The main shaft assembly runout of wind turbine is a kind of fault mode with high frequency and great fault harm in wind farms. At present

the method for detecting the runout displacement of the main shaft assembly in wind farm has the shortcomings of low detection efficiency and poor measurement accuracy. To solve this problem

an efficient on-line monitoring technology for detecting the runout displacement of the main shaft assembly is proposed

and the corresponding monitoring device and early warning system is developed to realize the real-time and on-line monitoring of the runout displacement of the main shaft assembly. The monitoring data of the main shaft assembly runout is obtained by installing laser displacement sensor reasonably

and then the data is analyzed by time series prediction method

and the ARMA model of the main shaft assembly runout is established.The actual application shows that

by predicting the development trend of the runout displacement of the main shaft assembly

the runout state of the main shaft assembly can be grasped in time and its abnormality or deterioration trend can be detected in advance

thus the early warning information can be given when it is judged that the main shaft assembly is likely to have runout fault. Therefore

the generation and spread of the runout fault can be avoided and the safe and stable operation of wind turbine can be ensured.