模糊神经网络在磨机负荷控制中的应用

Application of Fuzzy Neural Network in Mill Load Control

磨矿分级作业是选矿生产中的关键环节,其中的磨机负荷控制由于存在大惯性滞后、参数耦合性强和时变性等问题,所以难以建立精确的数学模型,因此常规PID控制很难达到预期的控制效果。针对上述情况,将模糊控制与人工神经网络控制相结合,既发挥了模糊控制鲁棒性强的优点,又可以通过数值运算的形式实现对结构性语言经验的综合推理,正向并联辨识的加入,极大地增加了磨机运行的稳定性。通过对现场运行情况的监控,表明该控制方法可以有效地消除运行过程中外部干扰带来的扰动。

Grinding-classification process is a key link in mineral processing production, the mill load are with large inertia lag and parameters strongly coupling and time-varying problems, it is difficult to establish accurate mathematical model, so the conventional PID control is difficult to achieve the desired control effect. According to the above situation, combining fuzzy control with artificial neural network control ,both play to the advantages of fuzzy control is robust, and through numerical computation of structural language experience, in the form of synthesis reasoning, positive identification in parallel, and greatly increases the stability of mill running processing. Through monitoring on the performance of the spot, which indicates that this control method can effectively avoid the disturbance caused by external disturbance during the operation.