An Efficient Approach for Short-Term Load Forecasting Using the Regression Learner Application

The regression method of forecasting allows businesses to use specific strategies so that those predictions, such as future sales, future needs for labor or supplies, or even future challenges, will yield meaningful information. In this paper, the use of the regression learner method for short-term load forecasting is performed. Its capabilities, advantages, and limitations are investigated using real load and weather data from the Hydro-Quebec database. Changes in electricity supply and demand have made their adjustment more complex. Our approach will be based on a predictive optimization method adapted to energy systems. This method is applied for the region of Mascouche (in Quebec, Canada) where the load varies between 60 megawatts and 140 megawatts.