Research on War Threat and Ground Force Defense Deployment Optimization Model Based on Dynamic Bayesian Network Model and Genetic Algorithm

This paper presents a study on the optimization of war threat assessment and ground force defense deployment using a dynamic Bayesian network model and genetic algorithm. Firstly, an evaluation index system for drone attacks in future warfare is established, and threat assessment analysis is conducted using a dynamic Bayesian network model. Fuzzy probability analysis is employed to assess the temporal threat to targets, yielding a comprehensive trend of threat value variation for each target across different time intervals. Experimental analysis of five targets over five-time intervals reveals that initially, target two poses a greater threat due to its proximity in height and distance. As time progresses, Target Two maintains a constant speed while Target Three accelerates during flight, resulting in a gradual increase in the threat posed by Target Three compared to Target Two. Subsequently, the paper addresses the issue of deploying our firepower in the case of clustered drone attacks. A ground force defense deployment optimization model based on a genetic algorithm is proposed to design intelligent combat strategies. Experimental results demonstrate the good convergence performance of the proposed solution algorithm, ultimately yielding an optimal allocation plan of 0, 1, 21, 11, 13, 4, 11, 14, 9, 16.