论文部分内容阅读
通过对冷板带轧机垂直振动过程的机理进行分析,结合轧机系统结构模型,建立含振动因素的冷轧机垂向系统动态轧制力模型.考虑复杂工况下,轧机在生产不同规格带钢时,由工艺参数波动等广义故障所致轧机垂直振动现象,基于工业现场数据进行数据驱动的故障诊断算法研究.采用集成经验模态分解算法对实测轧制力信号进行分解,选取有效的固有模态函数的能量作为特征向量,并将其输入到支持向量机分类器中,通过分类器对正常状态和故障状态进行区分,以实现轧机振动相关故障的准确诊断.
Based on the analysis of the mechanism of the vertical vibration process of the cold strip mill and the rolling mill structural model, the dynamic rolling force model of the vertical system of the cold rolling mill with vibration factors is established. Considering the complicated working conditions, , The data-driven fault diagnosis algorithm based on industrial field data is studied by the phenomenon of rolling mill vertical vibration caused by the generalized faults such as the fluctuation of process parameters, etc. The integrated empirical mode decomposition algorithm is used to decompose the measured rolling force signal, and the effective inherent mode The energy of the state function is used as the eigenvector and input into the SVM classifier. The normal state and the fault state are distinguished by the classifier so as to realize the accurate diagnosis of mill vibration related fault.