for people familiar with this technology to understand and read, and are not used to limit the conditions under which the present invention can be implemented. Therefore, they have no substantial technical significance. Any structural modification, change in proportion or adjustment of size should still fall within the scope of the technical contents disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention.

[0039] Example 1, please refer to Figure 2 This embodiment provides a combat decision support system based on big data analysis, the system includes a data acquisition module 1, a data preprocessing module 2, a multi-source intelligence fusion module 3, a situation prediction deep mining and analysis module 4, a force deployment module 5, and an auxiliary optimization module 6. The data acquisition module 1 is used to access multi-source battlefield data information in real time.

[0040] The data preprocessing module 2 is used to perform cleaning, denoising, format conversion, and spatiotemporal calibration preprocessing operations on multi-source battlefield data to ensure the quality, consistency, and timeliness of the data; the data preprocessing module 2 also includes a data integrity verification mechanism that can detect and fill in missing data, as well as a data consistency verification mechanism for resolving data conflicts between different data sources.

[0041] The multi-source intelligence fusion module 3 adopts data fusion algorithms, including Kalman filtering, Bayesian networks, deep clustering, etc., to integrate, remove redundancy, and resolve conflicts in fragmented intelligence information from different data sources, so as to achieve accurate identification, positioning, tracking, and attribute labeling of both enemy and friendly targets, and generate a unified and real-time updated target situation map; the data fusion algorithm includes a target identification confidence calculation submodule, which is used to quantify the credibility of each data source for target identification, and adjust the weight of the target attribute in the fusion result accordingly.

[0042] The situation prediction deep mining and analysis module 4 generates a prediction model through machine learning algorithms, which include deep neural networks, time series analysis, anomaly detection models, etc., and conducts deep mining and analysis of enemy combat information data. By deep mining and analysis of enemy historical action records, troop deployment changes, communication network traffic, social media information and other data, it predicts the enemy's possible action routes, time nodes, tactical actions, combat intentions and other key information.

[0043] The force deployment module 5 integrates regional data, force data and force status tasks to construct a multi-dimensional correlation model including strategic resources and combat tasks. By inputting the combat task requirements, the available force combination, configuration plan and action sequence suitable for the current battlefield situation can be recommended; the force deployment module 5 includes a force effectiveness evaluation submodule, a force demand prediction submodule, a force optimization configuration submodule, a force dynamic adjustment submodule, and a resource consumption tracking and supply planning submodule; the force effectiveness evaluation submodule is preset with a combat effectiveness evaluation model, and the combat capability of a single combat unit or the entire combat system is quantitatively evaluated in the combat effectiveness evaluation model in combination with the data characteristics of various strategic resources; the submodule considers factors including but not limited to force size, weapon and equipment performance, personnel training level, tactical coordination tacit understanding, logistics support capability, psychological endurance, battlefield adaptability, etc., and can dynamically optimize model parameters through machine learning algorithms to improve evaluation accuracy. The force demand prediction submodule integrates time series analysis, regression analysis, and neural network prediction methods, and predicts the type, quantity, configuration, and support requirements of troops required in different combat scenarios in the future based on historical combat data, enemy and friendly situation analysis, task planning information, and combat rules, providing a basis for force scheduling and configuration. The force optimization configuration submodule uses optimization algorithms, such as linear programming, integer programming, genetic algorithms, etc., to calculate the optimal force allocation plan based on the combat plan requirements and force demand prediction results, the existing strategic resource status, and battlefield environmental constraints, such as terrain, climate, and transportation networks, etc., including the force dispatch path, assembly location, task allocation, rotation plan, and reserve force setting, so as to maximize the overall combat effectiveness while taking into account the rational use of strategic resources and loss control. The dynamic force adjustment submodule adopts a feedback control mechanism, which triggers the dynamic adjustment of strategic resources by real-time monitoring of battlefield changes, combat progress, and force loss. The adjustment process includes but is not limited to re-evaluating force requirements, updating prediction models, adjusting and optimizing configuration plans, and executing emergency deployment instructions, etc., to ensure that the strategic resource allocation always matches the actual battlefield needs and maintains the effectiveness and adaptability of the combat plan. The resource consumption tracking and supply planning submodule is used to monitor the consumption of strategic resources in the combat process in real time, including casualties, equipment loss, ammunition consumption, material consumption, etc., and combines the remaining resource reserves, supply capacity, and supply line security factors. Supply chain optimization and inventory management methods are used to formulate supply strategies and emergency supply plans to ensure the continuous supply of strategic resources during the combat process.

[0044] The auxiliary optimization module adopts operations research methods combined with artificial intelligence algorithms, such as genetic algorithms, particle swarm optimization, reinforcement learning, etc., to calculate a variety of alternative combat plans that meet predetermined strategic goals, tactical principles and constraints, and through combat simulation, risk assessment, and benefit analysis, it provides commanders with intuitive plan comparison, advantages and disadvantages analysis, and decision-making suggestions, assisting commanders to quickly and scientifically select the combat plan that best suits battlefield reality and has the greatest combat advantages; the risk assessment includes risk factor identification, risk probability assessment, risk impact assessment, risk level classification, risk response strategy formulation, risk monitoring and early warning, risk assessment result application, and risk assessment update.

[0045] The risk factor identification includes systematically identifying various risk factors that may affect the execution of combat plans based on military theory, actual combat experience, enemy situation analysis, battlefield environment and other factors. Risk factors include strategic risk, tactical risk, technical risk, organizational risk, environmental risk, human risk, etc., forming a risk factor list; the risk probability assessment includes determining the probability of occurrence of each identified risk factor based on historical data, expert judgment, simulation experiments, intelligence analysis, etc., and can be assessed by qualitative scoring, quantitative calculation, probability distribution modeling, etc., and introduce sensitivity analysis or probability interval estimation for risk factors with strong uncertainty; the risk impact assessment includes assessing the possible negative impact of each risk factor on the achievement of combat plan goals, combat progress, casualties, equipment losses, combat costs, public opinion influence, alliance relations, etc. once it occurs, and using loss functions, utility functions, decision trees, scenario analysis and other methods to quantify the degree of risk impact, and consider the interaction and cumulative effect between risks; the risk level classification includes dividing risk factors into different levels, such as high risk, medium risk, and low risk, based on the risk probability and impact assessment results, and comprehensively using risk matrices, risk indexes, risk weights and other tools to give priority to and deal with risks that have a greater impact on combat plans; the risk response strategy formulation includes formulating a risk response strategy for each level of risk. The prevention measures include strengthening intelligence collection, improving technical equipment, strengthening training and exercises, etc., the mitigation measures include setting up backup systems, increasing redundant resources, optimizing combat procedures, etc., the transfer measures include signing insurance contracts, seeking external assistance, dispersing task responsibilities, etc., and the acceptance measures are to choose to accept risks and be prepared to bear losses when the risks are known and the impact is controllable; the risk monitoring and early warning include establishing a risk monitoring mechanism, monitoring the status changes of various risk factors in real time during the implementation of the combat plan, using data mining, artificial intelligence, Internet of Things and other technologies to timely discover risk signs, trigger early warning signals, and initiate corresponding response plans; the application of risk assessment results includes integrating risk assessment results and response strategies into the combat plan, which is reflected in task allocation, force deployment, action routes, time nodes, command and control, logistics support and other aspects to ensure that the combat plan has sufficient risk resistance and emergency response capabilities; the risk assessment update includes updating the risk assessment results regularly or on demand with changes in battlefield situation, enemy and friendly situation, technological progress and other factors, re-classifying risk levels, adjusting response strategies, and ensuring the timeliness and effectiveness of risk assessment.

[0046] Example 2, please refer to Figure 1 Based on Example 1, this embodiment provides a combat decision-making assistance method based on big data analysis, including the following steps S101 to S105.

[0047] S101. Data collection and preprocessing: battlefield data is collected in real time through collection equipment, including satellite remote sensing data, radar detection data, drone reconnaissance, electronic intelligence and image data, and the image data is denoised and standardized; the collection equipment includes sensors, radars, satellites, etc., and the collected data includes satellite remote sensing data, radar detection data, drone reconnaissance data, ground force reports, electronic signal intelligence, etc.; the collected raw data may have various potential problems, including but not limited to missing values, outliers, duplicate records and inconsistent formats. Missing values ​​may be caused by omissions or errors in the data collection process, while outliers may be caused by data entry errors, measurement errors or extreme events. In addition, duplicate records may be caused by improper operations during data integration, and inconsistent formats may be caused by compatibility issues between multiple data sources. This requires cleaning the collected raw data to remove duplicate, erroneous or irrelevant information. For missing values, appropriate filling strategies will be selected according to the characteristics of the data and analysis requirements, or missing values ​​will be estimated through prediction models. For abnormal values, professionals conduct a detailed review to determine whether they are real errors or anomalies, and then decide whether to retain, correct or delete these values. For duplicate records, deduplication is performed to ensure that each record in the data set is unique. At the same time, the format of the data is unified and standardized to ensure data consistency and comparability. Next, the data is denoised. Denoising image data is an important step to improve data quality, enhance image clarity and reduce analysis errors. Image data denoising aims to eliminate or reduce unnecessary random noise in the image. These noises first require in-depth analysis of the noise characteristics of the image data. Different types of noise, such as Gaussian noise, salt and pepper noise, Poisson noise, etc., have different performances on the image and have different effects on image quality. Therefore, it is crucial to choose the right denoising algorithm and parameters. For example, for Gaussian noise, Gaussian filtering or Wiener filtering is often used for smoothing; while for verification noise, median filtering or morphological filtering may be more preferred. It may come from sensor failure, interference during transmission, changes in ambient light and many other factors. When performing denoising, it is also necessary to balance the relationship between noise suppression and detail retention. Excessive denoising may cause loss of image details, affecting image clarity and recognizability; insufficient denoising cannot effectively eliminate noise, reducing image quality. Therefore, the selection of denoising algorithms and parameter adjustments need to be finely adjusted and optimized according to specific image data and application scenarios. In addition, by training deep learning models to automatically learn the noise distribution and characteristics in images, more accurate and efficient denoising can be achieved. Deep learning denoising methods can not only handle various complex noise types, but also better retain image details and edge information, improving denoising effects.The denoised image data needs to be quality assessed and verified. This includes using quantitative indicators (such as signal-to-noise ratio, peak signal-to-noise ratio, etc.) to evaluate the denoising effect, and visual inspection to confirm image clarity and detail retention. Through these evaluation methods, it can be ensured that the denoised image data meets the needs of subsequent analysis and application; finally, the format is converted through standardized processing to convert the data into a unified format and dimension to facilitate subsequent data fusion and analysis; in addition, in order to cope with the dynamic and real-time requirements of battlefield data, the data acquisition and data preprocessing module also uses streaming processing technology to achieve real-time data acquisition and preprocessing to ensure the timeliness and accuracy of the data.

[0048] S102, multi-source intelligence fusion to identify the target: use the data fusion algorithm to fuse the pre-processed data, integrate the complementary information and redundant information in the multi-source data, resolve the conflicting information, and use the target recognition confidence calculation to quantify the credibility of each data source for target recognition, and adjust the weight of the target attribute in the fusion result accordingly, so as to achieve accurate identification, positioning, tracking and attribute labeling of both enemy and friendly targets, and generate a unified and real-time updated target situation map; the multi-source intelligence fusion target identification module is responsible for fusing the pre-processed multi-source data to form a comprehensive and complete combat data warehouse. Then extract the target features from the data to finally identify the target. In the specific implementation, the module first uses the data fusion algorithm to fuse the pre-processed data. These algorithms include but are not limited to Kalman filtering, Bayesian network, deep clustering, etc., and the appropriate algorithm is selected for fusion according to the type and characteristics of the data. Through fusion processing, the complementary information and redundant information in multi-source data are integrated, and the conflicting information is resolved. At the same time, the target recognition confidence calculation is used to quantify the credibility of each data source for target recognition, and the weight of the target attribute in the fusion result is adjusted accordingly to achieve accurate identification, positioning, tracking and attribute labeling of both enemy and friendly targets, and generate a unified and real-time updated target situation map. After completing the multi-source information fusion of the preprocessed data, the target feature is extracted from the data by applying various algorithms and techniques, such as color histogram analysis, texture analysis, shape descriptor calculation and deep learning model application. The extracted feature information includes target type, target number, navigation speed, navigation direction, target position, target elevation, meteorological data and hydrological data. The extracted target feature information is compared with the target library. Use appropriate similarity or distance measurement methods, such as Euclidean distance, cosine similarity, Jaccard similarity, etc. to calculate the similarity or distance between the extracted feature information and each target object in the target library. Set a similarity threshold or distance threshold according to application requirements. If the similarity of a target object is higher than a threshold or the distance is lower than a threshold, the target object is considered to match the extracted feature information.

[0049] S103, accurately analyze and predict enemy intentions: collect enemy combat information data extensively and generate prediction models through machine learning algorithms, combine real-time combat data related to combat missions for in-depth analysis, such as enemy dynamics, battlefield environment, our resource status, etc., to build a comprehensive battlefield situation map to intuitively reveal the advantages, disadvantages, opportunities and threats of the current combat mission; the enemy combat information data includes historical action records, force deployment changes, communication network traffic, and social media information. This step is the core step of the technical solution, which is responsible for in-depth analysis and mining of combat data after integration and confirmation of targets. In the specific implementation, the current situation assessment is carried out based on the target information. This step involves extensive collection of real-time data related to the mission, such as enemy dynamics, battlefield environment, our resource status, etc. Through in-depth analysis of these data, a comprehensive battlefield situation map can be constructed to reveal the current advantages, disadvantages, opportunities and threats. In the evaluation process, professional analysis tools and methods need to be used to ensure the accuracy and reliability of the evaluation results. On the basis of situation assessment, action pattern recognition is further carried out. This requires extracting data related to enemy actions from a large amount of battlefield data, such as enemy historical action records, changes in force deployment, communication network traffic, social media information, etc. Then, using advanced technical means such as machine learning and data mining, these feature information are pattern recognized to reveal the enemy's common action patterns, including the enemy's possible action routes, time nodes, tactical actions, combat intentions and other key information. Finally, based on the results of action pattern recognition and the conclusions of the current situation assessment, the enemy's target intentions are predicted based on the prediction model established using machine learning and artificial intelligence algorithm-related technologies. At the same time, the prediction model also needs to dynamically adjust the prediction model parameters according to the newly acquired battlefield data to maintain the prediction accuracy so as to adjust and optimize the action plan and strategy in a timely manner.

[0050] S104, Multi-dimensional association recommendation of available forces: Use big data association analysis and optimization algorithms to integrate regional data, force data, and force status tasks to build a multi-dimensional association model that includes strategic resources and combat tasks, taking into account multi-dimensional factors including battlefield environment, mission requirements, troop status, and enemy and our strength comparison. After inputting combat mission requirements, the multi-dimensional association model intelligently recommends force combinations, configuration plans, and action sequences; the multi-dimensional association model covers the association between strategic resources and combat tasks, and also goes deep into the levels of force effectiveness evaluation, demand forecasting, optimal configuration, dynamic adjustment, and resource consumption tracking and supply planning. The core of this step is to build a multi-dimensional association model, which not only covers the association between basic strategic resources and combat tasks, but also goes deep into multiple levels such as force effectiveness evaluation, demand forecasting, optimal configuration, dynamic adjustment, and resource consumption tracking and supply planning. In terms of force effectiveness assessment, the model comprehensively considers multiple factors such as force size, weapon and equipment performance, personnel training level, tactical coordination, logistics support capabilities, psychological endurance and battlefield adaptability based on the preset combat effectiveness evaluation model, and continuously and dynamically optimizes model parameters through machine learning algorithms to achieve accurate quantitative assessment of the combat capabilities of a single combat unit or even the entire combat system. For force demand forecasting, the module uses a variety of methods such as time series analysis, regression analysis and neural network prediction based on rich historical combat data, real-time enemy and friendly situation analysis, detailed task planning information and established combat rules to accurately predict the type, quantity, configuration and support requirements of forces required in different future combat scenarios, providing solid data support for force scheduling and configuration. In the process of force optimization and configuration, the module cleverly uses advanced optimization algorithms such as linear programming, integer programming and genetic algorithms, and combines the specific needs of the combat plan, the results of force demand forecasting, the existing strategic resource status and the complex constraints of the battlefield environment, such as terrain, climate, and transportation network, to carefully formulate the optimal force allocation plan. These plans cover multiple aspects such as the dispatch path of troops, assembly location, task allocation, rotation plan and even the setting of reserve forces, aiming to maximize the overall combat effectiveness while taking into account the rational use and loss control of strategic resources. In addition, the module also has a strong ability to dynamically adjust troops. It can monitor battlefield changes, combat progress and troop losses in real time, and quickly trigger dynamic adjustments of strategic resources through a built-in feedback control mechanism. This process includes but is not limited to re-evaluating troop requirements, updating forecasting models, adjusting and optimizing configuration plans, and executing emergency deployment instructions, etc., to ensure that the strategic resource allocation is always highly matched with the actual needs of the battlefield, thereby maintaining the effectiveness and adaptability of the combat plan.Finally, in terms of resource consumption tracking and supply planning, the module can monitor in real time the consumption of strategic resources during the operation, such as casualties, equipment loss, ammunition and material consumption, etc., and combine factors such as remaining resource reserves, supply capacity and supply route security, and use advanced methods such as supply chain optimization and inventory management to formulate scientific and reasonable supply strategies and emergency supply plans. These measures implemented in this step ensure the continuous and stable supply of strategic resources during the operation, laying a solid material foundation for victory in the operation.

[0051] S105, assisting in the selection of combat plans: combining the above intelligence analysis results with the recommended force configuration, using operations research methods combined with artificial intelligence algorithms, calculate and generate a variety of combat plans that meet the predetermined strategic goals, follow tactical principles and meet battlefield constraints; the artificial intelligence algorithm includes one or more of genetic algorithms, particle swarm optimization, and reinforcement learning algorithms. This step simulates the combat scenarios under different plans through combat simulation technology, combines risk assessment and benefit analysis, and provides commanders with intuitive plan comparisons, detailed advantages and disadvantages analysis, and scientific decision-making suggestions, thereby assisting commanders to quickly and scientifically select the execution plan that best meets the actual battlefield and has the most combat advantages. In the risk assessment link of the combat plan, the module first conducts a comprehensive identification of risk factors. This process is based on military theory, actual combat experience, enemy situation analysis, and an in-depth understanding of the battlefield environment, and systematically lists all risk factors that may affect the smooth implementation of the combat plan, including strategic risks, tactical risks, technical risks, organizational risks, environmental risks, and human risks, etc., to form a detailed list of risk factors. Subsequently, the module conducts a risk probability assessment. By comprehensively using historical data, expert judgment, simulation experiments and intelligence analysis, the probability of occurrence of each identified risk factor is determined. This process may involve methods such as qualitative scoring, quantitative calculation or probability distribution modeling, and sensitivity analysis or probability interval estimation of risk factors with strong uncertainty to improve the accuracy of the assessment. Next is the assessment of risk impact, which targets each risk factor and deeply analyzes the negative impact it may have on the achievement of combat plan goals, combat progress, casualties, equipment losses, combat costs, public opinion influence and alliance relations once it occurs. The impact of risks is quantitatively assessed using methods such as loss function, utility function, decision tree or scenario analysis, and the interaction and cumulative effect between risks are considered to fully grasp the potential harm of risks. Based on the assessment results of risk probability and impact, the module further divides the risk level. By comprehensively using tools such as risk matrix, risk index and risk weight, risk factors are divided into different levels such as high risk, medium risk or low risk according to their potential harm degree, so that commanders can give priority to and deal with risks that have a greater impact on combat plans. For different levels of risks, the module formulates detailed response strategies. These strategies include preventive measures, such as strengthening intelligence collection, improving technical equipment, and strengthening training and exercises; mitigation measures, such as setting up backup systems, increasing redundant resources, and optimizing operational processes; transfer measures, such as signing insurance contracts, seeking external assistance, and dispersing task responsibilities; and acceptance measures, that is, when the risks are known and the impact is controllable, choose to accept the risks and be prepared to bear the losses. To ensure that risks are effectively monitored and responded to in a timely manner, the module also establishes a risk monitoring and early warning mechanism.By using advanced technologies such as data mining, artificial intelligence and the Internet of Things, the status changes of various risk factors during the execution of the combat plan can be monitored in real time, and risk signs can be discovered in time and early warning signals can be triggered so as to quickly initiate the corresponding response plan. Finally, the module integrates the risk assessment results and response strategies into the combat plan, which are reflected in various aspects such as task allocation, force deployment, action routes, time nodes, command and control, and logistics support. This ensures that the combat plan is not only highly feasible and scientific, but also has sufficient risk resistance and emergency response capabilities. At the same time, with the changes in factors such as the battlefield situation, enemy and our situation, and technological progress, the module can also update the risk assessment results regularly or on demand, reclassify the risk level, and adjust the response strategy to maintain its timeliness and effectiveness.

[0052] In summary, the present invention can significantly improve the scientificity, accuracy and response speed of military decision-making by efficiently integrating multi-source data, deeply analyzing battlefield situations, and intelligently generating and optimizing combat plans; this solution aims to introduce in detail how to use big data analysis technology to build an efficient and intelligent combat plan generation system to provide scientific and accurate tactical and strategic guidance for military decision makers; the system will integrate key technical links such as data collection, fusion, analysis, modeling and visualization to cope with complex and dynamic battlefield environments and maximize combat effectiveness. Therefore, the present invention effectively overcomes the various shortcomings of the prior art and has a high industrial utilization value.

[0053] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. A combat decision support system based on big data analysis, characterized in that: The system comprises: Data acquisition module, used to access multi-source battlefield data information in real time; Data preprocessing module, used for cleaning, denoising, format conversion, and spatiotemporal calibration preprocessing operations on multi-source battlefield data; A multi-source intelligence fusion module, which uses a data fusion algorithm to integrate, remove redundancy, and resolve conflicts in fragmented intelligence information from different data sources; the data fusion algorithm includes a target recognition confidence calculation submodule, which is used to quantify the credibility of each data source for target recognition and adjust the weight of the target attribute in the fusion result accordingly; A situation prediction deep mining and analysis module, which generates a prediction model through a machine learning algorithm and performs deep mining and analysis on enemy combat information data; A force deployment module, which integrates regional data, force data, and force status tasks to build a multi-dimensional correlation model including strategic resources and combat tasks. By inputting combat task requirements, it can obtain recommendations on available force combinations, configuration plans, and action sequences suitable for the current battlefield situation; The auxiliary optimization module uses operations research methods combined with artificial intelligence algorithms to calculate a variety of alternative combat plans that meet predetermined strategic goals, tactical principles and constraints, and provides commanders with intuitive plan comparisons, advantages and disadvantages analysis, and decision-making suggestions through combat simulation, risk assessment, and benefit analysis, assisting commanders to quickly and scientifically select the combat plan that best suits battlefield reality and has the greatest combat advantages.

2. The combat decision support system based on big data analysis according to claim 1 is characterized in that: The data preprocessing module also includes a data integrity verification mechanism for detecting and filling missing data, and a data consistency verification mechanism for resolving data conflicts between different data sources.

3. The combat decision support system based on big data analysis according to claim 1 is characterized in that: In the situation prediction deep mining and analysis module, the prediction model includes an adaptive moment estimation algorithm, which dynamically and adaptively updates the prediction model according to newly acquired battlefield parameters to maintain prediction accuracy.

4. The combat decision support system based on big data analysis according to claim 1 is characterized in that: The force deployment module includes a force effectiveness evaluation submodule, a force demand prediction submodule, a force optimization configuration submodule, a force dynamic adjustment submodule, and a resource consumption tracking and supply planning submodule; The force effectiveness evaluation submodule is pre-set with a combat effectiveness evaluation model, in which the combat effectiveness evaluation model combines the data characteristics of various strategic resources to quantitatively evaluate the combat capability of a single combat unit or the entire combat system; The force demand prediction submodule integrates time series analysis, regression analysis, and neural network prediction methods. Based on historical combat data, enemy and friendly situation analysis, mission planning information, and combat rules, it predicts the type, quantity, configuration, and support requirements of forces required in different combat scenarios in the future, providing a basis for force scheduling and configuration. The force optimization configuration submodule adopts an optimization algorithm, which includes linear programming, integer programming, and genetic algorithm, to calculate the optimal force allocation plan according to the combat plan requirements and force demand prediction results, the existing strategic resource status, and battlefield environment constraints; The dynamic force adjustment submodule adopts a feedback control mechanism, which triggers the dynamic adjustment of strategic resources by real-time monitoring of battlefield changes, combat progress, and force loss; The resource consumption tracking and supply planning submodule is used to monitor the consumption of strategic resources during combat in real time, including ammunition, oil, and material losses, and to formulate supply strategies and emergency supply plans based on factors such as remaining resource reserves, supply capabilities, and supply route security, using supply chain optimization and inventory management methods to ensure the continuous supply of strategic resources during combat.

5. The combat decision support system based on big data analysis according to claim 1 is characterized in that: In the auxiliary optimization module, the risk assessment includes risk factor identification, risk probability assessment, risk impact assessment, risk level classification, risk response strategy formulation, risk monitoring and early warning, risk assessment result application, and risk assessment update.

6. A combat decision support method based on big data analysis, based on the combat decision support system based on big data analysis according to any one of claims 1 to 5, characterized in that: The steps include: S101, data acquisition and preprocessing: collect battlefield data in real time through acquisition equipment, including satellite remote sensing data, radar detection data, drone reconnaissance, electronic intelligence and image data, and perform denoising and standardization on the image data; S102. Multi-source intelligence fusion to identify targets: Use data fusion algorithms to fuse pre-processed data, integrate complementary and redundant information in multi-source data, resolve conflicting information, and use target recognition confidence calculation to quantify the credibility of target recognition from each data source, and adjust the weight of target attributes in the fusion result accordingly, so as to achieve accurate recognition, positioning, tracking and attribute labeling of both enemy and friendly targets, and generate a unified and real-time updated target situation map; S103. Accurately analyze and predict enemy intentions: Collect enemy combat information data extensively and generate prediction models through machine learning algorithms. Combined with real-time combat data related to combat missions, conduct in-depth analysis to build a comprehensive battlefield situation map to intuitively reveal the advantages, disadvantages, opportunities and threats of the current combat mission and accurately predict the enemy's combat intentions based on the battlefield situation map; S104. Multi-dimensional correlation recommendation of available forces: Use big data correlation analysis and optimization algorithms to integrate regional data, force data, and force status tasks to build a multi-dimensional correlation model that includes strategic resources and combat tasks, taking into account multi-dimensional factors including battlefield environment, mission requirements, force status, and enemy and our strength comparison. After inputting combat mission requirements, the multi-dimensional correlation model intelligently recommends force combinations, configuration plans, and action sequences; S105. Assist in selecting optimal combat plans: Combine the above intelligence analysis results with the recommended force deployment, use operations research methods combined with artificial intelligence algorithms, and calculate and generate a variety of combat plans that meet predetermined strategic goals, follow tactical principles, and comply with battlefield constraints.

7. The combat decision-making assistance method based on big data analysis according to claim 6 is characterized in that: In step S101, the acquisition equipment includes sensors, radars, and satellites.

8. The combat decision-making assistance method based on big data analysis according to claim 6 is characterized in that: In step S103, the enemy combat information data includes historical action records, troop deployment changes, communication network traffic, and social media information.

9. The combat decision-making assistance method based on big data analysis according to claim 6 is characterized in that: In step S104, the multi-dimensional correlation model covers the relationship between strategic resources and combat tasks, and also goes deep into the levels of force effectiveness evaluation, demand forecasting, optimal configuration, dynamic adjustment, resource consumption tracking and supply planning.

10. The combat decision-making assistance method based on big data analysis according to claim 6 is characterized in that: In step S105, the artificial intelligence algorithm includes one or more of a genetic algorithm, a particle swarm optimization, and a reinforcement learning algorithm.

Images