Premium
Rule reduction in fuzzy logic for better interpretability in reservoir operation
Author(s) -
Sivapragasam C.,
Vasudevan G.,
Vincent P.,
Sugendran P.,
Marimuthu M.,
Seenivasakan S.
Publication year - 2007
Publication title -
hydrological processes
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.222
H-Index - 161
eISSN - 1099-1085
pISSN - 0885-6087
DOI - 10.1002/hyp.6488
Subject(s) - interpretability , fuzzy logic , fuzzy set operations , fuzzy number , defuzzification , computer science , neuro fuzzy , data mining , fuzzy classification , artificial intelligence , set (abstract data type) , fuzzy set , fuzzy associative matrix , machine learning , fuzzy control system , programming language
Decision‐making in reservoir operation has become easy and understandable with the use of fuzzy logic models, which represent the knowledge in terms of interpretable linguistic rules. However, the improvement in interpretability with increase in number of fuzzy sets (‘low’, ‘high’, etc) comes with the disadvantage of increase in number of rules that are difficult to comprehend by decision makers. In this study, a clustering‐based novel approach is suggested to provide the operators with a limited number of most meaningful operating rules. A single triangular fuzzy set is adopted for different variables in each cluster, which are fine‐tuned with genetic algorithm (GA) to meet the desired objective. The results are compared with the multi fuzzy set fuzzy logic model through a case study in the Pilavakkal reservoir system in Tamilnadu State, India. The results obtained are highly encouraging with a smaller set of rules representing the actual fuzzy logic system. Copyright © 2007 John Wiley & Sons, Ltd.