基于影子趋势对比的矿热炉炉况在线辨识及趋势预测  

Smelting Condition Identification and Prediction for Submerged Arc Furnace Based on Shadow-trend-comparison

作　　者：李沛[1] 阳春华[1] 贺建军[1] 桂卫华[1] LI Pei;YANG Chun-Hua;HE Jian-Jun;GUI Wei-Hua(School of Automation,Central South University,Changsha 410083)

机构地区：[1]中南大学自动化学院,长沙410083

出　　处：《自动化学报》2021年第6期1343-1354,共12页Acta Automatica Sinica

基　　金：国家自然科学基金委基础科学中心项目(61988101);国家自然科学基金国际(地区)合作与交流项目(61860206014)。

摘　　要：矿热炉埋弧冶炼炉况影响因素复杂且偶发迁移和跃变,炉况发展趋势难以把握,给冶炼过程控制带来挑战.对此,本文在深入分析埋弧冶炼机理的基础上,建立了可表征反应区内电弧热与电阻热交互耦合关系的反应区操作电阻模型,实现炉况发展趋势的在线跟踪.当炉况发生迁移或跃变时,利用前序炉况下所得模型生成影子趋势信息,并综合考虑冶炼工艺及电弧电阻与料层电阻的动态特性差异,辨析炉况变化的成因,形成规则化的待辨识参数在线选取方法,解决了炉况变化前后采样点少,传统辨识方法无法适用的问题.工业现场验证表明,所提出方法可在复杂条件下对冶炼炉况进行准确跟踪,并给出可靠的炉况发展趋势预测,为冶炼过程的精细化生产奠定了基础.Submerged arc smelting is an intermittent discharge process with complex influencing factors and the smelting condition would fluctuate in some cases,which makes the control of the smelting process extremely difficult.This paper proposed an operation resistance model that can well characterize the interactive relationship between arc resistance and burden resistance within reacting zones,which can help to track the trend of smelting condition.When smelting condition travels out of the predetermined trajectory,the model can also provide with shadow-trend of the smelting condition by predicting system dynamics under current control outputs,using parameters identified in pre-ordered smelting process.A selection rule of parameters to be re-identified is then proposed based on comparison between shadow-trend and measured data,guided by smelting mechanism.Thus it is practical to predict the trend of smelting condition with few measured data after changing point.Industrial data verification shows,that the proposed method can track and predict the trend of smelting condition well,which could definitely help to achieve the precise control of smelting process.

关 键 词：埋弧冶炼 操作电阻 炉况预测 影子趋势 

分 类 号：TF06[冶金工程—冶金物理化学]