Boning Yu’s research while affiliated with State Grid Electric Power Research Institute and other places

Accurately predicting electricity demand is of crucial importance for optimizing power resource allocation, improving grid operation safety and providing significant economic benefits. In recent years, deep learning models have been widely employed in time series analysis tasks, such as recurrent neural networks (RNNs), gradient boosting decision trees (GBDT), and transformer networks. However, these methods only focus on a single time dimension. For a specific periodic process, the temporal variations at each time point within the process are not only related to neighboring moments but also highly correlated with neighboring periods, exhibiting both intraperiod and interperiod temporal changes. In this study, we propose the TimesNet temporal forecasting model and conduct experiments using electricity consumption data from the A city, combined with a series of covariates such as temperature and holidays, to train the model and predict electricity consumption. We compare the performance of TimesNet with other methods, and the results demonstrate that the TimesNet model outperforms the other models. Overall, the hybrid model we propose provides a valuable framework for accurately predicting electricity demand and holds practical significance for managing and operating power grids in urban areas.

The accuracy of electricity demand forecasting is closely related to the correctness of decision-making in the power system, ensuring stable energy supply. Stable energy supply is a necessary guarantee for socioeconomic development and normal human life. Accurate electricity demand forecasting can provide reliable guidance for electricity production and supply dispatch, improve the power system's supply quality, and ultimately enhance the security and cost-effectiveness of power grid operation, which is crucial for boosting economic and social benefits. Currently, research on electricity demand forecasting mainly focuses on the single-factor relationship between power consumption and economic growth, industrial development, etc., while neglecting the study of multiple influencing factors and considering different time dependencies. To address this challenge, we propose a transformer-based forecasting model that utilizes transformer networks and fully connected neural networks (FC) for electricity demand forecasting in different industries within a city. The model employs the encoder part of the transformer to capture the dependencies between different influencing factors and uses FC to capture time dependencies. We evaluate our approach on electricity demand forecasting datasets from multiple cities and industries using various metrics. The experimental results demonstrate that our proposed method outperforms state-of-the-art methods in terms of accuracy and robustness. Overall, we provide a valuable framework in the field of electricity demand forecasting, which holds practical significance for stable power system operations.