摘要
船舶低温淡水系统作为保障船舶动力装置安全运行的动力系统,一旦发生故障仅依靠轮机员很难及时排除故障。针对支持向量机(support vector machine,SVM)在模式识别方面受自身参数选择影响较大的问题,提出了基于麻雀搜索算法(sparrow search algorithm,SSA)优化支持向量机的故障诊断方法。利用麻雀搜索算法(SSA)优化了支持向量机的惩罚参数和核参数,建立了基于SSA-SVM的船舶低温淡水系统故障诊断模型。结果表明,SSA-SVM诊断模型比传统的支持向量机(SVM)和粒子群算法(particle swarm optimization,PSO)优化的支持向量机诊断模型的准确率分别高28%和5%,且收敛速度更快。SSA-SVM算法对船舶低温淡水系统的常见故障进行有效地诊断,能为轮机设备的健康管理及轮机员的诊断决策提供一定的指导。
As a power system to ensure the safe operation of the ship's power plant,the ship's fresh water coolingsystem is difficult to troubleshoot in time only by the engineer crew once a failure occurs.Aiming at the problem that the support vector machine(SVM)is greatly affected by its own parameter selection in pattern recognition,a fault diagnosis method based on sparrow search algorithm(SSA)optimized support vector machine was proposed.The SSA was used to optimize the penalty parameters and kernel parameters of support vector machine,and the fault diagnosis model of ship lowfresh water coolingsystem based on SSA-SVM was established.The results show that the accuracy of SSA-SVM model is 28%and 5%higher than that of traditional SVM and particle swarm optimization(PSO),respectively,and the convergence rate is faster.The SSA-SVM algorithm can effectively diagnose the common faults of the ship's fresh water cooling system,and can provide certain guidance for the health management of marine equipment and the diagnostic decision of the engineer.
作者
尹衍楚
邹永久
杜太利
张跃文
YIN Yan-chu;ZOU Yong-jiu;DU Tai-li;ZHANG Yue-wen(Marine Engineering College,Dalian Maritime University,Dalian Liaoning 116026,China)
出处
《计算机仿真》
2024年第1期548-553,共6页
Computer Simulation
基金
国家自然科学基金(52101400,52101345)。
关键词
支持向量机
麻雀搜索算法
故障诊断
算法优化
Support vector machine
Sparrow search algorithm
Fault diagnosis
Algorithm to optimize
