Summary
The Indian power system is undergoing significant transformation due to technological advancement, environmental goals, evolving consumption patterns, renewable energy integration, emerging load types, weather diversity, and special events. As part of the national target of achieving 500 GW of non-fossil-fuel-based installed capacity by 2030, India has already surpassed 250 GW of renewable energy capacity [1]. Furthermore, electricity generation in India reached approximately 1829.7 BU in 2024–25, and the all-India peak demand met during the year was approximately 250 GW, underscoring the scale of the system and the need for reliable demand forecasting [2].
Read more Read lessThis paper presents a day-ahead forecasting framework for All-India electricity demand using a feed-forward artificial neural network, with forecasts generated for each 5-minute block of the day. The model incorporates historical demand lags, weather variables from multiple locations, and event indicators as input features. Historical demand data obtained from the national SCADA system and weather data from the Indian Meteorological Department for the year 2024 were used for model development. Forecasting performance was evaluated using unseen operational data for January 2025. The results show that the proposed data-driven ANNbased approach can effectively support short-term demand forecasting for high-resolution operational applications in the Indian power system.
Additional informations
| Publication type | Session Materials |
|---|---|
| Reference | D2_10524_2026 |
| Publication year | |
| Publisher | CIGRE |
| Country | India |
| Study committees | |
| File size | 886 KB |
| Price for non member | 30 € |
| Price for member | 30 € |
Authors
PATEL* Praveen - Grid-India INDIA; SUBHLAXMI Manisha - Grid-India INDIA; MUKHERJEE Subhendu - Grid-India INDIA; GUPTA Abhishek - Grid-India INDIA; ROY B S - Grid-India INDIA
Keywords
