Nonlinear System Identification Based on Adaptive Cloud C-means Clustering Algorithm

System modeling directly based on uncertain data is a hot research in recent years. T-S fuzzy model has been widely used in nonlinear system identification, and cloud model has a strong ability to deal with uncertainty. Combining the advantages of T-S model and cloud model, a T-S cloud model was proposed to identify nonlinear system among uncertain data. Traditional cloud c-means clustering algorithm has a shortcoming what was sensitive to initial cluster centers and the number of clusters. Then an adaptive strategy based on the density of sample data was used to solve this problem. The T-S cloud model adopted a new structure, whose premise structure was identified by the adaptive cloud c-means clustering algorithm and conclusion parameters were identified by least squares algorithm. The validity and availability of the proposed method are supported by simulation results.