公司LOGO
基于计算机视觉的锅炉燃烧优化系统设计及应用研究
摘要: 本文针对锅炉燃烧优化需求，提出一种基于计算机视觉的锅炉燃烧优化系统，旨在提升热效率并降低NOx排放。研究构建了包含感知、决策与执行层的技术框架，设计了多光谱成像硬件与边缘计算节点，开发了实时图像采集、火焰状态识别与优化决策算法，重点包括火焰图像分割网络、时空特征融合模型及强化学习优化策略。在600 MW燃煤锅炉平台上开展实验，结果表明，所提模型在分割精度和特征提取方面表现优异，实现了热效率提升与NOx减排，系统具备良好实时性与鲁棒性。研究为锅炉燃烧优化提供了新途径，具备重要理论与实践价值。
Abstract: To address the need for boiler combustion optimization, this paper proposes a computer vision-based system designed to enhance thermal efficiency and reduce Nitrogen Oxides (NOx) emissions. A technical framework comprising perception, decision-making, and execution layers was constructed. Key components included the design of multi-spectral imaging hardware and edge computing nodes, alongside the development of algorithms for real-time image acquisition, flame state recognition, and optimization decision-making. The study focused particularly on a flame image segmentation network, a spatiotemporal feature fusion model, and a reinforcement learning-based optimization strategy. Experiments conducted on a 600 MW coal-fired boiler demonstrated that the proposed model achieves excellent performance in segmentation accuracy and feature extraction. The system successfully improved thermal efficiency while reducing NOx emissions, also exhibiting good real-time performance and robustness. This research provides a novel approach for boiler combustion optimization, holding significant theoretical and practical value.
文章引用：郭恩山. 基于计算机视觉的锅炉燃烧优化系统设计及应用研究[J]. 智能技术与应用创新, 2025, 3(3): 1-14. DOI: https://doi.org/10.61784/stai2001
致谢：

无

基金项目：

无

参考文献

[1] CLIFTON J, LABER E. Q-learning: Theory and applications. Annual Review of Statistics and Its Application, 2020, 7(01): 279-301.

[2] 李会军. 分散型智能图像火焰检测系统的应用. 华东电力, 2003(10): 58-59.

[3] 刘培江, 董辉, 宋子刚, 等. 基于视频图像处理技术的火焰识别算法. 热能动力工程, 2021, 36(03): 64-71.

[4] Dong Q Z, Tan S K, Gersberg R M. Municipal solid waste management in China: status, problems and challenges. Journal of Environmental Management, 2010, 91(8): 1623-1633.

[5] 吕卓, 马永财. 机器视觉表型识别技术的研究现状. 现代化农业, 2024(3): 11-14.

[6] Nanda S. Municipal solid waste management and landfilling technologies: a review. Environmental Chemistry Letters, 2021, 19(2): 1433-1456.

[7] 邓拓宇, 田伯尧. 基于 CRITIC 权重法和多级模糊综合评判的燃烧状态诊断. 华北电力大学学报(自然科学版), 2022, 49(06): 20-27.

[8] Wang Z, Kamimoto T, Deguchi Y, et al. Two dimensional temperature measurement characteristics in pulverized coal combustion field by computed tomography-tunable diode laser absorption spectroscopy. Applied Thermal Engineering, 2020, 115066.

[9] Garami A, Csordás B, árpád P, et al. Reaction zone monitoring in biomass combustion. Control Engineering Practice, 2018, 74: 95-106.

[10] Bae H, Kim S, Wang B H, et al. Flame detection for the steam boiler using neural networks and image information in the Ulsan steam power generation plant. IEEE Transactions on Industrial Electronics, 2006, 53(1): 338-348.

[11] 周义, 张守玉, 郎森, 等. 煤粉炉掺烧生物质发电技术研究进展. 洁净煤技术, 2022, 28(6): 26-34.

[12] 曾旭华. 煤质下降对锅炉安全性和燃烧稳定性的影响及改进措施. 中国高新技术企业, 2010(7): 84-85.

[13] WOLANSKI P, WóJCICKI S. Stabilization of coal dust-air mixture by bluff bodies. Symposium on Combustion, 1975, 15(1): 1295-1302.

[14] ZHANG X, CHEN Z C, ZHANG M D, et al. Combustion stability, burnout and NO x emissions of the 300-MW down-fired boiler with bituminous coal: Load variation and low-load comparison with anthracite. Fuel, 2021, 295: 120641.

[15] 程福安, 魏舜昊, 徐帅, 等. 基于嵌入式技术的机器视觉系统设计与应用. 工业控制计算机, 2020, 33(2): 59-60.

[16] 谢能勇. 垃圾焚烧发电锅炉燃烧状态识别与主蒸汽温度预测方法. 工业炉, 2022, 44(6): 66-70.

[17] Vinicius M Sauer, Ingmar Schoegl. Numerical assessment of uncertainty and dynamic range expansion of multispectral image-based pyrometry. Measurement, 2019, 145.

[18] BLACKBURN L D, TUTTLE J F, ANDERSSON K, et al. Development of novel dynamic machine learning-based optimization of a coal-fired power plant. Computers & Chemical Engineering, 2022, 163: 107848.

[19] TUTTLE J F, BLACKBURN L D, ANDERSSON K, et al. A systematic comparison of machine learning methods for modeling of dynamic processes applied to combustion emission rate modeling. Applied Energy, 2021, 292: 116886.

[20] AFSHARI H, GADSDEN S A, HABIBI S. Gaussian filters for parameter and state estimation: A general review of theory and recent trends. Signal Processing, 2017, 135(06): 218-238.

[21] PARK S Y, HOLMQVIST K, NIEHORSTER D C, et al. How to improve data quality in dog eye tracking. Behavior Research Methods, 2023, 55(4): 1513-1536.

为你推荐