Predictive-Modeling-for-Renewable-Energy-Surplus-Detection

A predictive model for Brighton , East Sussex, England aiming to notify local clients of potential renewable energy surpluses, allowing them to opt-in for free energy slots. The project focused on creating a reliable system that predicts energy surpluses (wind and solar) at least 24 hours in advance to minimize financial risks associated with false positives. Historical energy data from 2000 to present was analyzed, cleaned, and modeled. Key tasks included data exploration, preprocessing, feature engineering, time series analysis, and implementing machine learning models. The final model was evaluated for its accuracy and reliability to ensure it met the company’s criteria for deployment in May 2024. The project also involved presenting insights and recommendations to the company’s CEO, focusing on the operational feasibility and potential future commercial opportunities based on data trends.