摘要
针对船舶压载水系统工作环境恶劣,相关设备易发生故障且传统人工故障检测方法耗时长、误判率高的问题,提出一种基于随机森林算法的船舶压载水系统故障诊断方法。以某大型集装箱船模拟器压载水系统的5种运行状态的180组运行数据为样本,随机选取150组数据作为训练集,30组数据作为测试集,使用MATLAB软件进行模型搭建。结果表明,模型的总体准确率达到95%以上,可以较为准确地对船舶压载水系统的常见故障进行判别。
Aiming at the problem that the working environment of ship ballast water system is poor,the related equipment is prone to failure,and the traditional manual fault detection method is time-consuming and has a high misjudgment rate,a fault diagnosis method of ship ballast water system based on random forest algorithm is proposed.Taking 180 groups of operation data of ballast water system in five operation states of a large container ship simulator as the most sample,150 groups of data are randomly selected as the training set,and 30 groups of data are selected as the test set.The mathematical software MATLAB is used to build the model.The results show that the overall accuracy of the model is more than 95%,which can accurately identify the common faults of ship ballast water system.
作者
林建宝
曹辉
杨碧涵
王萌萌
LIN Jianbao;CAO Hui;YANG Bihan;WANG Mengmeng(Marine Engineering College,Dalian Maritime University,Dalian 116026,Liaoning,China)
出处
《船舶工程》
CSCD
北大核心
2024年第11期92-97,共6页
Ship Engineering
基金
工信部装函“智能船舶综合测试与验证研究”资助项目([2018]473号)
2018年度辽宁省普通高等教育本科教学改革研究项目。
关键词
压载水系统
随机森林
故障诊断模型
CART算法
决策树
ballast water system
random forest
fault diagnosis model
Classification and Regression Tree(CART)algorithm
decision tree
