基于火焰图像时序规整的锅炉负荷监测方法

doi:10.16085/j.issn.1000-6613.2024-1743

摘要/Abstract

摘要：

燃煤发电厂的负荷信号往往基于蒸汽侧参数计算得到，而炉膛热力过程的热惯性会引起蒸汽参数的变化存在滞后性，从而影响运行人员的负荷监测和调整判断。因此，本文从炉膛火焰图像反映炉膛内部实际状况的特点出发，提出了一种基于火焰图像时序特征规整的改进锅炉负荷监测方法。该方法首先对锅炉火焰监控视频提取转换成连续图像帧，并经过滤波、分割等一系列自适应预处理步骤提取火焰时序特征，再通过降维得到火焰强度指标，最后将强度指标与负荷信号两时序信号进行动态时序规整计算，从而获得强度指标与负荷信号的规整对齐路径和时间延迟曲线。分析表明，动态时序规整算法在50%、80%、100%等各个典型负荷工况下使用图像监测指标对负荷信号的弹性关联提升效果明显，为燃煤发电厂等复杂工业系统的锅炉负荷变动监测和分析提供了一种高效准确的方法。

关键词: 燃煤发电厂, 火焰图像处理, 动态时间规整, 时延分析, 负荷监测

Abstract:

The load signal in coal-fired power plants is calculated based on steam parameters, but the thermal inertia of the furnace process causes a lag in steam parameter changes, affecting the load monitoring and adjustment decisions of the operators. Therefore, this paper proposed an improved load monitoring method based on the time series alignment of flame image features, leveraging the ability of flame images to reflect the actual conditions inside the furnace. This method firstly converted boiler flame monitoring videos into continuous image frames, then extracted flame temporal features through a series of adaptive preprocessing steps, including filtering and segmentation. Dimensionality reduction was then applied to obtain the flame image indicators, and finally, dynamic time warping (DTW) algorithm was used to align the flame image indicators with the load signal, generating an alignment path and time-delay curve between them. Analysis showed that the DTW algorithm significantly enhanced the correlation between flame image indicators and load signals under typical operating conditions (50%, 80%, and 100% load), providing an efficient solution for monitoring load fluctuations in coal-fired power plants.

Key words: coal-fired power plant, flame image processing, dynamic time warping, time-delay analysis, load monitoring

中图分类号:

TK315

王光亚, 董美蓉, 周杰恒, 陈翔, 陆继东. 基于火焰图像时序规整的锅炉负荷监测方法[J]. 化工进展, 2025, 44(4): 1934-1944.

WANG Guangya, DONG Meirong, ZHOU Jieheng, CHEN Xiang, LU Jidong. Boiler load monitoring method based on time series alignment of flame images[J]. Chemical Industry and Engineering Progress, 2025, 44(4): 1934-1944.

图/表 14

图1 图像处理流程

图2 动态时间规整流程

图3 DTW步进模式

图4 锅炉火焰监控采集系统1—炉膛B侧窥火孔；2—炉膛A侧窥火孔；3—炉墙；4—镜管；5—摄像机；6—电机；7—录像机

表1 运行工况参数

锅炉参数	50%负荷工况	80%负荷工况	100%负荷工况
负荷范围/MW	156.03~161.70	236.84~242.02	276.86~283.45
一次风量/t·h^-1	228.70~236.44	244.32~255.77	274.07~284.11
二次风量/t·h^-1	274.66~307.65	387.46~416.31	485.88~536.36
总给煤率/t·h^-1	74.90~80.13	103.98~111.16	125.61~133.70

图5 火焰图像原始数据集

图6 频域降噪前后效果

图7 A侧视角50%工况亮度分量图

图8 火焰亮度分量图

图9 锅炉负荷与图像特征强度时间序列

图10 图像监测指标箱线

图11 各步进模式平均绝对误差对比

图12 DTW重新规整对齐路径及对应时延曲线

图13 经DTW规整前后图像监测指标与锅炉负荷的瞬态时序关系

参考文献 29

1	黄晟, 王静宇, 郭沛, 等. 碳中和目标下能源结构优化的近期策略与远期展望[J]. 化工进展, 2022, 41(11): 5695-5708.
	HUANG Sheng, WANG Jingyu, GUO Pei, et al. Near-term strategies and long-term outlook for energy structure optimization under carbon neutrality goals[J]. Chemical Industry and Engineering Progress, 2022, 41(11): 5695-5708.
2	苗青青, 石春艳, 张香平. 碳中和目标下的光伏发电技术[J]. 化工进展, 2022, 41(3): 1125-1131.
	MIAO Qingqing, SHI Chunyan, ZHANG Xiangping. Photovoltaic power generation technology under carbon neutrality goals[J]. Chemical Industry and Engineering Progress, 2022, 41(3): 1125-1131.
3	李明扬, 蒋媛媛. 考虑煤耗率的火电机组灵活调峰对风电消纳的影响效果研究[J]. 热力发电, 2020, 49(2): 45-51.
	LI Mingyang, JIANG Yuanyuan. Study on the impact of flexible peak load regulation of thermal power units considering coal consumption rate on wind power accommodation[J]. Thermal Power Generation, 2020, 49(2): 45-51.
4	周怀春, 胡志方, 郭建军, 等. 面向智能发电的电站燃煤锅炉在线运行优化[J]. 分布式能源, 2019, 4(3): 1-7.
	ZHOU Huaichun, HU Zhifang, GUO Jianjun, et al. Online operation optimization of coal-fired boilers in power plants for intelligent power generation[J]. Distributed Energy, 2019, 4(3): 1-7.
5	王义松, 车帅, 宋延丽, 等. 燃煤锅炉火检系统应用中存在的问题及其优化方法综述[J]. 冶金能源, 2017, 36(S1): 105-107.
	WANG Yisong, CHE Shuai, SONG Yanli, et al. Review of issues and optimization methods in the application of flame detection systems for coal-fired boilers[J]. Metallurgical Energy, 2017, 36(S1): 105-107.
6	李欣宇, 唐德东, 张淑敏, 等. 发电厂锅炉燃烧状态监测技术综述[J]. 中国特种设备安全, 2024, 40(1): 28-33.
	LI Xinyu, TANG Dedong, ZHANG Shumin, et al. Review of monitoring technology for boiler combustion status in power plants[J]. China Special Equipment Safety, 2024, 40(1): 28-33.
7	周怀春, 韩曙东, 盛锋, 等. 炉膛燃烧温度场三维可视化监测方法模拟研究[J]. 动力工程, 2003, 23(1): 2154-2159.
	ZHOU Huaichun, HAN Shudong, SHENG Feng, et al. Simulation study on 3D visualization monitoring method for furnace combustion temperature field[J]. Power Engineering, 2003, 23(1): 2154-2159.
8	YAN Y, LU G, COLECHIN M. Monitoring and characterisation of pulverised coal flames using digital imaging techniques[J]. Fuel, 2002, 81(5): 647-655.
9	韩璞, 张欣, 王兵, 等. 基于神经网络的交互式炉膛火焰图像识别[J]. 中国电机工程学报, 2008, 28(20): 22-26.
	HAN Pu, ZHANG Xin, WANG Bing, et al. Interactive furnace flame image recognition based on neural networks[J]. Proceedings of the CSEE, 2008, 28(20): 22-26.
10	刘禾. 基于火焰图像和模糊神经网络的锅炉燃烧稳定性判别[J]. 仪器仪表学报, 2008, 29(6): 1280-1284.
	LIU He. Boiler combustion stability discrimination based on flame images and fuzzy neural networks[J]. Journal of Instruments, 2008, 29(6): 1280-1284.
11	CHEN Junghui, HSU Tong-Yang, CHEN Chih-chien, et al. Monitoring combustion systems using HMM probabilistic reasoning in dynamic flame images[J]. Applied Energy, 2010, 87(7): 2169-2179.
12	SUJATHA K, VENMATHI M, PAPPA N. Flame monitoring in power station boilers using image processing[J]. ICTACT Journal on Image and Video Processing, 2012, 2(4): 427-434.
13	QIU Tian, LIU Minjian, ZHOU Guiping, et al. An unsupervised classification method for flame image of pulverized coal combustion based on convolutional auto-encoder and hidden Markov model[J]. Energies, 2019, 12(13): 2585.
14	艾徐华. 基于炉膛火焰图像的锅炉燃烧稳定性判别[D]. 北京: 华北电力大学, 2019.
	AI Xuhua. Discrimination of boiler combustion stability based on furnace flame images[D]. Beijing: North China Electric Power University, 2019.
15	黄琪. 基于火焰图像的炉膛燃烧稳定性定量分析[D]. 北京: 华北电力大学, 2022.
	HUANG Qi. Quantitative analysis of furnace combustion stability based on flame images[D]. Beijing: North China Electric Power University, 2022.
16	黄埔, 楼波, 李森浩, 等. 锅炉燃烧火焰的脉动频率特征[J]. 锅炉技术, 2023, 54(6): 42-46.
	HUANG Pu, LOU Bo, LI Senhao, et al. Pulsation frequency characteristics of boiler combustion flames[J]. Boiler Technology, 2023, 54(6): 42-46.
17	王印松, 雷玉. 基于DCS数据和燃烧图像的垃圾焚烧炉主蒸汽温度预测[J]. 中国电机工程学报, 2023, 43(22): 8790-8801.
	WANG Yinsong, LEI Yu. Prediction of main steam temperature in waste incineration furnace based on DCS data and combustion images[J]. Proceedings of the CSEE, 2023, 43(22): 8790-8801.
18	唐振浩, 柴向颖, 曹生现, 等. 考虑时延特征的燃煤锅炉NO _x 排放深度学习建模[J]. 中国电机工程学报, 2020, 40(20): 6633-6643.
	TANG Zhenhao, CHAI Xiangying, CAO Shengxian, et al. Deep learning modeling of NO _x emissions from coal-fired boilers considering time delay characteristics[J]. Proceedings of the CSEE, 2020, 40(20): 6633-6643.
19	王文广, 王朔, 赵文杰. 基于最大信息系数变量选择的电站锅炉NO _x 排放量在线预估[J]. 华北电力大学学报(自然科学版), 2019, 46(6): 66-72.
	WANG Wenguang, WANG Shuo, ZHAO Wenjie. Online prediction of NO _x emissions from power plant boilers based on maximum information coefficient variable selection[J]. Journal of North China Electric Power University (Natural Science Edition), 2019, 46(6): 66-72.
20	张慧敏. 基于动态时间规整算法的语音识别技术研究[J]. 科技资讯, 2017, 15(26): 28-31.
	ZHANG Huimin. Research on speech recognition technology based on dynamic time warping algorithm[J]. Science and Technology Information, 2017, 15(26): 28-31.
21	刘洋, 于海东, 刘文彬, 等. 基于DTW-两阶四分位的分布式光伏发电异常数据辨识[J]. 热力发电, 2024, 53(7): 34-44.
	LIU Yang, YU Haidong, LIU Wenbin, et al. Identification of abnormal data in distributed photovoltaic power generation based on DTW-second order quartile[J]. Thermal Power Generation, 2024, 53(7): 34-44.
22	张一琛, 陈双全, 靳松, 等. 基于动态时间规整算法的纵波与转换波时间域匹配[J]. 石油科学通报, 2018, 3(2): 144-153.
	ZHANG Yichen, CHEN Shuangquan, JIN Song, et al. Time domain matching of longitudinal waves and converted waves based on dynamic time warping algorithm[J]. Petroleum Science Bulletin, 2018, 3(2): 144-153.
23	王瑞, 王英洲, 逯静. 基于ICEEMDAN-DTW和ISMA-WLSSVM的光伏发电功率预测[J]. 热能动力工程, 2023, 38(9): 131-140.
	WANG Rui, WANG Yingzhou, LU Jing. Photovoltaic power generation forecasting based on ICEEMDAN-DTW and ISMA-WLSSVM[J]. Journal of Engineering for Thermal Energy and Power, 2023, 38(9): 131-140.
24	GALBALLY Javier, GALBALLY David. A pattern recognition approach based on DTW for automatic transient identification in nuclear power plants[J]. Annals of Nuclear Energy, 2015, 81: 287-300.
25	刘健庄, 栗文青. 灰度图象的二维Otsu自动阈值分割法[J]. 自动化学报, 1993, 19(1): 101-105.
	LIU Jianzhang, LI Wenqing. Two-dimensional Otsu automatic threshold segmentation method for grayscale images[J]. Acta Automatica Sinica, 1993, 19(1): 101-105.
26	覃晓, 元昌安, 邓育林, 等. 一种改进的Ostu图像分割法[J]. 山西大学学报(自然科学版), 2013, 36(4): 530-534.
	QIN Xiao, YUAN Changan, DENG Yulin, et al. An improved Otsu image segmentation method[J]. Journal of Shanxi University (Natural Science Edition), 2013, 36(4): 530-534.
27	GIORGINO Toni. Computing and visualizing dynamic time warping alignments in R: The DTW package[J]. Journal of Statistical Software, 2009, 31(7): 1-24.
28	刘鲭洁, 陈桂明, 刘晓方, 等. FFT和小波变换在信号降噪中的应用[J]. 数据采集与处理, 2009, 24(S1): 58-60.
	LIU Qingjie, CHEN Guiming, LIU Xiaofang, et al. Applications of FFT and wavelet transform in signal denoising[J]. Journal of Data Acquisition and Processing, 2009, 24(S1): 58-60.
29	KIRCHHOFF Holger, LERCH Alexander. Evaluation of features for audio-to-audio alignment[J]. Journal of New Music Research, 2011, 40(1): 27-41.