Forecasting Bus Loads with a Combined Intelligent Prediction System

• Published in: 2018 IEEE International Conference on Environment and Electrical Engineering and 2018 IEEE Industrial and Commercial Power Systems Europe (EEEIC / I&CPS Europe) pp. 1 - 6
• Main Authors: Panapakidis, Ioannis P., Skiadopoulos, Nikolaos, Christoforidis, Georgios C.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2018
• Subjects: Artificial neural networks; bus loads; clustering; Clustering algorithms; Forecasting; Load forecasting; Load modeling; machine learning; Neurons; Predictive models; Training
• DOI: 10.1109/EEEIC.2018.8494376

Forecasting of load demand has been subject of continuing research over the last decades. Traditionally the focus has been mainly at system level, where local particularities disappear, however recently has been transferred to certain buses as well. One reason for that is the constant increase of renewable energy sources (RES) at the distribution level, and the technology advancements in microgrids, bringing forward the need for increased reliability of the demand coverage. In such cases forecasting is a more challenging task. Generally, the optimal design and operation of microgrids and distributed RES require reliable load forecasts. Bus load forecasting focuses on distribution system loads. The present study proposes a novel load Short-Term Load Forecasting (STLF) model specially tailored for loads with high volatility. The model refers to the hybridization of clustering and Feed-Forward Neural Network (FFNN). Experimental results and analysis indicate the robustness of the model.