2.765

2022影响因子

(CJCR)

  • 中文核心
  • EI
  • 中国科技核心
  • Scopus
  • CSCD
  • 英国科学文摘

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

数据与模型驱动的电熔镁群炉需量预报方法

杨杰 柴天佑 张亚军 吴志伟

杨杰, 柴天佑, 张亚军, 吴志伟. 数据与模型驱动的电熔镁群炉需量预报方法. 自动化学报, 2018, 44(8): 1460-1474. doi: 10.16383/j.aas.2017.c160597
引用本文: 杨杰, 柴天佑, 张亚军, 吴志伟. 数据与模型驱动的电熔镁群炉需量预报方法. 自动化学报, 2018, 44(8): 1460-1474. doi: 10.16383/j.aas.2017.c160597
YANG Jie, CHAI Tian-You, ZHANG Ya-Jun, WU Zhi-Wei. Data and Model Driven Demand Forecasting Method for Fused Magnesium Furnace Group. ACTA AUTOMATICA SINICA, 2018, 44(8): 1460-1474. doi: 10.16383/j.aas.2017.c160597
Citation: YANG Jie, CHAI Tian-You, ZHANG Ya-Jun, WU Zhi-Wei. Data and Model Driven Demand Forecasting Method for Fused Magnesium Furnace Group. ACTA AUTOMATICA SINICA, 2018, 44(8): 1460-1474. doi: 10.16383/j.aas.2017.c160597

数据与模型驱动的电熔镁群炉需量预报方法

doi: 10.16383/j.aas.2017.c160597
基金项目: 

国家自然科学基金 61403071

教育部项目基本科研业务费培育种子基金 N140804001

国家自然科学基金 61503066

中国博士后科学基金 2014M561246

中国博士后科学基金 2015M581355

辽宁省博士启动基金项目 201501151

教育部项目基本科研业务费培育种子基金 N160801001

详细信息
    作者简介:

    柴天佑  中国工程院院士, 东北大学教授, IEEE Fellow, IFAC Fellow, 欧亚科学院院士.1985年获得东北大学博士学位.主要研究方向为自适应控制, 智能解耦控制, 流程工业综合自动化理论、方法与技术.E-mail:tychai@mail.neu.edu.cn

    张亚军  东北大学博士后.主要研究方向为非线性模糊自适应控制理论, 广义预测控制, 多模型切换控制, 智能解耦控制, 数据驱动控制, 智能控制系统的大数据建模, 工业过程大数据建模及其应用.E-mail:zhangyajun79@gmail.com

    吴志伟  东北大学讲师.2015年获得东北大学博士学位.主要研究方向为复杂工业过程的运行控制和工业嵌入式控制系统开发.E-mail:wuzhiwei_2006@163.com

    通讯作者:

    杨杰  东北大学流程工业综合自动化国家重点实验室博士研究生.主要研究方向为工业过程数据驱动建模技术及应用.本文通信作者.E-mail:yjercou@126.com

Data and Model Driven Demand Forecasting Method for Fused Magnesium Furnace Group

Funds: 

National Natural Science Foundation of China 61403071

Fundamental Research Funds for the Central Universities of Ministry of Education N140804001

National Natural Science Foundation of China 61503066

China Postdoctoral Science Foundation 2014M561246

China Postdoctoral Science Foundation 2015M581355

Doctoral Research Foundation of Liaoning Province 201501151

Fundamental Research Funds for the Central Universities of Ministry of Education N160801001

More Information
    Author Bio:

     Academician of Chinese Academy of Engineering, professor at Northeastern University, IEEE Fellow, IFAC Fellow. He received his Ph. D. degree from Northeastern University in 1985. His research interest covers adaptive control, intelligent decoupling control, and integrated automation theory, method and technology of industrial process

     Postdoctoral at Northeastern University. His research interest covers nonlinear fuzzy adaptive control theory, generalized predictive control, multiple models and switching systems, intelligent decoupling control, data-based driven control, big data-driven modelling theory, method and technology of intelligent control systems and process industries, and their applications

     Lecturer at Northeastern University. He received his Ph. D. degree from Northeastern University in 2015. His current research interest covers operational control for complex industry process and industrial embedded control system

    Corresponding author: YANG Jie  Ph. D. candidate at the State Key Laboratory of Synthetical Automation for Process Industries, Northeastern University. His research interest covers data driven modeling technology and application for industrial process. Corresponding author of this paper
  • 摘要: 电熔镁群炉需量指当前时刻k和(k-1),…,(k-n+1)时刻群炉功率的平均值,用于度量高耗能电熔镁群炉用电量.(k+1)时刻群炉需量取决于功率变化率.本文建立了功率变化率与电流控制系统输出电流之间由线性项与未知非线性项组成的动态模型,其中线性项通过电流被控对象的参数和控制器的参数计算,未知非线性项采用基于偏自相关函数(Partial autocorrelation function,PACF)输入变量决策的径向基函数神经网络(Radial basis function neural network,RBFNN)来估计.本文提出了由当前k时刻的需量和功率,(k-n+1)时刻功率及k时刻功率变化率的估计组成的(k+1)时刻需量的计算模型.通过某电熔镁砂厂实际数据的仿真实验和工业实验表明所提方法可准确预报需量变化趋势,可以防止因原料变化引起需量尖峰导致错误切断电熔镁炉供电造成电熔镁砂质量降低.
    1)  本文责任编委 张卫东
  • 图  1  电熔镁群炉需量监控原理图

    Fig.  1  Schematic diagram of demand monitoring process for FMFG

    图  2  电熔镁群炉需量预报方法结构框图

    Fig.  2  The structure diagram of demand forecasting method for FMFG

    图  3  隐层节点数的交叉验证

    Fig.  3  Cross-validation of the number of hidden nodes

    图  4  高斯函数宽度的交叉验证

    Fig.  4  Cross-validation of the width of Gaussian function

    图  5  p(k)预报验证曲线

    Fig.  5  Forecast validation curves of ∆p(k)

    图  6  时段1需量曲线

    Fig.  6  Demand curve for the 1st time period

    图  7  时段2需量曲线

    Fig.  7  Demand curve for the 2nd time period

    图  8  时段3需量曲线

    Fig.  8  Demand curve for the 3rd time period

    图  9  需量预报软件界面

    Fig.  9  The interface of demand forecasting software

    图  10  时段1需量曲线

    Fig.  10  Demand curve for the 1st time period

    图  11  时段2需量曲线

    Fig.  11  Demand curve for the 2nd time period

    图  12  时段3需量曲线

    Fig.  12  Demand curve for the 3rd time period

    图  13  超限拉闸时段需量曲线

    Fig.  13  Demand curve for cut off time period

    图  14  超限拉闸时段需量预报误差变化曲线

    Fig.  14  Demand forecast error curve during cut off time period

    图  15  恢复供电动作下的需量预报

    Fig.  15  Demand forecast curve during restore operations

    图  16  恢复供电动作下的需量预报误差

    Fig.  16  Demand forecast error curve during restore operations

    图  17  工业实验需量预报误差白度分析

    Fig.  17  The whiteness analysis of demand forecast error in industrial experiment

    表  1  p(k)预报误差指标

    Table  1  Forecast error indicators of ∆p(k)

    方差 RMSE
    1.1481E+6 1 071.3
    下载: 导出CSV

    表  2  需量预报误差指标

    Table  2  Forecast error indicators of demand

    方法方差PB (%) RMSEMAPE (%)
    文献[5]1 533.097.0539.19210.0979
    本文 1 275.797.5535.71040.1054
    下载: 导出CSV

    表  3  超限拉闸时段需量预报误差

    Table  3  Demand forecast errors during cut off time period

    时间需量实际值(kW)需量预报值(kW)误差(kW)
    22 : 36 : 5021 62621 635-9
    22 : 36 : 5721 65421 6504
    22 : 37 : 0421 69221 6857
    22 : 37 : 1121 72821 7253
    22 : 37 : 1821 75421 7514
    22 : 37 : 2521 78821 7871
    22 : 37 : 3221 81221 829-17
    22 : 37 : 3921 83421 849-15
    22 : 37 : 4621 83521 839-4
    22 : 37 : 5321 83321 8294
    22 : 38 : 0021 82621 8224
    22 : 38 : 0721 69121 819-128
    22 : 38 : 1421 51021 46347
    22 : 38 : 2121 33521 31817
    22 : 38 : 2821 19121 17615
    22 : 38 : 3521 04921 056-7
    22 : 38 : 4220 90820 9071
    22 : 38 : 4920 75120 769-18
    22 : 38 : 5620 58120 5774
    22 : 39 : 0320 40720 38522
    下载: 导出CSV

    表  4  工业实验需量预报误差指标

    Table  4  Demand forecast error indicators of industrial experiment

    方差PB (%) RMSEMAPE
    1 049.897.6832.39810.0996
    下载: 导出CSV
  • [1] Paparoditis E, Sapatinas T. Short-term load forecasting:the similar shape functional time-series predictor. IEEE Transactions on Power Systems, 2013, 28(4):3818-3825 doi: 10.1109/TPWRS.2013.2272326
    [2] Ceperic E, Ceperic V, Baric A. A strategy for short-term load forecasting by support vector regression machines. IEEE Transactions on Power Systems, 2013, 28(4):4356-4364 doi: 10.1109/TPWRS.2013.2269803
    [3] Quan H, Srinivasan D, Khosravi A. Short-term load and wind power forecasting using neural network-based prediction intervals. IEEE Transactions on Neural Networks and Learning Systems, 2014, 25(2):303-315 doi: 10.1109/TNNLS.2013.2276053
    [4] Kebriaei H, Araabi B N, Rahimi-Kian A. Short-term load forecasting with a new nonsymmetric penalty function. IEEE Transactions on Power Systems, 2011, 26(4):1817-1825 doi: 10.1109/TPWRS.2011.2142330
    [5] Yang J, Chai T Y. Data-driven demand forecasting method for fused magnesium furnaces. In: Proceedings of the 12th World Congress on Intelligent Control and Automation. Guilin, China: IEEE, 2016. 2015-2022 http://ieeexplore.ieee.org/document/7578831/
    [6] Wu Z, Chai T Y, Sun J. Intelligent operational feedback control for fused magnesium furnace. In: Proceedings of the 19th World Congress on International Federation of Automatic Control. Cape Town, South Africa: IFAC, 2014. 8516-8521 http://www.sciencedirect.com/science/article/pii/S1474667016429575
    [7] Ozgun O, Abur A. Flicker study using a novel arc furnace model. IEEE Transactions on Power Delivery, 2002, 17(4):1158-1163 doi: 10.1109/TPWRD.2002.804013
    [8] 王其平.电器电弧理论.北京:机械工业出版社, 1991.

    Wang Qi-Ping. Arc Theory of Electrical Appliances. Beijing:Metallurgical Industry Press, 1991.
    [9] Shigley J E, Mischke C R, Budynas R G. Mechanical Engineering Design. New York, USA: McGraw-Hill, 1989.
    [10] 郭茂先.工业电炉.北京:冶金工业出版社, 2002.

    Guo Mao-Xian. Industry Furnace. Beijing:Metallurgical Industry Press, 2002.
    [11] Wu Z W, Wu Y J, Chai T Y, Sun J. Data-driven abnormal condition identification and self-healing control system for fused magnesium furnace. IEEE Transactions on Industrial Electronics, 2015, 62(3):1703-1715 doi: 10.1109/TIE.2014.2349479
    [12] Cecati C, Kolbusz J, Rózycki P, Siano P, Wilamowski B. A novel RBF training algorithm for short-term electric load forecasting and comparative studies. IEEE Transactions on Industrial Electronics, 2015, 62(10):6519-6529 doi: 10.1109/TIE.2015.2424399
    [13] Yu H, Xie T T, Paszczynski S, Wilamowski B M. Advantages of radial basis function networks for dynamic system design. IEEE Transactions on Industrial Electronics, 2011, 58(12):5438-5450 doi: 10.1109/TIE.2011.2164773
    [14] Xie T T, Yu H, Hewlett J, Rózycki P, Wilamowski B. Fast and efficient second-order method for training radial basis function networks. IEEE Transactions on Neural Networks and Learning Systems, 2012, 23(4):609-619 doi: 10.1109/TNNLS.2012.2185059
    [15] Du K L, Swamy M N S. Radial basis function networks. Neural Networks and Statistical Learning. London, UK: Springer, 2014. 299-335
    [16] Chen S, Cowan C F N, Grant P M. Orthogonal least squares learning algorithm for radial basis function networks. IEEE Transactions on Neural Networks, 1991, 2(2):302-309 doi: 10.1109/72.80341
    [17] Park J, Sandberg I W. Universal approximation using radial-basis-function networks. Neural Computation, 1991, 3(2):246-257 doi: 10.1162/neco.1991.3.2.246
    [18] Gomm J B, Yu D L. Selecting radial basis function network centers with recursive orthogonal least squares training. IEEE Transactions on Neural Networks, 2000, 11(2):306-314 doi: 10.1109/72.839002
    [19] Armstrong J S, Collopy F. Error measures for generalizing about forecasting methods:empirical comparisons. International Journal of Forecasting, 1992, 8(1):69-80 doi: 10.1016/0169-2070(92)90008-W
    [20] Dai W, Chai T Y, Yang S X. Data-driven optimization control for safety operation of hematite grinding process. IEEE Transactions on Industrial Electronics, 2015, 62(5):2930-2941 doi: 10.1109/TIE.2014.2362093
    [21] 代伟, 柴天佑.数据驱动的复杂磨矿过程运行优化控制方法.自动化学报, 2014, 40(9):2005-2014 http://www.aas.net.cn/CN/abstract/abstract18472.shtml

    Dai Wei, Chai Tian-You. Data-driven optimal operational control of complex grinding processes. Acta Automatica Sinica, 2014, 40(9):2005-2014 http://www.aas.net.cn/CN/abstract/abstract18472.shtml
    [22] 吴志伟, 柴天佑, 吴永建.电熔镁砂产品单吨能耗混合预报模型.自动化学报, 2013, 39(12):2002-2011 http://www.aas.net.cn/CN/abstract/abstract18239.shtml

    Wu Zhi-Wei, Chai Tian-You, Wu Yong-Jian. A hybrid prediction model of energy consumption per ton for fused magnesia. Acta Automatica Sinica, 2013, 39(12):2002-2011 http://www.aas.net.cn/CN/abstract/abstract18239.shtml
    [23] 黄宇斌, 袁景淇, 汪瑞清, 赵平伟.数据驱动的上海市日需水量预报建模研究.控制工程, 2010, 17(S2):58-60 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=QK201003398025

    Huang Yu-Bin, Yuan Jing-Qi, Wang Rui-Qing, Zhao Ping-Wei. Data-driven modeling for daily water demand forecast of Shanghai city. Control Engineering of China, 2010, 17(S2):58-60 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=QK201003398025
    [24] Landau I D, Zito G. Digital Control Systems: Design, Identification and Implementation. London, UK: Springer, 2006
  • 加载中
图(17) / 表(4)
计量
  • 文章访问数:  2422
  • HTML全文浏览量:  333
  • PDF下载量:  808
  • 被引次数: 0
出版历程
  • 收稿日期:  2016-08-20
  • 录用日期:  2017-02-03
  • 刊出日期:  2018-08-20

目录

    /

    返回文章
    返回