基于自组织模糊神经网络的污水处理过程溶解氧控制

doi:10.11949/j.issn.0438-1157.20161514

摘要/Abstract

摘要：

针对活性污泥污水处理过程溶解氧浓度控制问题，提出一种基于自组织模糊神经网络（SOFNN）的控制方法。该神经网络控制器依据激活强度和互信息理论在线动态增长和修剪规则层神经元，以满足实际工况的动态变化。同时，采用梯度下降算法在线优化隶属函数层中心、宽度和输出权值，以保证SOFNN的收敛性。进一步通过Lyapunov稳定性理论对SOFNN学习率进行分析，给出控制系统稳定性证明。最后在国际基准仿真平台BSM1上进行实验验证。实验结果显示，与PID、模糊逻辑控制（FLC）和固定结构FNN等控制策略相比，SOFNN在跟踪精度、控制平稳性和自适应能力上更具有优势。

关键词: 污水, 溶解氧, 过程控制, 神经网络, 自组织模糊神经网络, 互信息

Abstract:

A self-organizing fuzzy neural network (SOFNN) control method is proposed, and its application system is designed for controlling the dissolved oxygen (DO) concentration in the activated sludge wastewater treatment processes. The neurons of rule layer are grown or pruned adaptively based on firing strength and mutual information to meet dynamic change of the real operating condition. Meanwhile, the centers and widths of membership functions and weights of output layer are trained by gradient descent optimization algorithm to ensure the convergence of SOFNN. The stability of the control system is proved based on the analysis of the learning rates of parameters in SOFNN by applying the Lyapunov stability theory. This control strategy is investigated and evaluated based on international Benchmark Simulation Model No.1 (BSM1). Experimental results demonstrate that the SOFNN controller performs better than PID, fuzzy logical control (FLC), model predictive control (MPC), and some other existing control methods. Performance comparisons indicate that the proposed SOFNN control strategy obtains higher tracking accuracy, better control placidity and superior adaptive capability.

Key words: wastewater, dissolved oxygen, process control, neural network, self-organizing fuzzy neural network, mutual information

中图分类号:

TP273

周红标. 基于自组织模糊神经网络的污水处理过程溶解氧控制[J]. 化工学报, 2017, 68(4): 1516-1524.

ZHOU Hongbiao. Dissolved oxygen control of wastewater treatment process using self-organizing fuzzy neural network[J]. CIESC Journal, 2017, 68(4): 1516-1524.

参考文献 30

[1]	Shannon M A, Bohn P W, Elimelech M, et al. Science and technology for water purification in the coming decades[J]. Nature, 2008, 452(7185):301-310.
[2]	Hamilton R, Braun B, Dare R, et al. Control issues and challenges in wastewater treatment plants[J]. IEEE Control Systems, 2006, 26(4):63-69.
[3]	Piotrowski R, Brdys M A, Konarczak K, et al. Hierarchical dissolved oxygen control for activated sludge processes[J]. Control Engineering Practice, 2008, 16(1):114-131.
[4]	Carlsson B, Rehnström A. Control of an activated sludge process with nitrogen removal-a benchmark study[J]. Water Science and Technology, 2002, 45(4/5):135-142.
[5]	Chachuat B, Roche N, Latifi M A. Optimal aeration control of industrial alternating activated sludge plants[J]. Biochemical Engineering Journal, 2005, 23(3):277-289.
[6]	Harja G, Nascu I, Muresan C, et al. Improvements in dissolved oxygen control of an activated sludge wastewater treatment process[J]. Circuits, Systems, and Signal Processing, 2016, 35(6):2259-2281.
[7]	Åmand L, Olsson G, Carlsson B. Aeration control-a review[J]. Water Science and Technology, 2013, 67(11):2374-2398.
[8]	Yoo C K, Lee J M, Lee I B. Nonlinear model-based dissolved oxygen control in a biological wastewater treatment process[J]. Korean Journal of Chemical Engineering, 2004, 21(1):14-19.
[9]	Syafiie S, Tadeo F, Martinez E, et al. Model-free control based on reinforcement learning for a wastewater treatment problem[J]. Applied Soft Computing, 2011, 11(1):73-82.
[10]	Rojas J D, Flores-Alsina X, Jeppsson U, et al. Application of multivariate virtual reference feedback tuning for wastewater treatment plant control[J]. Control Engineering Practice, 2012, 20(5):499-510.
[11]	Han H G, Qiao J F, Chen Q L. Model predictive control of dissolved oxygen concentration based on a self-organizing RBF neural network[J]. Control Engineering Practice, 2012, 20(4):465-476.
[12]	Wahab N A, Katebi R, Balderud J. Multivariable PID control design for activated sludge process with nitrification and denitrification[J]. Biochemical Engineering Journal, 2009, 45(3):239-248.
[13]	Song X L, Zhao Y B, Song Z Y, et al. Dissolved oxygen control in wastewater treatment based on robust PID controller[J]. International Journal of Modelling, Identification and Control, 2012, 15(4):297-303.
[14]	Holenda B, Domokos E, Redey A, et al. Dissolved oxygen control of the activated sludge wastewater treatment process using model predictive control[J]. Computers & Chemical Engineering, 2008, 32(6):1270-1278.
[15]	Belchior C A C, Araújo R A M, Landeck J A C. Dissolved oxygen control of the activated sludge wastewater treatment process using stable adaptive fuzzy control[J]. Computers & Chemical Engineering, 2012, 37(2):152-162.
[16]	Qiao J F, Han H G. Identification and modeling of nonlinear dynamical systems using a novel self-organizing RBF-based approach[J]. Automatica, 2012, 48(8):1729-1734.
[17]	Wai R J, Chen M W, Liu Y K. Design of adaptive control and fuzzy neural network control for single-stage boost inverter[J]. IEEE Transactions on Industrial Electronics, 2015, 62(9):5434-5445.
[18]	Qiao J F, Han G, Han H G. Neural network on-line modeling and controlling method for multi-variable control of wastewater treatment processes[J]. Asian Journal of Control, 2014, 16(4):1213-1223.
[19]	Huang M Z, WAN J Q, MA Y W, et al. Control rules of aeration in a submerged biofilm wastewater treatment process using fuzzy neural networks[J]. Expert Systems with Applications, 2009, 36(7):10428-10437.
[20]	张伟, 乔俊飞, 李凡军. 溶解氧浓度的直接自适应动态神经网络控制方法[J]. 控制理论与应用, 2015, 32(1):115-121.ZHANG W, QIAO J F, LI F J. Direct adaptive dynamic neural network control for dissolved oxygen concentration[J]. Control Theory & Applications, 2015, 32(1):115-121.
[21]	Jeppsson U, Pons M N. The COST benchmark simulation model-current state and future perspective[J]. Control Engineering Practice, 2004, 12(3):299-304.
[22]	WU S Q, ER M J. Dynamic fuzzy neural networks-a novel approach to function approximation[J]. IEEE Transactions on Systems, Man, and Cybernetics, Part B:Cybernetics, 2000, 30(2):358-364.
[23]	许少鹏, 韩红桂, 乔俊飞. 基于模糊递归神经网络的污泥容积指数预测模型[J]. 化工学报, 2013, 64(12):4550-4556.XU S P, HAN H G, QIAO J F. Prediction of activated sludge bulking based on recurrent fuzzy neural network[J]. CIESC Journal, 2013, 64(12):4550-4556.
[24]	Han H G, Qiao J F. A self-organizing fuzzy neural network based on a growing-and-pruning algorithm[J]. IEEE Transactions on Fuzzy Systems, 2010, 18(6):1129-1143.
[25]	HAN H G, WU X L, QIAO J F. Nonlinear systems modeling based on self-organizing fuzzy-neural-network with adaptive computation algorithm[J]. IEEE Transactions on Cybernetics, 2014, 44(4):554-564.
[26]	Vergara J R, Estévez P A. A review of feature selection methods based on mutual information[J]. Neural Computing and Applications, 2014, 24(1):175-186.
[27]	Naghibi T, Hoffmann S, Pfister B. A semidefinite programming based search strategy for feature selection with mutual information measure[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(8):1529-1541.
[28]	Hacine-Gharbi A, Ravier P, Harba R, et al. Low bias histogram-based estimation of mutual information for feature selection[J]. Pattern Recognition Letters, 2012, 33(10):1302-1308.
[29]	Wei M, Chow T W S, Chan R H M. Heterogeneous feature subset selection using mutual information-based feature transformation[J]. Neurocomputing, 2015, 168:706-718.
[30]	Lin Y Y, Chang J Y, Lin C T. Identification and prediction of dynamic systems using an interactively recurrent self-evolving fuzzy neural network[J]. IEEE Transactions on Neural Networks and Learning Systems, 2013, 24(2):310-321.