基于振动信号测量的连铸下渣检测系统  

Continuous casting slag carry-over detection system based on vibration signal measurement

作　　者：谭大鹏[1] 李培玉[1] 潘晓弘[1] 

机构地区：[1]浙江大学机械与能源工程学院,浙江杭州310027

出　　处：《浙江大学学报（工学版）》2008年第8期1399-1403,共5页Journal of Zhejiang University：Engineering Science

基　　金：国家自然科学基金资助项目(50374061);浙江大学争创"全国百篇优秀博士论文"计划资助项目(07001A)

摘　　要：为了解决当前连铸下渣检测系统成本高、使用寿命短且难于安装维护等问题,提出了一种基于振动信号测量的下渣检测系统实现方法.通过分析连续下渣过程,根据钢水、钢渣因相对密度不同而产生的冲击振动差异判断下渣时间点.结合嵌入式系统技术,搭建了传感器远离钢水的下渣检测实验平台,有效地提高了系统使用寿命.利用自主设计的嵌入式实时数据采集系统进行数据采样与特征信号提取,通过人工神经网络(ANN)技术对经过预处理的实时信号进行训练与识别,进而判断钢水状态,并结合相关控制策略实现连铸下渣的自动控制.实验结果表明,该方法成本低,对现在有设备改造小,使用寿命长,下渣检出率在96%以上.A slag detection system realization method based on vibration signal measurement was put forward to resolve the problems of continuous casting slag carry-over detection system, such as high cost, short service life, and inconvenient installation and maintenance. Based on analysis of the physical process of continuous casting, the time point of slag carry-over was found out according to the shock vibration difference generated by molten steel and slag because of their relative density difference. Combined with the embedded system technology, an automatic control experiment platform with vibration sensor was established far away from steel water, which could effectively increase the system service life. By the selfdesigned real-time data collection system, vibration signal sampling and characteristic data extraction were finished, and trained to detect the slag by artificial neural network （ANN）. Then the status of molten steel was identified to realize the automatic control for continuous casting according to relevant control strategy. Experimental results proved that this method requires low cost and little re-building for current devices, and its slag detection ratio can reach above 96%.

关 键 词：连铸 下渣检测 振动 人工神经网络 

分 类 号：TB123[理学—工程力学]