The integration and application of fuzzy and grey modeling methods

An integrated fuzzy and grey model and its applications to the prediction control problems are presented. The basic grey model GM(1, 1) is accompanied with the adaptive fuzzy method to improve its prediction capability. The gradient descent scheme is applied to the fuzzy rules to determine whether the predicted results from the grey model should be adjusted, The quantity of adjustment is judged from the degrees of the correlation between the past data and the current input. We select a few most correlated patterns to decide the direction of adjustment. Due to the simplicity of the structure and its fast learning characteristics, this model is good as a real-time controller. Under the proposed methodology, the simulation results are shown to be superior to those systems which exploit complicated control variables and rules, A well-known difference equation and the weather forecast prediction problems are depicted to verify the superiority of the proposed method.