基于模糊神经网络的船用柴油机智能故障诊断

FNN-based Intelligent Fault Diagnosis for Ship-Diesel Engine

船用柴油机是整个船舶的动力装置,如果发生故障就会对船舶的安全性造成很大的威胁,因此,如何迅速判断故障发生的原因,进而有效排除故障,对保证船舶的正常航行具有特别重要的意义。针对船用柴油机故障多,且具有复杂性和模糊性的特点,以及模糊神经网络在解决非线性复杂问题方面具有的独特优势,提出了一种基于模糊神经网络的船用发动机智能故障诊断专家系统。该系统结合了模糊逻辑理论、人工神经网络以及专家系统的优点,推理速度快,容错能力强,且具有简单实用的人机界面。仿真结果表明,该系统不仅操作使用方便,而且故障诊断结果准确,具有广泛的应用价值。

Ship-diesel engine is power equipment of whole ship.Once fault occurrs,the security of ship will be threatened.Therefore,it is important for ship to sail properly how to find the cause of fault rapidly and remove the fault efficiently.According to the complexity and fuzziness of a variety of ship-diesel engine malfunctions and the advantage of FNN in solving nonlinear comlex problems,a kind of FNN-based intelligent fault diagnosis expert system is put forward in this paper.This system has simple interface,and it has the characteristic of fast inference speed and strong fault-torlerance ability by combining the advantages of the fuzzy logic theory,neural networks and expert system.Simulation result showes that the system has convenience for operation and high accuracy for fault diagnosis,it will have a bright future for further application.