基于LSTM神经网络算法的船舶柴油机热工故障诊断

Thermal Fault Diagnosis of Marine Diesel Engine Based on LSTM Neural Network Algorithm

柴油机作为船舶航行的动力源头,在航行过程中难免会产生冷却器周壁结垢、过滤器堵塞、阀门泄漏等热工故障,从而影响船舶航行的稳定,因此对热工故障的诊断变得至关重要。利用Simulink软件平台对柴油机的热工故障进行仿真建模,并选取7个热工参数作为数据集来源,将数据集输入到长短时记忆(LSTM)神经网络算法诊断模型中,输出得到包括柴油机涡轮喷嘴结碳、活塞壁温升高、过滤器污阻等典型故障模式,并在MATLAB中进行数据处理和图像绘制。LSTM神经网络算法相比于BP神经网络算法解决了长时依赖问题,并对预测数据有极高的解释度。研究结果表明,基于LSTM神经网络算法的故障诊断模型能够很好地对柴油机故障模式做出诊断。

As the power source of the ship sailing,the diesel engine inevitably produce some thermal faults in the course of sailing,such as cooler wall fouling obstruction,filter clogging,valve leakage and so on,which can affect the stability of the ship sailing,so the diagnosis of thermal faults becomes very important.Simulink software platform is used to simulate the thermal failure of diesel engine,and seven thermal parameters are selected as the source of data set.The data set is input into long short term memory(LSTM)neural network algorithm diagnosis model.The output includes the typical failure modes of diesel engine turbine nozzle carbon deposition,the rise of piston wall temperature,filter fouling blockage and so on.And the data processing and image drawing are carried out in MATLAB.Compared with BP neural network algorithm,LSTM neural network algorithm solves the long time dependence problem and has a high interpretation of the predicted data.The results show that the fault diagnosis model based on LSTM neural network algorithm can diagnose the diesel engine fault mode well.