基于XGBoost特征提取的数据驱动故障诊断方法

摘要/Abstract

摘要： 针对目前用于故障诊断领域的机器学习方法尚不能够充分挖掘数据中隐含故障特征信息，存在逼近精度不足的问题，提出一种基于XGBoost算法的隐含特征信息提取方法。根据故障数据与故障类型自定义XGBoost算法的损失函数，迭代构建故障分裂树；提取样本在故障树中的叶子节点位置索引向量并进行特征编码重构，得到隐含故障信息的智能化表征；基于该表征矩阵，使用SVM等机器学习算法建立故障诊断模型，实现多故障模式的识别诊断；最后，以某驱动器的故障诊断为例对方法进行了验证，结果表明：与原始特征下的故障诊断模型相比，基于XGBoost算法提取隐含特征下的诊断模型准确度更高，鲁棒性更好，同时能给出特征变量的重要性排序。

关键词: 故障诊断, 数据驱动, 特征提取, 机器学习, XGBoost算法

Abstract: Aiming at the problems that the current machine learning methods in fault diagnosis were not possible to fully exploit the hidden fault feature informations in the data and has insufficient approximation accuracy, an implicit feature information extraction method was proposed based on XGBoost. Firstly, the loss function of the XGBoost algorithm was customized according to the fault data and the fault types, and the fault splitting tree was constructed iteratively. Secondly, the leaf node position index vector of the sample in the fault tree was extracted and the feature code was reconstructed to obtain the intelligent representation of the implicit fault informations. Thirdly, based on the characterization matrix, a machine learning algorithm such as SVM was used to establish a fault diagnosis model to realize predictive diagnosis of multiple failure modes. Finally, the proposed method was validated by taking a driver fault diagnosis as an example. The results show that compared with the fault diagnosis model under the original features, the diagnostic model based on XGBoost extraction implicit features has higher accuracy and better robustness, which may give the order of importance of the feature variables.

Key words: fault diagnosis, data driven, feature extraction, machine learning, XGBoost algorithm

中图分类号:

TP391

姜少飞;邬天骥;彭翔;李吉泉;李治;孙涛. 基于XGBoost特征提取的数据驱动故障诊断方法[J]. 中国机械工程.

JIANG Shaofei;WU Tianji;PENG Xiang;LI Jiquan;LI Zhi;SUN Tao. Data Driven Fault Diagnosis Method Based on XGBoost Feature Extraction[J]. China Mechanical Engineering.

参考文献

[1]房汉鸣,税爱社,汪辉，等.基于大数据的控制系统故障诊断方法综述[J].单片机与嵌入式系统应用, 2016, 16(4):6-9.
FANG Hanming, RUI Aishe, WANG Hui,et al. Overview of Fault Diagnosis Methods for Control Systems Based on Big Data[J]. MCU and Embedded System Applications,2016, 16(4):6-9.
[2]WANG H, CHAI T Y, DING J L, et al. Data Driven Fault Diagnosis and Fault Tolerant Control: Some Advances and Possible New Directions[J]. Acta Automatica Sinica, 2009, 35(6):739-747.
[3]文成林, 吕菲亚,包哲静，等.基于数据驱动的微小故障诊断方法综述[J].自动化学报, 2016, 42(9):1285-1299.
Wen Chenglin, LYU Feiya, BAO Zhejing,et al. A Review of Micro-fault Diagnosis Methods Based on Data Drive[J].Acta Automatica Sinica, 2016, 42(9):1285-1299.
[4]RAPUR J S, TIWARI R. Automation of Multi-fault Diagnosing of Centrifugal Pumps Using Multi-class Support Vector Machine with Vibration and Motor Current Signals in Frequency Domain[J]. Journal of the Brazilian Society of Mechanical Sciences & Engineering, 2018, 40(6):278.
[5]QU X, LI B, YAO C, et al. Diagnosis Expert System Design of a Certain Ordnance Launch Circuit[J]. Lecture Notes in Electrical Engineering, 2015, 334:769-777.
[6]LI K , ZHANG Y L , LI Z X . Application Research of Kalman Filter and SVM Applied to Condition Monitoring and Fault Diagnosis[J]. Applied Mechanics and Materials, 2011, 121/126:268-272.
[7]FERNANDEZ-FRANCOS D, MARTNEZ-REGO D, FONTENLA-ROMERO O, et al. Automatic Bearing Fault Diagnosis Based on One-class V-SVM[J]. Computers and Industrial Engineering, 2013, 64(1): 357-365.
[8]NAMDARI M, JAZAYERI-RAD H. Incipient Fault Diagnosis Using Support Vector Machines Based on Monitoring Continuous Decision Functions[J]. Engineering Applications of Artificial Intelligence, 2014, 28: 22-35.
[9]XU Y, ZHANG Z, LU G, et al.Approximately Symmetrical Face Images for Image Preprocessing in Face Recognition and Sparse Representation Based Classification[J].Pattern Recognition,2016,54:68-82.
[10]LI C, HU S, GAO S, et al. Real-time Grayscale-thermal Tracking via Laplacian Sparse Representation[C]∥International Conference on Multimedia Modeling. Miami, 2016:54-65.
[11]ZHANG Z Y, WANG Y, WANG K S. Fault Diagnosis and Prognosis Using Wavelet Packet Decomposition, Fourier Transform and Artificial Neural Network[J]. Journal of Intelligent Manufacturing, 2013, 24(6): 1213-1227.
[12]WEI C H, TANG W H, WU Q H. A Hybrid Least-square Support Vector Machine Approach to Incipient Fault Detection for Oil Immersed Power Transformer[J]. Electric Power Components and Systems, 2014, 42(5): 453-463.
[13]TIAN Y, MA J, LU C, et al. Rolling Bearing Fault Diagnosis under Variable Conditions Using LMD-SVD and Extreme Learning Machine[J]. Mechanism and Machine Theory, 2015, 90: 175-186.
[14]YANG B S, OH M S, TAN A C C. Fault Diagnosis of Induction Motor Based on Decision Trees and Adaptive Neuro-fuzzy Inference[J]. Expert Systems with Applications, 2009, 36(2):1840-1849.
[15]雷亚国,贾峰,孔德同, 等. 大数据下机械智能故障诊断的机遇与挑战[J].机械工程学报,2018,54(5):94-104.
LEI Yaguo, JIA Feng, KONG Detong, et al. Opportunities and Challenges of Mechanical Intelligence Fault Diagnosis under Big Data[J].Journal of Mechanical Engineering,2018,54(5):94-104.
[16]CHEN T Q, GUESTRIN C. XGBoost: a Scalable Tree Boosting System[C]∥ACM SIGKDD Conference on Knowledge Discovery and Data Mining. San Francisco, 2016:785-794.
[17] FISHER R A. On the Mathematical Foundations of Theoretical Statistics[M]∥Breakthroughs in Statistics.New York :Springer, 1992:309-368.
[18]CORTES C, VAPNIK V. Support-vector Networks[J]. Machine Learning, 1995, 20(3): 273-297.
[19]HONG W C, PAI P F. Predicting Engine Reliability by Support Vector Machines[J]. International Journal of Advanced Manufacturing Technology, 2006, 28(1/2):154-161.