综合能源系统中热电联产机组故障预警现状

doi:10.3969/j.issn.2097-0706.2024.08.009

摘要/Abstract

摘要：

热电联产机组作为综合能源系统的重要组成部分之一，不仅承担着电能和热能的生产与传输等环节，也为系统消纳可再生能源提供了基础。风机和磨煤机作为热电联产机组的重要辅机设备，对热电联产机组的正常运行发挥着重要作用。介绍了故障预警技术背景，并对风机和磨煤机的故障类型进行了总结，随后基于人工智能算法将故障预警技术分为机器学习、深度学习和组合模型3种技术路线展开叙述。分析总结了各个技术的发展趋势和核心问题。最后对当前故障预警技术在综合能源系统中的发展应用进行了展望。

关键词: 综合能源系统, 热电联产机组, 机器学习, 组合模型, 深度学习, 故障预警, 可再生能源消纳, 磨煤机

Abstract:

As an important component of an integrated energy system， a CHP unit not only provides electric and thermal power，but also lays a foundation for renewable energy consumption. The fan and coal mill are significant devices for a cogeneration unit and play vital roles in its operation. In the research on fault diagnosis technologies，faults in fans and coal mills are summarized，and subsequently， based on artificial intelligence algorithms，the fault diagnosis technologies are categorized into three technical approaches： machine learning， deep learning， and hybrid models. The development trends and core issues of each technology are analysed. Finally， the prospects of fault diagnosis technologies applying in integrated energy systems are discussed.

Key words: integrated energy system, CHP unit, machine learning, hybrid model, deep learning, fault warning, ccommodation of renewable energy, coal mill

中图分类号:

TK01：TM621

邓振宇, 汪茹康, 徐钢, 云昆, 王颖. 综合能源系统中热电联产机组故障预警现状[J]. 综合智慧能源, 2024, 46(8): 67-76.

DENG Zhenyu, WANG Rukang, XU Gang, YUN Kun, WANG Ying. Current status of fault diagnosis for CHP units in integrated energy systems[J]. Integrated Intelligent Energy, 2024, 46(8): 67-76.

图/表 13

图1

图2

表1

图3

表2

表3

表4

图4

图5

图6

图7

图8

图9

参考文献 46

[1]	李秋燕, 王利利, 张艺涵, 等. 能源互联网多能流的耦合模型及动态优化方法综述[J]. 电力系统保护与控制, 2020, 48(19):179-186.
	LI Qiuyan, WANG Lili, ZHANG Yihan, et al. A review of coupling models and dynamic optimization methods for energy internet multi-energy flow[J]. Power System Protection and Control, 2020, 48(19):179-186.
[2]	严思韵, 王晨, 周登极. 含氢能气网掺混输运的综合能源系统优化研究[J]. 电力工程技术, 2021, 40(1):10-16,49.
	YAN Siyun, WANG Chen, ZHOU Dengji. Optimization of integrated electricity and gas system considering hydrogen-natural-gas mixture transportation[J]. Electric Power Engineering Technology, 2021, 40(1):10-16,49.
[3]	孙宏斌, 潘昭光, 郭庆来. 多能流能量管理研究:挑战与展望[J]. 电力系统自动化, 2016, 40(15):1-8,16.
	SUN Hongbin, PAN Zhaoguang, GUO Qinglai. Energy management for multi-energy flow:Challenges and prospects[J]. Automation of Electric Power Systems, 2016, 40(15):1-8,16.
[4]	文福拴, 鲁刚, 黄杰. 面向碳达峰、碳中和的综合能源系统[J]. 全球能源互联网, 2022, 5(2):116-117.
	WENG Fuquan, LU Gang, HUANG Jie. Integrated energy systems for carbon peaking and carbon neutrality[J]. Journal of Global Energy Interconnection, 2022, 5(2):116-117.
[5]	陶光跃, 钟文琪, 陈曦, 等. 以城市热电厂为核心的综合能源系统低碳设计优化[J]. 东南大学学报(自然科学版), 2022, 52(4):631-640.
	TAO Guangyue, ZHONG Wenqi, CHEN Xi, et al. Low carbon design optimization of integrated energy system with urban thermal power plant as core[J]. Journal of Southeast University(Natural Science Edition), 2022, 52(4):631-640.
[6]	吕凯文. 多源互补城市供热系统负荷调度实时优化研究[D]. 杭州: 浙江大学, 2018.
	LYU Kaiwen. Real-time load dispatch optimization of a multi-sources complementary urban heating system[D]. Hangzhou: Zhejiang University, 2018.
[7]	蒲星星. 基于模型的重型燃气轮机气路故障诊断研究[D]. 北京: 清华大学, 2013.
	PU Xingxing. Model-based gas path diagnostics for heavy-duty gas turbine[D]. Beijing: Tsinghua University, 2013.
[8]	王晓霞. 基于混合智能算法的热工动态过程故障诊断研究[D]. 北京: 华北电力大学, 2014.
	WANG Xiaoxia. Research on fault diagnosis of thermal dynamic process based on hybrid intelligent algorithms[D]. Beijing: North China Electric Power University, 2014.
[9]	刘涛, 刘吉臻, 吕游, 等. 基于多元状态估计和偏离度的电厂风机故障预警[J]. 动力工程学报, 2016, 36(6):454-460.
	LIU Tao, LIU Jizhen, LU You, et al. Early fault warning of power plant fans based on MSET and the deviation degree[J]. Journal of Chinese Society of Power Engineering, 2016, 36(6):454-460.
[10]	张海龙, 杨光锐, 井新经, 等. 中速磨煤机直吹式制粉系统节能优化研究[J]. 华电技术, 2020, 42(6):25-30.
	ZHANG Hailong, YANG Guangrui, JING Xinjing, et al. Research on energy-saving optimization for medium-speed mill pulverizing systems[J]. Huadian Technology, 2020, 42(6):25-30.
[11]	安吉振, 乔世超, 陈衡, 等. 基于多元状态估计和向量相似度的电站磨煤机故障智能预警研究[J]. 热力发电, 2022, 51(12):64-71.
	AN Jizhen, QIAO Shichao, CHEN Heng, et al. Research on intelligent early warning of coal mill failure in power station based on multi-state estimation and vector similarity[J]. Thermal Power Generation, 2022, 51(12):64-71.
[12]	陈波, 徐文韬, 黄亚继, 等. 基于小波包-LSTM神经网络磨煤机故障诊断[J]. 洁净煤技术, 2022, 28(5):211-220.
	CHEN Bo, XU Wentao, HUANG Yaji, et al. Fault diagnosis of coal mill based on wavelet packet-LSTM neural network[J]. Clean Coal Technology, 2022, 28(5):211-220.
[13]	周浩豪, 茅大钧, 李玉珍. 基于SSAPSO-LightGBM的火电厂引风机故障预警方法[J]. 热能动力工程, 2023, 38(2):153-160.
	ZHOU Haohao, MAO Dajun, LI Yuzhen. Fault early warning method for induced draft fan in thermal power plant based on improved SSAPSO-LightGBM[J]. Journal of Engineering For Thermal Energy and Power, 2023, 38(2):153-160.
[14]	杨婷婷, 张蓓, 吕游, 等. 基于MSET的电站风机故障预警技术研究[J]. 热能动力工程, 2017, 32(9):63-68,124.
	YANG Tingting, ZHANG Bei, LV You, et al. Research on early fault warning technology of fan in power plant based on MSET method[J]. Journal of Engineering for Thermal Energy＆Power, 2017, 32(9):63-68,124.
[15]	杨超. 热-电综合能源系统辨识与运行优化[D]. 杭州: 浙江大学, 2022.
	YANG Chao. Identification and operational optimization of integrated heat and electricity systems[D]. Hangzhou: Zhejiang University, 2022.
[16]	彭宇, 刘大同. 数据驱动故障预测和健康管理综述[J]. 仪器仪表学报, 2014, 35(3):481-495.
	PENG Yu, LIU Datong. Data-driven prognostics and health management:A review of recent advances[J]. Chinese Journal of Scientific Instrument, 2014, 35(3):481-495.
[17]	ZONTA T, COSTA C A, ROSA R R, et al. Predictive maintenance in the Industry 4.0:A systematic literature review[J]. Computers & Industrial Engineering, 2020, 150:106889.
[18]	刘耘彰. 电厂风机故障诊断与预警研究[D]. 杭州: 浙江大学, 2019.
	LIU Gengzhang. Identification and operational optimization of integrated heat and electricity systems[D]. Hangzhou: Zhejiang University, 2019.
[19]	刘海涛. 中速磨煤机状态监测与故障诊断方法研究[D]. 北京: 华北电力大学, 2015.
	LIU Haitao. Research on condition monitoring and fault diagnosis method of medium speed coal pulverizer[D]. Beijing: North China Electric Power University, 2015.
[20]	黄鑫. 基于典型工况样本的故障预警方法研究及应用[D]. 北京: 华北电力大学, 2020.
	HUANG Xin. Research and application of early fault detection method based on typical working condition[D]. Beijing: North China Electric Power University, 2020.
[21]	朱琼锋, 李家腾, 乔骥, 等. 人工智能技术在新能源功率预测的应用及展望[J]. 中国电机工程学报, 2023, 43(8):3027-3048.
	ZHU Qiongfeng, LI Jiateng, QIAO Ji, et al. Application and prospect of artificial intelligence technology in renewable energy forecasting[J]. Proceedings of the CSEE, 2023, 43(8):3027-3048.
[22]	欧阳刚. 基于LS-SVM的一次风机振动在线监测及故障预警[J]. 现代电子技术, 2017, 40(7):120-123.
	OUYANG Gang. LS-SVM based online monitoring and fault warning method for primary air fan vibration[J]. Modern Electronics Technique, 2017, 40(7):120-123.
[23]	刘定平, 叶向荣, 陈斌源, 等. 基于核主元分析和最小二乘支持向量机的中速磨煤机故障诊断[J]. 动力工程, 2009, 29(2):155-158.
	LIU Dingping, YE Xiangrong, CHEN Binyuan, et al. Fault diagnosis of medium speed mill based on KPCA and LSSVM[J]. Journal of Power Engineering, 2009, 29(2):155-158.
[24]	韩万里, 茅大钧, 印琪民. 基于PCA和多元状态估计的引风机故障预警[J]. 热能动力工程, 2020, 35(1):91-97.
	HAN Wanli, MAO Dajun, YIN Qimin. Induced draft fan fault warning based on PCA and multivariate state estimation technique[J]. Journal of Engineering For Thermal Energy and Power, 2020, 35(1):91-97.
[25]	余兴刚, 宾谊沅, 陈文, 等. 基于改进MSET的一次风机故障预警及诊断方法[J]. 机电工程, 2023, 40(4):535-541.
	YU Xinggang, CHEN Wen, et al. Early warning and diagnosis of primary fans based on improved MSET[J]. Journal of Mechanical ＆ Electrical Engineering, 2023, 40(4):535-541.
[26]	吴涛, 王毅, 刘震, 等. 基于多元状态估计-层次分析法的磨煤机故障预警[J]. 热力发电, 2023, 52(5):14-21.
	WU Tao, WANG Yi, LIU Zhen, et al. Coal mill fault warning based on multi-state estimation-analytic hierarchy process[J]. Thermal Power Generation, 2023, 52(5):14-21.
[27]	米江, 纪国宜. 改进的BP神经网络在风机故障诊断中的应用[J]. 噪声与振动控制, 2011, 31(2):94-98. doi: 10.3969/j.issn.1006-1355-2011.02.023
	MI Jiang, JI Guoyi. Application of improved BP neural network in fault diagnosis of fans[J]. Noise and Vibration Control, 2011, 31(2):94-98. doi: 10.3969/j.issn.1006-1355-2011.02.023
[28]	问姝雅, 卓旭升, 吴尔夫. 基于聚类分析的磨煤机断煤过渡状态识别[J]. 自动化与仪表, 2021, 36(11):31-34.
	WEN Shuya, ZHUO Xusheng, WU Erfu. Recognition of coal-breaking transitional state of coal mill based on cluster analysis[J]. Automation & Instrumentation, 2021, 36(11):31-34.
[29]	常玉清, 赵炜炜, 刘乐源, 等. 基于实例迁移的磨煤机过程监测建模[J]. 东北大学学报(自然科学版), 2021, 42(10):1369-1375. doi: 10.12068/j.issn.1005-3026.2021.10.001
	CHANG Yuqing, ZHAO Weiwei, LIU Leyuan, et al. Modeling of coal mill process monitoring based on instance-based transfer learning[J]. Journal of Northeastern University( Natural Science), 2021, 42(10):1369-1375.
[30]	罗婵纯, 李德忠, 杨柳, 等. 基于高斯混合模型的汽轮机转轴故障诊断方法[J]. 华电技术, 2017, 39(10):6-9,77.
	LUO Chanchun, LI Dezhong, YANG Liu, et al. Gaussian mixture model-based steam turbine shaft failure[J]. Huadian Technology, 2017, 39(10):6-9,77.
[31]	REZAEIANJOUYBARI B, SHANG Y. Deep learning for prognostics and health management:State of the art, challenges,and opportunities[J]. Measurement, 2020, 163:107929.
[32]	HELBING G, RITTER M. Deep learning for fault detection in wind turbines[J]. Renewable and Sustainable Energy Reviews, 2018,98:189-198.
[33]	李晓彬, 牛玉广, 葛维春, 等. 基于改进堆叠自编码网络的电站辅机故障预警[J]. 仪器仪表学报, 2019, 40(6):39-47.
	LI Xiaobin, NIU Yuguang, GE Weichun, et al. Early fault warning of power plant auxiliary engine based on improved stacked autoencoder network[J]. Chinese Journal of Scientific Instrument, 2019, 40(6):39-47.
[34]	刘瑶, 徐海平, 初宁, 等. 基于长短时记忆模型的风机故障诊断[J]. 工程热物理学报, 2020, 41(10):2437-2445.
	LIU Yao, XU Haiping, CHU Ning, et al. Fan fault diagnosis based on long-short term memory network[J]. Journal of Engineering Thermophysics, 2020, 41(10):2437-2445.
[35]	雷萌, 吕游, 魏玮, 等. 基于LSTM神经网络与贝叶斯优化的电站风机故障预警[J]. 热能动力工程, 2022, 37(8):213-220.
	LEI Meng, LYU You, WEI Wei, et al. Fault warning of power plant fans based on Long Short-Term Memory neural network and Bayesian optimization[J]. Journal of Engineering for Thermal Energy and Power, 2022, 37(8):213-220.
[36]	雷萌. 基于深度学习的电站风机故障预警研究[D]. 北京: 华北电力大学, 2021.
	LEI Meng. Research on fault warning of power station fans based on deep learning[D]. Beijing: North China Electric Power University, 2021.
[37]	潘岚川. 基于卷积神经网络的一次风机故障预警研究[D]. 北京: 华北电力大学, 2021.
	PAN Fengchuan. Research on fault early marning of primary air fan based on convolution neural network[D]. Beijing: North China Electric Power University, 2021.
[38]	王帝, 李治, 汪勇, 等. 基于改进最小二乘支持向量机的一次风机状态预测方法研究[J]. 动力工程学报, 2023, 43(1):74-82. doi: 10.19805/j.cnki.jcspe.2023.01.010
	WANG Di, LI Zhi, WANG Yong, et al. Research on state prediction method of primary air fan based on improved LSSVM[J]. Journal of Chinese Society of Power Engineering, 2023, 43(1):74-82. doi: 10.19805/j.cnki.jcspe.2023.01.010
[39]	张烨, 黄伟. 基于天牛群算法优化SVM的磨煤机故障诊断[J]. 机电工程, 2022, 39(3):411-418.
	ZHANG Ye, HUANG Wei. Fault diagnosis of coal mill based on BSO-SVM[J]. Journal of Mechanical & Electrical Engineering, 2022, 39(3):411-418.
[40]	张航, 周传杰, 张林, 等. 基于概率神经网络-小波神经网络-DS信息融合的电厂引风机故障诊断[J]. 发电技术, 2022, 43(6):951-958. doi: 10.12096/j.2096-4528.pgt.22162
	ZHANG Hang, ZHOU Chuanjie, ZHANG Lin, et al. Fault diagnosis of power plant induced draft fan based on PNN-WNN-DS information fusion[J]. Power Generation Technology, 2022, 43(6):951-958. doi: 10.12096/j.2096-4528.pgt.22162
[41]	杨婷婷, 高乾, 李浩千, 等. 基于卷积神经网络-长短时记忆神经网络的磨煤机故障预警[J]. 热力发电, 2022, 51(10):122-129.
	YANG Tingting, GAO Qian, LI Haoqian, et al. Coal mill fault early warning technology based on CNN-LSTM network[J]. Thermal Power Generation, 2022, 51(10):122-129.
[42]	孙栓柱, 江叶峰, 董顺, 等. 基于灰箱模型的中速磨煤机故障诊断方法[J]. 动力工程学报, 2018, 38(3):211-220.
	SUN Shuanzhu. JIANG Yefeng, DONG Shun, et al. Fault diagnosis of a medium-speed coal mill based on grey box model[J]. Journal of Chinese Society of Power Engineering, 2018, 38(3):211-220.
[43]	田亮, 袁存波. 基于LSTM和证据理论的引风机轴承故障诊断[J]. 动力工程学报, 2023, 43(5):614-621. doi: 10.19805/j.cnki.jcspe.2023.05.014
	TIAN Liang, YUAN Cunbo. Fault diagnosis of induced draft fan bearing based onlstm and evidence theory[J]. Journal of Chinese Society of Power Engineering, 2023, 43(5):614-621. doi: 10.19805/j.cnki.jcspe.2023.05.014
[44]	杨梅, 周喜超, 魏强, 等. 考虑经济效益目标的源网荷储综合能源系统动态最优能流优化研究[J]. 综合智慧能源, 2022, 44(11):63-69. doi: 10.3969/j.issn.2097-0706.2022.11.009
	YANG Mei, ZHOU Xichao, WEI Qiang, et al. Study on dynamic optimal energy flow of a source-network-load-storage integrated energy system considering its economic benefit[J]. Integrated Intelligent Energy, 2022, 44(11):63-69. doi: 10.3969/j.issn.2097-0706.2022.11.009
[45]	刘清龙. 火电厂风机状态监测与故障诊断系统研究[D]. 武汉: 武汉大学, 2005.
	LIU Qinglong. Researeh on the monitor anddiagnosis system for the air blower of therma1 power plant[D]. Wuhan: Wuhan University, 2005.
[46]	曾慧, 杜源, 李涛, 等. 考虑碳交易与绿证交易的电-热耦合园区低碳规划[J]. 综合智慧能源, 2023, 45(2): 22-29. doi: 10.3969/j.issn.2097-0706.2023.02.003
	ZENG Hui, DU Yuan, LI Tao, et al. Low-carbon planning of a park-level integrated electric and heating system considering carbon trading and green certificate trading[J]. Integrated Intelligent Energy, 2023, 45(2):22-29. doi: 10.3969/j.issn.2097-0706.2023.02.003

故障类型	影响后果
磨煤机落煤管堵塞	影响磨煤机的出力，进而影响锅炉内部的燃烧与运行
磨煤机入口风压增高	提高一次风压，导致一次风机的耗电量增大
磨煤机振动异常	导致煤粉泄漏，严重影响制粉系统和锅炉的安全稳定运行
磨煤机石子煤排量异常	对磨煤机的干燥出力、碾磨出力、通风出力等造成影响
磨煤机内部着火	磨煤机出口风粉混合物温度会急剧上升、同时磨煤机进出口压差会下降
磨煤机堵煤	堵煤的初期，磨煤机的电流、出口风粉混合物温度、进出口差压等参数都会发生变化，磨煤机堵煤会导致磨煤机出力严重下降

类别		基本原理
容积式	往复式	活塞通过曲柄连杆在气缸内往复运动，使气体的容积减小，压力增大
容积式	回转式	例如罗茨风机，通过2个转子逆时针的旋转，把气体吸入并压缩送入排气管道中
透平式	离心式	在离心力作用下，进入旋转的叶片的气体被压缩并抛送至叶轮外缘
	轴流式	沿轴向进入旋转叶片通道的气体，在与叶片的相互作用下，压缩后沿轴向排出
	混流式	介于轴流式和离心式之间，气体通过与主轴成一定角度的方向进入叶片通道
	横流式	气体横穿进入旋转叶道，受到叶片的作用而使压力升高
喷射式		经过管道中的喷嘴时，风机速度增加压力减小，低速气体在管中进行输送

故障类型	可能原因
轴承振动故障	底脚螺丝松动或混凝土基础损坏、风道损坏、轴承损坏、轴弯曲、转轴磨损、叶片磨损或积灰、叶片与外壳发生碰撞、引风机失速等
轴承温度异常	风机振动过大、风机过负荷运行时间长、润滑油系统冷却不到位、烟气温度过高、轴承冷却风机故障及备用冷却风机不联起、轴承质量差、轴承磨损严重等
旋转失速	风机振动过大、风机过负荷运行时间长、润滑油系统冷却不到位、烟气温度过高、轴承冷却风机故障及备用冷却风机不联起、轴承质量差、轴承磨损严重等
喘振	风机长期工作在低负荷、烟风道积灰堵塞、烟风道挡板开度不足等
动（静）叶片调节失灵	风机调节传动部分内部出现故障、不完全燃烧造成动叶碳垢或灰尘堵塞，使动叶调节执行机构故障等
系统油压低	油泵故障、油泵吸入口不充满、油箱油位过低、溢流阀失灵、液压缸阀芯间隙过大或工作状况不良（排油量大）等

类型	具体算法	特点
无监督学习	K均值聚类	简单直观，适用于处理大规模数据，但对异常值敏感且随着数据维度的升高，性能会逐渐下降
	高斯混合模型	不局限于特定的概率密度函数形式，模型的复杂度仅与所研究问题的复杂度有关，与样本集合的大小无关
	层次聚类	算法简单，实现速度快
	主成分分析	算法简单，不受参数限制
	多元状态估计	无需额外指定参数，只需参数之间相互关联即可，且无需完整的故障库，计算速度快，时效性高
有监督学习	人工神经网络	非线性学习能力较强，但模型构建需要大量的数据，采用的是经验风险最小化原则，容易陷入局部极小点，而且收敛速度慢，网络结构复杂
	支持向量机	适合用于小样本数据，当数据测点较多时，会出现运算效率较低问题
	随机森林	强抗噪力、可调参数少、适应力强