智能控制技术在黄磷生产中的电炉上的应用

APPLICATION OF INTELLIGENT CONTROL IN PHOSPHORUS FURNACES

黄磷生产所用的电炉具有大时滞、慢时变、非线性、强干扰等分布参数系统特性 ,而且无法建立精确的数学模型 ,单纯采用严重依赖精确数学模型的常规PID控制难以取得令人满意的控制效果。针对黄磷生产的实际情况 ,提出将模糊逻辑、人工神经元网络以及专家智能控制等三种智能控制方法相结合形成一种新型的智能控制系统。该系统采用分层递阶结构并通过计算机集散系统加以实现 ,其控制优点是不依赖于被控对象———黄磷电炉的严格的数学模型 ,鲁棒性强 ,并且对环境的变化有自适应和自学习能力。仿真实验结果表明 ,该设计方案是合理的和有效的。目前这套智能控制系统已经成功应用于鹏洋化工公司黄磷电炉炉内温度控制 ,控制结果令人满意 。

In this paper,the features of a phosphorus furnace are described.The characteristics of this kind of distributed parameter system such as big delay,time varying behavior,nonlinear,multi disturbance result in that it is hard to develop a practical mathematic model,and the phosphorus furnace is difficult to be controlled by only using PID control depended on precise models.In the light of present situation of phosphorus production,a new type of hierarchically intelligent control system integrating fuzzy neural network control with expert control is presented.Via simulation,a conclusion can be drawn that the system possesses is not only relative strong robustness but also superiority comparing with PID control because of its self learning and self adaptability.The system has been successfully used in operation of a phosphorus furnace of Pengyang Chemical Company.The results of the application showed the effectiveness of the proposed control system.