基于LSSVM和GMM的风电机组传动系统故障预测研究

Fault prediction for wind turbine drivetrain based on LSSVM and GMM

文章研究了一种基于最小二乘支持向量机(LSSVM)和高斯混合模型(GMM)的传动系统故障程度量化方法。在传动系统故障机理的基础上,利用风电机组的风速、环境温度、有功功率和4个相关的上一时刻温度监测值为输入,4个相关的温度监测值为输出建立LSSVM回归模型。模型的预测值与实际测量值的偏差向量定义为系统“偏离值”。采用GMM拟合正常运行时的偏离值分布,并用该模型实时计算系统的对数似然概率(LLP),实现系统故障程度的量化。结合实际的SCADA数据,验证了该方法对故障预测有较好的灵敏性和准确度。

A fault quantification method of transmission system based on least square support vector machines(LSSVM)and Gaussian mixture model(GMM)is researched.A LSSVM regression model is built based on failure mechanism of transmission system.Its inputs are wind speed,ambient temperature,active power and 4 related previous temperature monitoring amounts of transmission system and outputs are 4 related temperature monitoring amounts.Deviation of predictive value and actual value of the model is defined as“deviation value”of the system.The GMM can fit probability distribution of deviation values of normal transmission system and logarithm likelihood probability(LLP)of system can be calculated through the GMM,which is fault quantification index of system.The fault prediction method is proved to be sensitive and accurate with actual SCADA data.