【作者】 黄显明；

【导师】 易继锴；

【作者基本信息】 北京工业大学 ， 控制理论与控制工程， 2001， 硕士

【摘要】 随着科学技术的发展，现代工业生产过程的一个共同特征是控制系统的复杂性和不确定性日趋明显，即各子系统之间或其内部会有较强的关联性，参数的高维性、时变性和随机性，且系统和环境具有许多未知的和不确定的因素，这些因素还会随环境、工况和时间等发生不可预料的变化。因此已不可能采用那些基于定量数学模型的传统控制方法对其实现有效的控制，必须寻求新一类的控制策略。 模糊控制是一种不依赖于被控过程数学模型的仿人思维的控制技术。它可以利用领域专家的操作经验或知识建立被控系统的模糊规则，有较好的知识表达能力。但在工程实际应用中却缺乏自学习或自调整的能力。尽管神经网络是一类黑箱式的非线性映射，但它具有良好的自学习能力。将二者有机结合起来，取长补短，可以提高整个系统的学习能力和表达能力。目前这个方向的研究正方兴未艾。 本文首先对模糊控制、神经网络及模糊神经网络的发展、背景和原理等进行了综述。针对化纤工艺侧吹风的温、湿度控制系统生产过程的非线性特点，本文提出了一种新型的模糊神经网络控制策略FNNC，其基于联接机制，应用多层前馈网络构造模糊变量隶属函数和模糊推理控制模型，使神经网络不再表现为黑箱式映射，其所有节点和参数都具有模糊系统等价意义。FNNC采用了自组织学习和监督学习相结合的新型学习算法，该算法可根据领域专家的经验知识确定初始隶属函数和发现规则的存在，并可优化调整隶属函数，获得理想输出。本文利用该控制策略进行了仿真研究。结果表明，该控制策略可以使送风状态长期稳定的维持在要求的温、湿度参数上，几乎不会受到外界气象参数变化的影响。 本文的另一项重要工作，是针对工艺性空气调节过程中温、湿度多变量的非线性控制特征，设计开发了一类智能空调控制系统，包括有硬件选型、设计、工艺结构设计、软件设计等，文中给出了相应的设计电路、程序框图、电路板等结构图。系统实验证明该系统达到了设计要求。更多还原

【Abstract】 Along with the development of technology there are complexity and uncertainty of control systems in industry manufacture process. Many uncertain factors, such as correlation, randornicity, will change incidentally when environment and time change. So traditional control technique based on mathematical model is unuseful. New control strategy has being seeking. Fuzzy control is a kind of human imitating technique which is independent on the controlled plant mathematical model. It utilizes the knowledge and experience of experts to carry out rationalization. As a result, it has good robustness. But it is lack of the ability of self-learning or self-tuning after the fuzzy control rules have been set off-line. Neural network has the ability of self-learning in spite of its nonlinear mapping similar with Black-Box. The abilities of self-learning and expression of the whole system will be improved when they hand together. The research of this combination is in the ascendant. First, the backgrounds, improvements and principles of fuzzy control, neural networks and fuzzy neural networks are introduced. Facing the characteristics of multi-variable, nonlinear plant, this thesis established a fuzzy neural network which construct fuzzy subject function and fuzzy rational control model. As a result, the parameters of neural network have equivalent fuzzy meanings. New arithmetic combining self-organization with supervision is applied in FNNC which can detect initial- subject function and optimize rules. On the basis of fuzzy neural network technique, this thesis does research work of simulation on multi-variable side-wind for chemical fiber air conditioning temperature and humidity control system, and gives out the simulation curve of FNN control. The results of simulation show that this strategy is feasible in industry, so it has good perspective of industrial application. The other important work in this thesis is mainly about the design of intelligent air conditioning system. The background, significance, hardware selection, design, and software design are introduced in detail. The results of system experiment show that the whole system satisfy the design request.更多还原