Short-term Load Forecasting Using Method of Multiple Linear Regression

In this study, we use Multiple Linear Regression to forecast short-term load. This study obtains a day-ahead load forecasting. The regression coefficients were calculated using the Least Squares estimation method. Load forecasting has an effective role in economic operation of power utilities. In an electrical power system, load is affected by temperature, due point, and seasons, as well as previous load consumption (historical data) [1].Temperature, Due point, prior day's load, hours, and prior week's load are the input variables. The mean absolute percentage error is used to validate the model or assess its accuracy, and R squared is checked [2-5], which is shown in the results section. A weekly prediction is also obtained using day-ahead projected data.