Economic dispatch using fuzzy decision trees

Abstract Previous work (Elec. Power Sys. Res. 56 (2000) 225) has demonstrated the capability of decision trees (DT) in solving optimisation problems for the economic dispatch (ED) problem including environmental constraints. This paper explains a new improvement in the DT technique by adding fuzzy logic (FL) to the unit limits and load (FLDT). By doing so, the numerical convergence of the overall technique improves. Furthermore, the generating cost obtained via FLDT is lower than that obtained with the classical formulation due to the fuzzyfying of the generating unit limits. Also, including the uncertainty of the load as a fuzzy load, new results are obtained that account not only for minimum cost but also for uncertainty. A 10 unit Chilean test system is used to validate and highlight the performance of this proposition.