基于优化BP神经网络的船舶动力系统故障诊断  

Fault diagnosis of ship power system based on optimized BP neural network

作　　者：徐鹏 杨海燕[2] 程宁 杨元龙[2] 吴金祥 XU Peng;YANG Haiyan;CHENG Ning;YANG Yuanlong;WU jinxiang(The First Military Representative Office of Shenyang Bureau of Naval Armament Department in Dalian,Dalian 116005,China;China Ship Development and Design Center,Wuhan 430064,China)

机构地区：[1]海军装备部沈阳局驻大连地区第一军事代表室,辽宁大连116005 [2]中国舰船研究设计中心,湖北武汉430064

出　　处：《中国舰船研究》2021年第S01期106-113,共8页Chinese Journal of Ship Research

基　　金：国家自然科学基金资助项目(51709249)。

摘　　要：[目的]为实现船舶动力系统的故障诊断,基于优化的BP神经网络提出一种故障诊断方法。[方法]首先,采用附加动量-自适应学习速率调整算法来克服BP神经网络的缺陷;然后,运用"小网络集群"的思路分别构建网络以进行故障识别和故障溯源;接着,采用动力系统仿真平台生成的450组故障数据进行神经网络训练;最后,通过给水泵转速异常高这一故障案例展示故障诊断结果。[结果]通过对故障数据的学习,发现故障原因诊断准确率可高达99%以上。[结论]研究表明,基于优化BP神经网络的故障诊断方法能够精准实现船舶动力系统的故障诊断。[Objectives]In order to realize the fault diagnosis of a ship power system, this paper proposes a fault diagnosis method based on an optimized back-propagation(BP) neural network. [Methods]First, a momentum/adaptive learning rate adjustment algorithm is used to overcome the defects of the BP neural network. The idea of a "small network cluster" is then adopted to construct a separate network for fault identification and diagnosis. Next, 450 groups of fault data generated from a ship power system simulation platform are used for neural network training. Finally, the fault diagnosis results are demonstrated via a fault case in which the speed of a feed water pump is abnormal. [Results]Through training in fault data, the fault diagnosis accuracy level reaches more than 99%. [Conclusions]The proposed fault diagnosis method based on an optimized BP neural network can accurately realize the fault diagnosis of ship power systems.

关 键 词：优化BP神经网络 船舶动力系统 小网络集群 故障诊断 

分 类 号：U664.1[交通运输工程—船舶及航道工程]