Multi-level Optimization of Reactive Power Compensation in Industrial Nets with Heuristic Modelling Techniques

Abstract

The distortion of the parameters in electric supply nets that take place in industrial complexes, leads to the inefficient operation of the equipment, all kind of failures and, in summary, to reduced economic operational indicators of the net. For this, in industrial nets the deep reactive power compensation becomes an indispensable tool for reducing all kind of economic losses. It is needed to compensate the reactive component of the supplied energy and its distortion. The efficiency of the compensation depends on the adequate allocation of reactive power compensation units at the nodes of a network. Present work intends to formalize the compensation of the reactive power in industrial electric supply nets as a bi-level multiple objective optimization task, based in it systemic analysis, to propose a solution scheme based on the Random Exploration of the Extremes of Variable Codes Algorithm and the comparison of it behaviour with the one obtained from the application of the Localization of the Extremes of Variable Codes Algorithm. These results are applied to a study case devoted to the optimal multiple objective reactive power compensation in a concrete industrial electric supply net. The feasibility of solving the optimal selection of units of compensation of reactive power in great complexity industrial electric supply nets task demonstrated, based on the mathematical modeling and solution procedures developed.