基于t-SNE-VNWOA的船舶柴油机热工故障诊断  

Thermal fault diagnosis of marine diesel engine based on t-SNE-VNWOA

作　　者：陈家君 李蕾 李芷倩 陈诗文 肖金龙 尚前明[3] CHEN Jiajun;LI Lei;LI Zhiqian;CHEN Shiwen;XIAO Jinlong;SHANG Qianming(Changjiang Waterway Institute of Planning and Design,Wuhan 430063,China;National Engineering Research Center for Inland Waterway Regulation,Wuhan 430063,China;School of Naval Architecture,Ocean and Energy Power Engineering,Wuhan University of Technology,Wuhan 430063,China)

机构地区：[1]长江航道规划设计研究院,湖北武汉430063 [2]国家内河航道整治工程技术研究中心,湖北武汉430063 [3]武汉理工大学船海与能源动力工程学院,湖北武汉430063

出　　处：《舰船科学技术》2025年第3期82-88,共7页Ship Science and Technology

基　　金：国家重点研发计划资助项目(2019YFE0104600);国家自然科学基金资助项目(51909200)。

摘　　要：针对传统的船舶柴油机故障诊断方法难以快速准确定位故障的问题,提出一种基于流形学习结合智能算法的诊断模型。以MAN B&W 16 L/24型船用柴油机为研究对象,选用AVL-BOOST软件搭建仿真模型,对单缸喷油过多、喷油提前及气门正时故障进行模拟,再利用t-SNE对高维故障热工参数降维,并将新特征输入VNWOALSSVM分类模型。重复训练-测试结果表明,t-SNE-VNWOA-LSSVM故障诊断模型具有良好稳定性,且诊断精度可达98.67%。该智能诊断模型可作为船舶柴油机故障诊断的有效手段。In view of the difficulty of the traditional fault diagnosis method of diesel engine to locate the fault problem quickly and accurately,a diagnosis model based on manifold learning combined with intelligent algorithm was proposed.Taking the MAN B&W 16 L/24 marine diesel engine as the research object,the AVL-BOOST software was selected to build a simulation model,and the single-cylinder over-injection,early injection and valve timing faults were simulated,and then the t-SNE was used to reduce the dimensionality of the high-dimensional fault thermal parameters,and the new features were input into the VNWOA-LSSVM classification model.The results show that the classification accuracy of the t-SNE-VN-WOA-LSSVM fault diagnosis model is 98.67%,and it has good stability.The intelligent diagnosis model can be used as an effective means for fault diagnosis of marine diesel engine.

关 键 词：柴油机 热工参数 t-SNE 故障诊断 VNWOA LSSVM 

分 类 号：U664.121[交通运输工程—船舶及航道工程]