Optimal Load Prediction in a Smart Metering Network using Random Forest Algorithm

Smart meters are electronic devices used to measure and record real-time energy consumption. In this research, we utilized energy consumption data gathered from residential consumers in London, England. Our aim is to leverage various machine learning algorithms to establish connections between historical energy consumption data, allowing us to create and assess short-term load forecasts with low Root Mean Squared Error (RMSE) and Mean Absolute Percentage Error (MAPE). This research benefits customers by enabling them to plan their energy usage effectively, resulting in reduced overall energy bills. Our findings indicate that, among the tested algorithms, Random Forest stands out as the optimal choice, delivering the lowest RMSE and MAPE.