Artificial intelligence applied to the adjustment of the overall efficiency matrix from hydroelectric plants

• Grant number: 11/02100-7
• Support Opportunities: Scholarships abroad - Research
• Start date: June 01, 2011
• End date: August 31, 2011
• Field of knowledge: Engineering - Civil Engineering - Hydraulic Engineering
• Principal Investigator: Ieda Geriberto Hidalgo
• Grantee: Ieda Geriberto Hidalgo
• Host Investigator: Darrell G. Fontane
• Host Institution: Faculdade de Tecnologia (FT). Universidade Estadual de Campinas (UNICAMP). Limeira , SP, Brazil
• Institution abroad: Colorado State University, United States

Abstract

One of the great economic and environmental challenges facing humanity nowadays is the production of electricity. The Brazilian hydrothermal system has peculiar characteristics that distinguish it from other systems in the world and make the task of planning of the electric energy production difficult. In general, this task is done with the aid of computational tools that manipulate data in mathematical models. In Brazil, the input data to the software used for planning of the hydroelectric operation present problems of inconsistency, simplification and duplication that may compromise the reliability of the results. One of the most important information for these models is the function of overall efficiency from hydroelectric plants, represented graphically by a hill curve or by a matrix that relates, for example, power output, gross head and efficiency. Previous studies show that the quality of this information has great impact for the optimization and simulation models of the system operation. Therefore, this research aims to apply a technique from the Artificial Intelligence area, known as Genetic Algorithm, to adjust the overall efficiency matrix from hydroelectric plants. The objective is to improve the quality of this information, contributing to the choice of an operational policy that meets the Brazilian consumer market in an economic and safe way. (AU)