An Improved UKFNN Based on Square Root Filter and Strong Tracking Filter for Dynamic Evolutionary Modeling of Aluminum Reduction Cell
-
摘要: 铝电解过程具有多变量、强耦合、强干扰、参数时变等特征,故其模型开发是一个技术难点. 根据该过程的特点,本文提出强跟踪平方根无迹Kalman神经网络(Strong tracking square root unscented Kalman filter neural network,STR-UKFNN),并用其建立铝电解槽工艺能耗的动态演化模型. 该方法利用误差协方差矩阵的平方根代替UKFNN算法中的协方差阵,避免误差协方差矩阵可能出现负定而导致滤波发散,并在UKFNN算法中引入渐消因子和弱化因子,实时调整滤波增益,提高模型收敛速度和其对突变状态的跟踪能力. 通过某铝厂170kA预焙槽的日报样本验证表明,该方法提高了能耗模型的精度和对电解槽突变状态的实时跟踪能力,有助于指导铝电解过程操作参数的优化.Abstract: The aluminum electrolysis process has multiple characteristics including multivariate, strong coupling, strong interference and time-varying parameters. Therefore, its model development is technically difficult. According to the characteristics of the process, an improved unscented Kalman filter neural network based on strong tracking filter and square root filter (STR-UKFNN) is proposed in this paper. Then, the STR-UKFNN is used to create the dynamic evolutionary model for energy consumption of aluminum reduction cell. Firstly, the state covariance matrix of the UKFNN algorithm is replaced by its square root to participate in recursive operations; Secondly, the filter gain matrice in the algorithm of UKFNN is adjusted by introducing the time-varying fading factor and the diminishing factor. A series of experiments have been conducted by using the daily samples from the 170kA new pre-baked cell. The experimental results show that the method improves the precision of the energy model and the real-time tracking ability for the abrupt state change of the aluminum reduction cell. So the method is helpful to guide the optimization of operating parameters in the aluminum electrolysis process.
-
[1] Wu Juan-Ni, Wan Hong-Yan, Chen Wei-Qiang, Shi Lei. Quantifying energy consumption and greenhouse gas emissions of the primary aluminum industry in China. Journal of Tsinghua University (Science and Technology), 2010, 50(3): 407-410(武娟妮, 万红艳, 陈伟强, 石磊. 中国原生铝工业的能耗与温室气体排放核算. 清华大学学报(自然科学版), 2010, 50(3): 407-410) [2] Feng Nai-Xiang. A new cathode structure of aluminum reduction cell with the vertical and horizontal reduction wave function, China Patent 200820218402.9, September 2009(冯乃祥. 一种具有纵向和横向减波功能的新型阴极结构铝电解槽, 中国专利 200820218402.9, 2009-09) [3] Tian Ying-Fu. A method for producing perforated anode, China Patent 201010617877.7, May 2011(田应甫. 一种制作穿孔阳极的方法, 中国专利201010617877.7, 2011-05) [4] [4] Tang Chan-Yi. Research of Relevance Quantification of Process Parameters in Aluminum Electrolytic Based on Data Mining [Master dissertation], North China University of Technology, China, 2009 [5] Hu Bao-Gang, Wang Yong, Yang Shuang-Hong, Qu Han-Bing. How to add transparency to artificial neural networks? Pattern Recognition and Artificial Intelligence, 2007, 20(1): 72-83(胡包钢, 王泳, 杨双红, 曲寒冰. 如何增加人工神经元网络的透明度? 模式识别与人工智能, 2007, 20(1): 72-83) [6] Li Yong, Wu Min, Cao Wei-Hua, Lai Xu-Zhi, Wang Chun-Sheng. PSO-BP control algorithm of granulation process based on evaluation and optimization of granularity distribution. Acta Automatica Sinica, 2012, 38(6): 1007-1016(李勇, 吴敏, 曹卫华, 赖旭芝, 王春生. 基于粒度分布评估与优化的制粒过程PSO-BP控制算法. 自动化学报, 2012, 38(6): 1007-1016) [7] Zhang Li-Jun, Yang Li-Xin, Guo Li-Dong, Sun Li-Ning. Adaptive output feedback control for piezoactuator-driven stage. Acta Automatica Sinica, 2012, 38(9): 1550-1556(张利军, 杨立新, 郭立东, 孙立宁. 压电陶瓷驱动平台自适应输出反馈控制. 自动化学报, 2012, 38(9): 1550-1556) [8] Li He-Song, Yin Xiao-Bao, Huang Yong-Bo, Ding Li-Wei, Jiang Chang-Wei. Diagnosis system of different status of aluminum reduction cells based on anode current fluctuation. CIESC Journal, 2011, 62(6): 1770-1777(李贺松, 殷小宝, 黄涌波, 丁立伟, 姜昌伟. 基于阳极电流波动的铝电解槽槽况诊断系统. 化工学报, 2011, 62(6): 1770-1777) [9] Li Jie-Jia, Wu Cheng-Dong, Li Ying. Application research on neural network predictive control technology in aluminum electolysis process. Instrument Technique and Sensor, 2011, (8): 91-93(李界家, 吴成东, 李颖. 铝电解过程神经网络预测控制技术应用研究. 仪表技术与传感器, 2011, (8): 91-93) [10] Soares F M, Oliveira R C L. Modelling of temperature in the aluminium smelting process using neural networks. In: Proceedings of the 2010 International Joint Conference on Neural Networks. Barcelona: IEEE, 2010. 1-7 [11] Meghlaoui A, Thibault J, Bui R T, Tikasz L, Santerre R. Neural networks for the identification of the aluminium electrolysis process. Computers Chemical Engineering, 1998, 22(10): 1419-1428 [12] Wu Jian-Feng, He Xiao-Rong, Chen Bing-Zhen. Modeling nonlinear dynamic chemical process base on artificial neural networks. Computers and Applied Chemistry, 2001, 18(2): 105-110(吴建锋, 何小荣, 陈丙珍. 基于反馈神经网络的动态化工过程建模. 计算机与应用化学, 2001, 18(2): 105-110) [13] Salahshoor K, Hajisalehi M H, Sefat M H. Nonlinear model identification and adaptive control of CO2 sequestration process in saline aquifers using artificial neural networks. Applied Soft Computing, 2012, 12(11): 3379-3389 [14] Wu X D, Wang Y N. Extended and unscented Kalman filtering based feedforward neural networks for time series prediction. Applied Mathematical Modelling, 2012, 36(3): 1123-1131 [15] Julier S J, Uhlmann J K. Unscented filtering and nonlinear estimation. Proceedings of the IEEE, 2004, 92(3): 401-422 [16] Zhan Rong-Hui, Wan Jian-Wei. Neural network-aided adaptive unscented Kalman filter for nonlinear state estimation. IEEE Signal Processing Letters, 2006, 13(7): 445-448 [17] Kaminski P, Bryson A J, Schmidt S. Discrete square root filtering: a survey of current techniques. IEEE Transactions on Automatic Control, 1971, 16(6): 727-736 [18] Van der Merwe R, Wan E A. The square-root unscented kalman filter for state and parameter-estimation. In: Proceedings of the 2001 IEEE International Conference on Acoustics, Speech and Signal Processing. Salt Lake City, USA: IEEE, 2001. 3461-3464 [19] Zhang Shu-Chun, Hu Guang-Da. Square-root filtering and its applications in target tracking. Journal of Harbin Institute of Technology, 2008, 40(5): 700-704(张树春, 胡广大. 平方根滤波及其在目标跟踪方面的应用. 哈尔滨工业大学学报, 2008, 40(5): 700-704) [20] Zhou Dong-Hua, Xi Yu-Geng, Zhang Zhong-Jun. Suboptimal fading extended Kalman filtering for nonlinear systems. Control and Decision, 1990, 5(5): 1-6(周东华, 席裕庚, 张钟俊. 非线性系统带次优渐消因子的扩展卡尔曼滤波. 控制与决策, 1990, 5(5): 1-6) [21] Zhou Dong-Hua, Xi Yu-Geng, Zhang Zhong-Jun. A suboptimal multiple fading extended Kalman filter. Acta Automatica Sinica, 1991, 17(6): 689-695(周东华, 席裕庚, 张钟俊. 一种带多重次优渐消因子的扩展卡尔曼滤波器. 自动化学报, 1991, 17(6): 689-695) [22] Hu Wen-Jin, Zhou Dong-Hua. Adaptive control of DC-DC converters based on a strong tracking filter. Journal of Tsinghua University (Science and Technology), 2008, 48(S2): 1712-1717(胡文金, 周东华. 基于强跟踪滤波器的DC-DC变换器自适应控制. 清华大学学报(自然科学版), 2008, 48(S2): 1712-1717) [23] Press W H, Teukolsky S A, Vetterling W T, Flannery B P. Numerical Recipes in C: The Art of Scientific Computing (2nd edition). England: Cambridge University Press, 1992. 323-327 [24] Xu Bo, Zhu Huang-Qiu, Ji Wei, Pan Wei, Sun Xiao-Dong. Modified squareroot unscented Kalman filter and its application to speed sensorless control of bearingless permanent magnet synchronous motor. Control Theory Applications, 2012, 29(1): 53-58(许波, 朱熀秋, 姬伟, 潘伟, 孙晓东. 改进型平方根无迹卡尔曼滤波及其在无轴承永磁同步电机无速度传感器运行中的应用. 控制理论与应用, 2012, 29(1): 53-58)
点击查看大图
计量
- 文章访问数: 1754
- HTML全文浏览量: 96
- PDF下载量: 872
- 被引次数: 0