非线性状态估计法在风电齿轮箱故障预警中的应用

Fault Prediction in Gear Box of Wind Turbines Based on Nonlinear State Estimate Technology

摘要:

针对现阶段在齿轮箱的状态监测方法中遇到的问题,提出一种以易于采集的油温信号来对齿轮箱故障进行预警的分析方法;通过非线性状态估计法(nonlinear state estimate technology)建立齿轮箱的温度预测模型,之后根据统计温度预测残差的变化趋势,然后与通过验证集合确定出的残差均值和标准差的故障阈值进行对比,以实现及早发现齿轮箱的潜在故障,优化运行的效果;并且与神经网络预测方法进行对比分析,结果表明非线性状态估计法对风电齿轮箱进行故障预警有着较好的效果。

Abstract:

Aiming at the problem of condition monitoring method in the gearbox, it puts forward a new analysis method of gear box fault based on the oil temperature collected easily and establishes the temperature model of gear box by (nonlinear state estimate technology) NSET method to predict the temperature of the gear box. It is when gear box runs abnormally that the statistical features of the temperature prediction residual change largely to detect the fault of gearbox early. A statistical method is used to detect the changes of the residual mean value and standard deviation. When one of these parameters exceeds predefined thresholds, an incipient failure was flagged. And compared with neural network forecasting method, it highlights the advantages of this method.