基于EEMD小波阈值去噪和CS-BP神经网络的风电齿轮箱故障诊断

针对风电机组齿轮箱中齿面点蚀、齿轮磨损、断齿等故障的诊断问题,提出一种基于EEMD小波阈值去噪和布谷鸟算法优化BP神经网络的故障诊断方法。采用EEMD分解和小波阈值去噪方法对故障振动信号进行数据预处理,抑制原始振动信号中的噪声干扰。利用布谷鸟算法优化BP神经网络对预处理后的信号进行诊断。小波阈值能更好地对EEMD分解中的高频分量进行去噪处理,CS-BP神经网络具有准确的模式识别精度和出色的全局寻优能力。通过实例仿真表明,提出的故障诊断方法具有良好的诊断精度、速度和成功率,具有较高的应用价值。

Fault Diagnosis of Wind Turbine Gearbox based on EEMD Wavelet Threshold Denoising and CS-BP Neural Network

For the fault diagnosis of the pitting,gear wear and tooth breakage in the gearbox of wind turbines,a fault diagnosis method based on EEMD and wavelet threshold denoising and cuckoo search to optimize BP neural network is proposed.The EEMD decomposition and wavelet threshold denoising method are used to preprocess the fault vibration signals and suppress the noise interference in the original vibration signals.The cuckoo search is used to optimize the BP neural network to diagnose the preprocessed signals.The wavelet threshold can better denoise the high frequency components in the EEMD decomposition,and the CS-BP neural network has accurate pattern recognition accuracy and excellent global optimization ability.The simulation shows that the diagnostic method has good accuracy,speed and success rate,and is of high use significance.