Wind power main shaft monitoring system based on deep learning algorithm

The invention discloses a wind power main shaft monitoring system based on a deep learning algorithm, and relates to the technical field of fan safety monitoring. Vibration signals of a fan main shaft are collected through two displacement sensors and an absolute value encoder and are transmitted to a cabin controller for processing; the tower bottom controller transmits processed data to a remote upper computer monitoring system through an optical fiber ring network, the processed data are uploaded to an SCADA, the data are input into a built improved GASF-CA-Resnet50 model, and the GASF-CA-Resnet50 model is trained through measured values of two displacement sensors of the same model; during real-time monitoring, a GAF is used for converting a real-time signal into a two-dimensional image, and calculation is carried out in combination with a trained model, so that the accuracy of the monitoring system is further improved; a fan fault signal is determined in time, the fault type is judged as a fan spindle fa