Abstract:In order to solve the problem that deep neural networks are limited by hyperparameters and data volume, this paper proposes an improved deep forest model to realize efficient diagnosis of rotating machinery faults. First, use the multi-granularity scanning link to perform feature extraction on the initial input data to obtain probabilistic features, then add a stacking layer where it is cascaded with the multi-granularity scanning layer to perform corresponding feature extraction work on the input data, and finally process the multi-granularity scan and stacking layer. The data is input into the cascade forest to get the classification result. Experimental results show that the fault diagnosis accuracy of the improved deep forest model is 99.59% and 98.05%, which are better than the commonly used fault diagnosis models.
Keywords: fault diagnosis; deep learning; deep forest; stacking
刘东川,邓艾东,赵敏,卞文彬,许猛. 基于改进深度森林的旋转机械故障诊断方法[J]. 振动与冲击, 2022, 41(21): 19-27.
LIU Dongchuan, DENG Aidong, ZHAO Min, BIAN Wenbin, XU Meng. Fault diagnosis method of rotating machinery based on improved deep forest model. JOURNAL OF VIBRATION AND SHOCK, 2022, 41(21): 19-27.
[1] 唐海燕,陈潮宇.机械设备故障诊断与监测的常用方法及其发展趋势[J].内燃机与配件,2021(15):122-124.Tang Haiyan, Chen Chaoyu. Common methods and development trends of mechanical equipment fault diagnosis and monitoring [J]. Internal Combustion Engines and Accessories, 2021(15): 122-124.
[2] 王登峰,胡浩,吴建德.基于SAE与深度Q网络的旋转机械故障诊断[J].组合机床与自动化加工技术,2021(04):40-44. Wang Dengfeng, Hu Hao, Wu Jiande. Rotating machinery fault diagnosis based on SAE and deep Q network [J]. Modular Machine Tool and Automatic Processing Technology, 2021(04): 40-44.
[3] 潘继财.大数据样本与半监督环境下基于生成对抗网络的故障诊断[J].机械与电子,2021,39(05):20-25.Pan Jicai. Fault diagnosis based on generative confrontation network under big data samples and semi-supervised environment [J]. Machinery and Electronics, 2021, 39(05): 20-25.
[4] 李东东,谭涛.基于VMD和卷积深度信念网络的风电机组行星齿轮箱故障检测[J].水电能源科学,2021,39(03):145-148+77.Li Dongdong, Tan Tao. Fault detection of wind turbine planetary gearbox based on VMD and convolutional deep belief network[J]. Hydropower Energy Science, 2021, 39(03): 145-148+77.
[5] 邵怡韦,陈嘉宇,林翠颖,万程,葛红娟,石智龙.小训练样本下齿轮箱故障诊断:一种基于改进深度森林的方法[J/OL].航空学报:1-15[2021-10-18].Shao Yiwei, Chen Jiayu, Lin Cuiying, Wan Cheng, Ge Hongjuan, Shi Zhilong. Gearbox fault diagnosis under small training samples: a method based on improved deep forest[J/OL].Acta Aeronautica: 1-15[2021-10-18 ].
[6] 王玉静,王诗达,康守强,王庆岩,V.I.MIKULOVICH.基于改进深度森林的滚动轴承剩余寿命预测方法[J].中国电机工程学报,2020,40(15):5032-5043.Wang Yujing, Wang Shida, Kang Shouqiang, Wang Qingyan, V.I. MIKULOVICH. Remaining life prediction method of rolling bearing based on improved deep forest[J]. Proceedings of the Chinese Society of Electrical Engineering, 2020, 40(15): 5032-5043.
[7] 张玉良,马宏忠,朱昊,颜锦,蒋梦瑶,林元棣.基于稀疏深度森林的调相机轻微定子匝间短路故障诊断[J/OL].高电压技术:1-9[2021-08-01].Zhang Yuliang,Ma Hongzhong,Zhu Hao,Yan Jin,Jiang Mengyao,Lin Yuandi.Sparse deep forest based diagnosing the fault diagnosis of minor stator turn-to-turn short circuit of a dimming camera[J/OL].High Voltage Technology:1-9[2021-08-01].
[8] 周杨.基于CNN-RF滚动轴承故障诊断研究[J].自动化与仪表,2021,36(07):51-57.Zhou Yang.Research on Fault Diagnosis of Rolling Bearing Based on CNN-RF[J].Automation and Instrumentation,2021,36(07):51-57.
[9] ElNabawy Ala’a,Belal Nahla A.,ElBendary Nashwa. A Cascade Deep Forest Model for Breast Cancer Subtype Classification Using Multi-Omics Data[J]. Mathematics,2021,9(13):
[10] Mao Shasha,Yang Jinyuan,Gou Shuiping,Jiao Licheng,Xiong Tao,Xiong Lin. Multi-Scale Fused SAR Image Registration Based on Deep Forest[J]. Remote Sensing,2021,13(11):
[11] ZHOU Z H,FENG J. Deep forest: towards an alternative to deep neural networks[C]/ / Twenty-Sixth International Joint Conference on Artificial Intelligence,2017
[12] Yu Bin,Chen Cheng,Wang Xiaolin,Yu Zhaomin,Ma Anjun,Liu Bingqiang. Prediction of protein–protein interactions based on elastic net and deep forest[J]. Expert Systems With Applications,2021,176:
[13] Zhang Jiahuan,Song Hongjun. Multi-Feature Fusion for Weak Target Detection on Sea-Surface Based on FAR Controllable Deep Forest Model[J]. Remote Sensing,2021,13(4):
[14] Chu Yanyi,Kaushik Aman Chandra,Wang Xiangeng,Wang Wei,Zhang Yufang,Shan Xiaoqi,Salahub Dennis Russell,Xiong Yi,Wei Dong-Qing. DTI-CDF: a cascade deep forest model towards the prediction of drug-target interactions based on hybrid features[J]. Briefings in Bioinformatics,2021,22(1):
[15] Yao Nan,Cheng Ke. Electric Power Equipment Image Recognition Based on Deep Forest Learning Model with Few Samples[J]. Journal of Physics: Conference Series,2021,1732(1):