广义预测控制算法及应用研究

【作者】 周荔丹；

【导师】 童调生；

【作者基本信息】 湖南大学 ， 控制理论与控制工程， 2002， 硕士

【摘要】 由于控制实践中的许多复杂的工业系统的数学模型很难精确建立，而且工业对象的结构和参数往往具有一定的不确定性，因此，应用经典控制理论和现代控制理论很难达到期望的控制指标。广义预测控制是有代表性的预测控制算法之一，并被广泛应用于工业过程中。本文在研究了广义预测控制原理、模糊神经网络和径向基函数神经网络理论的基础上，针对广义预测控制在线计算量大，不适用于非线性对象的特点，提出了基于模糊径向基函数神经网络模型的模糊神经网络广义预测控制，并对线性和非线性被控对象进行了仿真，仿真证明所提出的控制方法有效。此外，本文对径向基函数神经网络的函数逼近理论以及基于径向基函数神经网络的自适应控制器的稳定性作了较深入的分析。更多还原

【Abstract】 Since the mathematical models of a lot of involved industrial systems in the control practices are difficult to be established, and that the structures’and parameters of industrial plants always have some uncertainties, it is hard to achieve the anticipated control indexes by using control theory and modern control theory. Generalized predictive control is one of the typical predictive control methods, and it is widely used in industry. Based on the research of the principle of generalized predictive control, the theory of fuzzy neural networks and radial basis function neural networks, and according to the complicated on-line calculation of generalized predictive control and that it is not suitable for nonlinear system, this paper proposes fuzzy neural networks generalized predictive control using a fuzzy radial basis function neural networks model. By simulating to the linear and the nonlinear system, the results certify that the proposed control method is in effect. In addition, the paper makes relatively in-depth analyses on the function approximation theory of radial basis function neural networks and the stability of the adaptive controller based on radial basis function neural networks.更多还原