IEEE Project Abstract

Load forecasting is critical for effective schedulingand operation of power systems, which are becoming increasingly complex and uncertain, especially with the penetration ofdistributed power. This paper proposes a data-driven deep learning framework to forecast the short-term power load. First, the loaddata is processed by Box-Cox transformation. The tail-dependenceof the power load on electricity price and temperature is then investigated by fitting the parametric Copula models and computingthe threshold of peak load. Next, a deep belief network is built toforecast the hourly load of the power system. One-year grid loaddata collected from urban areas in both Texas and Arkansas, inthe United States, is utilized in the case studies on short-term loadforecasting (day-ahead and week-ahead) is conducted for four seasons independently. The proposed framework is compared withclassical neural networks, support vector regression machine, extreme learning machine, and classical deep belief networks. Theload forecasting performance is evaluated using mean absolute percentage error, root mean square error, and hit rate. The proposedframework outperforms the tested state-of-the-art algorithms, withrespect to the accuracies of both day-ahead and week-ahead forecasting. Overall, the computational results confirm the effectiveness of the proposed data-driven deep learning framework.