A Comparative Study of Methods to Forecast Domestic Energy Consumption Aggregated with Photovoltaic and Heat Pumps System

Abstract

Rise in the usage of photovoltaic (PV) system at the residential sector brought challenges for the distribution network operator (DNO) and with a high demand of the Heat Pump (HP) system to fulfil the target of low carbon emission potentially brought far greater trials to predict energy at the domestic network. Prediction is very crucial for electrical distribution companies since their business largely relays on how to make the most out of their energy generation without making it go to waste. This study compares different methods (from machine learning to deep learning) to forecast domestic energy consumption aggregated with HP and PV system. The prediction tool proudly uses large residential energy measured data at a minute frequency for a year combines synthetically with the real measured data of HP and PV system. The forecasting methods is different for various data type, this study allows to compare which one would be more efficient in which type of data set and which one to predicts the finest.