Research On Network Modeling And Optimization Method For Course Of Action Planning Generation

"The warfare,which is the important affairs of the country,but also is related to the land of death and the way of survival,must not be ignored." Warfare is related to national security,so we must use scientific theories and methods to make correct decisions.The quality of combat operations plans is related to the success or failure of war.It must be studied in a scientific way.With the development of informationization and intelligence of weapon systems,the combat mode of modern combat has undergone earth-shaking changes.The interconnection between weapon equipment systems is more frequent,and the characteristics of combat systematization and networking are more obvious,especially the rise of unmanned equipment and cluster combat style,which makes the trend of combat systematization and networking particularly prominent.The development of unmanned equipment has improved the interaction style and frequency between equipment and the group intelligence generated by interaction,making the traditional course of action generation method unable to meet the needs of reality.The purpose of the generation,optimization and real-time adjustment of the course of action is to form the overall operational capability of the combat system of systems(CSo S).The composition of the combat So S is the carrier of the generation of the So S capability,and the So S combat capability is an innovative expression of the overall combat capability with the characteristics of the information age.As the name implies,So S combat capability can be understood as systematic combat capability.It is the overall combat capability formed by the interaction of various combat capabilities with the support of advanced integrated electronic information system,rather than the simple sum of the combat capabilities of single weapon equipment,single combat unit and single combat system.The generation of So S combat capability is a direct manifestation of the function and function of the combat So S.Under the support of the information foundation,the combat So S has obvious network characteristics,and it is not a single,single-layer network,but a multi-layer overlay network with different entities in different relationships in terms of links and mechanisms.Based on the analysis of the characteristics of the new combat So S and the formation of the combat capability of the So S,the paper comprehensively considers the generation,optimization and dynamic adjustment of the course of action,and proposes the network modeling and optimization method for the generation of the course of action from the solution ideas and key technical problems of each link.Firstly,the framework of action plan generation and optimization problem modeling and optimization is proposed,the necessity of network modeling of new combat So S is discussed,and the relevant technical route is given.On this basis,the network model of the combat So S is established,and the course of action is generated with the capability as the measurement index.Then multi-objective evolutionary algorithm is used to optimize the huge plan.According to the actual situation of the war,a fast and simple real-time adjustment method of the course of action is proposed.The main work and innovation achievements of this paper are as follows:(1)The research framework of generation and optimization of combat plan for capability generation is proposed.The generation,optimization and real-time adjustment of course of action are traditional but facing new challenges.Based on the analysis of the characteristics of the problem,this paper puts forward the generation and optimization framework of course of action,which is " combat So S network modeling(action plan generation)and action plan optimization and dynamic real-time adjustment".Combat operation generation and execution issues are solved from three combined aspects: functional chain-based course of action generation,action plan optimization based on evolutionary multi-objective optimization algorithm,and dynamic information fusion based on causal reasoning.Many studies are aimed at one aspect of the above problems.However,the adjustment of the operation plan from generation to implementation is closely related,separate research will carry out many assumptions and may lose a lot of important information.In this paper,from the perspective of capability generation,we propose a framework to solve the operational planning problem of the new combat So S,and put several problems together for research.(2)A mothod for modeling and generating course of action based on super network are proposed.With the rapid development of weapons and equipment,the equipment itself has already possessed certain intelligence and autonomy.The single equipment has multiple capabilities,and the interaction style between the equipments is more diverse.The cooperation of clusters and multiple equipment and other operational methods make the operation become a networked confrontation between the entire combat So S of the hostile parties.Starting from the system capability triangle,the combat So S is described by multi-layer and heterogeneous super network,and the module-function chain of the force network is used as the basic execution unit,which reduces frequent interaction between nodes and is not susceptible to communication interference.The impact of the various nodes can also be fully utilized.The generation of course of actions becomes the formation of the initial operational network in the So S super network and the mapping of the operation network to the functional network.(3)The optimization method of course of action based on the improved multi-objective evolutionary algorithm is proposed.With the change of combat scale and combat style,the selection of the plan is no longer aimed at a single goal,and is no longer limited to the time of mission completion.The availability of real-time resources and related consumption have become important winning factors in the battlefield.The traditional optimization method has a general performance in the face of multiple objectives.At the same time,the huge scheme space puts forward new requirements for the performance of the optimization algorithm.The third generation genetic algorithm is used as the basic algorithm.In order to avoid falling into the phenomenon of local optimization and precocity,the algorithm is changed into an adaptive evolution process by using the advantages of neural network function between linear and nonlinear.By comparing with linear operators and cosine operators,the advancement and effectiveness of the algorithm are verified.(4)The dynamic adjustment method of operation plan based on causal network transformation is proposed.War is a game between the two sides.Because it is related to the core interests of both sides of the war,there are bound to be unexpected situations in the course of combat.Aiming at the uncertainty in the combat process,the TIN,which is simple and easy to understand parameters,is used to evaluate the effectiveness of the operational action plan,and the specific modeling process of TIN is given.The algorithm of converting TIN to DBN is given.After the deterministic factors appear,the TIN is converted into a DBN for information fusion,and the parameters of the relevant action nodes are inferred,and then converted to TIN for evaluation of the scheme.The method combines the advantages of simple parameter setting and clear node relationship of TIN,and combines the advantages of DBN to easily integrate new information.