Power system frequency trajectory prediction method and device, medium and equipment

The invention discloses a power system frequency trajectory prediction method and device, a medium and equipment. Firstly, an electromagnetic simulation model is constructed, and a data set is obtained. And then, a data set is screened by using a clustering algorithm and a knee elbow method, the search range and speed of the algorithm are maximized, and the prediction model has higher robustness and reliability. Thirdly, calculating a correlation coefficient, determining a sensitive characteristic quantity, selecting an input variable strongly correlated with a response variable, improving the accuracy and stability of the prediction model, reducing the number of the input variables, and reducing the training time of the model; and then, optimizing the hyper-parameter combination by using a hyper-parameter optimization algorithm. And finally, constructing a prediction model based on the optimized hyper-parameter combination, and carrying out iterative training until convergence. Therefore, the training speed