基于T-S 模型的模糊预测控制研究

提出一种基于T—S模型的模糊预测控制策略．利用模糊聚类算法高线辨识T—S模型，采用带遗忘因子的递推最小二乘法进行模型参数的选择性在线学习；对模糊模型在每一采样点进行线性化，将T—S模型表示的非线性系统转化为线性时变状态空间模型，并将约束非线性优化问题转化为线性二次规划问题，解决了非线性预测控制中如何获得非线性模型和非线性优化在线求解的难题．将预测域内的线性模型序列作为预测模型，减小了模型误差，提高了控制性能．pH中和过程的仿真验证了该方法的有效性．

Fuzzy predictive control based on T-S model

A fuzzy model based predictive control of nonlinear system is presented. T-S fuzzy model is identified by fuzzy clustering algorithm, and its parameters are self-learning online by selective recursive least square method. (T-S) model is linearized to be time-varying system, and thus nonlinear optimization problem is turned to a quadratic programming problem. Consequently, two major difficulties in nonlinear predictive control to obtain accurate nonlinear model and to solve nonlinear optimization problem online are solved. The simulation result on pH neutralization process shows the effectiveness of the proposed method.