Research On Target Identification And Operational Effectiveness Evaluation Based On Neural Network

The proposal to build a "maritime power" places higher requirements on the field of marine operations.In the combat process,when the sonar finds a target,it is necessary to accurately discriminate between different targets.After determining that a combat action is required,it is necessary to estimate the effectiveness of the battle before the action is carried out.The combat effect is optimal.And after the end of the combat operation,the effectiveness of the combat operation needs to be evaluated to provide a basis for the design and use of weapons and the formulation of the combat plan.Based on the above ideas,this paper studies target discrimination and combat effectiveness evaluation based on neural network algorithms.This paper has completed the research of target discrimination based on neural network algorithm.Modeled and simulated the ship’s radiated noise and the radiated noise of high-speed moving small targets;expounded the principle of the MFCC(Mel Frequency Cepstral Coefficients)feature extraction method,extracted the MFCC features of the simulated signal;and BP(Back Propagation Algorithm)and RBF(Radial Basis Function Algorithm)neural network algorithms were described theoretically.The BP and RBF algorithms were used to analyze the discrimination ability of three types of target noises.Using public data,the neural network algorithm was completed.Of the four types of ship radiated noise and two types of high-speed moving small targetThis paper has completed the research of combat effectiveness evaluation based on neural network algorithm.First,the WSEIAC(Weapon System Effectiveness Industry Advisory Commission)method was used to study the combat effectiveness of the combat system.The theory of WSEIAC effectiveness evaluation method was expounded,the direct navigation torpedo system was modeled,and the factors affecting the combat effectiveness of the system were analyzed.Secondly,a combat effectiveness model based on the BP algorithm and the RBF algorithm is constructed,and simulation data is generated using the WSEIAC method.In the BP algorithm,the influence of various factors on the performance of the algorithm is discussed.As the BP algorithm is greatly affected by the initial value,it is easy to fall into For the local minimum problem,the genetic algorithm and particle swarm algorithm were used to optimize the initial weight of the BP algorithm;the problem of combat effectiveness estimation based on the RBF algorithm was studied.