• Gagan Deep
• Wesley Sam
• Pavan Raj
• Chethan P P
• Ganapathi Sharma

This project presents a comprehensive framework for simulating, processing, and forecasting building energy consumption, aimed at enhancing operational efficiency and sustainability in an academic environment. We generate a multiyear synthetic dataset with hourly resolution, capturing floor-wise and device-level energy usage, occupancy, and environmental factors such as temperature and humidity. A Python pipeline transforms raw JSON records into structured CSV files, facilitating seamless integration with analytics platforms. Leveraging Long Short-Term Memory (LSTM) networks, the system delivers accurate short- and long-term consumption forecasts, identifying peak load intervals and uncovering latent usage patterns. A companion Streamlit application provides interactive dashboards, enabling stakeholders to visualize trends, detect anomalies, and obtain tailored energy-saving recommendations. Experimental evaluation demonstrates that our LSTM model achieves over 90% prediction accuracy (RMSE < 0.5 kWh) and successfully identifies irregular consumption events with high precision. The end-to-end solution empowers facility managers to make data-driven decisions, reduce energy costs by up to 20%, and support institutional sustainability objectives.

Synthetic data generation, energy forecasting, LSTM, anomaly detection, Streamlit, Energy management.

"ENERGY CONSUMPTION ANALYSIS AND PREDICTOR ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 7, page no.805-809, July-2025, Available :http://www.jetir.org/papers/JETIRGX06152.pdf

"ENERGY CONSUMPTION ANALYSIS AND PREDICTOR ", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 7, page no. pp805-809, July-2025, Available at : http://www.jetir.org/papers/JETIRGX06152.pdf