【作者】 刘天键；

【导师】 王劭伯；

【作者基本信息】 福州大学 ， 控制理论与控制工程， 2001， 硕士

【摘要】 本文从工程实际出发，对模糊神经非模型控制系统设计理论和应用进行了研究。 文中着重研究了模糊系统与神经元非模型控制融合的可行性及融合方式，基于二者的 融合方式提出了自适应模糊神经元非模型控制方法，并与PID控制、模糊控制、单神 经元控制进行了仿真比较。最后，在电加热炉上进行实例设计。本文的主要研究的内 容有如下几个方面： 1．综述了自适应控制、模糊控制、神经元控制的发展和研究现状，并就模糊神经非 模型控制的研究提出了一些作者的观点。 2．介绍了神经元非模型控制的结构和基本原理及模糊控制器的设计的两种基本方 法。针对模糊控制系统和神经元控制系统的不足，探讨了神经元非模型控制与模 糊控制结合的可能性。 3．针对模糊系统和神经元系统的特点，提出了自适应模糊单神经元的控制算法。 4．介绍了基于数值仿真的方法，在MATLAB中对本文提出的自适应模糊单神经元 控制算法进行仿真实验，并与PID控制、模糊控制、单神经元控制进行仿真比较。 验证自适应模糊单神经元控制算法的动、静态特性、鲁棒性和抗干扰性。结果表 明本文提出的自适应模糊单神经元非模型控制方法不仅获得强鲁棒性、强抗干扰 性和满意的控制性能，而且系统设计方法简便，能方便地用于实际工业控制中。 5．针对实验室电加热炉系统，使用自适应模糊单神经元控制算法进行控制系统的设 计，得出阶跃响应曲线，取得了满意的控制效果。更多还原

【Abstract】 From the points of view for practical uses, the theory and applications of fuzzy neuronmodel-free control systems are researched in this dissertation. The feasibility and forms ofcombining a fuzzy system and a neuron model-free control system are discussed. The theoryof adaptive fuzzy-neuron model-free control system based on the combination of fuzzy andneuron system is put forward. Simulation experiments are made by comparison with the PIDcontrol, fuzzy control and neuron control. Finally, the examples are made on the electricalheater.The main research work and contributions of this dissertation are as follows:1. A survey of adaptive control, fuzzy control and neuron control is summarized, and theproblems existing in neuron model-free control systems and its integrating with fuzzysystems are also discussed.2. An introduction of the basic principles of the neuron model-free control and the basicdesign methods of the fuzzy controller is given. The iniegrating forms between a fuzzysystem and a neuron model-free control system are investigated and presented fordesigning the fuzzy neuron model-free controllers.3. The adapive fuzzy neuron model-free methods are proposed by the characteristics offuzzy system and neuron system4. The methods of simulation based on numeric integral are introduced. The simulationexperiments are made to the arithmetic of adaptive fuzzy neuron cotrol by MATLAB.Comparisons are made with PID control, fuzzy control and neuron control. And theperformances, the robustness, disturbance-rejection are validated. The simulation testresults with examples show their nice performances, strong robustness and disturbance-rejection. They are designed simply and very convenient to use in practice.5.The control systems of the heater are designed by the adaptive fuzzy neuron controlarithmetic. The step response curve is got. The results show their nice performances.更多还原