Method and apparatus for detecting fault, method and apparatus for training model, and device and storage medium

A method for training a temperature prediction model, comprising:

• acquiring training data of the temperature prediction model, wherein the training data comprises training samples, the training samples comprising history temperature of a target wind turbine and active power and turbine speed corresponding to the history temperature, the active power and turbine speed corresponding to the history temperature being intended to characterize a working state of the target wind turbine;
• acquiring training data after data cleaning by performing data cleaning on the training data, wherein the training data after data cleaning is intended to train the temperature prediction model;
• calculating a predicted temperature corresponding to the active power and turbine speed by the temperature prediction model;
• and
• training the temperature prediction model based on the predicted temperature and the history temperature.

A computer device for detecting a fault for a wind turbine, comprising:

• a data acquiring module, configured to acquire characteristic data and an actual temperature of a first wind turbine among n wind turbines, wherein the characteristic data of the first wind turbine is intended to characterize a working state of the first wind turbine, and n is an integer greater than 1;
• a temperature acquiring module, configured to acquire a prediction temperature set including a plurality of predicted temperatures by inputting the characteristic data of the first wind turbine into a plurality of temperature prediction models, each temperature prediction model corresponding to one of the n wind turbines;
• and
• a fault detecting module, configured to detect, based on the predicted temperature set and the actual temperature of the first wind turbine, whether the first wind turbine encounters a fault.

A computer device for training a temperature prediction model, comprising:

• a data acquiring module, configured to acquire training data of the temperature prediction model, wherein the training data comprises training samples, the training samples comprising history temperature of a target wind turbine and active power and turbine speed corresponding to the history temperature, the active power and turbine speed corresponding to the history temperature being intended to characterize a working state of the target wind turbine;
• a data preprocessing module, configured to acquire training data after data cleaning by performing data cleaning on the training data, wherein the training data after data cleaning is intended to train the temperature prediction model;
• a temperature calculating module, configured to calculate a predicted temperature corresponding to the active power and turbine speed by the temperature prediction model;
• and
• a model training module, configured to train the temperature prediction model based on the predicted temperature and the history temperature.