Abstract
Modern economies depend on electricity as the driving engine. Electricity production and consumption (load) always have to be in equilibrium, due to the fact that storing electricity in a substantial quantity, results in high costs. Load forecasting is a task of forecasting future electricity consumption or power load. However, the difficulty in making accurate forecasting for future consumption proves to be a difficult task. This is due to the role played by various external factors that affect the consumption behaviour of the users. These factors can be social, economic, environmental, etc. which has led to the development of several forecasting techniques. The study discovered that forecasting techniques can be broadly categorised into statistical and computational intelligence approach. In this study, various state-of-the-art methods found in literature are studied and presented with emphasis on techniques used for short-term electric load forecasting.