基于深度强化学习的火电机组制粉系统自启停智能决策Intelligent decision-making of start-up and shutdown for coal milling system in thermal power plants based on deep reinforcement learning
蔡佳辰,李军,高明,高林,高耀岿,昌鹏
摘要(Abstract):
针对目前人工决策基础上的制粉系统一键启停技术存在决策主观经验性强、操盘劳动强度大、节能优化潜力难以发掘等问题,提出了一种综合考虑制粉系统能耗与机组负荷跟踪性能的制粉系统启停决策评价模型,以安全引入网调负荷计划指令信号作为输入,研究了基于深度强化学习的制粉系统自启停智能决策方法,开发了制粉系统自启停决策闭环控制系统。研究结果通过仿真验证,并已在某超超临界1 000 MW机组常用磨煤机上成功应用,节能降耗效果显著。研究结果可为火电机组少人、无人化运行技术提供有效借鉴。
关键词(KeyWords): 深度强化学习;制粉系统;自启停控制;智能决策
基金项目(Foundation): 国家重点研发计划项目(2022YFB4100700);; 陕西省重点研发计划项目(2023-YBGY-274)~~
作者(Author): 蔡佳辰,李军,高明,高林,高耀岿,昌鹏
DOI: 10.19666/j.rlfd.202307118
参考文献(References):
- [1]刘吉臻,胡勇,曾德良,等.智能发电厂的架构及特征[J].中国电机工程学报,2017,37(22):6463-6470.LIU Jizhen,HU Yong,ZENG Deliang,et al.Architecture and feature of smart power generation[J].Proceedings of the CSEE,2017,37(22):6463-6470.
- [2]华志刚,郭荣,崔希,等.火电智慧电厂技术路线探讨与研究[J].热力发电,2019,48(10):8-14.HUA Zhigang,GUO Rong,CUI Xi,et al.Discussion and study on technical route of smart thermal power plant[J].Thermal Power Generation,2019,48(10):8-14.
- [3]方彦军,贺瑶,秦晓洁.直流锅炉快速变负荷磨煤机启停优化[J].热力发电,2013,42(9):10-15.FANG Yanjun,HE Yao,QIN Xiaojie.Optimization of start and stop of coal mills during once through boiler’s fast load varying[J].Thermal Power Generation,2013,42(9):10-15.
- [4]HUANG Y,CHEN Q,ZHANG L,et al.Dynamic optimization of variable load process for combined heat and power unit based on sequential quadratic programming and interior point method alternating solution method[J].Processes,2023,11(6):311-314.
- [5]赵俊杰,冯树臣,田景奇,等.基于燃煤智能发电ICS的主辅机一键启停APS技术应用分析[J].能源科技,2021,19(1):41-45.ZHAO Junjie,FENG Shuchen,TIAN Jingqi,et al.Application analysis of one-button start-up&shut-down technology of aps for main and auxiliary machines based on ICS for coal-fired power generation[J].Energy Technology,2021,19(1):41-45.
- [6]於晓博.大型火电机组一键启停方案设计研究[D].保定:华北电力大学(保定),2017:1.YU Xiaobo.Design and research of one-key start and stop scheme for large thermal power unit[D].Baoding:North China Electric Power University(Baoding),2017:1.
- [7]王志杰,彭梁,朱晓星,等.基于专家知识库的双进双出制粉系统自启停技术[J].中国电力,2018,51(1):133-138.WANG Zhijie,PENG Liang,ZHU Xiaoxing,et al.The automatic startup and shutdown technology for double inlet and outlet coal-pulverizing system based on expert knowledge base[J].Electric Power,2018,51(1):133-138.
- [8]李晓燕,郑卫东,熊志成,等.基于柔性顺控的制粉系统一键启停控制技术在1 000 MW超超临界火电机组的实践与应用[J].浙江电力,2019,38(12):39-43.LI Xiaoyan,ZHENG Weidong,XIONG Zhicheng,et al.Application and practice of one-click start-stop control technology for pulverized coal preparation system based on flexible sequential control in 1 000 MW ultrasupercritical thermal power generating units[J].Zhejiang Electric Power,2019,38(12):39-43.
- [9]CHEN X C,YAO L N,MCAULEY J L,et al.Deep reinforcement learning in recommender systems:a survey and new perspectives[J].Knowledge-Based Systems,2023,264:1-19.
- [10]刘全,翟建伟,章宗长,等.深度强化学习综述[J].计算机学报,2018,41(1):1-27.LIU Quan,ZHAI Jianwei,ZHANG Zongchang,et al.Asurvey on reinforcement learning[J].Chinese Journal of Computers,2018,41(1):1-27.
- [11]王永志.基于深度强化学习的机器人抓取及智能装配研究[D].沈阳:沈阳工业大学,2022:1.WANG Yongzhi.Research on robotic grasping and intelligent assembly based on deep reinforcement learning[D].Shenyang:Shenyang University of Technology,2022:1.
- [12]董瑶,葛莹莹,郭鸿湧,等.基于深度强化学习的移动机器人路径规划[J].计算机工程与应用,2019,55(13):15-19.DONG Yao,GE Yingying,GUO Hongyong,et al.Path planning for mobile robot based on deep reinforcement learning[J].Computer Engineering and Applications,2019,55(13):15-19.
- [13]王珂,姚建国,余佩遥,等.基于深度强化学习的电网前瞻调度智能决策架构及关键技术初探[J].中国电机工程学报,2022,42(15):5430-5439.WANG Ke,YAO Jianguo,YU Peiyao,et al.Architecture and key technologies of intelligent decision-making of power grid look-ahead dispatch based on deep reinforcement learning[J].Proceedings of the CSEE,2022,42(15):5430-5439.
- [14]梁涛,刘伟,曹欣,等.基于深度确定性策略梯度算法的可再生能源大规模制氢系统能量调度[J/OL].电网技术:1-16.(2023-04-11)[2023-07-23].https://doi.org/10.13335/j.1000-3673.pst.2023.0211.LIANG Tao,LIU Wei,CAO Xin,et al.Research on energy scheduling of renewable energy large-scale hydrogen production system based on deep deterministic strategy gradient algorithm[J/OL].Power System Technology,1-16.(2023-04-11)[2023-07-23].https://doi.org/10.13335/j.1000-3673.pst.2023.0211.
- [15]谭天宇.计及经济性与快速性的厂级负荷优化分配研究[D].北京:华北电力大学,2019:1.TAN Tianyu.Research on optimal load dispatch of thermal power plant considering economy and speediness[D].Beijing:North China Electric Power University,2019:1.
- [16]张凤军,戴国忠,彭晓兰.虚拟现实的人机交互综述[J].中国科学:信息科学,2016,46(12):1711-1736.ZHANG Fengjun,DAI Guozhong,PENG Xiaolan.Asurvey on human-computer interaction in virtual reality[J].Scientia Sinica (Informationis),2016,46(12):1711-1736.
- [17]WANG Y H,CHEN L,ZHOU H,et al.Flexible transmission network expansion planning based on DQNalgorithm[J].Energies,2021,14(7):1-21.
- [18]董永峰,杨琛,董瑶,等.基于改进的DQN机器人路径规划[J].计算机工程与设计,2021,42(2):552-558.DONG Yongfeng,YANG Chen,DONG Yao,et al.Robot path planning based on improved DQN[J].Computer Engineering and Design,2021,42(2):552-558.
- [19]刘建伟,高峰,罗雄麟.基于值函数和策略梯度的深度强化学习综述[J].计算机学报,2019,42(6):1406-1438.LIU Jianwei,GAO Feng,LUO Xionglin.Survey of deep reinforcement learning based on value function and policy gradient[J].Chinese Journal of Computers,2019,42(6):1406-1438.
- [20]刘朝阳,穆朝絮,孙长银.深度强化学习算法与应用研究现状综述[J].智能科学与技术学报,2020,2(4):314-326.LIU Zhaoyang,MU Chaoxu,SUN Changyin.An overview on algorithms and applications of deep reinforcement learning[J].Chinese Journal of Intelligent Science and Technology,2020,2(4):314-326.
- [21]HOMAYUN K,LUEN M T,CHOON Y C.Twin delayed DDPG based dynamic power allocation for mobility in Io RT[J].Journal of Communications Software and Systems,2023,19(1):19-29.
- [22]LI X,WU G,SUN Z,et al.Research on multi-agent D2Dcommunication resource allocation algorithm based on A2C[J].Electronics,2023,12(2):360-360.
- [23]GERPOTT F,LANG S,REGGELIN T,et al.Integration of the A2C algorithm for production scheduling in a two-stage hybrid flow shop environment[J].Procedia Computer Science,2022,200:585-594.
- [24]杨新民,高海东,陈丰.数字化电厂概念的解析及探讨[J].热力发电,2015,44(5):98-101.YANG Xinmin,GAO Haidong,CHEN Feng.Discussions on concept of digital power plant[J].Thermal Power Generation,2015,44(5):98-101.
- [25]高耀岿,王林,高海东,等.火电厂智能控制系统体系架构及关键技术[J].热力发电,2022,51(3):166-174.GAO Yaokui,WANG Lin,GAO Haidong,et al.Architecture and key technology of intelligent control system in thermal power plant[J].Thermal Power Generation,2022,51(3):166-174.
- [26]马天霆.磨煤机模糊与模糊PI复合控制系统[J].热力发电,2015,44(10):52-57.MA Tianting.A coal mill control system combining fuzzy control with fuzzy PID control[J].Thermal Power Generation,2015,44(10):52-57.