An adaptive network based fuzzy inference system–genetic algorithm clustering ensemble algorithm for performance assessment and improvement of conventional power plants

Abstract

Performance measurement and assessment are fundamental to management planning and control activitiesof complex systems such as conventional power plants. They have received considerable attentionby both management practitioners and theorists. There has been several ef?ciency frontier analysismethods reported in the literature. However, each of these methodologies has its strength and weakness.This study proposes a non-parametric ef?ciency frontier analysis methods based on adaptive networkbased fuzzy inference system (ANFIS) and genetic algorithm clustering ensemble (GACE) for performanceassessment and improvement of conventional power plants. The proposed ANFIS-GA algorithm is capableto ?nd a stochastic frontier based on a set of input–output observational data and do not require explicitassumptions about the functional structure of the stochastic frontier. Furthermore, it uses a similarapproach to econometric methods for calculating the ef?ciency scores. Moreover, the effect of the returnto scale of a power plant on its ef?ciency is included and the unit used for the correction is selected bynotice of its scale. GACE is used to cluster power plants to increase homogeneousness. The proposedapproach is applied to a set of actual conventional power plants to show its applicability and superiority.The superiority and advantages of the proposed algorithm are shown by comparing its results againstANN Fuzzy C-means Algorithm and conventional econometric method.