Tuning of Type-2 Fuzzy Systems by Simulated A nnealing to Predict Time Series
Majid Almaraashi, Robert John · 2011
In this paper, a combination of interval type-2 fuzzy system (IT2FS) models and simulated annealing are used to predict the Mackey-Glass time series by searching for the best configuration of the IT2FS. Simulated annealing is used to optimise the parameters of the antecedent and the consequent parts of the rules for a Mamdani model. Simulated annealing is combined with a method to reduce the computations associated with it using an adaptive step size. The results of the proposed methods are compared to results of a type-1 fuzzy system. Index Terms—Type-2-Fuzzy-Systems, Simulated-Annealing, Time-Series-Forecasting.