城市固废焚烧过程烟气含氧量自适应预测控制

孙剑; 蒙西; 乔俊飞

doi:10.16383/j.aas.c210935

城市固废焚烧过程烟气含氧量自适应预测控制

doi: 10.16383/j.aas.c210935

孙剑^{1, 2, 3,},
蒙西^{1, 2, 3,},
乔俊飞^{1, 2, 3,}

1.
北京工业大学信息学部北京 100124
2.
智慧环保北京实验室北京 100124
3.
智能感知与自主控制教育部工程研究中心北京 100124

基金项目: 国家自然科学基金(62021003, 61890930-5, 61903012), 科技创新2030 —— “新一代人工智能”重大项目(2021ZD0112301, 2021ZD0112302)资助

详细信息

作者简介:
孙剑：北京工业大学信息学部博士研究生. 主要研究方向为复杂工业过程数据驱动建模与智能控制. E-mail: sun8927@163.com

蒙西：北京工业大学信息学部副教授. 主要研究方向为神经网络, 学习系统和工业过程建模与控制. E-mail: mengxi@bjut.edu.cn

乔俊飞：北京工业大学信息学部教授. 主要研究方向为神经网络, 智能系统和复杂工业过程建模与优化控制. 本文通信作者. E-mail: adqiao@bjut.edu.cn

计量
- 文章访问数: 1161
- HTML全文浏览量: 510
- PDF下载量: 292
- 被引次数: 0
出版历程
- 收稿日期: 2021-09-29
- 录用日期: 2022-02-10
- 网络出版日期: 2022-05-08
- 刊出日期: 2023-11-22

Adaptive Predictive Control of Oxygen Content in Flue Gas for Municipal Solid Waste Incineration Process

SUN Jian^{1, 2, 3
,},
MENG Xi^{1, 2, 3
,},
QIAO Jun-Fei^{1, 2, 3
,}

1.
Faculty of Information Technology, Beijing University of Technology, Beijing 100124
2.
Beijing Laboratory of Smart Environmental Protection, Beijing 100124
3.
Engineering Research Center of Intelligence Perception and Autonomous Control, Ministry of Education, Beijing 100124

Funds: Supported by National Natural Science Foundation of China (62021003, 61890930-5, 61903012) and National Key Research and Development Program of China (2021ZD0112301, 2021ZD0112302)

More Information

Author Bio:
SUN Jian　Ph.D. candidate at the Faculty of Information Technology, Beijing University of Technology. His research interest covers data-driven modeling and intelligent control of complex industrial processes

MENG Xi　Associate professor at the Faculty of Information Technology, Beijing University of Technology. Her research interest covers neural networks, learning systems, and industrial process modeling and control

QIAO Jun-Fei　Professor at the Faculty of Information Technology, Beijing University of Technology. His research interest covers neural networks, intelligent systems, and modeling and optimal control of complex industrial processes. Corresponding author of this paper

摘要

摘要: 在城市固体废弃物焚烧(Municipal solid waste incineration, MSWI)过程中, 烟气含氧量是影响焚烧效果的重要工艺参数. 由于固废焚烧过程的复杂性, 在实际应用过程中, 难以实现烟气含氧量的有效控制. 面向城市固废焚烧过程烟气含氧量控制的实际需求, 提出一种基于数据驱动的烟气含氧量自适应预测控制方法. 首先, 采用自适应模糊C均值(Fuzzy C-means, FCM)算法辅助确定径向基函数(Radial basis function, RBF)神经网络隐含层神经元个数及初始中心, 建立基于FCM算法的径向基函数神经网络预测模型, 并在控制过程中通过自适应更新策略在线调节预测模型参数; 然后, 利用梯度下降算法求解控制律, 并基于李雅普诺夫理论分析了所提控制方法的稳定性; 最后, 基于城市固废焚烧厂实际数据, 验证了所提控制方法的有效性.
- 城市固体废物焚烧 /
- 烟气含氧量 /
- 自适应预测控制 /
- 径向基函数神经网络 /
- 梯度下降
Abstract: Oxygen content in flue gas is an important process parameter for incineration efficiency in municipal solid waste incineration (MSWI). Due to the complexity of municipal solid waste incineration process, it is difficult to achieve effective control of oxygen content in flue gas in practical application. A data-driven adaptive predictive control method for oxygen content in flue gas in MSWI process is proposed in this paper. Firstly, an adaptive fuzzy C-means (FCM) algorithm is used to determine the number of hidden layer neurons and the initial clustering center of radial basis function (RBF) neural network model, and the radial basis function neural network prediction model based on FCM algorithm is established. During the control process, the prediction model parameters are adjusted adaptively by an online updating strategy. Then, the gradient descent method is exploited to solve the control law, and the stability of the control system is analyzed based on the Lyapunov theory. Finally, the effectiveness of the proposed control method is verified based on the actual data of the municipal solid waste incineration plant.
- Municipal solid waste incineration (MSWI) /
- oxygen content in flue gas /
- adaptive predictive control /
- radial basis function neural network /
- gradient descent

HTML全文

图 1 MSWI炉排炉工艺流程

Fig. 1 MSWI process with grate furnace

下载: 全尺寸图片幻灯片

图 2 烟气含氧量自适应FCM-RBF-MPC策略

Fig. 2 Adaptive FCM-RBF-MPC strategy of oxygen content in flue gas

下载: 全尺寸图片幻灯片

图 3 RBF神经网络结构图

Fig. 3 RBF neural network framework

下载: 全尺寸图片幻灯片

图 4 不同建模方法的烟气含氧量预测效果

Fig. 4 Prediction effect of oxygen content in flue gas with different modeling methods

下载: 全尺寸图片幻灯片

图 5 FCM-RBF-MPC和自适应FCM-RBF-MPC控制效果

Fig. 5 Control results of FCM-RBF-MPC and adaptive FCM-RBF-MPC

下载: 全尺寸图片幻灯片

图 6 FCM-RBF-MPC和自适应FCM-RBF-MPC的操作变量变化情况

Fig. 6 Changes of manipulated variables of FCM-RBF-MPC and adaptive FCM-RBF-MPC

下载: 全尺寸图片幻灯片

图 7 反馈校正误差补偿变化情况

Fig. 7 Changes of feedback correction error compensation

下载: 全尺寸图片幻灯片

表 1 输入/输出变量变化范围

Table 1 Range of input and output variables

变量名	变量符号	变化范围
干燥炉排一次风流量(km³/h)	$x_{p1}$	10.75 ~ 14.76
燃烧炉排1一次风流量(km³/h)	$x_{p2}$	25.90 ~ 35.56
燃烧炉排2一次风流量 (km³/h)	$x_{p3}$	11.25 ~ 15.72
燃烬炉排一次风流量(km³/h)	$x_{p4}$	2.27 ~ 5.06
二次风流量(km³/h)	$x_{s}$	18.91 ~ 21.84
烟气含氧量 (%)	$y_o $	4.6 ~ 7.6

下载: 导出CSV

表 2 操作变量与烟气含氧量的皮尔森相关系数

Table 2 Pearson correlation coefficient between manipulated variables and oxygen content in flue gas

操作变量名	操作变量符号	$\gamma$
干燥炉排一次风流量	$x_{p1}$	−0.4303
燃烧炉排1一次风流量	$x_{p2} $	0.3015
燃烧炉排2一次风流量	$x_{p3} $	−0.1034
燃烬炉排一次风流量	$x_{p4} $	0.0697
二次风流量	$x_{s} $	0.1413

下载: 导出CSV

表 3 不同建模方法的烟气含氧量预测评价指标对比

Table 3 Comparison of prediction evaluation indexes of oxygen content in flue gas with different modeling methods

模型	MAE	MAPE	RMSE
MLP	0.0467	0.0086	0.0585
RBF	0.0357	0.0065	0.0449
FCM-RBF	0.0307	0.0056	0.0417

下载: 导出CSV

表 4 不同RBF神经网络预测控制器性能指标对比

Table 4 Comparison of evaluation indexes of different RBF neural network predictive controllers

控制器	IAE	ITAE	Dev^max
RBF-MPC	0.0056	4.3874	0.4991
FCM-RBF-MPC	0.0043	3.0002	0.4949
自适应RBF-MPC	0.0045	3.4617	0.4335
自适应FCM-RBF-MPC	0.0031	2.4152	0.4226

下载: 导出CSV

参考文献(36)

[1]	Maalouf A, Mavropoulos A, El-Fadel M. Global municipal solid waste infrastructure: delivery and forecast of uncontrolled disposal. Waste Management and Research, 2020, 38(9): 1028-1036 doi: 10.1177/0734242X20935170
[2]	中国统计年鉴. 北京: 中国统计出版社, 2019. China Statistical Yearbook. Beijing: China Statistics Press, 2019.
[3]	龙文, 梁昔明, 龙祖强. 基于混合PSO优化的LSSVM锅炉烟气含氧量预测控制. 中南大学学报(自然科学版), 2012, 43(3): 980-985 Long Wen, Liang Xi-Ming, Long Zu-Qiang. O₂ content in flue gas of boilers predictive control based on hybrid PSO and LSSVM. Journal of Central South University (Science and Technology), 2012, 43(3): 980-985
[4]	Zhang Y J, Jia Y, Chai T Y, et al. Data-driven PID controller and its application to pulp neutralization process. IEEE Transactions on Control Systems Technology, 2017, 26(3): 828-841
[5]	Moura J P, Neto J V F, Rêgo P H M. A neuro-fuzzy model for online optimal tuning of PID controllers in industrial system applications to the mining sector. IEEE Transactions on Fuzzy Systems, 2019, 28(8): 1864-1877
[6]	Nath U M, Dey C, Mudi R K. Desired characteristic equation based PID controller tuning for lag-dominating processes with real-time realization on level control system. IEEE Control Systems Letters, 2021, 5(4): 1255-1260 doi: 10.1109/LCSYS.2020.3030173
[7]	Li D P, Han H G, Qiao J F. Observer-based adaptive fuzzy control for nonlinear state-constrained systems without involving feasibility conditions. IEEE Transactions on Cybernetics, doi: 10.1109/TCYB.2021.3071336
[8]	Nguyen A T, Taniguchi T, Eciolaza L, et al. Fuzzy Control Systems: Past, Present and Future. IEEE Computational Intelligence Magazine, 2019, 14(1): 56-68 doi: 10.1109/MCI.2018.2881644
[9]	Shen D. Iterative learning control with incomplete information: a survey. IEEE/CAA Journal of Automatica Sinica, 2018, 5(5): 1-17 doi: 10.1109/JAS.2018.7511201
[10]	赵钢, 李斌. 燃气锅炉燃烧效率优化控制仿真研究. 计算机仿真, 2017, 34(1): 376-379 doi: 10.3969/j.issn.1006-9348.2017.01.081 Zhao Gang, Li Bin. Simulation on the optimal control of gas boiler combustion efficiency. Computer Simulation, 2017, 34(1): 376-379 doi: 10.3969/j.issn.1006-9348.2017.01.081
[11]	Zhang R D, Cao Z X, Li P, et al. Design and implementation of an improved linear quadratic regulation control for oxygen content in a coke furnace. IET Control Theory and Applications, 2014, 8(14): 1303-1311 doi: 10.1049/iet-cta.2013.1023
[12]	Thai S M, Wilcox S J, Chong A Z S, et al. Combustion optimisation of stoker fired boiler plant by neural networks. Journal of the Energy Institute, 2008, 81(3): 171-176. doi: 10.1179/174602208X339131
[13]	席裕庚, 李德伟, 林姝. 模型预测控制-现状与挑战. 自动化学报, 2013, 39(3): 222-236 doi: 10.1016/S1874-1029(13)60024-5 Xi Yu-Geng, Li De-Wei, Lin Shu. Model predictive control — status and challenges. Acta Automatica Sinica, 2013, 39(3): 222−236 doi: 10.1016/S1874-1029(13)60024-5
[14]	Karamanakos P, Liegmann E, Geyer T, et al. Model predictive control of power electronic systems: methods, results, and challenges. IEEE Open Journal of Industry Applications, 2020, 1: 95-114 doi: 10.1109/OJIA.2020.3020184
[15]	何德峰, 丁宝苍, 于树友. 非线性系统模型预测控制若干基本特点与主题回顾. 控制理论与应用, 2013, 30(3): 273-287 doi: 10.7641/CTA.2013.20737 He De-Feng, Ding Bao-Cang, Yu Shu-You. Review of fundamental properties and topics of model predictive control for nonlinear systems. Control Theory and Applications, 2013, 30(3): 273-287 doi: 10.7641/CTA.2013.20737
[16]	Zhang R D, Jin Q B. Design and Implementation of hybrid modeling and PFC for oxygen content regulation in a coke furnace. IEEE Transactions on Industrial Informatics, 2018, 14(6): 2335-2342 doi: 10.1109/TII.2018.2815717
[17]	Huang X Y, Wang J C, Zhang L W, et al. Data-driven modelling and fuzzy multiple-model predictive control of oxygen content in coal-fired power plant. Transactions of the Institute of Measurement and Control, 2017, 39(11): 1631-1642 doi: 10.1177/0142331216644498
[18]	Wang P, Yang C H, Tian X M, et al. Adaptive nonlinear model predictive control using an on-line support vector regression updating strategy. Chinese Journal of Chemical Engineering, 2014, 22(7): 774-781 doi: 10.1016/j.cjche.2014.05.004
[19]	Xiao H Z, Chen C L P. Incremental updating multi-robot formation using nonlinear model predictive control method with general projection neural network. IEEE Transactions on Industrial Electronics, 2019, 66(6): 4502-4512 doi: 10.1109/TIE.2018.2864707
[20]	陈进东, 潘丰. 基于在线支持向量回归的非线性模型预测控制方法. 控制与决策, 2014, 29(3): 460-464 Chen Jin-Dong, Pan Feng. Online support vector regression-based nonlinear model predictive control. Control and Decision, 2014, 29(3): 460-464
[21]	Vatankhah B, Farrokhi M. Nonlinear adaptive model predictive control of constrained systems with offset-free tracking behavior. Asian Journal of Control, 2018, 20(1): 1-13 doi: 10.1002/asjc.1548
[22]	Wu Z, Rincon D, Christofides P D. Real-time adaptive machine-learning-based predictive control of nonlinear processes. Industrial and Engineering Chemistry Research, 2020, 59: 2275-2290 doi: 10.1021/acs.iecr.9b03055
[23]	Lu J W, Zhang S K, Hai J, et al. Status and perspectives of municipal solid waste incineration in China: A comparison with developed regions. Waste Management, 2017, 69: 170-186 doi: 10.1016/j.wasman.2017.04.014
[24]	Tan S T, Ho W S, Hashim H, et al. Energy, economic and environmental (3E) analysis of waste-to-energy (WTE) strategies for municipal solid waste (MSW) management in Malaysia. Energy Conversion and Management, 2015, 102: 111-120 doi: 10.1016/j.enconman.2015.02.010
[25]	乔俊飞, 郭子豪, 汤健. 面向城市固废焚烧过程的二噁英排放浓度检测方法综述. 自动化学报, 2020, 46(6): 1063-1089 Qiao Jun-Fei, Guo Zi-Hao, Tang Jian. Dioxin emission concentration measurement approaches for municipal solid wastes incineration process: a survey. Acta Automatica Sinica, 2020, 46(6): 1063-1089
[26]	Xie S W, Xie Y F, Huang T W, et al. Generalized predictive control for industrial processes based on neuron adaptive splitting and merging rbf neural network. IEEE Transactions on Industrial Electronics, 2019, 66(2): 1192-1202 doi: 10.1109/TIE.2018.2835402
[27]	Zhou P, Guo D W, Chai T Y. Data-driven predictive control of molten iron quality in blast furnace ironmaking using multi-output LS-SVR based inverse system identification. Neurocomputing, 2018, 308: 101-110 doi: 10.1016/j.neucom.2018.04.060
[28]	Qiao J F, Meng X, Li W J, et al. A novel modular RBF neural network based on a brain-like partition method. Neural Computing and Applications, 2018, 32: 899-911
[29]	Ren M, Liu P Y, Wang Z H, et al. A self-adaptive fuzzy c-means algorithm for determining the optimal number of clusters. Computational Intelligence and Neuroscience, 2016: 2647389
[30]	Li Y, Yu F. A new validity function for fuzzy clustering. In: Proceedings of the International Conference on Computational Intelligence and Natural Computing. Wuhan, China: IEEE, 2009. 462−465
[31]	任俊超, 刘丁, 万银. 基于混合集成建模的硅单晶直径自适应非线性预测控制. 自动化学报, 2020, 6(5): 1004-1016 Ren Jun-Chao, Liu Ding, Wan Yin. Hybrid integrated modeling based adaptive nonlinear predictive control of silicon single crystal diameter. ACTA AUTOMATICA SINICA, 2020, 6(5): 1004-1016
[32]	Han H G, Wu X L, Qiao J F. Real-time model predictive control using a self-organizing neural network. IEEE Transactions on Neural Networks and Learning Systems, 2013, 24(9): 1425-1436 doi: 10.1109/TNNLS.2013.2261574
[33]	Noriega J R, Wang H. A direct adaptive neural-network control for unkonwn nonlinear systems and its application. IEEE Transactions on Neural Networks, 1998, 9(1): 27-34 doi: 10.1109/72.655026
[34]	Lu C H. Wavelet fuzzy neural networks for identification and predictive control of dynamic systems. IEEE Transactions on Industrial Electronics, 2011, 58(7): 3046-3058. doi: 10.1109/TIE.2010.2076415
[35]	Hou Z S, Xiong S S. On model-free adaptive control and its stability analysis. IEEE Transactions on Automatic Control, 2019, 64(11): 4555-4569. doi: 10.1109/TAC.2019.2894586
[36]	Han H G, Liu H X, Li J M, et al. Cooperative fuzzy-neural control for wastewater treatment process. IEEE Transactions on Industrial Informatics, 2021, 17(9): 5971-5981. doi: 10.1109/TII.2020.3034335