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VNWOA-LSSVM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-VNWOA-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.
2025,47(3): 82-88 收稿日期：2024-4-3
DOI：10.3404/j.issn.1672-7649.2025.03.014
分类号：U664.121
基金项目：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）
作者简介：陈家君(1998-),男,硕士,助理工程师,研究方向为智慧航运和智慧航道
参考文献：
[1] 郭晟江, 郑庆国. 智能船舶的轮机管理研究[J]. 船舶工程, 2023, 45(4): 44?49+57.
[2] KAZIMIERZ W. The increase of operational safety of ships by improving diagnostic methods for marine diesel engine[J]. International Journal on Marine Navigation and Safety of Sea Transportation, 2017, 11(2): 317?328.
[3] 张龙, 刘皓阳, 张号, 等. 改进共空间模式与多源特征融合的轴承智能诊断方法[J]. 西安交通大学学报, 2023, 57(8): 127?137.
[4] 魏东海, 王磊, 赵志超, 等. 随机森林算法在柴油机故障诊断中的应用[J]. 机械设计与制造, 2020, 7: 63?66.
[5] 刘治江, 曹玉佩. 基于改进AdaBoost算法的柴油机故障诊断研究[J]. 武汉理工大学学报(交通科学与工程版), 2021, 45(2): 264?269.
[6] XIN Y, LI S M, CHENG C. An intelligent fault diagnosis method of rotating machinery based on deep neural networks and time-frequency analysis[J]. Journal of Vibroengineering, 2018, 20(6): 2321?2335.
[7] JERZY K, BARTOSZ K, MICHA? W. Fault diagnosis of marine 4-stroke diesel engines using a one-vs-one extreme learning ensemble[J]. Engineering Applications of Artificial Intelligence, 2017, 57: 134?141.
[8] 尚前明, 杨安声, 陈辉, 等. 基于主成分分析的船舶柴油机故障监测方法[J]. 中国航海, 2018, 41(1): 19?23.
[9] WANG H L, WU F, ZHANG L. Application of variational mode decomposition optimized with improved whale optimization algorithm in bearing failure diagnosis[J]. Alexandria Engineering Journal, 2021, 60(5): 4689?4699.
[10] WANG Z, WANG S Y, CHENG Y. Fault feature extraction of parallel-axis gearbox based on IDBO-VMD and t-SNE[J]. Applied Sciences, 2023, 14(1): 289.
[11] WANG L J, PING D Z, WANG C G, et al. Fault diagnosis of rotating machinery bearings based on improved DCNN and WOA-DELM[J]. Processes, 2023, 11(7): 1928?1937.
[12] 肖林博, 陈辉, 管聪. 基于CAWOA-BP的船舶凝给水系统故障诊断[J]. 舰船科学技术, 2023, 45(6): 118?124.
XIAO L B, CHEN H, GUAN C. Fault diagnosis of ship condensate feed system based on CAWOA-BP[J]. Ship Science and Technology, 2023, 45(6): 118?124.
[13] HOSSEIN N, RASUL E, MAQSOOD M, et al. Frequency based feature selection method using whale algorithm[J]. Genomics, 2019, 111(6): 1946?1955.
[14] LIU L, LIU Z J, QIAN X F. Rolling bearing fault diagnosis based on generalized multiscale mean permutation entropy and GWO‐LSSVM[J]. IET Science, Measurement & Technology, 2023, 17(6): 243?256.
[15] TAIBI, AHMED, IKHLEF, et al. A novel intelligent approach based on WOAGWO-VMD and MPA-LSSVM for diagnosis of bearing faults[J]. The International Journal of Advanced Manufacturing Technology, 2022, 120(5?6): 3859?3883.
[16] 杨安声. 船舶柴油机热工故障仿真与诊断方法研究[D]. 武汉: 武汉理工大学, 2016.
[17] 赵志强, 张嘉锐, 罗炽恒, 等. 基于AVL BOOST的船用柴油机典型故障仿真及其数据分析[J]. 中国舰船研究, 2021, 16(2): 176?181+193.
[18] 黄加亮, 谢敢. 基于BOOST船用柴油机热工故障仿真研究[J]. 舰船科学技术, 2015, 37(1): 69?74.
HUANG J L, XIE G. Research on thermal fault simulation of marine diesel engine based on BOOST[J]. Ship Science and Technology, 2015, 37(1): 69?74.
[19] 王望望, 邓林峰, 赵荣珍. 集成KPCA与t-SNE的滚动轴承故障特征提取方法[J]. 振动工程学报, 2021, 34(2): 431?440.
[20] BRITO L, SUSTO G, BRITO J, et al. Fault detection of bearing: an unsupervised machine learning approach exploiting feature extraction and dimensionality reduction[J]. Informatics, 2021, 8(4): 85?100.
