论文部分内容阅读
针对多输入多输出(MIMO)热工过程的非线性、强耦合、变工况及参数时变等特点,提出了一种基于系统输入输出数据和模糊自适应竞争聚类的模型辨识新方法.该方法首先依据系统的各个典型运行工况,使用模糊自适应竞争聚类对输入输出数据进行聚类划分,并对T-S模糊模型进行结构辨识,以确定系统的模型结构和参数;然后采用最小二乘递推算法对模型后件参数进行辨识,同时对结构辨识参数进行精确修正.将所提出的模型辨识方法用于锅炉-汽轮机非线性系统的模型辨识,仿真结果验证了该方法的有效性.
Aiming at the nonlinearity, strong coupling, variable working conditions and time-varying parameters of a multiple-input multiple-output (MIMO) thermal process, a new model identification method based on system input and output data and fuzzy adaptive competition clustering is proposed. According to the typical operating conditions of the system, the method uses fuzzy adaptive competition clustering to classify the input and output data, and identifies the TS fuzzy model to determine the model structure and parameters of the system. Then, The recursive algorithm is used to identify the parameters of the model afterwards, and the structure identification parameters are corrected accurately.The proposed model identification method is applied to the model identification of the boiler-turbine nonlinear system, and the simulation results verify the effectiveness of the proposed method.