基于选择性集成模型库的选择性催化还原脱硝系统自适应建模
|
摘要
选择性催化还原(selectivecatalyticreduction,SCR)脱硝系统动态建模对优化喷氨控制、降低NOx排放和氨逃逸有着重要意义。基于机器学习的燃煤电站SCR脱硝系统建模方法在精度上具有优势,但大多数模型缺乏自学习或自适应更新机制,难以在长期运行中保持有效性。针对SCR脱硝系统动态特性随负荷与煤质参数变化的问题,提出选择性集成模型库算法,包含基于分时段数据的模型库构建方法、基于选择性实时误差权重法的结合策略和基于模型评价方法的模型库更新策略。采用某660MW燃煤发电机组SCR系统50天运行数据对所提出的模型进行训练、测试以及验证,并与传统自适应建模方法进行对比研究。当传统自适应模型失效时,所提出的模型库仍能保持较高精度。结果表明,选择性集成模型库在预测精度、鲁棒性和稳定性上均有明显优势。
Dynamic modeling of selective catalytic reduction(SCR) denitrification system is of great significance for optimization of ammonia injection, reducing of the NOx emission and ammonia slip. The modeling methods based on machine learning have advantages in accuracy, but most methods lack self-learning or adaptive strategy, and are difficult to keep good performance in long-term operation. In order to adapt to the dynamic characteristics of SCR denitrification system caused by load and coal quality change, selective ensemble model library was proposed including a construction method of library based on time-phased data, a combined strategy based on selective real-time error weighting method and an updating strategy based on model evaluation method. The model was trained, tested and verified by a 50-day operation data of SCR system in a 660 MW coal-fired boiler, and compared with the traditional adaptive model. When the traditional model failed, the proposed model can still keep high prediction accuracy. The results show that selective ensemble model library performs better in prediction accuracy, robustness and stability.
Dynamic modeling of selective catalytic reduction(SCR) denitrification system is of great significance for optimization of ammonia injection, reducing of the NOx emission and ammonia slip. The modeling methods based on machine learning have advantages in accuracy, but most methods lack self-learning or adaptive strategy, and are difficult to keep good performance in long-term operation. In order to adapt to the dynamic characteristics of SCR denitrification system caused by load and coal quality change, selective ensemble model library was proposed including a construction method of library based on time-phased data, a combined strategy based on selective real-time error weighting method and an updating strategy based on model evaluation method. The model was trained, tested and verified by a 50-day operation data of SCR system in a 660 MW coal-fired boiler, and compared with the traditional adaptive model. When the traditional model failed, the proposed model can still keep high prediction accuracy. The results show that selective ensemble model library performs better in prediction accuracy, robustness and stability.
引文
[1]毛剑宏．大型电站锅炉SCR烟气脱硝系统关键技术研究[D]．杭州：浙江大学，2011.Mao Jianhong.Key technology study of SCR flue gas De NOx system in large power plant boiler[D].Hangzhou:Zhejiang University,2011(in Chinese).
[2]赵乾．SCR烟气脱硝系统模拟优化及喷氨量最优控制[D]．重庆：重庆大学，2012.Zhao Qian.Simulation optimization and optimal control on spraying ammonia of SCR flue gas denitrification[D].Chongqing:Chongqing University,2012(in Chinese).
[3]刘吉臻，秦天牧，杨婷婷，等．基于自适应多尺度核偏最小二乘的SCR烟气脱硝系统建模[J]．中国电机工程学报，2015,35(23):6083-6088.Liu Jizhen,Qin Tianmu,Yang Tingting,et al.SCR denitration system modeling based on self-adaptive multi-scale kernel partial least squares[J].Proceedings of the CSEE,2015,35(23):6083-6088(in Chinese).
[4]秦天牧，林道鸿，杨婷婷，等．SCR烟气脱硝系统动态建模方法比较[J]．中国电机工程学报，2017,37(10):2913-2919.Qin Tianmu,Lin Daohong,Yang Tingting,et al.Comparative study on dynamic modeling methods of SCR denitration system[J].Proceedings of the CSEE,2017,37(10):2913-2919(in Chinese).
[5]周洪煜，赵乾，张振华，等．烟气脱硝喷氨量SA-RBF神经网络最优控制[J]．控制工程，2012,19(6):947-951.Zhou Hongyu,Zhao Qian,Zhang Zhenghua,et al.Sensitivity analysis radial basis function neural network control on spraying ammonia flow denitrification[J].Control Engineering of China,2012,19(6):947-951(in Chinese).
[6]王建国，彭雅轩，孙灵芳，等．电站锅炉烟气量在线软测量方法[J]．仪器仪表学报，2002,23(S2):756-757.Wang Jiangguo,Peng Yaxuan,Sun Lingfang,et al.A method of on-line soft-measuring for flue gas flow on power station boiler[J].Chinese Journal of Scientific Instrument,2002,23(S2):756-757(in Chinese).
[7]廖永进，陆继东，黄秋雄，等．选择性催化还原服役后催化剂活性及动力学实验研究[J]．中国电机工程学报，2013,33(17):37-44.Liao Yongjin,Lu Jidong,Huang Qiuxiong,et al.Experimental study on activity and kinetics of served SCR catalysts[J].Proceedings of the CSEE,2013,33(17):37-44(in Chinese).
[8] Khamassi I,Sayed-Mouchaweh M,Hammami M,et al.Discussion and review on evolving data streams and concept drift adapting[J].Evolving Systems,2018,9(1):1-23.
[9] Tresp V, Taniguchi M. Combining estimators using non-constant weighting functions[C]//Proceedings of the7th International Conference on Neural Information Processing Systems.Denver,Colorado:ACM,1995:419-426.
[10]刘天舒．BP神经网络的改进研究及应用[D]．哈尔滨：东北农业大学，2011.Liu Tianshu.The research and application on BP neural network improvement[D].Harbin:Northeast Agricultural University,2011(in Chinese).
[11]陶志富，葛璐璐，陈华友．基于滑动窗口的一类非负可变权组合预测方法[J/OL]．控制与决策，[2019-01-26].https://doi.org/10.13195/j.kzyjc.2018.1348.Tao Zhifu,Ge Lulu,Chen Huayou.A new non-negative variable weight combination forecasting method based on sliding window[J].Control and Decision,[2019-01-26].https://doi.org/10.13195/j.kzyjc.2018.1348(in Chinese).
[12]许冠英，韩萌，王少峰，等．数据流集成分类算法综述[J/OL]．计算机应用研究，[2019-04-02].https://doi.org/1019734/j.issn.1001-3695.2018.09.0510.Xu Guanying,Han Meng,Wang Shaofeng,et al.Summarization of data stream ensemble classification algorithm[J]. Application Research of Computers,[2019-04-02]. https://doi.org/10.19734/j.issn.1001-3695.2018.09.0510(in Chinese).
[13]唐伟，周志华．基于Bagging的选择性聚类集成[J]．软件学报，2005,16(4):496-502.Tang Wei,Zhou Zhihua.Bagging-based selective clusterer ensemble[J].Journal of Software,2005,16(4):496-502(in Chinese).
[14] Collobert R,Bengio S,Bengio Y.A parallel mixture of SVMs for very large scale problems[J]. Neural Computation,2002,14(5):1105-1114.
[15]张春霞，张讲社．选择性集成学习算法综述[J]．计算机学报，2011,34(8):1399-1410.Zhang Chunxia,Zhang Jiangshe.A survey of selective ensemble learning algorithms[J]. Chinese Journal of Computers.2011,34(8):1399-1410(in Chinese).
[16] Zhou Zhihua,Wu Jianxin,Tang Wei.Ensembling neural networks:Many could be better than all[J].Artificial Intelligence,2002,137(1-2):239-263.
[17]罗会兰，孔繁胜，李一啸．聚类集成中的差异性度量研究[J]．计算机学报，2007,30(8):1315-1324.Luo Huilan,Kong Fansheng,Li Yixiao.An analysis of diversity measures in clustering ensembles[J].Chinese Journal of Computers, 2007, 30(8):1315-1324(in Chinese).
[18]汤健，柴天佑，丛秋梅，等．基于EMD和选择性集成学习算法的磨机负荷参数软测量[J]．自动化学报，2014,40(9):1853-1866.Tang Jian,Chai Tianyou,Cong Qiumei,et al.Soft sensor approach for modeling mill load parameters based on EMD and selective ensemble learning algorithm[J].Acta Automatica Sinica,2014,40(9):1853-1866(in Chinese).
[19]李勇，刘战东，张海军．不平衡数据的集成分类算法综述[J]．计算机应用研究，2014,31(5):1287-1291.Li Yong,Liu Zhandong,Zhang Haijun.Review on ensemble algorithms for imbalanced data classification[J].Application Research of Computers,2014,31(5):1287-1291(in Chinese).
[20]潘吴斌，程光，郭晓军，等．基于信息熵的自适应网络流概念漂移分类方法[J]．计算机学报，2017,40(7):1556-1571.Pan Wubin,Cheng Guang,Guo Xiaojun,et al.An adaptive classification approach based on information entropy for network traffic in presence of concept drift[J].Chinese Journal of Computers,2017,40(7):1556-1571(in Chinese).
[21]汤健，柴天佑，刘卓，等．基于更新样本智能识别算法的自适应集成建模[J]．自动化学报，2016,42(7):1040-1052.Tang Jian,Chai Tianyou,Liu Zhuo,et al.Adaptive ensemble modelling approach based on updating sample intelligent identification[J].Acta Automatica Sinica,2016,42(7):1040-1052(in Chinese).
[2]赵乾．SCR烟气脱硝系统模拟优化及喷氨量最优控制[D]．重庆：重庆大学，2012.Zhao Qian.Simulation optimization and optimal control on spraying ammonia of SCR flue gas denitrification[D].Chongqing:Chongqing University,2012(in Chinese).
[3]刘吉臻，秦天牧，杨婷婷，等．基于自适应多尺度核偏最小二乘的SCR烟气脱硝系统建模[J]．中国电机工程学报，2015,35(23):6083-6088.Liu Jizhen,Qin Tianmu,Yang Tingting,et al.SCR denitration system modeling based on self-adaptive multi-scale kernel partial least squares[J].Proceedings of the CSEE,2015,35(23):6083-6088(in Chinese).
[4]秦天牧，林道鸿，杨婷婷，等．SCR烟气脱硝系统动态建模方法比较[J]．中国电机工程学报，2017,37(10):2913-2919.Qin Tianmu,Lin Daohong,Yang Tingting,et al.Comparative study on dynamic modeling methods of SCR denitration system[J].Proceedings of the CSEE,2017,37(10):2913-2919(in Chinese).
[5]周洪煜，赵乾，张振华，等．烟气脱硝喷氨量SA-RBF神经网络最优控制[J]．控制工程，2012,19(6):947-951.Zhou Hongyu,Zhao Qian,Zhang Zhenghua,et al.Sensitivity analysis radial basis function neural network control on spraying ammonia flow denitrification[J].Control Engineering of China,2012,19(6):947-951(in Chinese).
[6]王建国，彭雅轩，孙灵芳，等．电站锅炉烟气量在线软测量方法[J]．仪器仪表学报，2002,23(S2):756-757.Wang Jiangguo,Peng Yaxuan,Sun Lingfang,et al.A method of on-line soft-measuring for flue gas flow on power station boiler[J].Chinese Journal of Scientific Instrument,2002,23(S2):756-757(in Chinese).
[7]廖永进，陆继东，黄秋雄，等．选择性催化还原服役后催化剂活性及动力学实验研究[J]．中国电机工程学报，2013,33(17):37-44.Liao Yongjin,Lu Jidong,Huang Qiuxiong,et al.Experimental study on activity and kinetics of served SCR catalysts[J].Proceedings of the CSEE,2013,33(17):37-44(in Chinese).
[8] Khamassi I,Sayed-Mouchaweh M,Hammami M,et al.Discussion and review on evolving data streams and concept drift adapting[J].Evolving Systems,2018,9(1):1-23.
[9] Tresp V, Taniguchi M. Combining estimators using non-constant weighting functions[C]//Proceedings of the7th International Conference on Neural Information Processing Systems.Denver,Colorado:ACM,1995:419-426.
[10]刘天舒．BP神经网络的改进研究及应用[D]．哈尔滨：东北农业大学，2011.Liu Tianshu.The research and application on BP neural network improvement[D].Harbin:Northeast Agricultural University,2011(in Chinese).
[11]陶志富，葛璐璐，陈华友．基于滑动窗口的一类非负可变权组合预测方法[J/OL]．控制与决策，[2019-01-26].https://doi.org/10.13195/j.kzyjc.2018.1348.Tao Zhifu,Ge Lulu,Chen Huayou.A new non-negative variable weight combination forecasting method based on sliding window[J].Control and Decision,[2019-01-26].https://doi.org/10.13195/j.kzyjc.2018.1348(in Chinese).
[12]许冠英，韩萌，王少峰，等．数据流集成分类算法综述[J/OL]．计算机应用研究，[2019-04-02].https://doi.org/1019734/j.issn.1001-3695.2018.09.0510.Xu Guanying,Han Meng,Wang Shaofeng,et al.Summarization of data stream ensemble classification algorithm[J]. Application Research of Computers,[2019-04-02]. https://doi.org/10.19734/j.issn.1001-3695.2018.09.0510(in Chinese).
[13]唐伟，周志华．基于Bagging的选择性聚类集成[J]．软件学报，2005,16(4):496-502.Tang Wei,Zhou Zhihua.Bagging-based selective clusterer ensemble[J].Journal of Software,2005,16(4):496-502(in Chinese).
[14] Collobert R,Bengio S,Bengio Y.A parallel mixture of SVMs for very large scale problems[J]. Neural Computation,2002,14(5):1105-1114.
[15]张春霞，张讲社．选择性集成学习算法综述[J]．计算机学报，2011,34(8):1399-1410.Zhang Chunxia,Zhang Jiangshe.A survey of selective ensemble learning algorithms[J]. Chinese Journal of Computers.2011,34(8):1399-1410(in Chinese).
[16] Zhou Zhihua,Wu Jianxin,Tang Wei.Ensembling neural networks:Many could be better than all[J].Artificial Intelligence,2002,137(1-2):239-263.
[17]罗会兰，孔繁胜，李一啸．聚类集成中的差异性度量研究[J]．计算机学报，2007,30(8):1315-1324.Luo Huilan,Kong Fansheng,Li Yixiao.An analysis of diversity measures in clustering ensembles[J].Chinese Journal of Computers, 2007, 30(8):1315-1324(in Chinese).
[18]汤健，柴天佑，丛秋梅，等．基于EMD和选择性集成学习算法的磨机负荷参数软测量[J]．自动化学报，2014,40(9):1853-1866.Tang Jian,Chai Tianyou,Cong Qiumei,et al.Soft sensor approach for modeling mill load parameters based on EMD and selective ensemble learning algorithm[J].Acta Automatica Sinica,2014,40(9):1853-1866(in Chinese).
[19]李勇，刘战东，张海军．不平衡数据的集成分类算法综述[J]．计算机应用研究，2014,31(5):1287-1291.Li Yong,Liu Zhandong,Zhang Haijun.Review on ensemble algorithms for imbalanced data classification[J].Application Research of Computers,2014,31(5):1287-1291(in Chinese).
[20]潘吴斌，程光，郭晓军，等．基于信息熵的自适应网络流概念漂移分类方法[J]．计算机学报，2017,40(7):1556-1571.Pan Wubin,Cheng Guang,Guo Xiaojun,et al.An adaptive classification approach based on information entropy for network traffic in presence of concept drift[J].Chinese Journal of Computers,2017,40(7):1556-1571(in Chinese).
[21]汤健，柴天佑，刘卓，等．基于更新样本智能识别算法的自适应集成建模[J]．自动化学报，2016,42(7):1040-1052.Tang Jian,Chai Tianyou,Liu Zhuo,et al.Adaptive ensemble modelling approach based on updating sample intelligent identification[J].Acta Automatica Sinica,2016,42(7):1040-1052(in Chinese).
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |