Fuzzy modeling of the effects of irrigation and water salinity in harvest point of tomato crop. Part I: Description of the method

Abstract

It was used statistical techniques for the evaluation of agricultural experiments, but there are mathematical theories that allow finer adjustments, highlighting among them, the fuzzy logic. The objective of the study was characterizing a method of fuzzy modeling from an agronomic experiment. For this study it was used data from an experiment conducted at the School of Agriculture of São Paulo State University (UNESP) in Botucatu-SP. The system input variables based in fuzzy rules were soil water tension and doses of water salinity, being defined three fuzzy sets. The output variables was elected from the biometric and productivity analysis that showed statistically significant differences, namely, plant height, stem diameter, leaf area, green biomass, dry weight, number of fruits, average fruit weight and percentage of disabled fruits. For output variables 9 fuzzy sets were defined. From the adopted methodology, the model allowed extract directly from the data set a base of rules without the use of questionnaires to experts for its preparation. In addition, it will analyze intermediate regions at trial levels and weave other conclusions of the tomato growth and productivity, not limiting in this way only those observed with statistical analysis.