【作者】 宋勇； 荆丰伟； 孙树萌；

【Author】 Song Yong;Jing Fengwei;Sun Shumeng;Engineering Technology Research Institute of University of Science and Technology of Beijing;

【机构】 北京科技大学工程技术研究院；

【摘要】 在热连轧的产品质量过程控制中,轧件温度是一个至关重要的参数,其测量值的准确性尤为重要。在实际生产中,轧件温度测量一般采用的是非接触式的红外高温计,轧制过程的恶劣测量环境有可能对其测量值造成干扰。目前主要采用硬件冗余方式来保证轧件温度测量值可靠性,但只能解决温度测量值异常的问题,而无法有效应对现场环境偶然因素对测量值精度所造成的影响。本文利用聚类分析得到历史轧件生产的典型工况和温度测量数据,通过对当前轧件工况与历史典型工况的匹配分析,根据大量历史工况的测量数据分布规律,采用数据融合技术来排除外界干扰因素对温度测量精度的影响。通过测试数据对所构建的融合模型及融合系数确定方法进行了试验,分析了该方法的有效性。更多还原

【Abstract】 In the process of product quality control of hot rolling, the rolling temperature is a critical parameter, and the accuracy of the measured value is particularly important. In actual production, the non-contact infrared pyrometer is usually used in the strip temperature measurement. Therefore, rolling process of the poor measurement environment may cause interference to its measured value. At present, hardware redundancy is used to ensure the reliability of the temperature measurement of the rolling parts, but only the problem of abnormal temperature measurement can only be solved, and the influence of the accidental factors on the accuracy of the measurement cannot be effectively dealt with. In this paper, the typical working conditions and temperature measurement data of historical rolling production are obtained by cluster analysis. Through the matching analysis of the current rolling conditions and typical working conditions, according to the distribution of the measured data in a large number of historical conditions, to eliminate the external interference factors on the temperature measurement accuracy. The fusion model and the fusion coefficient method are tested by the test data, and the effectiveness of the method is analyzed.更多还原