Abstract
Abstract For unknown model of Continuous Stirred Tank Reactor (CSTR), a model identification based on the combination of principal component analysis (PCA) and a type-2 non-singular interval type-2 fuzzy logic system (Type-2 NSFLS-2) is proposed method. The PCA method can reduce the dimensionality of the high-dimensional input of the model. On this basis, type-2 NSFLS-2 identification model is established, and the backpropagation (BP) algorithm is used to update tthe parameters of antecedent and consequent membership function respectively. The method in this paper is applied to the example of CSTR, and it is compared with other types of FLS method and support vector machine (SVM) under the same conditions. The experimental results show that different types of interval type-2 FLS methods can achieve higher identification accuracy, among which Type-2 NSFLS-2 has higher identification accuracy.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call