基于云推理的叶丝干燥筒壁温度预测方法

Method of predicting cylinder wall temperature of cut strip drier based on cloud reasoning

为解决滚筒式烘丝机筒壁温度设定不准确而影响出口含水率等问题,基于云模型提出了一种叶丝干燥工序筒壁温度预测方法。通过对叶丝干燥关键工艺指标数据间的关系进行分析,证明脱水量和排潮罩负压是影响筒壁温度的关键因素。采用聚类分析法对筒壁温度进行聚类,得到云模型定性推理规则,并构建了筒壁温度的二维多规则预测模型。利用选取的在线监测样本数据,对筒壁温度预测模型进行了验证。结果表明,按照预测算法对叶丝干燥样本数据的筒壁温度进行计算,预测误差低于0.7%,预测精度符合要求。所建立的云预测模型可精确设置烘丝机的脱水量和筒壁温度,有效提高叶丝干燥各项工艺参数的稳定性。

The setting of cylinder wall temperature significantly influences the moisture content in output cut strip,therefore a method for predicting cylinder wall temperature was proposed based on cloud model. The results of analyzing the relationships between key parameters in cut strip drying demonstrated that,the dehydration amount and negative pressure in moisture exhaust hood were the key factors influencing cylinder wall temperature. By clustering the cylinder wall temperatures with cluster analysis,a qualitative reasoning rule of cloud model was obtained,and a two-dimensional multi-rule prediction model for cylinder wall temperature was established. The model was validated with the selected online monitoring sample data. The results showed that,the prediction error of the model was less than 0.7%,which satisfied the requirement for prediction precision. The established model could set the dehydration amount and cylinder wall temperature of cut strip drier precisely,it effectively improves the stability of technical parameters in cut strip drying.