Wargame deduction method and system combined with knowledge graph

Abstract

The invention discloses a Wargame deduction system combined with a knowledge graph, and belongs to the technical field of Wargame deduction. Comprising a war type and complexity judgment module, a war game rule and scene configuration module, a knowledge graph construction and management module, a plan making module, a plan generation module, an intelligent deduction engine module, a risk prediction and dynamic resource optimization module and a deduction result analysis module. The war type and complexity judgment module, the war game rule and scene configuration module, the knowledge graph construction and management module, the plan making module, the plan generation module, the intelligent deduction engine module, the risk prediction and dynamic resource optimization module and the deduction result analysis module are in communication connection in sequence; according to the Wargame deduction method and system combined with the knowledge graph, deduction can more comprehensively and more appropriately reflect the real battlefield condition.

Description

Technical Field

[0001] The present invention belongs to the field of war game simulation technology, and specifically relates to a war game simulation method and system combined with a knowledge graph. Background Art

[0002] Wargames, as an important means of military strategic analysis and tactical research, have a long history and profound development context. Its origins can be traced back to the early 19th century. Initially, they were mainly used for teaching in military academies and training military commanders. They aimed to improve military personnel's strategic thinking, decision-making ability, and understanding of the complexity of war by simulating battlefield environments and combat processes.

[0003] Although computer technology has been incorporated into existing war game simulations, they are still not comprehensive enough. The war game simulations in existing technologies do not construct knowledge graphs based on the complexity of the war. The considerations for military operations in simple and complex scenarios are completely different. Failure to construct a targeted knowledge graph will make it difficult for the simulation system to obtain appropriate knowledge to assist in decision-making during reasoning, and constructing a complex knowledge graph in a simple battle is also a waste of resources. In addition, when conducting risk simulation predictions, soldiers' emotions are not taken into account as risk factors, and soldiers' emotions will also affect the simulation. Therefore, a war game simulation method and system combined with a knowledge graph are proposed. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a war game simulation method and system combined with a knowledge graph, so that the war game simulation is more in line with the actual war scene and effectively improves its accuracy and scientificity.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A war game deduction method and system combined with a knowledge graph, comprising a war type and complexity determination module, a war game rule and scenario configuration module, a knowledge graph construction and management module, a plan formulation module, a plan generation module, an intelligent deduction engine module, a risk prediction and dynamic resource optimization module, and a deduction result analysis module, wherein the war type and complexity determination module, the war game rule and scenario configuration module, the knowledge graph construction and management module, the plan formulation module, the plan generation module, the intelligent deduction engine module, the risk prediction and dynamic resource optimization module, and the deduction result analysis module are sequentially connected in communication;

[0007] The war type and complexity determination module is used to determine the specific war type and complexity of the war game simulation;

[0008] The war game rules and scenario configuration module provides a graphical interface for defining war game rules, including movement rules, attack rules, and resource management rules, and supports users to customize war game scenarios, including terrain settings, troop deployment, and target settings;

[0009] The knowledge graph construction and management module constructs a knowledge graph based on known war types, war complexity, and war game rules. A complex knowledge graph is constructed for wars with high complexity, and a simple knowledge graph is constructed for wars with low complexity.

[0010] The plan-making module is used to make war plans, including offensive plans, defensive plans, retreat plans and mixed plans, and calculate the predicted winning rate of the plans, and finally select the top three plans with the highest comprehensive scores from the plans;

[0011] The plan generation module is used to receive three generated plans for the user to choose;

[0012] The intelligent game engine module uses the plan to conduct war games and record key events and decisions during the game;

[0013] The risk prediction and dynamic resource optimization module predicts risks according to the progress of real-time deduction, identifies potential risks, optimizes solutions in advance, and optimizes solutions according to the real-time dynamics of deduction;

[0014] The simulation result analysis module conducts an in-depth analysis of the simulation results and automatically generates a detailed simulation report, including a tactical summary and decision-making suggestions. When the user approves the combat plan, the comprehensive winning rate is evaluated and the actual plan is output. When the user does not approve the plan, it returns to the plan generation module for secondary simulation.

[0015] Furthermore, the war type and complexity determination module is used to determine the specific war type and war complexity of the war game simulation. The process is as follows:

[0016] War types include traditional ground warfare, modern hybrid warfare, strategic command, special operations, and historical battles;

[0017] Classifying war types into traditional ground warfare, modern hybrid warfare, strategic command, special operations and historical battles will help to more accurately analyze the characteristics and needs of different types of war, so that targeted tactics and strategies can be formulated.

[0018] The war complexity conditions include the level of detail, real-time requirements, and scale of participation. In terms of the level of detail: highly detailed tactical-level simulations, including individual soldier operations and weapon performance judgments, are considered high complexity; force-level and strategic-level simulations, which focus on tactical formations and tactical operations and emphasize the overall battle situation and strategic decision-making, are considered low complexity. In terms of real-time requirements: those that require quick response and real-time decision-making are considered high complexity; those that allow decision-making and deduction over a long period of time are considered low complexity. In terms of the scale of participation: those involving a large number of troops and frequent tactical operations are considered high complexity, while those involving a small number of troops and fewer tactical operations are considered low complexity;

[0019] Let the level of detail be A, the real-time requirement be B, and the scale of participation be C. When more than two or two of A, B, and C are of high complexity, the war complexity is judged to be of high complexity; otherwise, it is judged to be of low complexity.

[0020] Furthermore, the above-mentioned setting where the level of detail is A, the real-time requirement is B, and the scale of participation is C. When more than two or two of A, B, and C are of high complexity, the war complexity is judged to be of high complexity; otherwise, it is judged to be of low complexity specifically includes:

[0021] Combinations of high complexity:

[0022] A: high, B: high, C: high;

[0023] A: high, B: high, C: low;

[0024] A: high, B: low, C: high;

[0025] A: low, B: high, C: high;

[0026] Combinations of low complexity:

[0027] A: high, B: low, C: low;

[0028] A: low, B: high, C: low;

[0029] A: low, B: low, C: high;

[0030] A: low, B: low, C: low.

[0031] Furthermore, the knowledge graph construction and management module constructs a knowledge graph based on the known war types, war complexities, and wargame rules. For wars with high complexity, a complex knowledge graph is constructed, and for wars with low complexity, a simple knowledge graph is constructed. The processing process is as follows:

[0032] Design the node types of the knowledge graph based on known types of wars and knowledge of war game rules, design the relationship types of the knowledge graph, and design the relationship types of the knowledge graph, and design attributes for each node and relationship;

[0033] For wars of high complexity, construct a detailed and complex knowledge graph; for wars of low complexity, construct a concise and simple knowledge graph.

[0034] For wars of high complexity, construct a complex knowledge graph that can provide detailed and comprehensive knowledge support; for wars of low complexity, construct a simple knowledge graph that can ensure the efficiency and practicality of the deduction.

[0035] Furthermore, the pre-plan formulation module is used to formulate war pre-plans, including offensive pre-plans, defensive pre-plans, retreat pre-plans, and mixed pre-plans, calculate the predicted winning rates of the pre-plans, and finally screen the top three pre-plans with the highest comprehensive scores from the pre-plans. The processing process is as follows:

[0036] Formulate combat pre-plans based on the constructed knowledge graph;

[0037] Collect historical data, battlefield intelligence, and environmental data, construct a model for predicting winning rates based on these data, predict the winning rates of the formulated pre-plans, and display the calculated predicted winning rate values in percentage form;

[0038] Conduct risk level scoring and resource requirement scoring on the obtained pre-plans, and screen the top three pre-plans with the highest comprehensive scores from the pre-plans. The specific formula is:

[0039] G = α·W + β·D + γ·R;

[0040] Where G is the final comprehensive score obtained, W is the predicted result of the winning rate, R is the resource requirement score, and α, β, and γ are the weights of each index;

[0041] Sort the pre-plans according to the comprehensive scores, and screen the top three pre-plans with the highest comprehensive scores from the sorted results.

[0042] Furthermore, the processing process of collecting historical data, battlefield intelligence, and environmental data, constructing a model for predicting winning rates based on these data, predicting the winning rates of the formulated pre-plans, and displaying the calculated predicted winning rate values in percentage form is as follows:

[0043] Extract key features from the collected data and normalize the extracted features;

[0044] Use a linear regression model to predict the winning rate, and use the weighted average method to combine multiple features to predict the winning rate. The specific formula is:

[0045]

[0046] Among them, n is the number of features, and f i is the value of the i-th feature after normalization, and W is the predicted winning rate obtained;

[0047] When the weights of the force ratio, equipment configuration, terrain advantage, climate condition, enemy strategy, our strategy, and logistics support are w 1 、w 2 、w 3 、w 4 、w 5 、w 6 、w 7 respectively, and the feature values are f 1 、f 2 、f 3 、f 4 、f 5 、f 6 、f 7 respectively, the prediction formula is:

[0048]

[0049] The predicted winning rate obtained is displayed in percentage form.

[0050] The percentage form plays a role in standardization and normalization. Using percentages can unify the predicted values within a standard range of 0 - 100%, and at the same time, it can make people more intuitively and quickly understand the meaning represented by the data.

[0051] Furthermore, according to the process of real-time deduction, risk prediction is carried out, potential risks are identified, and the plan is optimized in advance. At the same time, the process of optimizing the plan according to the real-time dynamics of the deduction is as follows:

[0052] Obtain the current battlefield situation data and real-time battlefield intelligence, and use the real-time data to update the battlefield situation model;

[0053] Count the number of battle failures in the real-time deduction, construct a combat emotion prediction model, and use the combat emotion prediction model to predict the change of soldiers' combat emotions. The specific formula is:

[0054]

[0055] Among them, E N The finally obtained combat emotion index, E 1 is the initial value of the combat emotion at the beginning of the deduction, N is the number of battle failures, m i is the different weight of the i-th battle failure, and ΔE i is the emotion change amount;

[0056] The preset emotion standard value, and the calculated combat emotion index EN ;

[0057] Establish a risk assessment model, incorporate the combat emotion score into the risk assessment model, and dynamically adjust the risk assessment model according to the changes in combat emotions;

[0058] Use the risk assessment model to predict future risks, and formulate countermeasures in advance based on the risk prediction results;

[0059] During real-time deduction, continuously monitor the battlefield situation, and adjust the plan in a timely manner according to the real-time dynamics and the risks emerging in real time.

[0060] Furthermore, the deduction result analysis module deeply analyzes the deduction results, and simultaneously automatically generates a detailed deduction report, including tactical summaries and decision-making suggestions. When the user approves the combat plan, evaluate the comprehensive winning rate and output the actual plan. When the user does not approve the plan, return to the pre-plan generation module for secondary deduction. The process is as follows:

[0061] Collect the overall deduction data, clean and organize the collected data, and analyze the implementation effects of various tactics;

[0062] Automatically generate a deduction report based on the collected deduction data, and output the deduction report and the deduction combat plan for the user to view. When the user approves the combat plan, conduct a comprehensive evaluation of this plan and output the plan. When the user does not approve the plan, return to the pre-plan customization module for secondary deduction.

[0063] A war game deduction method combining a knowledge graph:

[0064] S1. Judge the specific war type and war complexity of the war game deduction;

[0065] S2. Configure war game rules and scenarios, and construct a knowledge graph based on the war type, war complexity, war game rules, and scenarios;

[0066] S3. Formulate a war plan, calculate the predicted winning rate of the plan, and finally screen the top three plans with the highest comprehensive scores from the plans for the user to choose;

[0067] S4. Use the plan for war game deduction, and record the key events and decisions during the deduction process;

[0068] S5. Conduct risk prediction according to the progress of real-time deduction, incorporate the soldiers' emotions into the risk prediction, identify potential risks, optimize the plan in advance according to the identified potential risks, and simultaneously optimize the plan according to the real-time dynamics of the deduction;

[0069] S6. Deeply analyze the deduction results, and at the same time automatically generate a detailed deduction report for the user to view. If the user approves, the plan is output; if the user does not approve, a second deduction is carried out.

[0070] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0071] By setting up a war type and complexity determination module, it is possible to judge the war type and war complexity in advance, and accurately formulate a suitable strategic framework according to the obtained determination results. At the same time, for different war types such as conventional wars, local conflicts or asymmetric wars, corresponding military resources can be quickly allocated, including troops, weapons and equipment, and logistics support forces, to ensure the maximization of resource utilization. Moreover, after obtaining the war type and war complexity, it can provide strong data support for the subsequent construction of the knowledge graph, enabling the construction of the knowledge graph to be more in line with the war ontology framework, and the war complexity can also be used as a standard for constructing simple and complex knowledge graphs.

[0072] By setting up a pre-plan customization module, it is possible to predict the winning rate and comprehensively evaluate the pre-plan in advance, and screen the top three pre-plans with the highest comprehensive scores according to the comprehensive evaluation results, which can greatly save time and energy in terms of decision-making efficiency and speed up the efficiency of military deduction.

[0073] By setting up a risk prediction and dynamic resource optimization module, it is possible to predict risks according to the real-time combat situation and make early adjustments. At the same time, the factor of soldiers' emotions is added to the risk prediction. The factor of soldiers' emotions is also an important factor affecting the deduction. When the battle fails, the soldiers' emotions will be low. At this time, the combat effectiveness is weakened, which will bring great risks and uncertainties to the combat plan of one's own side. Therefore, incorporating the factor of soldiers' emotions into the risk prediction can make the deduction results more accurately reflect the real battlefield situation and provide a more practical and forward-looking basis for decision-making.

[0074] After the deduction is over, when the user is not satisfied with the final plan, a second deduction can be carried out. The second deduction can re-examine the application of strategies and tactics and the allocation of resources based on the lessons learned from the first deduction, so as to improve the quality and feasibility of the final plan. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 It is a block diagram of a war game deduction system combined with a knowledge graph according to the present invention.

[0076] Figure 2 It is a schematic diagram of the war complexity according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0077] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0078] As Figure 1 shown, a wargame deduction system combined with a knowledge graph includes a war type and complexity determination module, a wargame rule and scenario configuration module, a knowledge graph construction and management module, a pre-plan formulation module, a pre-plan generation module, an intelligent deduction engine module, a risk prediction and dynamic resource optimization module, and a deduction result analysis module. The war type and complexity determination module, the wargame rule and scenario configuration module, the knowledge graph construction and management module, the pre-plan formulation module, the pre-plan generation module, the intelligent deduction engine module, the risk prediction and dynamic resource optimization module, and the deduction result analysis module are sequentially communicatively connected;

[0079] The war type and complexity determination module is used to judge the specific war type and war complexity of the wargame deduction;

[0080] In this embodiment, the process of the war type and complexity determination module for judging the specific war type and war complexity of the wargame deduction is as follows:

[0081] The war types include traditional ground warfare, modern hybrid warfare, strategic command, special operations, and historical battles;

[0082] The war complexity conditions include the level of detail, real-time requirements, and scale of participation. In the level of detail: highly detailed tactical-level simulations, including individual soldier operations and weapon performance judgments, are considered high complexity; simulations at the unit level and strategic level, focusing on tactical formations and tactical operations, and emphasizing the overall battle situation and strategic decision-making, are considered low complexity; in real-time requirements: the need for quick response and real-time decision-making is considered high complexity; allowing decision-making and deduction over a long period of time is considered low complexity; in the scale of participation: involving a large number of troops and high-frequency tactical operations is considered high complexity, and involving a small number of troops and fewer tactical operations is considered low complexity;

[0083] Let the level of detail be A, the real-time requirement be B, and the scale of participation be C. When more than two or two of A, B, and C are of high complexity, the war complexity is judged to be high complexity, otherwise it is judged to be low complexity.

[0084] It should be noted that judging the type and complexity of a war before wargaming can significantly improve the accuracy, practicality, and strategic guidance of the wargame. By judging the type and complexity of the war, a strong basis can be provided for the subsequent construction of the knowledge graph.

[0085] In this embodiment, the level of detail is set as A, the real-time requirement is B, and the scale of participation is C. When more than two or both of A, B, and C are of high complexity, it is judged that the complexity of the war is of high complexity; otherwise, it is judged as of low complexity. Specifically, it includes:

[0086] Combinations of high complexity:

[0087] A: high, B: high, C: high;

[0088] A: high, B: high, C: low;

[0089] A: high, B: low, C: high;

[0090] A: low, B: high, C: high;

[0091] Combinations of low complexity:

[0092] A: high, B: low, C: low;

[0093] A: low, B: high, C: low;

[0094] A: low, B: low, C: high;

[0095] A: low, B: low, C: low.

[0096] The wargame rules and scenario configuration module provides a graphical interface for defining the rules of wargaming, including movement rules, attack rules, and resource management rules, and at the same time supports users to customize wargaming scenarios, including terrain setting, troop deployment, and target setting;

[0097] The knowledge graph construction and management module constructs a knowledge graph based on the known war type, war complexity, and wargame rules. For wars with high complexity, a complex knowledge graph is constructed, and for wars with low complexity, a simple knowledge graph is constructed;

[0098] In this embodiment, the knowledge graph construction and management module constructs a knowledge graph based on the known war type, war complexity, and wargame rules. For wars with high complexity, a complex knowledge graph is constructed, and for wars with low complexity, a simple knowledge graph is constructed. The processing process is as follows:

[0099] Design the node types of the knowledge graph based on known types of wars and knowledge of wargame rules, design the relationship types of the knowledge graph, and design the relationship types of the knowledge graph, and design attributes for each node and relationship;

[0100] For wars of high complexity, construct a detailed and complex knowledge graph; for wars of low complexity, construct a concise and simple knowledge graph.

[0101] It should be noted that for wars of high complexity, more nodes, relationships, and attributes can be included. For example, for modern hybrid warfare, "information warfare nodes", "cyber warfare nodes", etc. can be added, and detailed relationships and attributes can be designed; for wars of low complexity, unnecessary nodes, relationships, and attributes can be reduced. For example, for traditional ground warfare, it can be simplified to "ground combat nodes", "tactical deployment nodes", etc., and the relationships and attributes can be simplified;

[0102] The pre-plan formulation module is used to formulate war pre-plans. The pre-plans include offensive pre-plans, defensive pre-plans, retreat pre-plans, and hybrid pre-plans, calculate the predicted winning rates of the pre-plans, and finally screen out the top three pre-plans with comprehensive scores from the pre-plans.

[0103] In this embodiment, the pre-plan formulation module is used to formulate war pre-plans. The pre-plans include offensive pre-plans, defensive pre-plans, retreat pre-plans, and hybrid pre-plans, calculate the predicted winning rates of the pre-plans, and finally screen out the top three pre-plans with comprehensive scores from the pre-plans. The processing procedure is as follows:

[0104] Formulate combat pre-plans based on the constructed knowledge graph;

[0105] Collect historical data, battlefield intelligence, and environmental data, construct a model for predicting winning rates based on these data, predict the winning rates of the formulated pre-plans, and display the calculated predicted winning rate values in percentage form;

[0106] Conduct risk level scoring and resource requirement scoring on the obtained pre-plans, and screen out the top three pre-plans with comprehensive scores from the pre-plans. The specific formula is:

[0107] G = α·W + β·D + γ·R;

[0108] Among them, G is the finally obtained comprehensive score, W is the predicted result of the winning rate, R is the resource requirement score, and α, β, and γ are the weights of each index;

[0109] Sort the pre-plans according to the comprehensive scores, and screen out the top three pre-plans with comprehensive scores from the sorting results.

[0110] It should be noted that calculating the comprehensive scores of the contingency plans and screening the top three contingency plans with the highest comprehensive scores can systematically and comprehensively evaluate the performance of each contingency plan in terms of victory rate prediction, risk level, resource requirements, etc., so as to ensure that the selected contingency plans achieve the best balance between scientificity and practicality.

[0111] The said contingency plan generation module is used to receive the three generated contingency plans for users to select;

[0112] The said intelligent deduction engine module uses the contingency plan for wargame deduction and records the key events and decisions during the deduction process;

[0113] In this embodiment, the process of collecting historical data, battlefield intelligence and environmental data, constructing a model for predicting the victory rate based on these data, predicting the victory rate of the formulated contingency plans, and displaying the calculated victory rate prediction value in percentage form is as follows:

[0114] Extract key features from the collected data and perform normalization processing on the extracted features;

[0115] Use a linear regression model to predict the victory rate and use the weighted average method to combine multiple features to predict the victory rate. The specific formula is:

[0116]

[0117] where n is the number of features, f i is the value of the i-th feature after normalization, and W is the obtained victory rate prediction value;

[0118] When the weights of force ratio, equipment configuration, terrain advantage, climate condition, enemy strategy, our strategy, and logistics support are w 1 , w 2 , w 3 , w 4 , w 5 , w 6 , w 7 , and the feature values are f 1 , f 2 , f 3 , f 4 , f 5 , f 6 , f 7 , then the prediction formula is:

[0119]

[0120] Display the obtained victory rate prediction value in percentage form.

[0121] By systematically collecting and analyzing historical data, battlefield intelligence, and environmental data, a win rate prediction model can be constructed based on the actual situation, providing a scientific basis for military decision-making and reducing subjectivity and blindness. The win rate prediction model can convert complex battlefield situations and the effects of pre-plans into specific win rate prediction values, which will be intuitively displayed in the form of percentages, facilitating decision-makers to compare and evaluate different pre-plans.

[0122] According to the process of real-time deduction, conduct risk prediction, identify potential risks, and optimize the plan in advance. At the same time, the process of optimizing the plan according to the real-time dynamics of the deduction is as follows:

[0123] Obtain the current battlefield situation data and real-time battlefield intelligence, and use the real-time data to update the battlefield situation model;

[0124] Count the number of campaign failures in the real-time deduction, construct a combat emotion prediction model, and use the combat emotion prediction model to predict the changes in soldiers' combat emotions. The specific formula is:

[0125]

[0126] Where E N The finally obtained combat emotion index, E 1 Is the initial value of combat emotion at the start of the deduction, N is the number of campaign failures, m i Is the different weight of the i-th campaign failure, ΔE i Emotion change amount;

[0127] Preset the emotion standard value, and calculate the combat emotion index E N ;

[0128] Establish a risk assessment model, incorporate the combat emotion score into the risk assessment model, and dynamically adjust the risk assessment model according to the changes in combat emotion;

[0129] Use the risk assessment model to predict future risks, and formulate countermeasures in advance according to the risk prediction results;

[0130] When conducting real-time deduction, continuously monitor the battlefield situation, and adjust the plan in a timely manner according to the real-time dynamics and the risks that occur in real time.

[0131] It should be noted that when conducting war game deductions, more emphasis is usually placed on strategic techniques, and the emotional states of soldiers are usually ignored during the deductions. As the direct participants in the war, the emotional fluctuations of soldiers actually have an inestimable potential impact on combat operations, and the number of campaign failures is usually also an important factor affecting soldiers' emotions and the next strategic deployment. Therefore, appropriately incorporating the factor of soldiers' emotional states in war game deductions can make the deduction results more in line with the real war scenario, providing a more comprehensive and accurate reference basis for strategic decision-making.

[0132] The deduced result analysis module deeply analyzes the deduced result, and at the same time automatically generates a detailed deduction report, including tactical summary and decision-making suggestions. When the user approves the combat plan, it evaluates the comprehensive winning rate and outputs the actual plan. When the user does not approve the plan, it returns to the pre-plan generation module for secondary deduction; the process of the deduced result analysis module deeply analyzing the deduced result, and at the same time automatically generating a detailed deduction report, including tactical summary and decision-making suggestions, evaluating the comprehensive winning rate when the user approves the combat plan, and outputting the actual plan, and returning to the pre-plan generation module for secondary deduction when the user does not approve the plan is as follows:

[0133] Collect the overall deduction data, clean and sort the collected data, and analyze the implementation effects of various tactics;

[0134] Automatically generate a deduction report based on the collected deduction data, and output the deduction report and the deduced combat plan for the user to view. When the user approves the combat plan, comprehensively evaluate this plan and output the plan. When the user does not approve the plan, return to the pre-plan customization module for secondary deduction. A wargame deduction method combining a knowledge graph

[0135] S1. Judge the specific war type and war complexity of the wargame deduction;

[0136] S2. Configure the wargame rules and scenarios, and construct a knowledge graph based on the war type, war complexity, wargame rules and scenarios;

[0137] S3. Formulate a war plan, calculate the predicted winning rate of the plan, and finally screen the top three plans with comprehensive scores from the plans for the user to choose;

[0138] S4. Use the plan for wargame deduction, and record the key events and decisions during the deduction process;

[0139] S5. Conduct risk prediction according to the real-time deduction process, add the soldiers' emotions to the risk prediction, identify potential risks, optimize the plan in advance according to the identified potential risks, and optimize the plan according to the real-time dynamics of the deduction;

[0140] S6. Deeply analyze the deduced result, and at the same time automatically generate a detailed deduction report for the user to view. If the user approves, the plan is output. If the user does not approve, conduct secondary deduction.

[0141] It should be noted that when conducting secondary deduction, any one of the three plans selected during pre-plan generation can be used. When the plan used is different from the plan in the initial deduction, the deduction process and results will also be different.

[0142] In the embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation; the modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the method of this embodiment.

[0143] The above embodiments are only used to illustrate the technical methods of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical methods of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A war game simulation system combined with knowledge graph, characterized by: It includes a war type and complexity determination module, a war game rule and scenario configuration module, a knowledge graph construction and management module, a plan formulation module, a plan generation module, an intelligent deduction engine module, a risk prediction and dynamic resource optimization module, and a deduction result analysis module. The war type and complexity determination module, the war game rule and scenario configuration module, the knowledge graph construction and management module, the plan formulation module, the plan generation module, the intelligent deduction engine module, the risk prediction and dynamic resource optimization module, and the deduction result analysis module are sequentially connected in communication; The war type and complexity determination module is used to determine the specific war type and complexity of the war game simulation; The war game rules and scenario configuration module provides a graphical interface for defining war game rules, including movement rules, attack rules, and resource management rules, and supports users to customize war game scenarios, including terrain settings, troop deployment, and target settings; The knowledge graph construction and management module constructs a knowledge graph based on known war types, war complexity, and war game rules, constructs a complex knowledge graph for wars with high complexity, and constructs a simple knowledge graph for wars with low complexity; The plan-making module is used to make war plans, including offensive plans, defensive plans, retreat plans and mixed plans, and calculate the predicted winning rate of the plans, and finally select the top three plans with the highest comprehensive scores from the plans; The plan generation module is used to receive three generated plans for the user to choose; The intelligent game engine module uses the plan to conduct war games and record key events and decisions during the game; The risk prediction and dynamic resource optimization module predicts risks according to the progress of real-time deduction, identifies potential risks, optimizes solutions in advance, and optimizes solutions according to the real-time dynamics of deduction; The simulation result analysis module conducts an in-depth analysis of the simulation results and automatically generates a detailed simulation report, including a tactical summary and decision-making suggestions. When the user approves the combat plan, the comprehensive winning rate is evaluated and the actual plan is output. When the user does not approve the plan, it returns to the plan generation module for secondary simulation.

2. According to claim 1, a war game simulation system combined with a knowledge graph is characterized in that: The war type and complexity determination module is used to determine the specific war type and war complexity of the war game simulation. The process is as follows: War types include traditional ground warfare, modern hybrid warfare, strategic command, special operations, and historical battles; The complexity of war includes the level of detail, real-time requirements, and scale of participation. In the level of detail: highly detailed tactical-level simulation, including individual actions and weapon performance, is judged as high complexity; simulation at the troop level and strategic level, focusing on tactical formations and tactical actions, and focusing on the overall battle situation and strategic decision-making is judged as low complexity; in the real-time requirements: the need for rapid response and real-time decision-making is judged as high complexity; allowing decision-making and deduction over a long period of time is judged as low complexity; in the scale of participation: involving a large number of troops and high-frequency tactical actions is judged as high complexity, involving a small number of troops and fewer tactical actions is judged as low complexity; Assume that the level of detail is A, the real-time requirement is B, and the scale of participation is C. When more than two of A, B, and C or more than two of them are of high complexity, the complexity of the war is judged to be high, otherwise it is judged to be low.

3. The war game simulation system combined with knowledge graph according to claim 2, characterized in that: Assuming the level of detail is A, the real-time requirement is B, and the scale of participation is C, when more than two of A, B, and C are of high complexity, or both are of high complexity, the complexity of the war is judged to be high, otherwise it is judged to be low. Specifically, the following are included: High complexity combinations: A: high, B: high, C: high; A: high, B: high, C: low; A: high, B: low, C: high; A: low, B: high, C: high; Low complexity combinations: A: high, B: low, C: low; A: low, B: high, C: low; A: low, B: low, C: high; A: low, B: low, C: low.

4. The war game simulation system combined with knowledge graph according to claim 1, characterized in that: The knowledge graph construction and management module constructs a knowledge graph based on known war types, war complexity, and war game rules. A complex knowledge graph is constructed for wars with high complexity, and a simple knowledge graph is constructed for wars with low complexity. The processing process is as follows: Design node types of knowledge graphs based on known war types and war game rules, design relationship types of knowledge graphs, and design attributes for each node and relationship; For highly complex warfare, build detailed and complex knowledge graphs; For low-complexity warfare, build concise and simple knowledge graphs.

5. The war game simulation system combined with knowledge graph according to claim 1, characterized in that: The plan making module is used to make war plans, including offensive plans, defensive plans, retreat plans and mixed plans, and calculate the predicted winning rate of the plans. Finally, the process of selecting the top three plans with the most comprehensive scores is as follows: Formulate operational plans based on the constructed knowledge graph; Collect historical data, battlefield intelligence and environmental data, and build a model to predict the winning rate based on these data, and predict the winning rate of the formulated plan, and display the calculated winning rate prediction value in the form of percentage; The obtained plans are scored for risk level and resource requirements, and the top three plans with the highest comprehensive scores are selected from the plans. The specific formula is: G = α·W + β·D + γ·R; Among them, G is the final comprehensive score, W is the win rate prediction result, R is the resource demand score, and α, β, and γ are the weights of each indicator; The plans are sorted according to the comprehensive score, and the top three plans with the highest comprehensive scores are selected from the sorting results.

6. The war game simulation system combined with knowledge graph according to claim 5, characterized in that: The process of collecting historical data, battlefield intelligence and environmental data, and building a model for predicting the winning rate based on these data, predicting the winning rate of the formulated plan, and displaying the calculated winning rate prediction value in percentage form is as follows: Extract key features from the collected data and normalize the extracted features; Use the linear regression model to predict the winning rate, and use the weighted average method to combine multiple features to predict the winning rate. The specific formula is: Where n is the number of features, f i is the normalized value of the i-th feature, and W is the obtained winning rate prediction value; When the weights of force comparison, equipment configuration, terrain advantage, climate conditions, enemy strategy, our strategy, and logistics support are w1, w2, w3, w4, w5, w6, and w7 respectively, and the eigenvalues ​​are f1, f2, f3, f4, f5, f6, and f7 respectively, the prediction formula is: The obtained winning rate prediction value is displayed as a percentage.

7. The war game simulation system combined with knowledge graph according to claim 1, characterized in that: The process of predicting risks, identifying potential risks, and optimizing solutions in advance based on the progress of real-time simulation is as follows: Obtain current battlefield situation data and real-time battlefield intelligence, and use real-time data to update the battlefield situation model; The number of battle failures in real-time simulations is counted, a combat emotion prediction model is constructed, and the combat emotion prediction model is used to predict the changes in soldiers' combat emotions. The specific formula is: Where E N The combat sentiment index obtained at the end, E1 is the initial value of combat sentiment at the beginning of the simulation, N is the number of battle failures, m i is the different weights of the failure of the i-th battle, ΔE i Amount of emotional change; The preset emotional standard value will be calculated as the combat emotional index E N ; Establish a risk assessment model, incorporate combat sentiment scores into the risk assessment model, and dynamically adjust the risk assessment model based on changes in combat sentiment; Use risk assessment models to predict future risks and formulate response measures in advance based on the risk prediction results; When conducting real-time simulations, continuously monitor the battlefield situation and adjust the plan in a timely manner based on real-time dynamics and risks that arise in real time.

8. The war game simulation system combined with knowledge graph according to claim 1, characterized in that: The simulation result analysis module conducts in-depth analysis of the simulation results and automatically generates a detailed simulation report, including tactical summary and decision-making suggestions. When the user approves the combat plan, the comprehensive winning rate is evaluated and the actual plan is output. When the user does not approve the plan, the plan generation module is returned to perform a secondary simulation. The processing process is as follows: Collect overall simulation data, clean and organize the collected data, and analyze the implementation effects of various tactics; A simulation report is automatically generated based on the collected simulation data, and the simulation report and simulation combat plan are output for the user to review. When the user approves the combat plan, a comprehensive evaluation is conducted on the plan and the plan is output. When the user does not approve the plan, the user returns to the plan customization module for a second simulation.

9. A war game simulation method combined with knowledge graph, characterized by: The processing process is as follows: S1. Determine the specific war type and complexity of the war game simulation; S2. Configure war game rules and scenarios, and build a knowledge graph based on war types, war complexity, war game rules and scenarios; S3, formulate a war plan, calculate the predicted winning rate of the plan, and finally select the top three plans with the highest comprehensive scores from the plans for the user to choose; S4. Use the plan to conduct war games and record key events and decisions during the game; S5. Perform risk prediction based on the progress of real-time simulation, add soldier emotions to risk prediction, identify potential risks, optimize solutions in advance based on the identified potential risks, and optimize solutions based on the real-time dynamics of simulation; S6. Conduct an in-depth analysis of the simulation results and automatically generate a detailed simulation report for the user to review. If the user approves, the solution is output. If the user disagrees, a second simulation is performed.

Images