基于BWO-SVM的AUV推进系统液压故障诊断  

Hydraulic Fault Diagnosis of AUV Propulsion System Based on BWO-SVM

作　　者：彭浩 李维波 黄康政 高俊卓 PENG Hao;LI Weibo;HUANG Kangzheng;GAO Junzhuo(School of Automation,Wuhan University of Technology,Wuhan Hubei 430070,China)

机构地区：[1]武汉理工大学自动化学院,湖北武汉430070

出　　处：《机床与液压》2025年第3期199-203,共5页Machine Tool & Hydraulics

基　　金：国家重点研发计划课题(2020YFB1506802);国家自然科学基金面上项目(51977164)。

摘　　要：自主水下机器人(AUV)作为海洋探测的重要工具,在执行复杂任务时,其推进系统的可靠性至关重要。针对AUV推进系统液压故障诊断困难、经常误判和漏判等问题,提出基于白鲸优化算法(BWO)优化支持向量机(SVM)的故障诊断算法BWO-SVM,对AUV推进系统液压故障进行智能化诊断。在AMESim中搭建AUV推进系统,然后模拟不同故障并采集相关数据,并采用BWO优化SVM的超参数,最后实现了液压故障诊断分类预测。实验结果表明:BWO-SVM能够区分不同的故障,有效诊断AUV推进系统潜在的液压故障。为了验证算法的优越性,与6种基准算法进行对比,BWO-SVM的准确率至少提升7.35%。BWO-SVM算法在故障诊断方面具有更高的准确率,有效提升了AUV的安全性和可靠性。The autonomous underwater vehicle(AUV)is a crucial tool for ocean exploration,and the reliability of its propulsion system is essential when performing complex tasks.In view of the problems with difficult diagnosis,frequent misjudgment and missed judgment in hydraulic fault diagnosis within the AUV propulsion system,the BWO-SVM fault diagnosis algorithm based on the beluga whale optimization(BWO)optimized support vector machine(SVM)was proposed for intelligent diagnosis of hydraulic faults in the AUV propulsion system.The AUV propulsion system was modeled in AMESim,and different faults were simulated to collect relevant data.Then,BWO was used to optimize the hyperparameters of the SVM,and hydraulic faults diagnosis and classification prediction were realized.Experimental results show that BWO-SVM can differentiate various faults and effectively diagnose potential hydraulic faults in the AUV propulsion system.In order to verify the superiority of the proposed algorithm,it was compared with six benchmark algorithms.The accuracy of BWO-SVM is improved by at least 7.35%.The BWO-SVM algorithm achieves higher fault diagnosis accuracy,significantly enhancing the safety and reliability of the AUV.

关 键 词：自主水下机器人(AUV) 推进系统 故障诊断 白鲸优化算法(BWO) 支持向量机(SVM) 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]