Long Term Electrical Load Forecasting: An Empirical Study across Techniques and Domains

Objective: The rapid development of human population, buildings, industries and technology has caused electric consumption to grow rapidly. So it is crucial to have accurate load forecasts for resource planning, rate cases, designing rate structures, financial planning etc. Long-term load forecasting&consumption is used for planning of power system expansion for developing and developed countries or states. In the past years, research mostly focused on short term forecasting instead of than long term. Methods: So here more emphasis is given to the methodologies used for long-term forecasting. Parametric methods like trend analysis, end use&econometric technique and computational intelligence based methods like artificial neural network; fuzzy logic model; wavelet networks and genetic algorithm/programming are explained in detail. Findings: The results from the comparative study show that the ANN model is a superior technique for long term electric load forecasting due to its ability to give satisfactory results where the availability of past data is low. It also gives lower error than the other models discussed. Application/Improvement: The findings will help the researchers to go for hybrid models for improving the error in long term forecasting. It can be applied for finding out the appropriate methodologies of long term forecasting of cities, states, countries or islands with respect to different economic scenarios.