Wattsahead - An LSTM-Based Predictive Framework For Energy Consumption Forecasting In Suburban India
DOI:
https://doi.org/10.64252/bppjwk48Keywords:
Energy Consumption Forecasting, Long Short-Term Memory (LSTM), Time Series Prediction, Power Outage Management, Smart Grid, Blackout-Aware Prediction, Suburban Energy Demand, Machine Learning, RMSE and MAPE, Deep Learning for Energy, Energy Optimization, Sustainable Energy Systems, Web-Based Energy Monitoring, MERN Stack Integration, Real-Time Prediction.Abstract
This thesis presents our research on energy consumption forecasting in suburban regions, focusing on the development of an LSTM-based prediction model capable of addressing power cut scenarios. The core focus of the research is to accurately forecast short-term electricity demand using time-series data and to provide intelligent insights during blackout conditions to support proactive energy management. To achieve this, we utilize open-source suburban datasets to ensure legal compliance while maintaining real-world applicability. Our model leverages Long Short-Term Memory (LSTM) networks to capture complex temporal dependencies in electricity consumption patterns. Evaluation using RMSE and MAPE demonstrates strong predictive performance. Furthermore, we deploy the model through a MERN stack-based web platform, allowing users to access real time forecasts, visualize consumption trends, and receive actionable recommendations during outages. This research contributes to the advancement of smart energy systems by filling the gap in blackout-aware forecasting and offering a practical, user-friendly tool for sustainable energy planning and grid reliability.
