【摘要】 提出了一种改进的T-S模糊模型,固定了前件输入变量的形式以及后件线性系统的初始结构,使前后件中的输入变量相互独立。在此基础上利用非线性系统的输入输出数据进行T-S模糊建模的步骤为首先寻找最佳聚类数,然后利用模糊聚类划分输入空间,最后用RLS辨识后件参数并依据结果消去无关后件输入项。在SISO和MIMO非线性系统上的仿真实验尤其是阶跃响应测试结果验证了算法的有效性。更多还原

【Abstract】 An improved T-S fuzzy model is proposed, which fixes the form of premise input variables and the initial structure of the consequent linear system, and eventually makes the premise input and consequence input independent of each other. On the basis of the previous work, the steps of T-S fuzzy modeling using the input-output data of the nonlinear system are as follows: Firstly, the appropriate number of clusters is fixed. Then the fuzzy clustering is applied to partition the input space. Finally, the recursive least square algorithm is employed to identify the parameter of the consequent part, and the unrelated input is eliminated according to the result. The simulation experiment on the SISO and MIMO nonlinear system, especially the result of step response test illustrates the effectiveness of the proposed method.更多还原