A flexible long-term load forecasting approach based on new dynamic simulation theory — GSIM

A non-linear multivariable regression model for midterm energy forecasting of power systems

“…More specifically, the improvement being achieved for the mean forecasting value is greater than 0.2% for the second order regression model and greater than 2.1% for all the other models. It should be noted that the accuracy of the proposed method is very satisfactory when compared with the results obtained by the annual energy forecasting methods in [4,11,26,29,30,32,35,36,40], where the respective MAPE varies between 0.60% and 8.4%. Only Kandil et al [32] have reported smaller MAPE (0.60%) than the proposed model.…”

“…[31][32][33] a knowledge-based expert system has been implemented to support the choice of the most suitable long-term peak load forecasting model giving better results than the classical ones. Alternatively, decision support systems [34], support vector machines using simulated annealing algorithms [26] or genetic ones [35], models based on dynamic simulation theory (GSIM) [36] and Grey methodology [37] have also been proposed. Probabilistic forecasts of the peak electricity demand have been presented [38], giving the magnitude and the timing of the peak [39].…”

Analysis and comparison of various methods available for load forecasting: An overview

“…These trend curves, once extended, are utilized to forecast future load. Another technique for long term forecasting is correlation, [8] where load consumption is correlated with various demographic and economic factors such as population, industrial development, weather conditions etc. Three types of parametric models are discussed: trend analysis, end use models and econometric models 1) Trend analysis or Statistical Techniques: This technique based on previous changes in load consumption to predict changes in future load consumption.…”

“…The difference between long and medium term forecast is of time horizon. The time range for medium forecast is from a couple of months to a year [8]. The various available methods can be classified in [10]:…”

Forecasting the load of electrical power systems in mid‐ and long‐term horizons: a review

“…Jia et al [101][102] developed a dynamic simulation approach for LTLF a decade ago called General Simulations theory (GSIM) based on the limitations of both parametric and AI-methods in context to interaction between load and load-impacting factors. The technique was implemented in Tokyo area while comparing the results with traditional regression-based method.…”