摘要
研究了舰艇航行时声纳部位接收到的全艇自噪声的预报问题.分析了影响声纳部位自噪声的各种参数,将BP神经网络和遗传算法相结合用于舰艇声纳部位自噪声预报.通过使用遗传算法对神经网络的初始权值进行优化,可以在解空间中定位出一个较好的搜索空间,然后采用BP算法在这个小的解空间中搜索出最优解.通过整理大量的测试数据,对神经网络进行训练,训练好的神经网络能够迅速而精确地对舰艇各种航行状态的自噪声进行预报.结果表明,该方法不仅具有足够的精度,而且实用方便、适用性强.
The problem to forecast integrated sonar self-noise of sailing naval vessels was researched. All the factors which influence the naval vessel sonar self-noise are analyzed, and the genetic algorithms and BP neural network are combined to forecast naval vessel sonar self-noise. Through the use of genetic algorithms to optimize original weights and biases of neural network, a better searching space in solution space could be obtained. Then the optimal solution could be achieved using BP neural network. The NN that have been trained With large numbers of test data can predict sonar self-noise of naval vessel in all kinds of sailing states swiftly and accurately. The result shows that the proposed technique not only has enough precision for engineering, but also is convenient and applicable.
出处
《华中科技大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2006年第8期46-47,83,共3页
Journal of Huazhong University of Science and Technology(Natural Science Edition)
基金
十五国防科技预研跨行业综合技术项目
关键词
声纳自噪声
BP神经网络
遗传算法
sonar self-noise
BP neural network
genetic algorithms
