李蔚,吴懿范,毛静宇,常增军,李仲博,王方舟
LI Wei,WU Yifan,MAO Jingyu,CHANG Zengjun,LI Zhongbo,WANG Fangzhou

• 1:浙江大学能源工程学院
• 2:北京热力集团有限责任公司
• 3:北京京能能源技术研究有限责任公司

摘要(Abstract)：

随机森林算法具有抗噪声和计算能力强的优点，被广泛应用于旋转机械的振动故障诊断中，但在工业场景中存在样本较少、无法引入先验知识、准确度较低等问题。对此，基于层次分析思想，利用信息熵引入先验知识优化决策树，提出了基于IRF（改进随机森林算法）的汽轮机振动故障诊断方法。为验证所提方法的有效性和可靠性，采用某百万火电机组数据中心的真实运行数据集进行评估。计算结果表明，相较于经典随机森林算法，IRF能够在降低33%决策树数目的情况下具有更高的精确度和低漏报率，同时运行时间缩短至经典随机森林算法的11.4%，在火电机组实时精确振动故障诊断方面有较高的实用价值。
The random forest algorithm, known for its robustness against noise and powerful computational capabilities, is widely employed in diagnosing vibration faults in rotating machinery. However, when applied in industrial settings, the algorithm encounters challenges such as limited sample sizes, the inability to integrate prior knowledge, and comparatively lower accuracy. To tackle these issues, a method for diagnosing vibration faults in steam turbines using an improved random forest(IRF) algorithm is proposed. This approach incorporates prior knowledge to optimize decision trees, utilizing analytical hierarchy process(AHP) and information entropy. Genuine operational datasets from the data center of a million-kW thermal power plant are utilized to validate the efficacy and reliability of the proposed method. Computational findings indicate that, in comparison to the traditional random forest algorithm, IRF achieves higher accuracy and a reduced miss rate, with a 33% decrease in the number of decision trees. Moreover, the operational time is slashed to just 11.4% of that taken by the traditional random forest algorithm. These results suggest that IRF holds significant practical value for real-time, precise vibration fault diagnosis in thermal power units.

关键词(KeyWords)： 汽轮机;振动故障诊断;改进随机森林算法;层次分析;信息熵
steam turbine;vibration fault diagnosis;IRF;AHP;information entropy

Abstract:

Keywords:

基金项目(Foundation): 国家重点研发计划（2019YFE0126000）

作者(Author): 李蔚,吴懿范,毛静宇,常增军,李仲博,王方舟
LI Wei,WU Yifan,MAO Jingyu,CHANG Zengjun,LI Zhongbo,WANG Fangzhou

DOI: 10.19585/j.zjdl.202409012

参考文献(References)：

• [1]廖景雯，关向雨，林建港，等.基于CycleGAN和CNN的GIS振动信号去噪与机械缺陷识别[J].电力工程技术，2023,42(5):37-45.LIAO Jingwen,GUAN Xiangyu,LIN Jiangang,et al.GIS vibration signal denoising and mechanical defect identification based on CycleGAN and CNN[J].Electric Power Engineering Technology,2023,42(5):37-45.
• [2]尤渺，刘育，葛嘉，等.基于振摆信息融合的水电机组状态预测及符号化表征研究[J].电网与清洁能源，2023,39(7):127-133.YOU Miao,LIU Yu,GE Jia,et al.Research on prediction and expression for running states of hydroelectric generating set based on the information fusion of vibration signal[J].Power System and Clean Energy,2023,39(7):127-133.
• [3] YUAN M,WU Y T,LIN L.Fault diagnosis and remaining useful life estimation of aero engine using LSTM neural network[C]//2016 IEEE International Conference on Aircraft Utility Systems(AUS).October 10-12,2016.Beijing,China.IEEE,2016:135-140.
• [4] LI K,WANG Q X.Study on signal recognition and diagnosis for spacecraft based on deep learning method[C]//2015 Prognostics and System Health Management Conference(PHM).October 21-23,2015.Beijing,China.IEEE,2015.
• [5] TAMILSELVAN P,WANG P F.Failure diagnosis using deep belief learning based health state classification[J].Reliability Engineering&System Safety,2013,115:124-135.
• [6] KUREMOTO T,KIMURA S,KOBAYASHI K,et al.Time series forecasting using a deep belief network with restricted Boltzmann machines[J].Neurocomputing,2014,137:47-56.
• [7]马绿洲，周涛，王建飞，等.机组启动过程振动故障诊断与消除[J].山东电力技术，2022,49(2):79-82.MA Lyuzhou,ZHOU Tao,WANG Jianfei,et al.Vibration fault diagnosis and elimination during unit start up[J].Shandong Electric Power,2022,49(2):79-82.
• [8]孙海蓉，曹瑶佳，张雨晴.基于WKFCM-SMOTE和随机森林的风电机组故障诊断[J].山东电力技术，2022,49(3):65-70.SUN Hairong,CAO Yaojia,ZHANG Yuqing. Wind turbine fault diagnosis based on WKFCM-SMOTE and random forest[J]. Shandong Electric Power,2022,49(3):65-70.
• [9]马绿洲，刘健，周涛，等.轴系稳定性下降引发振动故障的诊断处理[J].山东电力技术，2023,50(5):55-60.MA Lyuzhou,LIU Jian,ZHOU Tao,et al.Diagnosis and treatment of vibration fault caused by weakening of shafting stability[J].Shandong Electric Power,2023,50(5):55-60.
• [10]李俊卿，马亚鹏，胡晓东，等.基于CBAM-InceptionV2-双流CNN的风电机组轴承故障诊断[J].智慧电力，2023,51(6):28-33.LI Junqing,MA Yapeng,HU Xiaodong,et al. Wind turbine bearing fault diagnosis based on CBAM-InceptionV2-two-stream CNN[J].Smart Power,2023,51(6):28-33.
• [11] ASSAAD M,BONéR,CARDOT H.A new boosting algorithm for improved time-series forecasting with recurrent neural networks[J].Information Fusion,2008,9(1):41-55.
• [12] SUN W J,SHAO S Y,ZHAO R,et al. A sparse autoencoder-based deep neural network approach for induction motor faults classification[J]. Measurement,2016,89:171-178.
• [13]王培良，夏春江.基于PCA-PDBNs的故障检测与自学习辨识[J].仪器仪表学报，2015,36(5):1147-1154.WANG Peiliang,XIA Chunjiang.Fault detection and selflearning identification based on PCA-PDBNs[J].Chinese Journal of Scientific Instrument,2015,36(5):1147-1154.
• [14] TRAN V T,ALTHOBIANI F,BALL A.An approach to fault diagnosis of reciprocating compressor valves using Teager-Kaiser energy operator and deep belief networks[J].Expert Systems with Applications,2014,41(9):4113-4122.
• [15]李文业，杨帆，周亚星，等.基于改进EMD与PNN的汽轮机通流部分故障诊断[J].振动测试与诊断，2018,38(6):1138-1142.LI Wenye,YANG Fan,ZHOU Yaxing,et al.Research on fault diagnosis of steam turbine flow passage based on improved EMD and PNN[J].Journal of Vibration,Measurement&Diagnosis,2018,38(6):1138-1142.
• [16] DUAN Y J,LV Y S,LIU Y L,et al.An efficient realization of deep learning for traffic data imputation[J].Transportation Research Part C:Emerging Technologies,2016,72:168-181.
• [17] LI C,SáNCHEZ R V,ZURITA G,et al.Fault diagnosis for rotating machinery using vibration measurement deep statistical feature learning[J].Sensors,2016,16(6):895.
• [18] HRASKO R,PACHECO A G C,KROHLING R A.Time series prediction using restricted boltzmann machines and backpropagation[J]. Procedia Computer Science,2015,55:990-999.
• [19] Chen Zhiqiang,Li Chuan,René-Vinicio Sánchez. Multilayer neural network with deep belief network for gearbox fault diagnosis[J].Journal of Vibro-engineering,2015,17(5):2379-2392.
• [20]李兵，韩睿，何怡刚，等.改进随机森林算法在电机轴承故障诊断中的应用[J].中国电机工程学报，2020,40(4):1310-1319.LI Bing,HAN Rui,HE Yigang,et al.Applications of the improved random forest algorithm in fault diagnosis of motor bearings[J].Proceedings of the CSEE,2020,40(4):1310-1319.
• [21]黄良韬，赵志诚，赵亚群.基于随机森林的密码体制分层识别方案[J].计算机学报，2018,41(2):382-399.HUANG Liangtao,ZHAO Zhicheng,ZHAO Yaqun. A two-stage cryptosystem recognition scheme based on random forest[J]. Chinese Journal of Computers,2018,41(2):382-399.
• [22]刘镇武，隋然，张铮，等.基于信息熵与软件复杂度的软件多样性评估方法[J].信息工程大学学报，2020,21(2):207-213.LIU Zhenwu,SUI Ran,ZHANG Zheng,et al.Software diversity evaluation method based on information entropy and software complexity[J].Journal of Information Engineering University,2020,21(2):207-213.
• [23]梁成玉.基于信息熵融合的PSO-SVC涡旋压缩机故障诊断[D].兰州：兰州理工大学，2020.LIANG Chengyu. Fault diagnosis of PSO-SVC scroll compressor based on information entropy fusion[D].Lanzhou:Lanzhou University of Technology,2020.
• [24]李粉红，李超.基于AHP-信息熵的行人不文明交通行为影响因素分析[J].系统仿真技术，2020,16(1):27-31.LI Fenhong,LI Chao.Analysis of influencing factors of pedestrian uncivilized traffic behavior based on AHPinformation entropy[J]. System Simulation Technology,2020,16(1):27-31.
• [25]张力寰.基于工业大数据的汽轮机故障深度分析方法研究[D].杭州：浙江大学，2018.ZHANG Lihuan. Deep fault diagnosis for steam turbines based on industrial big data[D].Hangzhou:Zhejiang University,2018.
• [26]肖辉，薛亚丽，刘培，等.基于运行数据的火电机组最佳效率分析[J].中国电机工程学报，2019,39（增刊1）:164-170.XIAO Hui,XUE Yali,LIU Pei,et al.Optimal efficiency analysis of thermal power units based on operation data[J].Proceedings of the CSEE,2019,39(S1):164-170.
• [27] ZHOU Q C,SHEN H H,ZHAO J,et al.Health management of mechanical equipment based on deep learning:review and prospect[J]. Modern Machinery. 2018,(4):19-27.
• [28]李蔚，吴恺逾，陈坚红，等.基于非线性自回归神经网络和随机森林算法的核电汽轮机组出力优化[J].中国电机工程学报，2021,41(2):409-416.LI Wei,WU Kaiyu,CHEN Jianhong,et al.Output optimization of nuclear power steam turbine based on nonlinear autoregressive neural network and random forest algorithm[J].Proceedings of the CSEE,2021,41(2):409-416.
• [29]耿丽娟，李星毅.用于大数据分类的KNN算法研究[J].计算机应用研究，2014,31(5):1342-1344.GENG Lijuan,LI Xingyi. Improvements of KNN algorithm for big data classification[J]. Application Research of Computers,2014,31(5):1342-1344.
• [30]吴忠强，尚梦瑶，申丹丹，等.基于神经网络和MS-AUKF算法的蓄电池荷电状态估计[J].中国电机工程学报，2019,39(21):6336-6344.WU Zhongqiang,SHANG Mengyao,SHEN Dandan,et al.SOC estimation of battery by MS-AUKF algorithm and BPNN[J].Proceedings of the CSEE,2019,39(21):6336-6344.
• [31]赵波，廖坤，邓春宇，等.基于卷积神经学习的光伏板积灰状态识别与分析[J].中国电机工程学报，2019,39(23):6981-6989.ZHAO Bo,LIAO Kun,DENG Chunyu,et al.Image convolutional neural learning based image recognition and analysis method for dust on photovoltaic panel[J]. Proceedings of the CSEE,2019,39(23):6981-6989.
• [32]王丰华，曾全昊，郑一鸣，等.基于Bayes估计相空间融合和CM-SVDD的有载分接开关机械故障诊断[J].中国电机工程学报，2020,40(1):358-368.WANG Fenghua,ZENG Quanhao,ZHENG Yiming,et al. A mechanical fault diagnosis of on-load tap-changers based on phase space fusion of Bayes estimation and CMSVDD[J]. Proceedings of the CSEE,2020,40(1):358-368.
• [33]李蔚，俞芸萝，盛德仁，等.基于动态数据挖掘的热力参数传感器故障诊断[J].振动测试与诊断，2016,36(4):694-699.LI Wei,YU Yunluo,SHENG Deren,et al.The fault diagnosis of thermal parameter sensor based on dynamic data mining[J].Journal of Vibration,Measurement&Diagnosis,2016,36(4):694-699.