基于支持向量机回归的T-S模糊模型自组织算法及应用

结合模糊聚类算法和支持向量机回归算法提出了一种新的T-S模糊模型自组织算法. 该算法首先利用一种改进模糊聚类算法提取模糊规则和辨识前件参数，然后将T-S模糊模型后件变换为标准线性支持向量机回归模型，并利用支持向量机回归算法辨识后件参数. 仿真结果表明，相比现有的自组织算法，本文提出的T-S模糊模型自组织算法在规则数较少的情况下，仍然具有较高的辨识精度和较好的泛化能力. 最后，利用提出的T-S模糊模型自组织算法较好地建立了直拉硅单晶炉加热器和空气预热器的温度模型.

A Self-organizing Algorithm for T-S Fuzzy Model Based on Support Vector Machine Regression and Its Application

A new self-organizing algorithm for T-S fuzzy model is proposed by combining the fuzzy clustering algorithm and the support vector machine (SVM) regression algorithm. This algorithm firstly uses an improved fuzzy clustering algorithm to extract fuzzy rules and identify antecedent parameters. Then the T-S fuzzy model consequent is transformed into a standard linear support vector machine regression model, thus its parameters are identified using the support vector machine regression algorithm. Simulation results show that the self-organizing algorithm for T-S fuzzy model in this paper still has higher approximation accuracy and better generalization ability in the case of a small number of rules compared with the existing self-organizing algorithm. Finally, a heater temperature model of Czochralski single crystal furnace and an air preheater temperature model are better established using the proposed self-organizing algorithm for T-S fuzzy model.