Grey and Linear Regression Models in Load Forecasting for Enhanced Smart Grid Management- Smart Grid Modelling

Abstract

Load forecasting is the estimation of future electric load based on historical electric load data and it is important in power system expansion planning and management. To economically and reliably operate the grid, a power utility company requires accurate demand forecasting models. Therefore, developing and refining the existing forecasting models in the field of load forecasting is an interesting research concept. In this paper we compare advanced Grey Model (GM) with the linear trend regression method in load forecasting for enhancing smart grid management systems. The results show that the Grouped Grey Model, GGM(1,1), which is an improvement from the original Grey Model, GM(1,1), is more accurate and reliable in load forecasting compared with the linear trend regression method. The power utility company will find it useful to adopt the GGM(1,1) in load forecasting for power system planning such as fuel supplies scheduling, maintenance operation and unit management. Hence, smart grid modelling for enhanced grid management.