A system and method for generating combat planning scheme template based on knowledge graph

Abstract

A system and method for generating a battle plan template based on a knowledge graph, belonging to the field of data processing technology. In the system, the battle area map module is configured to present a battle area map window of the first size when the user selects the system; the knowledge graph module is configured to construct all entity information participating in the battle into a dynamic knowledge graph, and generate a buoy corresponding to the entity superimposed on the battle area map window according to the entity's geographical location information; the battle plan template is configured to obtain all entity information participating in the battle from the dynamic knowledge graph, and formulate the best attack plan according to all entity information obtained participating in the battle, and when the entity is included in the plan, the buoy corresponding to the entity included in the plan presents a first color, and when the entity is not included in the plan, the buoy corresponding to the entity not included in the plan presents a second color different from the first color. The present invention can obtain the best attack plan through Hopfield, has a high degree of intelligence, and can adapt to the ever-changing needs.

Description

Technical Field

[0001] The present invention relates to a system and method for generating a battle plan template based on a knowledge graph, and belongs to the technical field of electrical data processing. Background Art

[0002] A Chinese invention patent application with publication number CN116718198A discloses a path planning method and system for an offensive equipment cluster based on a temporal knowledge graph. The method includes: taking each state in the flight environment of the offensive equipment cluster as a node in the knowledge graph, and taking the relationship between multiple states as edges to form a knowledge graph of the flight environment of the offensive equipment cluster; predicting possible changes in the flight environment in real time and updating the knowledge graph; taking the offensive equipment cluster as an intelligent agent, taking the current state of the offensive equipment cluster and the action selected in the current state as input, and taking environmental feedback as output, the environmental feedback including the next action and reward, taking maximizing the total reward as the optimization goal, and performing iterative training through reinforcement learning to find the optimal flight path on the knowledge graph. The invention can make full use of environmental information, real-time feedback and prediction results to achieve effective path planning for the offensive equipment cluster.

[0003] However, there is no report in the prior art that discloses how to obtain the best attack plan through deep neural networks under the constraints of manpower, weapons, cost and area. Summary of the invention

[0004] The technical problem to be solved by the present invention is to propose a system and method for generating a combat planning scheme template based on a knowledge graph, which can obtain the best attack plan through a neural network, has a high degree of intelligence, and can adapt to ever-changing needs.

[0005] To achieve the above-mentioned purpose, the present invention provides a system for generating a battle plan template based on a knowledge graph, which includes a user instruction receiving and sending unit, a battle area map module, a knowledge graph module, a battle plan template and a display control unit. The user instruction receiving and sending unit is configured to make the system perform an operation corresponding to the user instruction when receiving the user instruction; the battle area map module is configured to present a battle area map window of a first size when the user selects the system; the knowledge graph module is configured to construct all entity information participating in the battle into a dynamic knowledge graph, and generate a buoy corresponding to the entity superimposed on the battle area map window according to the entity's geographic location information; the battle plan template is configured to obtain all entity information participating in the battle from the dynamic knowledge graph, and formulate the best plan based on the obtained information of all entity participating in the battle. Attack plan, when the entity is included in the plan, the buoy corresponding to the entity included in the plan presents a first color, and when the entity is not included in the plan, the buoy corresponding to the entity not included in the plan presents a second color different from the first color; the display control unit is at least used to coordinate the combat area map module, the knowledge graph module and the combat planning scheme template, so that the combat area map module, the knowledge graph module and the combat planning scheme template are activated at the same time, so that the buoy is superimposed on the combat area map window, and determines whether to expand the buoy into a second size of the expanded floating window superimposed on the combat area map window according to the trigger condition; the combat planning scheme template includes a calculation module and a deep learning neural network model, and the calculation module is configured to obtain the interception equipment in the entity from the knowledge graph, and count the type a of the interception equipment, the number R of each type a interception equipment a , type is The probability g that the interception equipment intercepts the attack of the attacking equipment and completely protects the resource s sa and the type of interception equipment A; the computing module is configured to obtain the offensive equipment in the entity, the type d of the offensive equipment, and the number R of each type d of offensive equipment from the knowledge graph. d , is the probability of an attacking device of type d attacking a resource s and resisting an interception device of type a successfully attacking resource s and all types of offensive equipment D; the deep learning neural network model is configured to attack S resources to resist all types of interception equipment intercepted by the blue side against the red side's D type offensive equipment, and the number of types d is Distribute offensive equipment to obtain the best attack plan.

[0006] To achieve the above-mentioned purpose, the present invention also provides a method for generating a battle plan template based on a knowledge graph, which comprises the following steps:

[0007] S01: When the user selects the system, a combat area map window of a first size is presented through the combat area map module;

[0008] S02: constructing all the entity information participating in the battle into a dynamic knowledge graph through the knowledge graph module, and generating a buoy corresponding to the entity superimposed on the combat area map window according to the geographical location information of the entity;

[0009] S03: Obtain all the entity information participating in the battle from the dynamic knowledge graph through the battle planning template, and formulate the best attack plan based on the acquired information of all the entities participating in the battle. When the entity is included in the plan, the buoy corresponding to the entity included in the plan presents a first color, and when the entity is not included in the plan, the buoy corresponding to the entity not included in the plan presents a second color different from the first color;

[0010] S04: Coordinate the combat area map module, the knowledge graph module and the combat planning scheme template, so that the combat area map module, the knowledge graph module and the combat planning scheme template are activated at the same time, so that the buoy is superimposed on the combat area map window, and determine whether to expand the buoy into an expanded floating window of a second size superimposed on the combat area map window according to the trigger condition, wherein the combat planning scheme template includes a calculation module and a deep learning neural network model, and the calculation module is configured to obtain the interception equipment in the entity from the knowledge graph, and count the type a of the interception equipment, the number R of each type a of interception equipment a , type is The probability g that the interception equipment intercepts the attack of the attacking equipment and completely protects the resource s sa and the type of interception equipment A; the computing module is configured to obtain the offensive equipment in the entity, the type d of the offensive equipment, and the number R of each type d of offensive equipment from the knowledge graph. d , is the probability of an attacking device of type d attacking a resource s and resisting an interception device of type a successfully attacking resource s and all types of offensive equipment D; the deep learning neural network model is configured to attack S resources to resist all types of interception equipment intercepted by the blue side against the red side's D type offensive equipment, and the number of types d is Distribute offensive equipment to obtain the best attack plan.

[0011] Beneficial Effects

[0012] Compared with the prior art, the system and method for generating a battle plan template based on knowledge graph provided by the present invention have the following beneficial effects:

[0013] The best attack plan can be obtained through neural networks, which are highly intelligent and can adapt to ever-changing needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1It is a block diagram of the composition of the knowledge graph-based combat planning scheme template generation system provided in the first embodiment of the present invention.

[0015] Figure 2 It is a workflow diagram of the knowledge graph construction module provided in the first embodiment of the present invention.

[0016] Figure 3 It is a block diagram of the composition of the attack situation construction module provided by the first embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] First embodiment

[0019] Figure 1 1 is a block diagram of a system for generating a battle plan template based on a knowledge graph according to a first embodiment of the present invention. Figure 1 As shown, the battle plan template generation system based on knowledge graph provided by the present invention includes a user instruction receiving and sending unit, a battle area map module, a knowledge graph module, a battle plan template and a display control unit. The user instruction receiving and sending unit is configured to make the system perform an operation corresponding to the user instruction when receiving the user instruction; the battle area map module is configured to present a battle area map window of a first size when the user selects the system; the knowledge graph module is configured to construct all entity information participating in the battle into a dynamic knowledge graph, and generate a buoy corresponding to the entity superimposed on the battle area map window according to the geographical location information of the entity; the battle plan template is configured to display a buoy from the dynamic knowledge graph. The display control unit is used to obtain all entity information participating in the battle from the knowledge graph module, and formulate the best attack plan based on the obtained information of all entities participating in the battle. When the entity is included in the plan, the buoy corresponding to the entity included in the plan presents a first color, and when the entity is not included in the plan, the buoy corresponding to the entity not included in the plan presents a second color different from the first color; the display control unit is at least used to coordinate the combat area map module, the knowledge graph module and the combat planning scheme template, so that the combat area map module, the knowledge graph module and the combat planning scheme template are activated at the same time, so that the buoy is superimposed on the combat area map window, and it is determined whether to expand the buoy into an expanded floating window of a second size superimposed on the combat area map window according to the trigger condition.

[0020] In the first embodiment, the buoy includes an opaque area or a semi-transparent area that covers or partially covers the combat area map window; when the user's instructions are transmitted to the combat area map window, the combat area map module performs operations corresponding to the user's instructions, and the buoy is always consistent with its geographic information on the combat area map window; when the user's instructions are transmitted to the buoy, the buoy performs operations corresponding to the user's instructions.

[0021] The display control unit is also configured to expand the buoy to an expanded floating window of a second size superimposed on the combat area map window, and display entity-related data corresponding to the selected buoy.

[0022] In the first embodiment, the display control unit further includes a closing unit, which is configured to receive a user's instruction and shrink the second expanded floating window of the second size to a float superimposed on the combat area map window.

[0023] In the first embodiment, the second-sized expanded floating window superimposed on the combat area map window includes an opaque area or a semi-transparent area that covers or partially covers the combat area map window. When the user's instruction is transmitted to the combat area map window, the combat area map executes the operation corresponding to the user's instruction, and the expanded floating window is always consistent with its geographic information on the combat area map window; when the user's instruction is transmitted to the second-sized expanded floating window, the second-sized expanded floating window performs the operation corresponding to the user instruction.

[0024] Figure 2 is a workflow diagram of the knowledge graph construction module provided by the first embodiment of the present invention, such as Figure 2 As shown, the dynamic knowledge graph construction module is used to implement the following process:

[0025] S01: Cluster the existing entities into multiple groups to obtain the cluster set Cr=[Cr 1 , …, Cr a , …, Cr A ], A is a positive integer greater than or equal to 3, and the group includes a target group, a drone swarm and an obstacle group;

[0026] S02: In each group Cr a In the example, the group Cr is constructed with the entities in the group as nodes and the relationships between the nodes as edges. a Knowledge graph;

[0027] S03: Connect related entities in two adjacent groups with edges to form an initial knowledge graph;

[0028] S04: Calculate the newly discovered entity v new With each group Cr in the existing group set Cra If the repulsion of a If both are mutually exclusive, execute S06; if there are non-exclusive groups, the non-exclusive group set is recorded as Cr 吸 =[Cr 1 , …, Cr b , …, Cr B ], B is a positive integer greater than or equal to 1, then calculate the newly discovered entity v new The set of non-repulsive groups Cr 吸 The similarity Sim(v NEW ,Cr b );

[0029] S05 If you believe Sim(v NEW ,Cr b ) is less than the threshold, execute S06; if the confidence Sim(v NEW ,Cr b ) is greater than or equal to the threshold, the newly discovered entity V NEW Placed in similarity Sim(v NEW ,Cr b ) in the largest group, calculate the newly discovered entity v new Topological association with existing entities in the group, connected by edges;

[0030] S06: Based on the newly discovered entity v NEW Create a new group Cr NEW , and group Cr NEW Join the group set Cr;

[0031] S07: Repeat S04, S05 and S06 until all newly discovered entities have v NEW Clustering is completed and a dynamic knowledge graph is formed.

[0032] The first embodiment of the present invention achieves the technical effect of not requiring a large amount of computing power and saving the construction time of the dynamic knowledge graph as the newly discovered entities are updated through the above-mentioned technical solution.

[0033] Figure 3 is a block diagram of the attack situation building module provided by the first embodiment of the present invention, such as Figure 3 As shown, the present invention provides an attack situation construction module including a calculation module and a deep learning neural network model. The calculation module is configured to obtain interception equipment in the entity from the knowledge graph, and count the type a of the interception equipment, the number R of each type a interception equipment a , type is The probability g that the interception equipment intercepts the attack of the attacking equipment and completely protects the resource s saand the type of interception equipment A; the computing module is configured to obtain the offensive equipment in the entity, the type d of the offensive equipment, and the number R of each type d of offensive equipment from the knowledge graph. d , is the probability of an attacking device of type d attacking a resource s and resisting an interception device of type a successfully attacking resource s and all types of offensive equipment D; the deep learning neural network model is configured to attack S resources to resist all types of interception equipment intercepted by the blue side against the red side's D type offensive equipment, and the number of types d is Distribute offensive equipment to obtain the best attack plan.

[0034] The deep learning neural network model is a continuous Hopfield neural network model, and the continuous Hopfield neural network is configured to attack S resources to resist all types of interception equipment intercepted by the blue side against the D type of offensive equipment of the red side, the number of which is The D-type offensive equipment is distributed to obtain the best attack plan, including:

[0035] S03-1: Use the following formula to calculate the activation equation of the neuron :

[0036] ,

[0037] In the formula, is a neuron constrained by the energy function and neurons The connection strength, For neurons under the energy function constraint External input; is the neuron under the constraint of weapon availability and neurons The connection strength, For neurons external inputs subject to weapon availability constraints; For neurons under human constraints and neurons The connection strength, For neurons under human constraints External input; Neurons under cost constraints and neurons The connection strength, is a neuron under the cost constraint External input; is a neuron under region constraint and neurons The connection strength, is a neuron under region constraint External input; for A is the total number of interception equipment types; for The total number of offensive equipment of this type; D is the total number of interception equipment types; S is the total number of resources; , , , is the coefficient, For neurons The learning factor;

[0038] S03-2: Use the first-order Euler method to calculate Input variables of neurons :

[0039] + ;

[0040] S03-3: Calculate the Output variables of neurons :

[0041] ;

[0042] S03-4: Assign a value to t and calculate the energy function according to the following formula:

[0043] ,

[0044] In the formula, is the value of resource s; For a type The probability that the interception equipment successfully intercepts the attack of the attacking equipment and completely protects the resource s; For type The interception equipment protects the resource plan; The probability of an attacking device of type d attacking a resource s and resisting an interception device of type a successfully attacking resource s; The total number of Class A interception equipment; is the total number of type d offensive equipment; D is the total number of offensive equipment types; S is the total number of resources;

[0045] S03-5: Determine the energy function Whether it reaches the minimum value, if the energy function E does not reach the minimum value, then return to step S01; if the energy function E reaches the minimum value, then the iteration ends and the output is .

[0046] The first embodiment of the present invention can obtain the best attack plan through the above technical solution, has a high degree of intelligence, fast and quasi-real-time calculation speed, and can adapt to ever-changing needs.

[0047] Second embodiment

[0048] The second embodiment of the present invention only describes the contents that are different from the first embodiment, and the same contents are not repeated.

[0049] The second embodiment of the present invention provides a method for generating a battle plan template based on a knowledge graph, which comprises the following steps:

[0050] S01: When the user selects the system, a combat area map window of a first size is presented through the combat area map module;

[0051] S02: constructing all the entity information participating in the battle into a dynamic knowledge graph through the knowledge graph module, and generating a buoy corresponding to the entity superimposed on the combat area map window according to the geographical location information of the entity;

[0052] S03: Obtain all the entity information participating in the battle from the dynamic knowledge graph through the battle planning template, and formulate the best attack plan based on the acquired information of all the entities participating in the battle. When the entity is included in the plan, the buoy corresponding to the entity included in the plan presents a first color, and when the entity is not included in the plan, the buoy corresponding to the entity not included in the plan presents a second color different from the first color;

[0053] S04: Coordinate the combat area map module, the knowledge graph module and the combat planning scheme template, so that the combat area map module, the knowledge graph module and the combat planning scheme template are activated at the same time, so that the buoy is superimposed on the combat area map window, and determine whether to expand the buoy into an expanded floating window of a second size superimposed on the combat area map window according to the trigger condition, wherein the combat planning scheme template includes a calculation module and a deep learning neural network model, and the calculation module is configured to obtain the interception equipment in the entity from the knowledge graph, and count the type a of the interception equipment, the number R of each type a of interception equipment a , type is The probability g that the interception equipment intercepts the attack of the attacking equipment and completely protects the resource s sa and the type of interception equipment A; the computing module is configured to obtain the offensive equipment in the entity, the type d of the offensive equipment, and the number R of each type d of offensive equipment from the knowledge graph. d , is the probability of an attacking device of type d attacking a resource s and resisting an interception device of type a successfully attacking resource s and all types of offensive equipment D; the deep learning neural network model is configured to attack S resources to resist all types of interception equipment intercepted by the blue side against the red side's D type offensive equipment, and the number of types d is Distribute offensive equipment to obtain the best attack plan.

[0054] The second embodiment of the present invention can obtain the best attack plan through the above technical solution, has a high degree of intelligence, fast and quasi-real-time calculation speed, and can adapt to ever-changing needs.

[0055] Third embodiment

[0056] The third embodiment of the present invention further provides a program product, which includes a computer program code, and the computer program code can be called by a processor to execute the method described in the second embodiment.

[0057] The beneficial effects of the method provided by the third embodiment are the same as those of the first and second embodiments, and will not be described again here.

[0058] Fourth embodiment

[0059] A fourth embodiment of the present invention provides a computer storage medium storing a computer program code, wherein the computer program code can be called by a processor to execute the method described in the second embodiment. The storage medium includes a memory. The processor includes a CPU.

[0060] The beneficial effects of the computer storage medium provided by the fourth embodiment are the same as those of the first or second embodiment, and will not be described again here.

[0061] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined. The meaning of "several" is one or more, unless otherwise clearly and specifically defined.

[0062] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A system for generating battle plan templates based on knowledge graph, characterized in that: It includes a user instruction receiving and sending unit, a combat area map module, a knowledge graph module, a combat planning scheme template and a display control unit. The user instruction receiving and sending unit is configured to make the system perform an operation corresponding to the user instruction when receiving the user instruction; The combat area map module is configured to present a combat area map window of a first size when the system is selected by a user; The knowledge graph module is configured to construct all the entity information participating in the war into a dynamic knowledge graph, and generate a buoy corresponding to the entity superimposed on the combat area map window according to the geographical location information of the entity; the combat planning scheme template is configured to obtain all the entity information participating in the war from the dynamic knowledge graph, and formulate the best attack plan according to the obtained information of all the entities participating in the war, when the entity is included in the plan, the buoy corresponding to the entity included in the plan presents a first color, and when the entity is not included in the plan, the buoy corresponding to the entity not included in the plan presents a second color different from the first color; The display control unit is at least used to coordinate the combat area map module, the knowledge graph module and the combat planning scheme template, so that the combat area map module, the knowledge graph module and the combat planning scheme template are activated at the same time, so that the buoy is superimposed on the combat area map window, and whether to expand the buoy into an expanded floating window of a second size superimposed on the combat area map window is determined according to the trigger condition; the combat planning scheme template includes a calculation module and a deep learning neural network model, and the calculation module is configured to obtain the interception equipment in the entity from the knowledge graph, and count the type a of the interception equipment and the total number R of the interception equipment of type a a ,type The probability g that the interception equipment intercepts the attack of the attacking equipment and completely protects the resource s sa and the total number of interception equipment types A; the calculation module is also configured to obtain the offensive equipment in the entity, the type d of the offensive equipment, and the total number B of the offensive equipment of type d from the knowledge graph d , the probability of successful interception of a type a interception device by an attacking device of type d attacking a resource s and the total number of offensive equipment types D; the deep learning neural network model is configured to attack a total number of S resources to resist all types of interception equipment of the blue side against the offensive equipment of the red side of type D, the total number of The attack equipment of type d is allocated to obtain the best attack plan, 1≤a≤A, A is a positive integer greater than or equal to 2; 1≤d≤D, D is a positive integer greater than or equal to 2; The deep learning neural network model is a continuous Hopfield neural network model, and the working process of the continuous Hopfield neural network model includes: S03-1: Use the following formula to calculate the activation equation of the neuron : ， In the formula, is a neuron constrained by the energy function and neurons The connection strength, For neurons under the energy function constraint External input; is the neuron under the constraint of weapon availability and neurons The connection strength, For neurons external inputs subject to weapon availability constraints; For neurons under human constraints and neurons The connection strength, For neurons under human constraints External input; Neurons under cost constraints and neurons The connection strength, is a neuron under the cost constraint External input; is a neuron under region constraint and neurons The connection strength, is a neuron under region constraint External input; for The total number of interception equipment of type; A is the total number of interception equipment types; for The total number of offensive equipment of the type; D is the total number of offensive equipment types; S is the total number of resources; , , , is the coefficient, For neurons The learning factor; S03-2: Use the first-order Euler method to calculate Input variables of neurons : + ; S03-3: Calculate the Output variables of neurons : ； S03-4: Assign a value to t and calculate the energy function according to the following formula: ， In the formula, is the value of resource s; For a type The probability that the interception equipment successfully intercepts the attack of the attacking equipment and completely protects the resource s; For Type The interception equipment protects the resource plan; The probability of successful interception by an attacking device of type d against a resource s against an intercepting device of type a; The total number of interception equipment of type a; is the total number of attack equipment of type d; D is the total number of attack equipment types; S is the total number of resources; A is the total number of interception equipment types; S03-5: Determine the energy function If the energy function E does not reach the minimum value, then return to step S03-1; if the energy function E reaches the minimum value, then the iteration ends and the output is ; Neurons Output value When it is 1, it means that the i-th attack equipment of type d needs to defend against the attack resource s of the j-th interception equipment of type a; neuron Output value When it is 0, it means that no attack plan is assigned.

2. The system for generating a battle plan template based on knowledge graph according to claim 1 is characterized in that: The dynamic knowledge graph construction module is used to implement the following process: S01: Cluster the existing entities into multiple groups to obtain the cluster set Cr = [Cr1, ..., Cr i , …, Cr I ], the group includes a target group, an interception equipment group of type a, and an attack equipment group of type d; S02: In each group Cr i In the example, the group Ar is constructed with the entities in the group as nodes and the relationships between the nodes as edges. i Knowledge graph; S03: The two adjacent groups Ar i The related entities in are connected with edges to form an initial knowledge graph; S04: Calculate the newly discovered entity v new With each group Cr in the existing group set Cr i If the repulsion of each group Ar in the group set Cr i If all are mutually exclusive, execute S04; if there are non-exclusive groups, the set of non-exclusive groups is denoted as Cr=[Cr1,…,Cr k , …, Cr K ], then calculate the newly discovered entity v new The similarity Sim(v NEW ,Cr k ); S05 If the similarity Sim(v NEW ,Cr k ) is less than the threshold, execute S06; if the similarity Sim(v NEW ,Cr k ) is greater than or equal to the threshold, the newly discovered entity S NEW Placed in similarity Sim(v NEW ,Cr k ) in the largest group, calculate the newly discovered entity v new Topological association with existing entities in the group, connected by edges; S06: Based on the newly discovered entity v NEW Create a new group Cr NEW , and group Cr NEW Join the group set Cr; S07: Repeat S04, S05 and S06 until all newly discovered entities have v NEW Clustering is completed and a dynamic knowledge graph is formed.

3. The system for generating a battle plan template based on a knowledge graph according to any one of claims 1 to 2, characterized in that: The buoy includes an opaque area or a semi-transparent area covering or partially covering the combat area map window; when a user's instruction is transmitted to the combat area map window, the combat area map module performs an operation corresponding to the user's instruction, and the buoy is always consistent with its geographic information on the combat area map window; When the user's instruction is transmitted to the buoy, the buoy performs an operation corresponding to the user's instruction.

4. The system for generating a battle plan template based on knowledge graph according to claim 3 is characterized in that: It also includes a display control unit, which is configured to expand the buoy to an expanded floating window of a second size superimposed on the combat area map window, and display entity-related data corresponding to the buoy.

5. The system for generating a battle plan template based on knowledge graph according to claim 4 is characterized in that: The display control unit also includes a closing unit, which is configured to receive a user's instruction and shrink the second expanded floating window of the second size to a float superimposed on the combat area map window.

6. The system for generating a battle plan template based on knowledge graph according to claim 5 is characterized in that: The expanded floating window of the second size superimposed on the combat area map window includes an opaque area or a semi-transparent area covering or partially covering the combat area map window. When a user's instruction is transmitted to the combat area map window, the combat area map performs an operation corresponding to the user's instruction, and the expanded floating window is always consistent with its geographic information on the combat area map window; When the user's instruction is transmitted to the expanded floating window of the second size, the expanded floating window of the second size operates corresponding to the user's instruction.

7. A method for generating a battle plan template according to the battle plan template generation system based on knowledge graph according to any one of claims 1 to 6, characterized in that: The steps include: S01: When the user selects the system, a combat area map window of a first size is presented through the combat area map module; S02: constructing all the entity information participating in the battle into a dynamic knowledge graph through the knowledge graph module, and generating a buoy corresponding to the entity superimposed on the combat area map window according to the geographical location information of the entity; S03: Obtain all the entity information participating in the battle from the dynamic knowledge graph through the battle planning template, and formulate the best attack plan based on the acquired information of all the entities participating in the battle. When the entity is included in the plan, the buoy corresponding to the entity included in the plan presents a first color, and when the entity is not included in the plan, the buoy corresponding to the entity not included in the plan presents a second color different from the first color; S04: Coordinate the combat area map module, the knowledge graph module and the combat planning scheme template, so that the combat area map module, the knowledge graph module and the combat planning scheme template are activated at the same time, so that the buoy is superimposed on the combat area map window, and determine whether to expand the buoy into an expanded floating window of a second size superimposed on the combat area map window according to the trigger condition; The combat planning scheme template includes a calculation module and a deep learning neural network model. The calculation module is configured to obtain the interception equipment in the entity from the knowledge graph and count the type a of the interception equipment and the total number R of interception equipment of type a. a ,type The probability g that the interception equipment intercepts the attack of the attacking equipment and completely protects the resource s sa The computing module is also configured to obtain the offensive equipment in the entity, the type d of the offensive equipment, and the total number R of the offensive equipment of type d from the knowledge graph. d , the probability of a type d offensive device attacking a resource s against a type a interception device intercepting successfully and the total number of all offensive equipment types D; the deep learning neural network model is configured to attack S resources to resist all types of interception equipment of the blue side against the D type offensive equipment of the red side, the total number is The type d offensive equipment is allocated to obtain the best attack plan; The deep learning neural network model is a continuous Hopfield neural network model, and the working process of the continuous Hopfield neural network model includes: S03-1: Use the following formula to calculate the activation equation of the neuron : ， In the formula, is a neuron constrained by the energy function and neurons The connection strength, For neurons under the energy function constraint External input; is the neuron under the constraint of weapon availability and neurons The connection strength, For neurons external inputs subject to weapon availability constraints; For neurons under human constraints and neurons The connection strength, For neurons under human constraints External input; Neurons under cost constraints and neurons The connection strength, is a neuron under the cost constraint External input; is a neuron under region constraint and neurons The connection strength, is a neuron under region constraint External input; for A is the total number of interception equipment types; A is the total number of interception equipment types; for The total number of attack equipment of the same type; D is the total number of interception equipment types; S is the total number of resources; , , , is the coefficient, For neurons The learning factor; S03-2: Use the first-order Euler method to calculate Input variables of neurons : + ; S03-3: Calculate the Output variables of neurons : ， S03-4: Assign a value to t and calculate the energy function according to the following formula: ， In the formula, is the value of resource s; For a type The probability that the interception equipment successfully intercepts the attack of the attacking equipment and completely protects the resource s; For Type The interception equipment protects the resource plan; The probability of successful interception of a type a interception device against a resource s by an attack device of type d; The total number of interception equipment of type a; is the total number of attack equipment of type d; D is the total number of attack equipment types; S is the total number of resources; A is the total number of interception equipment types; S03-5: Determine the energy function If the energy function E does not reach the minimum value, then return to step S03-1; if the energy function E reaches the minimum value, then the iteration ends and the output is ; Neurons Output value When it is 1, it means that the i-th attack equipment of type d needs to defend against the attack resource s of the j-th interception equipment of type a; neuron Output value When it is 0, it means that no attack plan is assigned.

Images