NCBC’s Sustainable Energy Informatics Lab develops STLF Tool for CPPA

Short Term Load Forecasting is essential for any power system operations. It is key to monitoring, planning and despatch of power supply. Not only is it necessary in ensuring stability in grid, it also helps good financial governance by using the resources efficiently, keeping the demand side in view.

Short term load forecasting (STLF) is one of the essential tools for managing energy markets. It uses a variety of inputs to come up with the most accurate load forecast for a given locale. It is available for any granularity, especially forecasts that may be needed at the DISCO scale. Although, the tool works at the level of DISCOs but could work at the level of circles, divisions, and sub-divisions for providing better accuracy and area-specific load forecasts to forecast the electricity demand on day-to-day intervals. SEIL team is working with Central Power Purchasing Agency  (CPPA) daily on the improvement of the STLF tool to achieve an average daily MAPE of < 3% . It uses a neural network based on historical load (daily lag, weekly lag, yearly lag), city’s temperature, seasonality, day of the week label. Its output consists of load forecast values for various time horizons. It is capable of forecasting district-level to national grid demand of Pakistan.

Leave a Comment

Your email address will not be published. Required fields are marked *