【作者】 侯录；

【导师】 秦树凯；

【作者基本信息】 东北大学 ， 控制理论与控制工程， 2008， 硕士

【摘要】 隧道窑作为耐火材料制品生产的重要设备,其温度控制水平是制约耐火材料质量提高的一个重要因素。因此温度控制策略是隧道窑控制系统中的重要研究内容。如何设计满足生产要求的隧道窑控制系统及制定先进的温度控制策略来提高产品质量、降低能耗、减少环境污染,成为工业窑炉控制领域一个迫切需要解决的问题。本文以实际项目为背景,根据隧道窑生产过程的要求,在详细分析隧道窑工艺特点的基础上,设计出一套集回路控制、过程监控和管理信息系统于一体的隧道窑控制系统。该系统采用西门子PLC实现系统的回路控制,采用WinCC组态软件实现系统的监控管理功能。为了研究先进的隧道窑温度控制策略,首先必须建立隧道窑温度模型。由于隧道窑工况复杂,难以用精确的数学模型来描述。本文以现场数据为基础,采用BP神经网络对窑温模型进行辨识。仿真结果表明基于BP神经网络的隧道窑温度模型辨识效果良好。隧道窑是一个复杂的控制对象,具有大滞后、时变及非线性的特性,这些特性对于隧道窑温度的控制都是十分不利的。采用传统PID控制策略来控制隧道窑的温度,由于参数不易整定、抗干扰能力差、自适应性差、严重依赖操作人员的现场经验等一系列的缺点,达不到理想的控制效果。本文以BP神经网络辨识的窑温模型为被控对象,根据隧道窑温度控制的要求,设计一种基于模糊神经网络的温度控制器。该控制器具有清晰的网络结构,它利用神经网络的学习能力在线调整网络的参数,从而增强模糊控制的在线学习能力。通过MATLAB仿真实验可以看出：与基于PID的窑温控制系统相比,基于模糊神经网络的窑温控制系统对窑温的变化具有更好的跟随性,调整时间较短,而且具有较强的抗扰性和鲁棒性,能够满足隧道窑温度控制的要求。更多还原

【Abstract】 The tunnel kiln is an important equipment which is used for burning refractory materials, and the control level of the temperature has the direct effect on the quality of products. So, how to design control system and good temperature control strategy for tunnel kiln which can satisfy the produce requirements, enhance the quality of the product, reduce the energy consumption and reduce the pollution, becomes a question which is need to be solved imminency.The paper takes the practical project as the background, at first, according to the requirement of the produce process of tunnel kiln and detailed analysis of the demand and the characteristic of tunnel kiln, we design control system for tunnel kiln which contain loop control, process supervising and management information system. We use the PLC of Simense to take the loop control of the system and use the configuration software of WinCC to carry out the supervisory function of the system.We need to establish the tunnel kiln temperature model before researching the advanced control strategy. Because the tunnel kiln is a complicated object which is hard to establish an exact mathematic model, so this paper identifies the dynamic model of the tunnel kiln temperature object through BP neural network based on the data from field. After making the matlab simulative experiment, we get the conclusion that the effect of the tunnel kiln temperature model identified by BP neural network is good.The tunnel kiln is a complicated object which has some characteristics, such as, hysteretic property, time-varying and non-linear, these charactertisics causing as to its temperature control more difficult. The conventional PID control technique can’t get the perfect control effect because the three parameters of PID can’t be set easily, the anti-interference and adapting ability of PID are week, and it depends on the operators’ experience. In this paper, we take the tunnel kiln temperature model identified by neural network as an object to be controlled. According to the requirement of the temperature control, we design a kind of new controller based on fuzzy neural network. The controller has a distinct network structure. The learning ability of the neural network is used to revise the parameters of neural to enhance the learning ability of the fuzzy control. At last, making the matlab simulative experiment, we can make the conclusion that kiln temperature control system based on fuzzy neural network has better performance than the kiln temperature control system based on PID in the following aspects:better following performance for temperature varying, less adjusting time, greater anti-interference ability and greater robustness, so, the fuzzy neural network controller can meet the requirements of tunnel kiln temperature control.更多还原