融合深度学习算法的炉内燃烧温度场分布在线重建

doi:10.16085/j.issn.1000-6613.2024-1478

摘要/Abstract

摘要：

在锅炉恶劣测量环境下，为了保证热辐射成像技术对炉内燃烧温度场分布在线检测的持续有效性及可靠性，融合使用深度学习算法以获取炉内温度场分布。在对某电厂350MW四角切圆燃煤锅炉进行数据提取及计算后，获取包含机组运行参数和炉内燃烧温度场分布的数据集并进行划分及预处理，进而分别建立并训练基于多层感知器（MLP）、长短时记忆（LSTM）和转置卷积神经网络（TCNN）的燃烧温度场预测模型。使用3种模型对不同负荷工况进行了炉内温度场预测及误差分析，并使用测试集对3种模型进行了评价指标计算及对比。结果表明：在变负荷运行范围内，TCNN模型对炉内温度场的泛化能力在3种模型中最佳，能够更准确预测炉内燃烧温度场分布；在3种模型中，TCNN模型对测试集的平均绝对误差和均方根误差降低至45.51K和59.73K，并且平均预测相对误差小于3.6%，满足工程应用需求，论证了该模型可用于弥补图像探头清洁期间不能获得炉内温度场的不足，进而确保其在炉内恶劣测量环境下在线检测炉内温度场的连续性及可靠性。

关键词: 燃煤锅炉, 燃烧温度场, 深度学习算法, 转置卷积神经网络, 热辐射成像

Abstract:

In the harsh measurement environment of the boiler, in order to ensure the continuous effectiveness and reliability of the thermal radiation imaging technology for online detection of the temperature field distribution in the furnace, the fusion of deep learning algorithms was used to obtain the temperature field distribution in the furnace. After data extraction and calculation for a 350MW coal-fired boiler with a quadrangular cut-circle design, a dataset containing operating parameters and temperature field distribution in the furnace was obtained and divided and preprocessed, and then MLP, LSTM, and TCNN models for temperature field prediction were respectively established and trained based on the dataset. The temperature field prediction and error analysis were carried out for different load conditions using the three models, and the evaluation indicators were calculated and compared using the test set. Results illustrated that in the variable load operation range, the TCNN model had the best generalization ability of the furnace temperature field among the three models, and could more accurately predict the distribution of the furnace combustion temperature field. Among the three models, the mean absolute error (MAE) and root mean square error (RMSE) of the TCNN model for the test set were reduced to 45.51K and 59.73K, and the average prediction relative error was less than 3.6%, which met the requirements of engineering applications. It was demonstrated that the model could be used to make up for the deficiency of failure in acquiring the temperature field in the furnace during the cleaning of the image probe, thereby ensuring the continuity and reliability of online detection of the temperature field under the harsh measurement environment in the furnace.

Key words: coal-fired boiler, combustion temperature field, deep learning algorithms, transpose convolutional neural networks (TCNN), thermal radiation imaging

中图分类号:

TP183

任世鹏, 安元, 娄春, 梅晟东, 刘凯, 陈新建. 融合深度学习算法的炉内燃烧温度场分布在线重建[J]. 化工进展, 2025, 44(4): 1923-1933.

REN Shipeng, AN Yuan, LOU Chun, MEI Shengdong, LIU Kai, CHEN Xinjian. Online reconstruction of combustion temperature field distribution in furnace by integrating deep learning algorithm[J]. Chemical Industry and Engineering Progress, 2025, 44(4): 1923-1933.

图/表 18

参考文献 23

1	王树东, 吕为智. 深度调峰形势下燃煤机组的价值量化评估[J]. 动力工程学报, 2020, 40(9): 701-706.
	WANG Shudong, Weizhi LYU. Quantitative valuation of coal-fired units under deep peak regulation conditions[J]. Journal of Chinese Society of Power Engineering, 2020, 40(9): 701-706.
2	刘辉, 周科, 解冰, 等. 基于火焰温度场在线测量的燃煤锅炉深度调峰试验[J]. 热力发电, 2019, 48(8): 49-54.
	LIU Hui, ZHOU Ke, XIE Bing, et al. Experimental investigation on deep peak load regulation of coal-fired boiler based on on-line measurement of flame temperature field[J]. Thermal Power Generation, 2019, 48(8): 49-54.
3	娄春, 张鲁栋, 蒲旸, 等. 基于自发辐射分析的被动式燃烧诊断技术研究进展[J]. 实验流体力学, 2021, 35(1): 1-17.
	LOU Chun, ZHANG Ludong, PU Yang, et al. Research advances in passive techniques for combustion diagnostics based on analysis of spontaneous emission radiation[J]. Journal of Experiments in Fluid Mechanics, 2021, 35(1): 1-17.
4	张杰, 程勇, 黄志锋. 三维燃烧火焰辐射成像模型及其构建算法[J]. 燃烧科学与技术, 2022, 28(3): 347-354.
	ZHANG Jie, CHENG Yong, HUANG Zhifeng. Radiative imaging model of three-dimensional flame and its construction method[J]. Journal of Combustion Science and Technology, 2022, 28(3): 347-354.
5	闫慧博, 唐广通, 李路江, 等. 热辐射成像法测量大型炉膛内三维温度场的算法新进展[J]. 洁净煤技术, 2022, 28(5): 97-108.
	YAN Huibo, TANG Guangtong, LI Lujiang, et al. New progress of algorithm for three-dimensional temperature field in large scale furnace measured by thermal radiative imaging[J]. Clean Coal Technology, 2022, 28(5): 97-108.
6	周怀春, 李框宇, 安元, 等. 燃煤电站锅炉及工业窑炉三维燃烧温度分布监测研究进展[J]. 洁净煤技术, 2022, 28(10): 1-15.
	ZHOU Huaichun, LI Kuangyu, AN Yuan, et al. Research progress of measuring three-dimensional temperature distributions in coal-fired boilers and industrial furnaces[J]. Clean Coal Technology, 2022, 28(10): 1-15.
7	HOLSTEIN Peter, RAABE Armin, Roland MÜLLER, et al. Acoustic tomography on the basis of travel-time measurement[J]. Measurement Science Technology, 2004, 15(7): 1420-1428.
8	张雅琪, 王飞, 崔海滨. 基于固定波长法吸收光谱技术的CO₂温度测量[J]. 激光与光电子学进展, 2019, 56(19): 278-283.
	ZHANG Yaqi, WANG Fei, CUI Haibin. Application of TDLAS to the measurement of temperature and concentration of gas components in a power station boiler furnace temperature measurement of carbon dioxide using fixed-wavelength absorption spectroscopy technique[J]. Laser & Optoelectronics Progress, 2019, 56(19): 278-283.
9	Manli LYU, ZHAO Jianping, CAO Shengxian, et al. Prediction of temperature distribution in a furnace using the incremental deep extreme learning machine[J]. PeerJ Computer Science, 2023, 9: 1218.
10	高正阳, 郭振, 胡佳琪, 等. 基于支持向量机与数值法的W火焰锅炉多目标燃烧优化及火焰重建[J]. 中国电机工程学报, 2011, 31(5): 13-19.
	GAO Zhengyang, GUO Zhen, HU Jiaqi, et al. Multi-objective combustion optimization and flame reconstruction for W shaped boiler based on support vector regression and numerical simulation[J]. Proceedings of the CSEE, 2011, 31(5): 13-19.
11	郭子申, 董美蓉, 叶托, 等. 基于k近邻和数值模拟的锅炉炉膛温度场在线重建[J]. 工业炉, 2021, 43(6): 1-5.
	GUO Zishen, DONG Meirong, YE Tuo, et al. Reconstruction of furnace temperature field based on k-nearest neighbor and numerical simulation[J]. Industrial Furnace, 2021, 43(6): 1-5.
12	ZHOU Lei, SONG Yuntong, JI Weiqi, et al. Machine learning for combustion[J]. Energy and AI, 2022, 7： 100128.
13	贾永会, 杜建桥, 汪潮洋, 等. 基于BP神经网络的燃煤锅炉温度分布预测[J]. 热能动力工程, 2020, 35(7): 130-138.
	JIA Yonghui, DU Jianqiao, WANG Chaoyang, et al. Prediction model of temperature distribution in combustion zone of coal-fired boiler based on BP neural network[J]. Journal of Engineering for Thermal Energy and Power, 2020, 35(7): 130-138.
14	XUE Wenyuan, TANG Zhenhao, CAO Shengxian, et al. A novel online method incorporating computational fluid dynamics simulations and neural networks for reconstructing temperature field distributions in coal-fired boilers[J]. Energy, 2024, 286: 129568.
15	唐广通, 许烨烽, 闫慧博, 等. 基于深度学习与热辐射成像耦合的炉内温度场在线测量[J]. 动力工程学报, 2022, 42(10): 960-966.
	TANG Guangtong, XU Yefeng, YAN Huibo, et al. Research of on-line measurement of temperature field in furnaces based on deep learning coupled thermal radiative imaging[J]. Journal of Chinese Society of Power Engineering, 2022, 42(10): 960-966.
16	HOCHREITER Sepp, Jürgen SCHMIDHUBER. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780.
17	马双忱, 林宸雨, 周权, 等. 基于深度神经网络的脱硫系统预测模型及应用[J]. 化工进展, 2021, 40(3): 1689-1698.
	MA Shuangchen, LIN Chenyu, ZHOU Quan, et al. Prediction model of FGD system based on deep neural network and its application[J]. Chemical Industry and Engineering Progress, 2021, 40(3): 1689-1698.
18	张杰, 齐琪, 韩哲哲, 等. 基于深度学习和光场成像的火焰三维温度场重建算法[J]. 东南大学学报(自然科学版), 2021, 51(6): 1060-1067.
	ZHANG Jie, QI Qi, HAN Zhezhe, et al. Reconstruction algorithm of flame 3D temperature distribution based on deep learning and light field imaging[J]. Journal of Southeast University (Natural Science Edition), 2021, 51(6): 1060-1067.
19	CAI Weiwei, HUANG Jianqing, DENG Andong, et al. Volumetric reconstruction for combustion diagnostics via transfer learning and semi-supervised learning with limited labels[J]. Aerospace Science and Technology, 2021, 110: 106487.
20	GOWDA Vishruth B, GOPALAKRISHNA M T, MEGHA J, et al. Foreground segmentation network using transposed convolutional neural networks and up sampling for multiscale feature encoding[J]. Neural Networks, 2024, 170: 167-175.
21	曹永杰. 基于CFD和机器学习的煤粉燃烧过程重建[D]. 北京: 华北电力大学, 2023.
	CAO Yongjie. Reconstruction of pulverized coal combustion process based on CFD and machine learning[D]. Beijing: North China Electric Power University, 2023.
22	LOU Chun, ZHOU Huaichun. Deduction of the two-dimensional distribution of temperature in a cross section of a boiler furnace from images of flame radiation[J]. Combustion and Flame, 2005, 143(1/2): 97-105.
23	李天宇. 基于CFD与POD的煤粉锅炉燃烧动力场快速预测及防超温优化研究[D]. 南京: 东南大学, 2022.
	LI Tianyu. Rapid prediction of combustion dynamic field and optimization of anti-overtemperature in coal-fired boiler based on CFD and POD[D]. Nanjing: Southeast University, 2022.

参数	数值范围
负荷/MW	233~345
给煤量/t·h^-1	140~195
给水量/t·h^-1	695~1119
主蒸汽流量/t·h^-1	750~1121
总一次风量/m³·h^-1	325~463
总二次风量/m³·h^-1	467~892

参数	数值范围
负荷/MW	233~345
给煤量/t·h^-1	140~195
给水量/t·h^-1	695~1119
主蒸汽流量/t·h^-1	750~1121
总一次风量/m³·h^-1	325~463
总二次风量/m³·h^-1	467~892

隐藏层层数	MLP神经网络	LSTM神经网络	TCNN
1	全连接层（256）	LSTM层（64）	全连接层（64）
2	全连接层（256）	LSTM层（128）	全连接层（128）
3	全连接层（256）	LSTM层（256）	全连接层（25）
4	全连接层（256）	LSTM层（400）	转置卷积层（128）
5	全连接层（512）	全连接层（400）	转置卷积层（256）
6	全连接层（512）	全连接层（400）	转置卷积层（256）
7	全连接层（512）	全连接层（400）	转置卷积层（256）
8	全连接层（512）	全连接层（400）	转置卷积层（256）

隐藏层层数	MLP神经网络	LSTM神经网络	TCNN
1	全连接层（256）	LSTM层（64）	全连接层（64）
2	全连接层（256）	LSTM层（128）	全连接层（128）
3	全连接层（256）	LSTM层（256）	全连接层（25）
4	全连接层（256）	LSTM层（400）	转置卷积层（128）
5	全连接层（512）	全连接层（400）	转置卷积层（256）
6	全连接层（512）	全连接层（400）	转置卷积层（256）
7	全连接层（512）	全连接层（400）	转置卷积层（256）
8	全连接层（512）	全连接层（400）	转置卷积层（256）

神经网络模型	最大相对误差/%
神经网络模型	250MW	280MW	310MW	340MW
MLP	14.07	10.35	13.04	8.46
LSTM	9.68	9.56	9.56	7.56
TCNN	8.56	7.94	6.92	9.57