Optimization to improve the Physical and Mechanical Properties of the Electric Power Transmission Wires made from waste using a Genetic Algorithm

Full Text PDF PDF
Author(s) Hussein Ali Alwan | Haidar A.H. al-Jubouri | Nabil L. Al Saffar
Pages 1036-1049
Volume 3
Issue 12
Date December, 2013
Keywords genetic algorithm, recycling, optimization, Homogenizing treatment, Thermo – mechanical treatment, Electrical conductivity, Vickers hardness, Tensile, Thermal conductivity.

Abstract

The current study is one of the recent studies prospects in Iraq, where the exploitation of waste empty soft drink cans and scrap damaged electric wires for the manufacture of electric power transmission wires for voltage (11,33,66,132 kv). Practically been improved physical and mechanical properties of the alloy from through thermo-mechanical treatment where the percentage of improvement (electrical conductivity - thermal conductivity - Tensile strength - yield strength - hardness) of the alloy (C8) compared with the alloy base (A) (58%,54%,158.7%,226%,70.7%) respectively. This paper provides method to reach to the optimum alloy using the hybrid method, which is represented by the statistical parameters and genetic algorithms, where the use of statistical data obtained from practical results to determine the optimum properties of alloys (ie, in this research have been identified six of the properties of alloys), accordingly, the database was built describe alloys depending on their properties, then, the evolution algorithm of type breeder genetic algorithm to procedure genetic clustering process and provides a number of required clusters, to avoid the overlapping between clusters with other, one of the clustering validity measures called "Davies-Bouldin index" as fitness function of that algorithm was used. Then was extracted two types of properties for each cluster namely mechanical properties (Tensile strength - yield strength – hardness-elongation) and physical properties (electrical conductivity - thermal conductivity). The proposed methodology achieved 95% accuracy when compare process results with the results of the optimization algorithm.

< Back to December Issue