Title: A greedy randomised adaptive search procedure - genetic algorithm for electricity consumption estimation and optimisation in agriculture sector with random variation

Authors: Ali Azadeh; Seyed Mohammad Asadzadeh; Rana Jalali; Samira Hemmati

Addresses: Department of Industrial Engineering, Center of Excellence for Intelligent Based Experimental Mechanics, Karegar Shomali, P.O. Box 11155-4563, Tehran, Iran; Department of Engineering Optimization Research, College of Engineering, University of Tehran, Karegar Shomali, P.O. Box 11155-4563, Tehran, Iran ' Department of Industrial Engineering, Center of Excellence for Intelligent Based Experimental Mechanics, Karegar Shomali, P.O. Box 11155-4563, Tehran, Iran; Department of Engineering Optimization Research, College of Engineering, University of Tehran, Karegar Shomali, P.O. Box 11155-4563, Tehran, Iran ' Department of Industrial Engineering, Center of Excellence for Intelligent Based Experimental Mechanics, Karegar Shomali, P.O. Box 11155-4563, Tehran, Iran; Department of Engineering Optimization Research, College of Engineering, University of Tehran, Karegar Shomali, P.O. Box 11155-4563, Tehran, Iran ' Department of Industrial Engineering, Center of Excellence for Intelligent Based Experimental Mechanics, Karegar Shomali, P.O. Box 11155-4563, Tehran, Iran; Department of Engineering Optimization Research, College of Engineering, University of Tehran, Karegar Shomali, P.O. Box 11155-4563, Tehran, Iran

Abstract: This study presents a flexible algorithm for electricity energy consumption estimation and optimisation in agriculture sector based on greedy randomised adaptive search procedure (GRASP) and genetic algorithm (GA) with variable parameters using stochastic procedures. The standard economic indicators used in this paper are price, value added, number of customers and electricity consumption in the previous period. The proposed algorithm may be used to estimate energy demand in the future by optimising parameter values. The proposed algorithm uses analysis of variance (ANOVA) to select from GA, GRASP or conventional regression for future demand estimation. Furthermore, if the null hypothesis in ANOVA F-test is rejected, the least significant difference (LSD) method is used to identify which model is closer to actual data at α level of significance. To show the applicability and superiority of the proposed algorithm the data for electricity consumption in Iranian agriculture sector from 1979 to 1999 is used and applied to the proposed algorithm. This is the first study that introduces and uses an integrated GRASP-GA-regression for electricity consumption estimation and optimisation in agriculture sector.

Keywords: greedy randomised adaptive search procedure; GRASP; genetic algorithms; electricity consumption; analysis of variance; ANOVA; conventional regression; agriculture; random variation; energy consumption; energy demand forecasting; Iran.

DOI: 10.1504/IJISE.2014.062539

International Journal of Industrial and Systems Engineering, 2014 Vol.17 No.3, pp.285 - 301

Published online: 25 Jul 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article