基于BP算法的船舶机舱报警监控系统的故障诊断分析

摘要

The Engine Room Alarm and Monitoring System (AMS) plays a vital role during the entire ship operation as an important point in the field of ship control.Being the most fundamental section of the whole system,AMS data acquisition determines whether the whole system can maintain its continual stable operation.BP algorithm,as an advanced network algorithm,is widely applied to data analysis in various fields.In this paper,a BP network model is established for data fitting on the basis of the data of the integrated engine room alarm and monitoring system of a 7 000-cubic meter mud barge belonging to Zhenjiang YIHUA System Integration Co.,Ltd.During fault diagnosis,the BP network algorithm model is used to predict the exhaust gas temperature of the diesel engine to decide whether to replace the diesel engine with possible failures.The results show that the fitting precision of the BP algorithm meets the requirement,and the prediction accuracy is basically good enough to meet the ship requirement.%机舱报警与监控系统(AMS)作为船舶控制领域里重要的一环,在整个船舶运行期间起到了至关重要的作用.AMS数据的采集作为整个系统里最基础的一节,决定着整个系统能否持续稳定地工作.BP算法作为一种先进的网络算法,被广泛的运用在各个领域的数据分析.模型以镇江市亿华系统集成有限公司的7 000方泥驳机舱综合报警监测系统的数据为基础,建立BP网络模型,进行数据拟合.在故障诊断时,利用BP网络算法的模型预测柴油机的排烟温度来决定是否更换会有故障发生的柴油机.结果表明BP算法的拟合精度符合要求,预测精度基本可行,达到船用需求.