电加热炉温度控制策略与仿真.doc

电加热炉温度控制策略与仿真 [摘要] 电加热炉作为一类具有非线性和大滞后性的工业加热设备，其应用极为广泛，尤其随着社会的飞速发展，煤矿资源的减少以及环境因素的制约，工业电加热炉由于污染小，加上电网的普及，它越来越受到客户青睐，所以研究电加热炉的控制策略显得相当重要。 由于传统的PID控制在时滞系统中的应用有一定局限性，而且快速行与稳定性本身就存在一定矛盾，对于大时滞系统，常规PID控制在参数选取上往往显得非常困难。所以，可以考虑模糊控制方法来对PID控制器的参数进行实时修改，它无须建立被控对象的数学模型，对被控对象的时滞、非线性、时变性具有一定的适应能力，尤其是神经网络能从数据样本中直接自动推导出规则，而不必利用领域知识。 由于模糊控制方法参数不易确定，本文采用BP神经网络调整PID参数，利用神经网络所具有的自学习、自适应、容错性和并行性相结合，进一步完善了PID控制的自适应性能。最后通过仿真结果表明，它能发挥模糊控制鲁棒性能、动态响应好，上升时间快，超调小的特点，又具有PID控制器的动态跟踪品质和稳态精度，取得较好效果。 关键词：PID控制；模糊控制；BP神经网络PID控制；仿真 Electric furnace temperature control strategy and simulation Abstract: Electric furnace as a class of nonlinear and large time delay of industrial heating equipment, their using is extremely broad. Especially with the rapid development of society, the limited reduction of coal resource environmental factors, because of pollution-free, coupled with the popularity of grid, the electric furnace is increasingly popular during the clients, So the research of electric heating furnace control strategy is important. As the traditional PID control have some limitations in time-delay system application. Plus the express bank and the stability exist in itself contradictory, for large time-delay system, Conventional PID control parameters it is often very difficult. Therefore, we can consider fuzzy control approach to the parameters of PID controller for real-time modification, it does not establish a mathematical model of the object, the object of time-delay, nonlinear, time-varying with a certain degree of adaptability , especially the neural network directly from data samples derived from the rules automatically, without having to make use of domain knowledge. As the Fuzzy control of the parameters are uncertain, using BP neural network to adjust to PID parameters in this paper , which is self-learning, adaptive, fault tolerant and parallel combined , to further improve the performance of adaptive PID control. Finally, these simulation results show that it can not only play a rob