Optimization of Underwater Cluster Operational Effectiveness Evaluation Based on Support Vector Machine

In modern naval warfare, the development of underwater combat groups is an inevitable trend of networking, unmanned and intelligent naval warfare. Therefore, it is very important to evaluate the effectiveness of underwater combat cluster accurately and quickly. At present, most of the system effectiveness values are the sum of the effectiveness of the subsystems, ignoring the overall emergence and nonlinearity of the system. From the point of view of system theory, this paper constructs an underwater unmanned cluster combat effectiveness evaluation model based on the improved cuckoo search algorithm and optimizes the support vector machine (SVM), and uses the SVM to solve the problems of small sample, non-linearity, high dimension and so on. The improved cuckoo search (ICS) algorithm is used to find the optimal parameters, which avoids the blindness of artificially setting penalty factors and kernel function parameters. The simulation results show that the model can evaluate the combat effectiveness of underwater unmanned cluster quickly and effectively.