SIA OpenIR  > 工业控制网络与系统研究室
氧气底吹铜熔炼过程关键参数软测量及智能控制研究
其他题名Research on Soft Sensor and Intelligent Control for Key Parametres in Bottom Blown Oxygen Copper Smelting Process
王斌1,2
导师于海斌
分类号TF811.04
关键词氧气底吹铜熔炼过程 软测量 智能控制
索取号TF811.04/W31/2017
页数135页
学位专业机械电子工程
学位名称博士
2017-06-02
学位授予单位中国科学院沈阳自动化研究所
学位授予地点沈阳
作者部门工业控制网络与系统研究室
摘要本文对氧气底吹铜熔炼过程机理进行了分析,然后在总结采用不同工艺的铜熔炼过程建模和优化控制的研究现状基础上,研究了关键参数软测量和智能控制策略,并开发了智能控制系统。该系统已成功应用于底吹炉的实际生产,取得了明显的成效。主要研究成果如下: 1、在深入分析氧气底吹铜熔炼过程工艺机理的前提下,对熔炼过程进行合理简化,采用“三传一反”基本原理并结合底吹喷射过程传质方面的研究成果,根据质量平衡、能量平衡、结合冶金反应动力学和流体动力学,从理论上建立了氧气底吹铜熔炼的动态机理数学模型,并应用现场运行数据验证了模型的有效性,从而使得模型可以用于五个关键参数的软测量。 2、提出了一种基于PAPE (PLS-Adaboost-PLS-Ensemble)的熔体温度集成软测量建模方法。首先,对传统的单一软测量模型进行改进,采用Adaboost算法生成多个有差异的子模型,子模型采用PLS方法建立;接着,对Adaboost算法进行了改进,采用PLS方法合成各个子模型,进而得到熔体温度的在线预测结果。仿真结果表明,与传统方法相比,熔体温度集成软测量模型具有更好的预测精度。 3、提出了氧气底吹铜熔炼过程的智能控制方法。该方法由石英石流量、冷料流量和氧气流量的预设定模块和反馈补偿模块组成。预设定模块根据铜锍品位、炉渣铁硅比和熔体温度三个关键参数的目标值,以及给混合铜精矿成分、混合铜精矿配比流量、冷料成分等边界条件,采用案例推理方法,给出氧气底吹铜熔炼过程石英石流量、冷料流量和氧气流量的预设定值。反馈补偿模块根据铜锍品位、炉渣铁硅比和熔体温度的化验值或软测量值与目标值之间的偏差,通过规则推理,产生石英石流量、冷料流量和氧气流量的补偿值,对预设定值进行校正,从而产生石英石流量、冷料流量和氧气流量的设定值。同时,根据总体液位和铜锍液位的目标值与软测量值之间的偏差,发出排放炉渣和铜锍的操作建议。从而保证氧气底吹铜熔炼过程的五个关键参数的实际值基本位于目标值范围内。 4、设计开发了上述以软测量和智能控制方法为核心的氧气底吹铜熔炼过程智能控制系统,并应用于底吹炉的实际生产过程。智能控制系统由基础控制系统、信息集成平台和先进控制软件组成。应用结果表明,智能控制系统实现了将五个关键参数基本控制在目标范围内,减小了它们的波动,改善了经济技术指标。
其他摘要The mechanism of bottom blown oxygen copper smelting process is analyzed. On the basis of summarizing the research status of modeling and optimizing control of copper smelting process using different technologies, the key parameters of soft sensing and intelligent control strategy are studied, and intelligent control system is developed. The system has been successfully applied to the actual production of the bottom blowing furnace and has achieved remarkable results. The main research results are as follows: 1. Based on the deep analysis of the mechanism of oxygen bottom blowing copper smelting process, the smelting process is simplified reasonably. The basic principle of "three transfers and one reaction" is adopted, and the research results of mass transfer in bottom blowing injection process are combined, a mathematical model of oxygen bottom blowing copper smelting is established theoretically based on mass balance, energy balance, metallurgical reaction kinetics and fluid dynamics. The effectiveness of the model is validated by using field operational data, so that the model can be used for soft measurement of five key parameters. 2. A melt temperature ensemble soft sensor based on PAPE (PLS-Adaboost-PLS-Ensemble) is proposed and applied in the practical training and validation. First, the traditional single soft sensing model is improved, and the Adaboost algorithm is used to generate several different sub models established by PLS. Then, Adaboost is improved by using PLS to synthesize the sub models, thus the online prediction result of the melt temperature is obtained. The simulation results show that the melt temperature ensemble soft sensor model has better prediction accuracy compared with the traditional methods. 3. An intelligent control method for oxygen bottom blowing copper smelting process is proposed. The method is composed of quartz flow, cold charge flow and oxygen flow pre-setting module and feedback compensation module. The preset module according to the target values of matte grade, slag Fe/SiO2 and melt temperature, the boundary condition such as composition of mixed copper concentrate, the flow rate of mixed copper concentrate, and the composition of cold material, gives the pre-setting values of the quartz flow, cold charge flow, the volume of cooling material and the oxygen flow based on CBR(Case-based Reasoning). The feedback compensation module produces quartz flow, cold charge flow and oxygen flow compensation value according to the deviation between test value or soft measurement and target value of copper matte grade, slag Fe/SiO2 and melt temperature based on RBR(Rule-based Reasoning). The compensation value corrects the preset values so as to produce a set value of the quartz flow, the cold charge flow and oxygen flow. At the same time, according to the deviation between the target scope and the soft measurement of the total level and the copper matte level, the operation suggestion of the slag and matte charging is given. Thus, the actual value of the five key parameters of the oxygen bottom blowing copper smelting process is basically within the target value range. 4. The intelligent control system of oxygen bottom blowing copper smelting process based on soft measurement and intelligent control method is designed and applied to the actual production process of bottom blowing furnace. The intelligent control system is composed of basic control system, information integration platform and advanced control software. The application results show that the intelligent control system can control the five key parameters nearly in the target range, reduce their fluctuation, and improve the economic and technical indexes.
语种中文
产权排序1
文献类型学位论文
条目标识符http://ir.sia.cn/handle/173321/20570
专题工业控制网络与系统研究室
作者单位1.中国科学院沈阳自动化研究所
2.中国科学院大学
推荐引用方式
GB/T 7714
王斌. 氧气底吹铜熔炼过程关键参数软测量及智能控制研究[D]. 沈阳. 中国科学院沈阳自动化研究所,2017.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
氧气底吹铜熔炼过程关键参数软测量及智能控(6145KB)学位论文 开放获取CC BY-NC-SA请求全文
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[王斌]的文章
百度学术
百度学术中相似的文章
[王斌]的文章
必应学术
必应学术中相似的文章
[王斌]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。