基于温度定性趋势分析的锌冶炼焙烧炉实时状态评价  

Real-time state evaluation of zinc smelting roaster based on qualitative trend analysis of temperature

作　　者：陈宇 金文兵 谢晓能[1] 邱玮杰 蒋梦轩 CHEN Yu;JIN Wenbing;XIE Xiaoneng;QIU Weijie;JIANG Mengxuan(School of Internet of Things,Hangzhou Polytechnic,Hangzhou 311402,China;School of Automation,Central South University,Changsha 410083,China)

机构地区：[1]杭州科技职业技术学院物联网技术学院,浙江杭州311402 [2]中南大学自动化学院,湖南长沙410083

出　　处：《冶金自动化》2022年第4期86-94,共9页Metallurgical Industry Automation

基　　金：浙江省教育厅科技项目(Y202248621,Y202145960)。

摘　　要：在锌冶炼焙烧过程中,准确识别焙烧炉当前状态对焙烧氛围及产物质量至关重要。针对传统的基于温度阈值对焙烧炉进行状态划分不准确的问题,建立了一种基于温度定性趋势分析(qualitative trend analysis,QTA)的焙烧状态评价模型。首先,提出了改进的QTA方法对温度参数进行趋势提取,得到定性的温度趋势信息。然后,将当前温度与趋势信息结合,通过分类决策树得到当前焙烧炉状态。最后,通过仿真和实际运行进行验证。结果表明,1号炉的沸腾层标温的标准差相对原系统降低了57.9%,2号炉的标温平均值误差相对原系统降低了64.5%。该方法可以有效地评价焙烧炉当前状态,并能提前预警异常状态,对于焙烧过程稳定控制和提高焙烧产品质量具有重要意义。In the roasting process of zinc smelting,it is very important to accurately identify the current state of roasting furnace for roasting atmosphere and product quality.Aiming at the problem that the traditional state division of roasting furnace based on temperature threshold is inaccurate,a roasting state evaluation model based on qualitative trend analysis(QTA) was established.Firstly,an improved QTA method was proposed to extract the trend of temperature parameters and obtain the qualitative temperature trend information.Then,combining the current temperature with the trend information,the current baking furnace state was obtained through the classification decision tree.Finally,the simulation research and actual operation results show that the standard deviation of boiling layer standard temperature of No.1 boiler is reduced by 57.9%,and the average error of standard temperature of No.2 boiler is reduced by 64.5% compared with the original system.This method can effectively evaluate the current state of the roasting furnace and warn the abnormal state in advance.It is of great significance for the stable control of the roasting process and improving the quality of roasted products.

关 键 词：锌冶炼 焙烧过程 定性趋势分析 分类决策树 状态评价 

分 类 号：TF813[冶金工程—有色金属冶金]