Early Fault Diagnosis for Wind Turbine Gearbox Based on Improved Multivariate Outlier Detection

Abstract

Abstract: An improved method of multivariate outlier detection was used in early fault diagnosis for wind turbine gearboxes, which might extract the early fault features under fluctuation working conditions. First, the primitive vibration signals were preprocessed by order resampling, and the processed results were analyzed by dimensionless parameter analysis. Second, a model of early fault diagnosis was created based on Mahalanobis distance, which was used for turbine gearboxes. Finally, the actual data was analyzed by the method of multivariate outlier detection, which was improved by multiple linear regression. The results show that the new method may detect the gearbox faults earlier than original one.

Key words: order resampling, dimensionless parameter analysis, multiple linear regression(MLR), multivariate outlier detection

摘要： 针对风电机组运行工况波动性以及机组早期故障特征不易提取的特点，提出一种基于改进的多元离群监测方法来实现风机齿轮箱故障的早期诊断。运用阶比重采样方法对原始振动信号进行预处理，并对处理结果进行量纲一因子分析；通过马氏距离建立风电齿轮箱的早期故障识别模型；利用多元线性回归改进多元离群检测算法进行实际数据的分析计算。结果表明，该方法较原始方法能够更早地察觉出风电齿轮箱早期故障。

关键词: 阶比重采样, 量纲一因子分析, 多元线性回归, 多元离群检测

CLC Number:

TK83

Gu Yujiong, Jia Ziwen, Wang Rui, Ren Yuting. Early Fault Diagnosis for Wind Turbine Gearbox Based on Improved Multivariate Outlier Detection[J]. China Mechanical Engineering.

顾煜炯, 贾子文, 王瑞, 任玉亭. 基于改进的多元离群检测方法的风机齿轮箱早期故障诊断[J]. 中国机械工程.

References

[1]刘德顺，戴巨川，胡燕平，等.现代大型风电机组现状与发展趋势[J].中国机械工程，2013，24(1)：125-134.

Liu Deshun, Dai Juchuan, Hu Yanping, et al. Status and Development Trends of Modern Large-scale Wind Turbines[J]. China Mechanical Engineering, 2013，24(1)：125-134.

[2]张亮.风力发电机组齿轮箱早期故障诊断方法研究??[D]. 大连：大连理工大学，2010.

[3]何邵灿，高宏力，许明恒.基于隐马尔科夫模型的机床部件故障预警技术[J].机械设计与制造，2012 (8)：159-161.

He Shaocan, Gao Hongli, Xu Mingheng. Components of Machine Tools’ Failure Warning Based on Hidden Markov Model[J]. Machinery Design and Manufacture, 2012 (8)：159-161.

[4]李蓉，于德介，陈向民，等.基于阶次分析与循环平稳解调的齿轮箱复合故障诊断方法[J]. 中国机械工程，2013，24（10）：1320-1327.

Li Rong, Yu Dejie, Chen Xiangmin, et al. A Compound Fault Diagnosis Method for Gearbox Based on Order Tracking and Cyclostationary Demodulation[J]. China Mechanical Engineering, 2013,24（10）：1320-1327.

[5]祁丽婉，梁庚，童国炜.基于果蝇算法优化 BP 神经网络的齿轮箱故障诊断[J]. 电网与清洁能源，2014，30(9)：31-42.

Qi Liwan, Liang Geng, Tong Guowei. A Gearbox Diagnosis Method Based on Fruit Fly Optimization Algorithm to Optimize the BP Neural Network[J]. Power System and Clean Energy, 2014，30(9)：31-42.

[6]马怀祥，徐明新，金珏.应用油液分析法和振动法诊断往复机械故障[J].石家庄铁道学院学报，1998,11（2）：39-43.

Ma Huaixiang, Xu Mingxin, Jin Jue. Breakdown Diagnosis of Reciprocating Machinery by Means of Oil Analysis and Vibration Analysis[J].Journal of Shijiazhuang Railway Institute, 1998,11（2）：39-43.

[7]郭厚明，行志刚. 量纲一参数在矿用低速重载齿轮故障诊断中的应用[J]. 煤炭科学技术，2006,34（8）：28-31.

Guo Houming, Xing Zhigang. Dimensionless Parameters Applied to Fault Diagnosis of Mine Low Speed Heavy Loaded Gear[J]. Coal Science and Technology, 2006,34（8）：28-31.

[8]岑少起，于卫东，李瑞珍. 动静压滑动轴承基本方程和特性系数量纲一化研究[J].机械传动，2008,32（3）：4-7.

Cen Shaoqi, Yu Weidong, Li Ruizhen. The Study of Dimensionless Basic Equation and Characteristics Coefficients of Hydrodynamic Hydrostatic Bearings[J]. Journal of Mechanical Transmission, 2008,32（3）：4-7.

[9]Hochman D，Sadok M. Theory of Synchronous Averaging[C]//Proceedings of the 2004 IEEE Aerospace Conference. Big Sky, Montana，USA，2004：3636-3653.

[10]Wonlen K，Manson G，Fieller N J. Damage Detection Using Outlier Analysis[J]. Journal of Sound and Vibration, 2000, 229(3):647-667.

[11]扈静，刘明周，龚任波，等. 基于主成分分析的汽车操纵力舒适性分析与评价[J]. 中国机械工程，2011,22（20）：2456-2459.

Hu Jing, Liu Mingzhou, Gong Renbo, et al. Evaluation of Vehicle Operating Force Comfort Based on Principal Component Analysis[J]. China Mechanical Engineering, 2011,22（20）：2456-2459.

[12]申志刚，何宁，李亮. 高速硬铣削加工刀具磨损监测研究[J].中国机械工程,2009,20(13)：1582-1586.

Shen Zhigang, He Ning, Li Liang. Monitoring of Tool Wear in Hard Milling Process[J]. China Mechanical Engineering, 2009,20(13)：1582-1586.