Long-term electrical energy consumption formulating and forecasting via optimized gene expression programming

This study formulates the effects of two different historical data types on electrical energy consumption of ASEAN-5 counties. On this basis, optimized GEP (gene expression programming) is applied to precisely formulate the relationships between historical data and electricity consumption. The optimized GEP is a more recent extension of GEP with high probability of finding closed-form solution in mathematical modeling without prior knowledge about the nature of the relationships between variables. This merit is provided by balancing the exploration of solution structure and exploitation of its appropriate weighting factors through use of a robust and efficient optimization algorithm in learning process of GEP. To assess the applicability and accuracy of the proposed method, its estimates are compared with those obtained from ANN (artificial neural network), SVR (support vector regression), ANFIS (adaptive neuro-fuzzy inference system), rule-based data mining algorithm, GEP, linear and quadratic models optimized by PSO (particle swarm optimization), CSA (cuckoo search algorithm) and BSA (backtracking search algorithm). The simulation results are validated by actual data sets observed from 1971 until 2011. The results confirm the higher accuracy of the proposed method as compared with other artificial intelligence based models. Future estimations of electrical energy consumption in ASEAN-5 countries are projected up to 2030 according to rolling-based forecasting procedure. © 2017 Elsevier Ltd. All rights reserved.