Water Acoustic Countermeasure System Performance Analysis And Evaluation

Nowadays, underwater acoustic countermeasure has been given sufficient attention in the navy of every country. Because of the problem that can not be ignored of improving the own survivability in the countermeasure of various warship and submarine. And underwater acoustic countermeasure is just one of the important measures to improve this survivability. The main research emphases of the thesis includes the efficiency analysis of underwater acoustic countermeasure system, chooses the exact, reasonable evaluating indicator, builds the relative system of efficiency evaluating indicator and chooses an evaluating method to reflect the relationships among the main performance indicators synthetically, builds the scientific, reliable evaluating model.It is an attempt of using the neural network to evaluate the efficiency of underwater acoustic countermeasure system. A method of efficiency evaluating is proposed which is based on the improved Back Propagation (BP) algorithm. In the design process of BP network, the relationship and influence of some parameters are discussed, such as the number of hidden layer in the network, initial weights and the learning speed etc. The fuzzy neural network is applied in the acquirement of training sample of the evaluating model, and the training sample contains endpoint value and middle value, can reflect the characteristic of the evaluating indicator, has considered the variety of every factor, and makes the evaluating process simply and clear. A evaluating model is set up which commix training and neural network, combing genetic algorithm and neural network. Finally, the neural network evaluating model is applied to evaluate the efficiency of underwater acoustic countermeasure system, the results indicate the model is efficient and feasible. The generalization of the model makes it convenient to have an efficiency evaluation in similar weapon system with only a few modulation, so a great deal work of building the analytical model is leaved out and a new way is provided to evaluate the equipment efficiency exactly and quickly.