基于主成分和神经网络的风力发电机主轴承故障预警

Study on early warning the fault of wind turbine’s main bearing based on principal component and neural network

主轴承作为风力发电机组的重要部件,一旦发生故障,会影响风力发电机组整机工作的发电性能,严重时故障甚至会造成停机,不仅影响发电量,更会产生高昂的维修费用。通过运用相关性分析,根据Pearson相关系数矩阵对原有的多个指标进行分析。然后运用主成分分析,首先对数据的原始特征预处理,得到6个主成分,然后将这6个主成分作为BP神经网络的输入,运用神经网络对风力发电机的主轴承进行预警。神经网络模型结果表明,该模型对风力发电机主轴承故障预警具有非常好的识别效果,基于主成分和神经网络对风力发电机主轴承故障预警对实现机组智能故障诊断,提高机组的运行效率具有十分重要的意义。

As an important component of wind turbine,the main bearing was once in the event of failure,it will affect the power generation performance of the whole wind turbine.Serious faults even can cause downtime and leading to high maintenance costs.Using the correlation analysis and according to Pearson correlation coefficient matrix,the original multiple indicators are screened and important indicators are extracted.Inducting principal component analysis method to pre-analyze the original character of the data,and using the six principal components of original character as the input of BP neural network.Then study on early warning the Fault of Wind Turbine’s Main Bearing.The simulation results show that the model has a good recognition effect on early warning the fault of the main bearing.This method is of great significance for achieving the intelligent diagnosis of Wind Turbine and improving the efficiency of Wind Turbine.