Please wait a minute...
Journal of ZheJiang University (Engineering Science)  2023, Vol. 57 Issue (7): 1428-1438    DOI: 10.3785/j.issn.1008-973X.2023.07.018
    
Optimization of boiler real-time operation based on pattern-matching of agent model
Wei ZHONG1,2(),Xue-ru LIN1,2,Xiao-jie LIN1,3,*(),Yi ZHOU1
1. Key Laboratory of Clean Energy and Carbon Neutrality of Zhejiang Province, Zhejiang University, Hangzhou 310027, China
2. College of Energy Engineering, Zhejiang University, Hangzhou 310027, China
3. Jiaxing Research Institute, Zhejiang University, Jiaxing 314024, China
Download: HTML     PDF(1664KB) HTML
Export: BibTeX | EndNote (RIS)      

Abstract  

A novel framework for modeling coal-fired power plant boiler operations was proposed based on pattern matching with an agent model (PMAM) in order to enhance the effectiveness and real-time performance. A new method for calculating the lag of the main steam flow rate was proposed. An improved pattern-matching optimization model was introduced to calculate the optimal operational database for historical optimization. A three-level scheme optimization mechanism was incorporated in order to ensure the effectiveness of the pattern-matching approach. The mechanism includes attention parameters, state parameter interval frequency and regulation minimum. An agent model for boiler operation optimization was constructed offline by using a neural network algorithm, and pattern-matching steps were represented based on the agent model to enable online applications. The case results show that the proposed pattern-matching optimization model can effectively find the optimized boiler operation scheme, and the similarity of working conditions is more than 95%, which can improve the boiler efficiency by 1.92% in practice. The mean square error of the trained agent model is less than 0.35%. The method avoids the influence of generalization error caused by optimization solutions compared with traditional methods, and has high reliability and real-time performance while improving boiler efficiency.



Key wordsfuzzy C-means clustering      data driven      pattern matching      operation optimization      online optimization      power plant boiler     
Received: 09 August 2022      Published: 17 July 2023
CLC:  TK 227  
Fund:  国家重点研发计划资助项目(2019YFE0126000);国家自然科学基金资助项目(51806190)
Corresponding Authors: Xiao-jie LIN     E-mail: zhongw@zju.edu.cn;xiaojie.lin@zju.edu.cn
Cite this article:

Wei ZHONG,Xue-ru LIN,Xiao-jie LIN,Yi ZHOU. Optimization of boiler real-time operation based on pattern-matching of agent model. Journal of ZheJiang University (Engineering Science), 2023, 57(7): 1428-1438.

URL:

https://www.zjujournals.com/eng/10.3785/j.issn.1008-973X.2023.07.018     OR     https://www.zjujournals.com/eng/Y2023/V57/I7/1428


基于代理模型模式匹配的锅炉实时操作优化

为了提高电厂燃煤锅炉操作优化的有效性与实时性,提出新的基于代理模型模式匹配(PMAM)的建模框架. 提出主蒸汽流量的滞后性计算方法. 采用改进的模式匹配优化模型,计算历史优化操作库. 引入工况注意力机制参数、状态参数区间频率法及调控最小的3层方案优化机制,确保模式匹配方案的有效性. 采用神经网络算法预建模构建锅炉操作优化的代理模型,基于代理模型表征模式匹配步骤,使得本文方法可以适用于在线应用. 案例结果表明,利用提出的模式匹配优化模型,能够有效地寻找优化的锅炉操作方案,工况相似度大于95%,可以使得锅炉效率提升1.92%;训练的代理模型均方误差小于0.35%. 与传统方法相比,本文方法避免了优化求解带来的泛化误差影响,在提升锅炉效率的同时,具有高可靠性及实时性.


关键词: 模糊C均值聚类,  数据驱动,  模式匹配,  操作优化,  在线优化,  电厂锅炉 
Fig.1 Boiler real-time operation optimization calculation framework based on pattern matching of agent model
Fig.2 Schematic diagram of data reconstruction in hysteresis calculation
序号 参数类型 参数
1 工况参数 主蒸汽质量流量qmsteam/(t·h?1)
2 主蒸汽温度θsteam/℃
3 主蒸汽压力psteam/MPa
4 应用基水分Cwat/%
5 应用基灰分Cash/%
6 应用基挥发分Cvdaf/%
7 收到基低位发热量Qnet/(kJ·kg?1)
8 给水温度θwater/℃
9 目标参数 锅炉效率Eboiler/%
10 关键状态参数 炉膛温度θfurnace/℃
11 烟气含氧体积分数φoxygen/%
12 一二次风比例Rair
13 一次风体积流量qVprimaryAir/(m3·h?1)
14 二次风体积流量qVsecondaryAir/(m3·h?1)
15 操作参数 1#给煤机频率f1#GMJ/Hz
16 2#给煤机频率f2#GMJ/Hz
17 3#给煤机频率f3#GMJ/Hz
18 4#给煤机频率f4#GMJ/Hz
19 1#引风机频率f1#YFJ/Hz
20 2#引风机频率f2#YFJ/Hz
21 一次风机频率f1#SFJ/Hz
22 二次风机频率f2#SFJ/Hz
Tab.1 Table of parameters related to boiler operation optimization calculation
Fig.3 Normal operation data of 1 # boiler efficiency of thermal power plant
Fig.4 Curve between main steam mass flow rate and frequency of coal feeder
Fig.5 Hysteresis of main steam mass flow rate relative to coal feeder frequency
c m $ {{\varepsilon }} $ L lr
3 2 10 8.101 22
3 3 10 8.820 22
3 4 10 9.548 38
3 5 10 9.941 85
Tab.2 Influence of fuzzy index on error value and iteration number of clustering objective
c m $ {{\varepsilon }} $ L lr
3 2 10 8.101 22
4 2 10 9.086 81
5 2 10 9.576 19
6 2 10 9.648 23
Tab.3 Influence of clustering numbers on final error value and iteration times
Fig.6 Clustering effect of different cluster number
c m $ {{\varepsilon }} $ L lr
3 2 0.01 0.009 47
3 2 0.1 0.081 39
3 2 1 0.927 30
3 2 10 8.101 22
Tab.4 Influence of update iteration thresholds on final error and iteration times
工况 qmsteam/(t·h?1) Sxm/% Sxo/% EdeBoiler/% ftotalGM/Hz ftotalSF/Hz ftotalYF/Hz
工况1(原始) 191.16 79.30 58.44 58.23
工况1(优化后) 199.06 98.28 97.27 1.92 78.22 58.51 59.45
工况1(仅考虑工况相似) 197.73 98.83 96.19 2.10 79.74 57.66 58.24
工况1(仅考虑工况相似与操作相似) 189.96 99.84 99.89 0.06 79.19 58.42 58.24
工况2(原始) 170.66 71.39 56.01 52.98
工况2(优化后) 162.98 98.47 95.14 0.41 67.23 54.01 51.96
工况3(原始) 140.58 55.29 49.97 42.28
工况3(优化后) 140.25 99.44 95.13 0.64 55.29 49.97 42.28
工况4(原始) 123.29 52.55 46.14 39.46
工况4(优化后) 120.31 97.00 93.76 0.22 52.15 45.73 39.68
工况5(原始) 103.55 47.47 45.77 39.30
工况5(优化后) 111.00 98.00 95.82 0.93 46.91 44.95 38.67
工况6(原始) 192.68 78.25 59.84 58.24
工况6(优化后) 201.55 97.40 94.46 0.73 74.23 61.74 63.34
工况7(原始) 194.62 78.06 61.89 62.31
工况7(优化后) 199.60 97.99 96.92 0.60 74.13 61.76 63.33
工况8(原始) 195.16 83.82 61.10 61.51
工况8(优化后) 201.70 97.51 92.63 0.76 74.20 61.74 63.34
工况9(原始) 201.01 82.22 60.30 61.51
工况9(优化后) 199.60 99.09 93.32 0.66 74.13 61.76 63.33
Tab.5 Optimization results of pattern matching under multiple working conditions-1# boiler
Fig.7 Batch optimization results of pattern-matching method
Fig.8 Neural network structure diagram adopted by PMAM agent model
Fig.9 Error results of optimal agent model
序号 参数 e/%
第1次 第2次 第3次 第4次 第5次 第6次 第7次 第8次 第9次 第10次
1 Eboiler 0.33 0.36 0.37 0.33 0.31 0.29 0.38 0.32 0.35 0.33
2 θfurnace 0.16 0.16 0.16 0.16 0.16 0.16 0.16 0.16 0.17 0.16
3 φoxygen 0.20 0.20 0.20 0.21 0.21 0.20 0.20 0.20 0.20 0.20
4 Rair 0.18 0.18 0.18 0.19 0.17 0.18 0.19 0.17 0.17 0.16
5 qVprimaryAir 0.18 0.19 0.19 0.19 0.19 0.19 0.18 0.19 0.18 0.18
6 qVsecondaryAir 0.35 0.35 0.36 0.35 0.34 0.35 0.33 0.35 0.33 0.34
7 f1#GMJ 0.19 0.19 0.19 0.19 0.19 0.20 0.20 0.19 0.20 0.19
8 f2#GMJ 0.24 0.24 0.25 0.24 0.25 0.24 0.23 0.24 0.24 0.23
9 f3#GMJ 0.24 0.24 0.25 0.24 0.24 0.24 0.24 0.24 0.24 0.24
10 f4#GMJ 0.20 0.21 0.21 0.22 0.20 0.20 0.20 0.22 0.21 0.21
11 f1#YFJ 0.15 0.14 0.15 0.15 0.15 0.15 0.15 0.15 0.16 0.15
12 f2#YFJ 0.15 0.15 0.16 0.16 0.16 0.16 0.16 0.16 0.16 0.15
13 f1#SFJ 0.18 0.18 0.19 0.19 0.18 0.19 0.18 0.18 0.18 0.18
14 f2#SFJ 0.21 0.22 0.20 0.20 0.21 0.22 0.21 0.21 0.21 0.21
Tab.6 Cross validation-error percentage of test set
数据源 Eboiler/% f1#GMJ/Hz f2#GMJ/Hz f3#GMJ/Hz f4#GMJ/Hz f1#SFJ/Hz f2#SFJ/Hz f1#YFJ/Hz f2#YFJ/Hz tc/s
原始数据 90.22 18.54 17.08 23.05 20.63 36.05 22.39 29.09 29.14
由模式匹配计算 91.95 18.46 17.88 22.17 19.71 35.51 23.00 29.72 29.73 12.00
由代理模型计算 91.56 17.90 17.20 21.99 19.23 35.63 23.00 29.38 29.05 2.00
Tab.7 Examples of online application results
Fig.10 Calculation time of different methods
Fig.11 Online application results of PMAM method under different loads
Fig.12 Comparison between PMAM and regression model
工况 qmsteam/(t·h?1) Sxm/% Sxo/% EdeBoiler/% ftotalGM/Hz ftotalSF/Hz ftotalYF/Hz
工况1(原始) 113.57 40.69 48.87 41.42
工况1(优化后) 119.95 97.57 97.57 0.65 41.42 50.29 40.61
工况2(原始) 133.93 46.12 52.96 45.06
工况2(优化后) 131.23 98.88 98.73 1.24 46.26 51.77 44.25
工况3(原始) 144.49 50.14 53.93 47.06
工况3(优化后) 141.24 98.67 99.86 0.04 50.02 53.92 47.04
工况4(原始) 158.02 55.72 55.68 52.96
工况4(优化后) 157.98 99.68 99.80 0.05 55.72 55.68 52.96
工况5(原始) 181.26 62.85 61.92 62.81
工况5(优化后) 189.13 97.61 99.13 0.77 64.41 62.57 62.02
工况6(原始) 197.58 70.14 66.28 65.64
工况6(优化后) 198.51 99.60 99.91 0.05 70.03 66.30 65.65
Tab.8 Optimization results of pattern matching under multiple working conditions-2#boiler
[1]   樊宇琦, 丁涛, 孙瑜歌, 等 国内外促进可再生能源消纳的电力现货市场发展综述与思考[J]. 中国电机工程学报, 2021, 41 (5): 1729- 1752
FAN Yu-qi, DING Tao, SUN Yu-ge, et al Review and cogitation for worldwide spot market development to promote renewable energy accommodation[J]. Proceedings of the Chinese Society of Electrical Engineering, 2021, 41 (5): 1729- 1752
[2]   刘连德, 何江, 周家旭, 等 含高比例风光发电的电力系统中抽蓄电站的优化控制策略[J]. 储能科学与技术, 2022, 11 (7): 2197- 2205
LIU Lian-de, HE Jiang, ZHOU Jia-xu, et al Optimization control strategy of pumped storage station in power system with high proportion wind/photovoltaic power[J]. Energy Storage Science and Technology, 2022, 11 (7): 2197- 2205
[3]   李晖, 刘栋, 姚丹阳 面向碳达峰碳中和目标的我国电力系统发展研判[J]. 中国电机工程学报, 2021, 41 (18): 6245- 6259
LI Hui, LIU Dong, YAO Dan-yang Analysis and reflection on the development of power system towards the goal of carbon emission peak and carbon neutrality[J]. Proceedings of the Chinese Society of Electrical Engineering, 2021, 41 (18): 6245- 6259
[4]   韩肖清, 李廷钧, 张东霞, 等 双碳目标下的新型电力系统规划新问题及关键技术[J]. 高电压技术, 2021, 47 (9): 3036- 3046
HAN Xiao-qing, LI Ting-jun, ZHANG Dong-xia, et al New issues and key technologies of new power system planning under double carbon goals[J]. High Voltage Engineering, 2021, 47 (9): 3036- 3046
[5]   国家能源局. 国家能源局发布1-5月份全国电力工业统计数据 [EB/OL]. (2022-06-16) [2022-11-01]. http://www.nea.gov.cn/2022-06/16/c_1310624653.htm.
National Energy Administration. National Energy Administration released statistical data of national power industry from January to May [EB/OL]. (2022-06-16) [2022-11-01]. http://www.nea.gov.cn/2022-06/16/c_1310624653.htm.
[6]   于晨龙. 混合多目标优化方法研究及在电厂锅炉负荷分配中的应用 [D]. 杭州: 浙江大学, 2013.
YU Chen-long. Studies on hybrid multi-objective optimization method and its applications to boiler load distribution in power plant [D]. Hangzhou: Zhejiang University, 2013.
[7]   姜忠生 热电厂锅炉-汽轮机的优化调度[J]. 山东工业技术, 2015, 190 (8): 44
JIANG Zhong-sheng Optimal scheduling of boiler-turbine in thermal power plant[J]. Shandong Industrial Technology, 2015, 190 (8): 44
[8]   夏燕萍, 吕泽华, 赵士杭 热电厂锅炉-汽轮机的优化调度[J]. 清华大学学报: 自然科学版, 2000, 40 (11): 59- 66
XIA Yan-ping, LV Ze-hua, ZHAO Shi-hang Load assignment optimization in combined heat and power plant[J]. Journal of Tsinghua University: Science and Technology, 2000, 40 (11): 59- 66
[9]   曹英华. 济南北郊热电厂热电机组与热水锅炉供热优化运行调度 [D]. 哈尔滨: 哈尔滨工业大学, 2016.
CAO Ying-hua. Optimal operation regulation between cogeneration unit and hot water boiler in Jinan Beijiao thermal power plant [D]. Harbin: Harbin Institute of Technology, 2016.
[10]   SONG Z, KUSIAK A Constraint-based control of boiler efficiency: a data-mining approach[J]. IEEE Transactions on Industrial Informatics, 2007, 3 (1): 73- 83
doi: 10.1109/TII.2006.890530
[11]   贾继宁, 吕勇哉 基于数据驱动与模式挖掘的循环流化床锅炉运行优化研究[J]. 热力发电, 2020, 49 (5): 33- 39
JIA Ji-ning, LV Yong-zai Operation optimization of CFB boiler based on data-driven and pattern mining[J]. Thermal Power Generation, 2020, 49 (5): 33- 39
[12]   孙吉贵, 刘杰, 赵连宇 聚类算法研究[J]. 软件学报, 2008, 19 (1): 48- 61
SUN Ji-gui, LIU Jie, ZHAO Lian-yu Clustering algorithms research[J]. Journal of Software, 2008, 19 (1): 48- 61
doi: 10.3724/SP.J.1001.2008.00048
[13]   吴家标, 刘兴高, 闫正兵 循环流化床锅炉节能的操作优化[J]. 江南大学学报: 自然科学版, 2012, 11 (4): 396- 405
WU Jia-biao, LIU Xing-gao, YAN Zheng-bing Operation optimization of circulating fluidized bed boiler for energy saving[J]. Journal of Jiangnan University: Natural Science Edition, 2012, 11 (4): 396- 405
[14]   TANG Z H, WU X Y, CAO S X Adaptive nonlinear model predictive control of the combustion efficiency under the NOx emissions and load constraints[J]. Energies, 2019, 12 (9): 1738
doi: 10.3390/en12091738
[15]   XU X Y, CHEN Q, REN M F, et al Combustion optimization for coal fired power plant boilers based on improved distributed ELM and distributed PSO[J]. Energies, 2019, 12 (6): 1036
doi: 10.3390/en12061036
[16]   ZHANG W, LIU J Z, GAO M M, et al A novel operation cost optimization system for mix-burning coal slime circulating fluidized bed boiler unit[J]. Applied Thermal Engineering, 2019, 148: 620- 631
doi: 10.1016/j.applthermaleng.2018.11.087
[17]   NIKULA R P, RUUSUNEN M, LEIVISKA K Data-driven framework for boiler performance monitoring[J]. Applied Energy, 2016, 183: 1374- 1388
doi: 10.1016/j.apenergy.2016.09.072
[18]   李传金. 基于机器学习的锅炉多工况优化建模研究 [D]. 大连: 大连海事大学, 2018.
LI Chuan-jin. The research of heating boiler's multi-mode optimal modeling based on machine learning [D]. Dalian: Dalian Maritime University, 2018.
[19]   ZOU L B, CHENG Y, ZHUANG Z W, et al. An optimization control of thermal power combustion based on reinforcement learning [C]// Proceedings of the 37th Chinese Control Conference. Wuhan: IEEE, 2018: 3553-3558.
[20]   JIANG H, CAI Z, ZHANG T, et al. Prediction of boiler combustion energy efficiency via deep reinforcement learning [C]// Proceedings of the 40th Chinese Control Conference. Shanghai: IEEE, 2021: 2658-2662.
[21]   TANG Z H, ZHANG Z J The multi-objective optimization of combustion system operations based on deep data-driven models[J]. Energies, 2019, 182: 37- 47
[22]   李啸晨, 苏宏业, 邵寒山, 等 基于梯度信息的实时优化与控制集成策略[J]. 浙江大学学报: 工学版, 2019, 53 (5): 843- 851
LI Xiao-chen, SU Hong-ye, SHAO Han-shan, et al Gradient information-based strategy for real time optimization and control integration[J]. Journal of Zhejiang University: Engineering Science, 2019, 53 (5): 843- 851
[23]   李达, 林雪茹, 胡城煌, 等 蒸气动力系统的运行优化研究与应用[J]. 自动化仪表, 2018, 39 (8): 40- 46
LI Da, LIN Xue-ru, HU Cheng-huang, et al Research and application on the operation optimization of steam power system[J]. Process Automation Instrumentation, 2018, 39 (8): 40- 46
[24]   ABDI H, WILLIAMS L J Principal component analysis[J]. Wiley Interdisciplinary Reviews: Computational Statistics, 2010, 2 (4): 433- 459
doi: 10.1002/wics.101
[25]   DIDAY E, ESCOUFIER Y. Selected contributions in data analysis and classification [M]. [S. l.]: Springer , 2007: 161-172.
[1] Hai-dong FAN,Zhu CHEN,Zhong-yang ZHAO,Cheng-si LIANG,Cheng-hang ZHENG,Xiang GAO. Operating optimization of oxidation subsystem of wet flue gas desulfurization system of 1 000 MW coal-fired unit[J]. Journal of ZheJiang University (Engineering Science), 2023, 57(4): 675-682.
[2] Xiang-yu KONG,Xiao-bing WANG,Hong-zeng LI,Jia-yu LUO. On-line fault monitoring technology for industrial stationary/nonstationary complex system[J]. Journal of ZheJiang University (Engineering Science), 2021, 55(10): 1856-1866.
[3] Chang CAI,Yi-xiang HUANG,Hong-wen XING. State recognition for fuselage join based on gradient boosting tree[J]. Journal of ZheJiang University (Engineering Science), 2019, 53(7): 1274-1281.
[4] ZHANG Ya nan, CHEN De yun, WANG Ying jie, LIU Yu peng. Incremental graph pattern matching based dynamic recommendation method for cold-start user[J]. Journal of ZheJiang University (Engineering Science), 2017, 51(2): 408-415.
[5] ZHANG Ya nan,QU Ming cheng,LIU Yu peng. Recommendation method based on social topology for cold start users[J]. Journal of ZheJiang University (Engineering Science), 2016, 50(5): 1001-1008.
[6] YE Xiao-wei, ZHANG Xiao-ming, NI Yi-qing, WONG Kai-yuen, FAN Ke-qing. Bridge deflection measurement method based on machine vision technology[J]. Journal of ZheJiang University (Engineering Science), 2014, 48(5): 813-819.
[7] HE Hui-mei, HOU Di-bo, ZHAO Hai-feng, HUANG Ping-jie, ZHANG Guang-xin. Multi-parameters fusion algorithm for detecting anomalous water quality[J]. Journal of ZheJiang University (Engineering Science), 2013, 47(4): 735-740.