Short-term Load Forecasting Method Based on Artificial Intelligence Highway Neural Network

Comprehensively considering power load and real-time weather influence factors, this paper proposes a short-term load forecasting method based on artificial intelligence highway neural network (HNN). Firstly, through the Pearson correlation analysis, the weather influencing factors that affect the most can be obtained, which are considered as part of the input variable of the model. Then, the principle component analysis (PCA) method is used to reduce the dimension of the daily feature matrix. And the matrix is used as the input of K-means clustering to divide the operating days into three categories. Next and most importantly, the HNN models corresponding to each type of operating days are established separately. Finally, through comprehensive case studies, it is proved that the HNN load forecasting model proposed in this paper can achieve better forecasting results than support vector machine (SVM), long short-term memory (LSTM) and non-clustering HNN method.