船舶主动力装置故障诊断系统设计与实现  

Fault Diagnosis Software for Marine Main Propulsion Plant

作　　者：郑元璋[1] 陈锋[1] 

机构地区：[1]上海船舶运输科学研究所,上海200135

出　　处：《中国航海》2009年第1期34-37,67,共5页Navigation of China

摘　　要：船舶主动力装置的故障诊断对船舶的安全运行具有重大意义。目前的研究集中于使用人工智能、模式识别等进行故障诊断算法的设计和仿真,而缺乏把故障诊断作为与船舶其他系统有机结合的一个系统来研究。为此,设计了基于船舶网络平台的诊断系统,把来源于平台的基础数据进行格式转换,采用神经网络诊断,并以直观的形式实时显示训练过程,包括网络输出的动态曲线和权值阈值的网格表示,并以文字的形式给出明确的诊断结果。利用此系统,可以充分发挥网络平台的优势,对装置故障进行有效诊断和预测,为船舶管理提供辅助决策。The diagnosis of marine main propulsion plant is important for ship＇s safe operation.While most of current research focuses on design and simulation of fault diagnosis algorithm,this design puts fault diagnosis as a system which integrates with other vessel systems.The features of designed diagnosis system are as follows.It is based on network platform,the basic data which is for neural network diagnosis comes from historical data and real-time data of marine network platform;The diagnosis system in real-time displays training procedure in the intuitive form,including dynamic curve of network output and grid display of weight and threshold.The diagnosis system gives a clear diagnostic result in the form of text.Using this system,it can fully take advantages of network platform and effectively diagnose and forecast main propulsion plant failures,it can provide decision-making aids for ship Management.

关 键 词：船舶、舰船工程 主动力装置 故障诊断 神经网络 网络平台 系统结构 数据输入 格式转换 

分 类 号：U664.1[交通运输工程—船舶及航道工程]