Abstract
Demand side load prediction is one of the most challenging tasks in smart grid environment due to uncertainties between demand and supply. Hence, in order to overcome this issue, this paper presents a scheme based on machine learning and deep learning for energy load forecasting by considering the weather condition of the area. We propose a two-tier Ensemble model, which ensembles the results of machine learning model (Support Vector Machine) and deep learning models (Convolu- tional Neural Network One Dimensional and Long Term Short memory) with a simple neural network to predict the load and demand gap. Then, we train and test the model with the UMass Smart* Dataset - 2017 release by taking the readings of appliances and weather conditions. The experimental results demonstrate that the proposed scheme has a significant improvement over the existing load forecasting methods having short-term and long- term load prediction models with an overall accuracy of 95.6%.