A floating-point genetic algorithm for solving the unit commitment problem

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalpeer-review

55 Scopus Citations
View graph of relations



Original languageEnglish
Pages (from-to)1370-1395
Journal / PublicationEuropean Journal of Operational Research
Issue number3
Publication statusPublished - 16 Sept 2007


This paper proposes a floating-point genetic algorithm (FPGA) to solve the unit commitment problem (UCP). Based on the characteristics of typical load demand, a floating-point chromosome representation and an encoding-decoding scheme are designed to reduce the complexities in handling the minimum up/down time limits. Strategic parameters of the FPGA are characterized in detail, i.e., the evaluation function and its constraints, population size, operation styles of selection, crossover operation and probability, mutation operation and probability. A dynamic combination scheme of genetic operators is formulated to explore and exploit the FPGA in the non-convex solution space and multimodal objective function. Experiment results show that the FPGA is a more effective technique among the various styles of genetic algorithms, which can be applied to the practical scheduling tasks in utility power systems. © 2006 Elsevier B.V. All rights reserved.

Research Area(s)

  • Dynamic genetic strategy, Electrical power generation, Floating-point genetic algorithm, Generators scheduling and economic dispatch, Unit commitment