随着数据挖掘技术的发展，深度置信网络（DBN）这类深度学习算法被越来越多运用到工程领域。在故障诊断领域，结合DBN强大的自适应特征提取和非线性映射能力，可以摆脱以往对专家经验的依赖。基于此，本文为有效地监测柴油机气缸运行状态，提出一种基于改进深度学习算法的船舶柴油机故障诊断技术。先将原始信号的频域形式输入DBN当中，采用蚱蜢优化算法（GOA）搜索DBN的最优参数组合，并建立起最佳的柴油机气缸故障诊断模型。经测试验证，本文提出的诊断模型能够准确识别柴油机气缸运行状态并进行故障诊断，诊断率可以达到99.5%以上，具有较好的工程实用价值。
With the development of data mining technology, deep learning algorithms such as deep belief network (DBN) are widely used in engineering. In the field of fault diagnosis, the dependence on expert experience could be avoidable with the strong adaptive feature extraction and nonlinear mapping ability of DBN. Based on this, in order to find out the abnormal operation of diesel engine cylinder efficiently and timely as well as diagnose its fault accurately, a marine diesel engine fault diagnosis technology based on improved deep learning algorithm is proposed in this paper: Inputting the original signal in frequency domain into DBN, searching the optimal parameter combination of DBN through Grasshopper optimization algorithm (GOA), and establishing the optimal cylinder fault diagnosis model of diesel engine. The test results indicate that the diagnosis model is able to identify the running state of diesel engine cylinder accurately, and is able to carry out fault diagnosis with the diagnosis rate up to more than 99.5%, which proves to be of good engineering practical value.
2021,43(4): 131-134 收稿日期：2020-11-11
DOI：10.3404/j.issn.1672-7649.2021.04.026
分类号：TH113.1
作者简介：黄金娥(1975-),女,高级工程师,从事装备总体及关键系统和装备六性论证、可靠性试验与评估研究工作
参考文献：
[1] 蔡豪杰. 基于DSP的柴油机状态监测与故障诊断系统研究[D]. 哈尔滨: 哈尔滨工程大学, 2011.
CAI Hao-jie. DSP-based diesel engine condition monitoring and troubleshooting system research[D]. Harbin: Harbin Engineering University, 2011.
[2] 雷亚国, 贾峰, 周昕, 等. 基于深度学习理论的机械装备大数据健康监测方法[J]. 机械工程学报, 2015, 51(21): 49-56
LEI Ya-guo, JIA Feng, ZHOU Xin, et al. A deep learning-based method for machinery health monitoring with big data[J]. Journal of Mechanical Engineering, 2015, 51(21): 49-56
[3] 李小娟, 徐增丙, 熊文, 等. 基于深度度量学习的轴承故障诊断方法[J]. 振动与冲击, 2020, 39(15): 25-31
LI Xiao-juan, XU Zeng-bing, XIONG Wen, et al. Bearing fault diagnosis method based on deep metric learning[J]. Journal of Vibration and Shock, 2020, 39(15): 25-31
[4] SAREMI S, MIRJALILI S, LEWIS A. Grasshopper optimisation algorithm: Theory and application[J]. Advances in Engineering Software, 2017, 105: 30-47
[5] 李俊卿, 李斯璇, 陈雅婷. 基于深度置信网络的同步发电机故障诊断方法[J/OL]. 华北电力大学学报(自然科学版): 1−8[2020-08-23].
LI Jun-qing, LI Si-xuan, CHEN Ya-ting. DBN-based fault diagnosis method of synchronous generator[J]. Journal of North China Electric Power University(Natural Science Edition): 1−8[2020-08-23].
[6] 杨星. 基于深度置信网络的行星齿轮箱故障诊断方法研究[D]. 重庆: 重庆交通大学, 2019.
YANG Xing. Study on planetary gearbox troubleshooting methods based on deep confidence network[D]. Chongqing: Chongqing Jiaofu University, 2019.
[7] ZHANG Y F. A research based on hybrid methods GOA and LSSVM for electric load forecasting[D]. Lanzhou: Lanzhou University, 2018.
[8] HINTON G E, OSINDERO S, TEH Y W. A fast learning algorithm for deep belief nets[J]. Neural Computation, 2006, 18(7): 1527-1554
[9] 邹岩崑, 马孝江, 朱泓, 等. 局域波法在柴油机气缸磨损故障诊断中的研究[J]. 中国机械工程, 2004, (20): 27-30
YAN Yan-xuan, MA Xiao-jiang, ZHU Wei, et al. Study on diagnosis of diesel engine piston-liner wearing faults based on partial wave method[J]. China Mechanical Engineering, 2004, (20): 27-30
