Paper Title

Abstract - The amount of energy used by buildings is increasing as a consequence of increased urbanization and social advancement. Predicting a building's energy needs is essential for promoting sustainable growth and energy efficiency, which in turn reduces energy costs and has a lesser impact on the environment. This research focuses on the topic of applying deep learning (DL) techniques to forecast energy use across time series using actual data. The performance of statistical and DL algorithms was evaluated using data collected in real time from a smart grid installed in an experimental building. Usage of energy in ensemble and single situations was examined using well-known artificial intelligence techniques. The models which combine prediction and optimization approaches, is examined in-depth. The thorough comparative analysis demonstrated that the hybrid model was excellent in performance than the single and ensemble models in terms of accuracy. These models are thought to be suitable for usage and accurate enough to provide predictions, which can help users plan their energy management strategies. Keywords - Energy consumption; Artificial intelligence; Data mining; Time-series forecasting; Machine learning; Residential building