A deep learning model for short-term power load and probability density forecasting. (1st October 2018)

• Record Type: Journal Article
• Authors: Guo, Zhifeng; Zhou, Kaile; Zhang, Xiaoling; Yang, Shanlin
• Is Part Of: Energy. Volume 160(2018)
• Page Start: 1186
• Page End: 1200
• Publication Date: 2018-10-01
• Subjects: Deep learning -- Probability density forecasting -- Feature engineering -- Power load forecasting; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2018.07.090
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: Accurate load forecasting is critical for power system planning and operational decision making. In this study, we are the first to utilize a deep feedforward network for short-term electricity load forecasting. Our results are compared to those of popular machine learning models such as random forest and gradient boosting machine models. Then, electricity consumption patterns are explored based on monthly, weekly and temperature-based patterns in terms of feature importance. Also, a probability density forecasting method based on deep learning, quantile regression and kernel density estimation is proposed. To verify the efficiency of the proposed methods, three case studies based on daily electricity consumption data for three Chinese cities for 2014 are conducted. The empirical results demonstrate that (1) the proposed deep learning-based approach exhibits better forecasting accuracy in terms of measuring electricity consumption relative to the random forest and gradient boosting model; (2) monthly, weekly and weather-related variables are key factors that have a great influence on household electricity consumption; and (3) the proposed probability density forecasting method is capable of forecasting high-quality prediction intervals via probability density forecasting. Highlights: Deep neural network with multilayer perceptron for short-term load forecasting. Electricity consumption patterns are explored from different perspectives. Probability density forecasting method based on deep learning is proposed. Three case studies in China are presented to demonstrate the methods.