Transmission line dynamic thermal constant value probability prediction method based on BILSTM-MDN

The invention provides a transmission line dynamic thermal constant value probability prediction method based on BILSTM-MDN, and belongs to the field of power system line operation state evaluation. The method comprises the following steps: firstly, carrying out normalization processing on transmission line surrounding environment data and DTR sequence data, and calculating an autocorrelation coefficient; secondly, sliding window features and labels are generated from the data according to the common input length, a sliding window data set is generated in a rolling mode, and a training set and a prediction set are divided; thirdly, a BILSTM-MDN neural network is built, after hyper-parameters are set, the training set is input for learning, and the hyper-parameters are adjusted until the error reaches the minimum. And finally, the prediction accuracy of the model is evaluated through the prediction set, and a final prediction result and a final prediction error are obtained. According to the method, the front