针对柴油机气缸失火故障诊断需求，提出一种密度-距离自适应的改进K近邻（KNN）诊断方法。以曲轴转速传感器采集的瞬时转速信号为基础，构建包含时域、频域和非线性特征的特征集。分类过程中引入自适应邻居数以匹配不同数据密度分布，同时结合类内马氏距离与“距离 - 密度”联合加权机制，有效提升特征相关性建模精度与边界判别效能。实验结果表明，该方法在多类工况下均能保持较高的诊断精度，平均准确率超过93%，性能优于传统K近邻（KNN）、支持向量机（SVM）及随机森林（RF）等主流诊断方法，在识别船舶柴油机气缸失火方面具有实时性强、结构简洁和应用前景良好的优势。
To address the fault diagnosis requirements of diesel engine cylinder misfire, an improved K-Nearest Neighbors (KNN) diagnosis strategy with adaptive density-distance is proposed. Based on the instantaneous speed signal collected by the crankshaft speed sensor, a feature set including time-domain, frequency-domain, and nonlinear features is constructed. In the classification process, an adaptive neighbor number is introduced to match different data density distributions, and a joint weighting mechanism combining within-class Mahalanobis distance and "distance-density" is employed to effectively improve the modeling accuracy of feature correlation and boundary discrimination performance. Experimental results show that this method maintains high diagnostic accuracy under multiple operating conditions, with an average accuracy rate exceeding 93%. Its performance is superior to mainstream diagnosis methods such as traditional KNN, Support Vector Machine (SVM), and Random Forest (RF). It has the advantages of strong real-time performance, simple structure, and good application prospects in identifying ship diesel engine cylinder misfire.
2025,47(24): 115-119 收稿日期：2025-3-17
DOI：10.3404/j.issn.1672-7649.2025.24.017
分类号：U664.121
基金项目：国家自然科学基金资助项目（51812345）
作者简介：刘学强(1992-),男,博士,讲师,研究方向为船舶动力装置故障诊断
参考文献：
[1] 柯赟, 宋恩哲, 姚崇, 等. 船舶柴油机故障预测与健康管理技术综述[J]. 哈尔滨工程大学学报, 2020, 41(1): 125-131.
KE Y, SONG E Z, YAO C, et al. A review: ship diesel engine prognostics and health management technology[J]. Journal of Harbin Engineering University, 2020, 41(1): 125-131.
[2] SHI Q G, HU Y H, YAN G H. Fault diagnosis of ME marine diesel engine fuel injector with novel IRCMDE method[J]. Polish Maritime Research, 2023, 30(3): 96-110
[3] 王瑞涵, 陈辉, 管聪. 基于机器学习的船舶机舱设备状态监测方法[J]. 中国舰船研究, 2021, 16(1): 158-167.
WANG R H, CHEN H, GUAN C. Condition monitoring method for marine engine room equipment based on machine learning[J]. Chinese Journal of Ship Research, 2021, 16(1): 158-167.
[4] LUO C, ZHAO M, FU X, et al. Thermodynamic simulation-assisted random forest: towards explainable fault diagnosis of combustion chamber components of marine diesel engine[J]. Measurement, 2025, 251: 117-252
[5] TANG C, BI F R, HUANG M, et al. Diesel engine fault diagnosis method based on WACGAN and IRCNN[J]. Neiranji Xuebao/Transactions of CSICE (Chinese Society for Internal Combustion Engines), 2025, 43(3): 270-278
[6] FAN Z T, SHANG Y L, ZHANG Y D, et al. An angle-domain conversion method based on the instantaneous rotational speed signal of a diesel engine[J]. Beijing Huagong Daxue Xuebao (Ziran Kexueban)/Journal of Beijing University of Chemical Technology (Natural Science Edition), 2017, 44(6): 70-75
[7] LIN T R, TAN CC A, MA L, et al. Condition monitoring and fault diagnosis of diesel engines using instantaneous angular speed analysis[J]. Proceedings of the Institution of Mechanical Engineers Part C-Journal of Mechanical Engineering Science, 2015, 229(2): 304-315
[8] YANG K, OUYANG G, LI A, et al. Diesel engine misfire fault diagnosis based on instantaneous speed[C]//Proceedings of the 2015 International Conference on Mechatronics, Electronic, Industrial and Control Engineering, 2015
[9] 祝存玺. 基于遗传算法的船用中速机转速控制策略研究[D]. 哈尔滨: 哈尔滨工程大学, 2023.
[10] COSTA M A, MARTINS S S C H D, LOPES S G, et al. Fault classification in a fuel injection system of a marine diesel engine using machine learning[J]. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 2025, 47(4): 192-192
[11] JIANG Y B, GUO M Z, BI X Y, et al. Diesel engine fault diagnosis method based on fusion data augmentation[J]. Neiranji Gongcheng/Chinese Internal Combustion Engine Engineering, 2024, 45(6): 71-80
[12] 蒋佳炜, 胡以怀, 方云虎, 等. 船舶动力装置智能故障诊断技术的应用与展望[J]. 中国舰船研究, 2020, 15(1): 56-67.
JIANG J W, HU Y H, FANG Y H, et al. Application and prospects of intelligent fault diagnosis technology formarine power system[J]. Chinese Journal of Ship Research, 2020, 15(1): 56-67.
[13] 李少康. 基于多源信息融合的船舶柴油机故障诊断研究[D]. 武汉: 武汉理工大学, 2023.
[14] ZHANG J H, DONG E Y, ZHANG K, et al. Fault diagnosis strategy for MMC current sensors based on fault characteristics recognition[J]. IEEE Journal of Emerging and Selected Topics in Power Electronics, 2025, 13(3): 1.
[15] 闾城, 陈时熠, 华心果, 等. 基于改进KNN算法的汽轮机通流故障诊断方法及应用[J]. 热力发电, 2021, 50(7): 84-90.
LV C, CHEN S Y, HUA X G, et al. Fault diagnosis method of turbine flow passage based on improved KNN algorithm and its application[J]. Thermal Power Generation, 2021, 50(7): 84-90.
[16] 曾学文, 陈高超, 付名江, 等. 基于投票加权GS-KNN的离心风机故障诊断[J]. 节能, 2024, 43(1): 47-50.
ZENG X W, CHEN G C, FU M J, et al. Fault diagnosis of centrifugal fans based on voting weighted GS-KNN[J]. Energy Conservation, 2024, 43(1): 47-50.
[17] QIAN W W, LI S M, LU J J. Adaptive nearest neighbor reconstruction with deep contractive sparse filtering for fault diagnosis of roller bearings[J]. Engineering Applications of Artificial Intelligence, 2022, 111.
