This paper presents a Genetic Algorithm (GA) for long-term natural Electricity consumption prediction. Six models are proposed to forecast annual Electricity demand. 27 GA have been constructed and tested in order to finding best GA for Electricity consumption. Four parameters have been considered in constructing and examination of plausible GA models. Type of selection, crossover, mutation and crossover fraction are four mentioned parameters. Five different selection operators are considered in building GA as follows: stochastic uniform, remainder, uniform, roulette, and tournament. Three different mutation operators are used as follows: gaussian, uniform, and adaptive feasible. Also scattered, single point, two points, intermediate, heuristic, and arithmetic are six different crossover operators that are considered in present study. Finally, the value between 0 and 1 is considered for crossover fraction. The proposed models consist of input variables such as: Gross Domestic Product (GDP) and Population (POP). All of trained GA are then compared with respect to mean absolute percentage error (MAPE). To show the applicability and superiority of the GA, actual Electricity consumption in U.S.A, Canada, Germany, United Kingdom, Japan, France and Italy from 1980 to 2007 are considered.
کلید واژگان :Electricity Consumption; Long-Term prediction; Genetic Algorithm .
ارزش ریالی : 300000 ریال
با پرداخت الکترونیک