Fuzzy controller in the selection of sugarcane and energy cane ideotypes

Abstract

In response to the new requirements of the sugar-energy sector, sugarcane breeding programs are addressing the selection of clones for different purposes, e.g., sucrose production, ethanol production for first-generation (1G) and second-generation (2G) biofuels. Consequently, agronomic variables such as fiber content, sucrose content and biomass yield, become more relevant in the selection process, for underlying the definition of conventional, high biomass and multipurpose ideotypes. The relation between fiber and sucrose contents and biomass quantity should be used to differentiate thee different clone types: conventional clones, with high sucrose content and high biomass; high biomass, with high fiber content and high biomass; or multipurpose, with high fiber and sucrose contents and high biomass. In view of the difficulty of selecting different ideotypes from the same population, the objective of this study was to develop a method of selecting sugarcane clones for the different purposes. A population with 220 clones derived from crosses involving parents of different Saccharum species were subjected to the new methodology of clone selection with a fuzzy controller programmed to classify experimental clones in three ideotypes. Apart from the selection involving real data, the fuzzy controller was tested on 26 simulated populations. The controller proved to be efficient in the classification and selection of clones from the three ideotypes with real and simulated data.