Uncertainty Modeling Design with a Probabilistic Fuzzy Neural Network

In this paper, a probabilistic fuzzy neural network (PFNN) is proposed to handle dynamic uncertainties. The probabilistic fuzzy logic system (PFLS) is capable to process the stochastic and fuzzy information together. When the PFLS and neural networks are integrated in a unified framework, the PFNN can adaptively capture and model the probabilistic uncertainties from the measured variables to improve its modeling capability. Finally, the simulation result shows the proposed PFNN is effective for uncertainty modeling.