ALO-CNN-LSTM Short-term Load Forecasting by Considering Feature Correlation
In view of the fact that the short-term load forecasting model does not fully consider the timing and nonlinearity of the load and the high redundancy of the historical load,an ALO-CNN-LSTM short-term load forecasting model considering feature correlation is proposed.The convolutional neural network(CNN)is used to obtain high-dimensional spatial features of load time series.The Copula function is used to analyze the correlation between the sequence of meteorological factors such as weather and humidity and the high-dimensional spatial features,and the characteristic parameters with high correlation are se-lected.At the same time,combined with the ant lion optimization algorithm(ALO)to train the model and determine the opti-mal parameters to improve the convergence speed and prediction accuracy of the model.The simulation analysis is carried out by taking the load of the electrical mathematical modeling competition as an example.Different optimization algorithms and pre-diction models are compared.The simulation results show that the model has faster convergence speed and higher prediction ac-curacy,which verifies the validity and practicability of the proposed model.
万方数据
维普
