Modelling Multiobjective Bilevel Programming for Environmental-Economic Power Generation and Dispatch Using Genetic Algorithm

Abstract

This article describes a multiobjective bilevel programming (MOBLP) model to solve environmental-economic power generation and dispatch (EEPGD) problem through genetic algorithm (GA) based fuzzy goal programming (FGP) in a thermal power plant operational system. In MOBLP approach, first the objectives of problem are divided into two sets of objectives, and they are separately included at two hierarchical decision levels (top-level and bottom-level), where each level contains one or more controls variables associated with power generation decision system. Then, optimization problems of both the levels are described fuzzily to accommodate the impression arises with regard to optimizing them. In FGP model formulation, the membership functions associated with defined fuzzy goals are designed, and then they are converted into membership goals by assigning highest membership value (unity) as achievement level and introducing under- and over-deviational variables to each of them. In achievement function, minimization of under-deviational variables of membership goals according to weights of importance is considered to achieve optimal solution in decision environment. In the process of solving FGP model, a GA scheme is adopted at two stages, direct optimization of individual objectives at the first stage for fuzzy representation of them and, at the second stage, evaluation of goal achievement function to reach optimal power generation decision. The use of the proposed method is demonstrated via IEEE 30-bus system.