Evaluación mediante indicadores clave de rendimiento del despacho económico hidrotérmico resuelto por medio de técnicas heurísticas

Abstract

This paper solves the hydrothermal scheduling problem by using heuristic techniques as an alternative to exact optimization methods. The techniques used are the particle swarm method, genetic algorithms with its variants of selection by roulette and tournament and the novel and recent grey wolf algorithm. The objective of the proposed optimization model lies in the minimization of the objective function regarding fuel costs of thermal power plants, also considering the valve point effect that gives a more realistic touch to the problem. The proposed solution methodology includes penalties in the objective function related to the violations of constraints of the power balance and dynamic balance of reservoirs. The economic dispatch is executed for a test system composed of multiple thermal power plants and several hydroelectric power plants with cascaded reservoirs. The algorithms developed were implemented in the Matlab software. Additionally, the results achieved by each heuristic technique are innovatively evaluated using different key performance indicators.