带钢轧制过程产品质量优化控制研究  

Research on Optimal Control of Strip Steel Qualityin Rolling Strip Steel

作　　者：徐盛 刘毅敏[1] XU Sheng;LIU Yi-min(College of Information Science and Engineering,Wuhan University of Science and Technology,Wuhan Hubei 430081,China)

机构地区：[1]武汉科技大学信息科学与工程学院,湖北武汉430081

出　　处：《计算机仿真》2018年第9期349-352,357,共5页Computer Simulation

摘　　要：在带钢轧制过程中,带钢张力一旦发生较大的波动会影响带钢产品的质量,对轧制线带钢张力故障进行分类并进行及时有效的处理,可以提高带钢连轧生产的稳定性和可靠性。首先提出针对带钢张力信号特点而采用的局部均值分解方法的原理,并给出了支持向量机的设计方法。其次结合现场生产过程中故障发生较少的情况,对从现场采集的带钢张力信号采用LMD-SVM方法进行仿真研究,验证了LMD-SVM方法能有效识别带钢生产中的故障。通过对比在不同训练样本个数下的仿真结果,验证了LMD-SVM方法在小样本下仍然能对故障具有较高的辨识率,可以为带钢质量优化控制与故障处理提供一定的参考。In the strip steel rolling process, if the tension of strip steel fluctuates greatly, the quality of strip prod- ucts will be affected. Thus, the classification, timely and effective treatment of tension faults in rolling line can im-prove the stability and reliability of strip continuous roiling. Firstly, for the characteristics of strip tension signal, a fault classification method based on Local Mean Decomposition （LMD） was proposed. Subsequently, the design method of support vector machines （SVM） was given. According to the actual situation, the analysis results from the actual strip tension signal which were simulated to demonstrate that the LMD method can be applied to the fault classi- fication of strip rolling effectively. Furthermore, compared with the training simples in different numbers, the simula- tion results indicate that the LMD method still has the high recognition rate with small sample size. Therefore, the LMD-SVM method can provide some references for quality optimization control and fault handling of strip steel.

关 键 词：带钢张力 局部均值分解 支持向量机 故障分类 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术] TE345[自动化与计算机技术—计算机科学与技术]