Published Paper Details:

Accurate forecasting of electrical appliance usage and peak demand is crucial for effective planning, maintenance, and the development of automation in electrical power systems. Discrepancies between actual appliance consumption and energy demand may arise from various factors, such as losses in lines and appliances, as well as mismanagement of energy demand. A groundbreaking approach for forecasting household electric appliance consumption and peak demand through a hybrid machine learning (ML) framework is presented here. To address these variations, a thorough examination of smart meter data is essential to pinpoint the key attributes and primary causes of differences between electrical appliance consumption and customers' peak demand. Understanding these intricacies is vital for optimizing power system operations and implementing strategies to enhance efficiency and reliability. The study employs a comprehensive dataset spanning multiple households, ensuring the generalizability of the proposed methodology. Results demonstrate superior predictive capabilities compared to traditional models, offering significant advancements in energy demand forecasting precision. This hybrid approach not only captures nuanced relationships within the data but also adapts dynamically to changing environmental and behavioral factors. As the energy landscape evolves, our innovative methodology stands poised to revolutionize how we understand and anticipate household electricity consumption, providing valuable insights for policymakers, utility providers, and consumers alike.