卷取张力控制过程的故障诊断

在冶金工业生产中，带材的卷取过程会直接影响产品的最终质量.卷取张力是保证卷取过程稳定的重要因素.因此卷取张力控制过程的故障诊断具有现实意义.采用偏最小二乘（Partial Least Squares，PLS）对卷取张力控制过程的故障进行诊断.PLS是一种利用统计原理提取过程数据中的有用信息建立过程模型的降维技术.它不仅可以完成数据降维和特征提取，还考虑了输入、输出数据之间的回归关系.通过PLS分解可以降低空间的维数，使提取的主元具有变化度和可区分性，最后根据T²和Q统计量实现故障检测，通过累计贡献图实现故障识别.仿真结果验证了方法的可行性.

Fault Diagnosis of the Winding Tension Control Process

In procedure of metallurgical production,the strip winding process directly affects the the quality of the final product.Winding tension is one of the important factors for the guarantee of the stability of the coiling process.The winding tension control process is the key to ensure the quality of the product.Therefore,the fault diagnosis of the winding tension control process is of realistic significance.In this paper,a fault diagnosis method based on partial least square (PLS) is used to diagnose the faults in winding tension control process.PLS analysis method is a data dimensionality reduction technology,which extracts useful data information and builds process model by the statistical principle.This technology can not only reduce the dimensions of process data and extract the data feature,but also consider the regression relationship between the input and output variables.By PLS,dimensionality reduction is achieved and principal component is variational and separable.Hence,the fault can be detected by monitoring T² and Q which are calculated as relative principal components and the fault can be recognized according to the accumulation contribution chart. Simulation results have verified the validity of the presented approach.