Abstract: | In this paper, a short-term power network load forecasting algorithm based on evolutionary neural network is proposed. In this paper, an improved artificial bee colony algorithm is combined with BP neural network to generate an evolutionary neural network, and then the bias and weight of the evolutionary neural network are optimized by using the improved artificial bee colony algorithm. The algorithm takes the historical load data of thermal power as input, and uses evolutionary neural network to train the forecasting model to predict the power grid load in the future. First, historical load data is obtained. Then, the obtained data are input into the evolutionary neural network model for training. In the training process, the improved artificial bee colony algorithm is used to optimize the weight and bias of evolutionary neural network to improve the prediction accuracy of the model. As a global search algorithm, artificial bee colony algorithm can effectively explore the model parameter space and find the optimal model parameter combination, thus improving the prediction accuracy of the model. In order to verify the validity of the proposed load forecasting method, we use the load data of thermal power network to test. The experimental results show that the evolutionary neural network proposed in this paper shows good forecasting accuracy and practicability in short-term power grid load forecasting. Compared with traditional prediction methods, the prediction error of this algorithm is smaller and the prediction result is more accurate and reliable. |