Variable Neighborhood Search Guided Differential Evolution for Non Convex Economic Load Dispatch

Abstract

This article addresses a novel and effective algorithm for solving the economic load dispatch (ELD) problem of generating units. Generator constraints, such as valve point loading, ramp rate limits and prohibited operating zones are taken into account in the problem formulation of ELD. The cost function of the generating units exhibits nonconvex characteristics, as valve-point effects are modeled and included as rectified sinusoid components in its conventional formulation. The paper investigates the application of variable neighborhood search (VNS) to tune Differential Evolution (DE) algorithm for solving ELD problem considering non-smooth characteristics (NSELD). The basic idea of VNS is to perform a systematic change of neighborhood within a local search. The hybrid method incorporates the DE as the main optimizer and VNS as the local optimizer to fine-tune the solution region discovered by DE during its progress of run. Thus, the VNS guides PSO for better performance in the complex solution space. To demonstrate its efficiency and feasibility, the VNS guided DE is applied to solve NSELD problem of power systems with 6 and 13 units. The simulation results obtained from the VNS guided DE was compared to those from previous literature in terms of solution quality and computational efficiency. It is shown that, the proposed technique for non-smooth ELD problem generates quality solutions reliably.