Abstract:The power industry plays a pillar role in the national industrial system. The smooth operation of electricity is related to the lifeline of the national economy. In power system management, power load forecasting is crucial. Accurate power load forecasting can provide important basis for the smooth operation of the system and real-time power dispatching. Especially in the economic field, power load forecasting can play a significant role in rationally deploying resources, optimizing power generation plans, and achieving optimal social and economic benefits. However, with the rapid economic development of our country, the demand for electricity is increasing day by day, and the power load itself is also affected by the date, weather, climate, market, and policies, which greatly increases the difficulty of accurately predicting the power load. People have always been committed to improving the accuracy of power load forecasting. Artificial neural networks have the advantages of self-learning and generalization ability. They have been widely used in power load forecasting and have achieved satisfactory results. In recent years, the field of artificial neural networks has made gratifying breakthroughs, and a new research field of deep learning has emerged. This article based on the latest development of artificial neural network, combined with the actual data on the short-term powerload forecasting issues related research.