Scheduling economic power generators under insulation conditions from national land using multiobjective optimization

Johanna Amaya, Luceny Guzman, Yezid Donoso

Research output: Contribution to journalArticlepeer-review

Abstract

This paper presents a new multiobjective evolutionary algorithm for Economic Power Dispatch Problem (EPD). The EPD problem is formulated as a nonlinear constrained multiobjective optimization problem. A new strength Pareto evolutionary algorithm (SPEA) based approach is proposed to handle the EPD as a mixed combinatorial and continuous optimization problem with competing and noncommensurable objectives. The evaluated objective functions were: Power Production Costs and Losses on Transmission Lines. Performance of this algorithm was compared with the results obtained using traditional method E - Constraint and those are shown in graphics of Pareto Optimal fronts in two dimensions. Here we show the application of SPEA algorithm to determine the economic operation of power subsystem of Colombian Atlantic Coast considering only the generator units of this region of the country; this can be the solution when failures in the National Transmission System occur and the Caribbean coast is disconnected from the rest of national land. In addition, we apply this meta-heuristic to determine the schedule to one specific operation day looking for the combination that minimize start, stop and generation costs in each stage. The results demonstrate the capabilities of the proposed approach to generate well-distributed Pareto-optimal solutions of the multiobjective EPD problem in just five generations. The comparison with the classical techniques demonstrates and confirms its potential to solve the multiobjective EPD problem. The main contributions of this paper are: the multiobjective model, the development and solution of algorithm with computational complexity in polynomial time and contribution to energetic sector.

Original languageEnglish (US)
Pages (from-to)802-810
Number of pages9
JournalWSEAS Transactions on Mathematics
Volume6
Issue number8
StatePublished - Aug 2007

All Science Journal Classification (ASJC) codes

  • Endocrinology, Diabetes and Metabolism
  • Algebra and Number Theory
  • Statistics and Probability
  • Discrete Mathematics and Combinatorics
  • Management Science and Operations Research
  • Control and Optimization
  • Computational Mathematics
  • Applied Mathematics

Fingerprint

Dive into the research topics of 'Scheduling economic power generators under insulation conditions from national land using multiobjective optimization'. Together they form a unique fingerprint.

Cite this