基于RBF神经网络的船用低速柴油机故障诊断系统  

Fault diagnosis system for marine low speed diesel engine based on RBF neural network

作　　者：娄松涛 LOU Song-tao(Henan Fault-tolerant Server Engineering Technology Research Center,Zhengzhou 450046,China;Henan Polytechnic,Zhengzhou 450046,China)

机构地区：[1]河南省容错服务器工程技术研究中心,河南郑州450046 [2]河南职业技术学院,河南郑州450046

出　　处：《舰船科学技术》2022年第4期108-111,共4页Ship Science and Technology

摘　　要：为确保船舶海上运输的安全性与稳定性,设计基于RBF神经网络的船用低速柴油机故障诊断系统。使用多传感器采集船用低速柴油机各关键构件信号,并对信号进行预处理,运用过限判断模块获得故障信号,RBF神经网络依据故障信号特征进行船用低速柴油机故障诊断和故障程度判断。实验结果表明,该系统能有效滤除信号中的无用高频信号,故障诊断结果与故障实际结果相同,能明显减少柴油机故障发生次数。In order to ensure the safety and stability of marine transportation, a fault diagnosis system of marine lowspeed diesel engine based on RBF neural network is designed. Multi sensors are used to collect the signals of key components of marine low-speed diesel engine, and the signals are preprocessed. The fault signals are obtained by using the over limit judgment module. RBF neural network carries out fault diagnosis and fault degree judgment of marine low-speed diesel engine according to the characteristics of fault signals. The experimental results show that the system can effectively filter the useless high-frequency signals in the signal, the fault diagnosis results are the same as the actual fault results, and can significantly reduce the number of diesel engine faults.

关 键 词：RBF神经网络 船用低速 柴油机 故障诊断系统 传感器 信号调理 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]