EMD-ESN模型在热牵伸辊温控系统中的应用

热牵伸辊是一种被广泛应用在化纤纺织行业的设备。热辊表面温度的稳定性好坏直接影响纺丝的品质，因此本文对如何提高热辊表面温度控制的效果进行了研究。文章首先将经验模态分解（EMD）方法与回声状态网络（ESN）方法进行结合，组成EMD-ESN组合预测模型，对热辊表面温度进行预测。并将该预测方法与模糊PID控制方法结合，将热辊表面温度的预测值与实际值的差值作为模糊PID控制系统的一组输入，并利用模糊PID系统热辊温度进行控制，组成热辊温度预测控制系统。实验结果证明，该方法取得了较好的温度控制效果。

Use of EMD-ESN model in temperature control system of hot roll

Draw Roll is a kind of equipment which has been widely used in chemical textile industry. The temperature stability on the surface of draw roll directly affects the quality of spinning. So the mean point of this paper is how to improve the temperature stability on the surface of draw roll. First, this paper combined empirical mode decomposition(EMD) and echo state network(ESN) as the EMD-ESN combination forecast model. Then predicted the surface temperature of draw roll by using the model. After these, combine EMD-ESN prediction method with fuzzy PID method, use the actual value minus the predicted values and put the difference as one of the input value of fuzzy PID control system. Last, used fuzzy PID system to forecast the draw roll temperature and form the draw roll temperature prediction control system. The experimental result shows that this paper’s method obtains a good effect on temperature control.