AP-Transco is deploying an electricity forecasting model using Artificial Intelligence (AI) and Machine Learning (ML) to forecast next day’s consumption, including the day-ahead electricity demand every 15 minute basis.
This enables taking right decisions on electricity demand and supply, management of grid and minimising power purchase cost.
According to an official release, tech giant Google offered to jointly develop the forecasting model with Andhra Pradesh State Load Dispatch Centre (APSLDC). It will be a major step towards making Andhra a preferred destination for cost-effective power.
The model consumes different kinds of data, including climate data, holidays, seasonal information, weather forecasts etc, flowing into its servers on a real-time basis and achieves a high degree of accuracy with an error of less than three percent in forecasting.
According to APSLDC officials, AP-Transco also intends to develop four more forecast models for wind energy, solar energy, market prices, Central generating stations' surplus and frequency.
Besides, AP-Transco is developing a least-cost electricity dispatch model which will tell how much electricity should be dispatched every 15 minutes next day from each generating station.
The load dispatch centres in India are now using manual forecast mechanism which is not accurate and leads to excess or under drawing of power which entail heavy penalties and sometimes leads to power cuts, said APSLDL officials.
On the other hand, APSLDC has planned to transfer the knowledge about electricity demand forecasting to start-up companies willing to commercialise the product.
Energy secretary N. Srikant stated that the government aimed at building an efficient and financially robust power sector to address the future power demand and supply cost effective power to consumers with zero interruptions.
Power generators and DISCOMs rely on price forecasting information to develop their corresponding bidding strategies. If a generator has an accurate forecast of the prices, it can develop a bidding strategy to maximise its profit.
