基于自适应区域权重混合模型的燃烧场温度和气体浓度二维重建方法

陈楚戈; 石顶峰; 丛洲洋; 黄安; 许振宇; 聂伟; 夏晖晖; 郭浩帆

doi:10.7498/aps.74.20250988

摘要

燃烧场温度与气体组分浓度的二维分布对发动机燃烧效率以及性能评估具有重要意义. 本文提出一种基于自适应区域权重混合模型的燃烧场温度和气体组分浓度二维重建方法, 提高复杂突变燃烧场重建精度. 通过区域权重机制将多项式模型与高斯径向基函数模型结合为混合模型, 并自适应迭代计算区域权重矩阵. 一方面通过区域权重矩阵保证了混合模型在兼顾全局特征的同时, 提高混合模型细节特征的描述能力; 另一方面, 在残差函数中加入区域权重正则化方法, 提升算法的精度. 数值模拟了三种燃烧场分布, 通过对比验证了混合模型的表征能力和重建精度, 结果表明, 混合模型算法重建误差低于单一模型及传统ART算法, 其温度、浓度分布重建最大误差分别为3.31%和7.13%. 并在标准McKenna燃烧器上搭建了扫描式TDLAS测量平台及热电偶测量平台对该方法进行实验验证, 重建结果与实际分布一致性较好, 1800 K下中心温度与热电偶测量结果偏差为10 K, 验证了该方法的有效性, 可为发动机燃烧场测量分析提供有效的参考.

关键词:

Abstract

Diagnosis of combustion flow fields in aeroengines, scramjets, and related systems plays a crucial role in understanding combustion mechanisms, evaluating combustion stability and performance, and and is also a major challenge in the development of advanced propulsion technologies. Among the non-intrusive diagnostic approaches, laser absorption spectroscopy has become one of the most representative techniques. In particular, tunable diode laser absorption spectroscopy (TDLAS) offers advantages such as a compact system architecture, easy miniaturization, strong environmental adaptability, and the capability of simultaneous temperature and concentration measurements. By employing multiple laser beams intersecting at different angles and collecting absorption spectra along various paths, the two-dimensional distribution of flow-field parameters can be reconstructed using computed tomography (CT) algorithms.However, traditional nonlinear tomographic algorithms based on polynomial models encounter difficulties in reconstructing flow fields with steep gradients. To solve this problem, we propose a hybrid reconstruction method that integrates a regional weighting mechanism. In this framework, the polynomial model is combined with a Gaussian radial basis function (RBF) model, and a regional weight matrix is iteratively updated in an adaptive manner. The regional weight matrix is determined by introducing perturbations into the current temperature field and jointly considering its temperature gradient. This design allows the hybrid model to capture global features while enhancing its ability to resolve local details. In addition, a regional weight regularization term is incorporated into the residual function to further improve reconstruction accuracy.To validate the proposed approach, numerical simulations are conducted on three representative combustion field distributions, and comparisons are made between polynomial model, RBF model, and traditional algebraic reconstruction technique (ART) algorithms. The results demonstrate that the hybrid model achieves higher representational capability and reconstruction accuracy, with maximum temperature and concentration errors reduced to 3.31% and 7.13% (for the Top-Hat case), respectively. A scanning TDLAS measurement platform and a thermocouple measurement platform are built on a standard McKenna burner to experimentally verify the method. The reconstructed distribution has good consistency with the experimental results, and the deviation between the reconstructed 1800 K central temperature and the thermocouple measurement value is only 10 K. These findings verify the effectiveness of the proposed method and highlight its potential as a reliable tool for combustion field diagnostics in propulsion systems.

Keywords:

作者及机构信息

1.
中国科学院合肥物质科学研究院, 安徽光学精密机械研究所, 环境光学与技术重点实验室, 合肥　230031

2.
中国科学技术大学研究生院科学岛分院, 合肥　230026

通信作者: 黄安, ahuang@aiofm.ac.cn

基金项目: 国家重点研发计划(批准号: 2023YFF0716400)和中国科学院合肥物质科学研究院院长基金(批准号: YZJJ202302-CX)资助的课题.

Authors and contacts

1.
Key Laboratory of Environmental Optics and Technology, Anhui Institute of Optics and Fine Mechanics, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei 230031, China

2.
Science Island Branch, Graduate School of University of Science and Technology of China, Hefei 230026, China

Corresponding author: HUANG An, ahuang@aiofm.ac.cn

Funds: Project supported by the National Key R&D Program of China (Grant No. 2023YFF0716400) and the HFIPS Director’s Fund (Grant No. YZJJ202302-CX).

文章全文

参考文献

[1]	Mitani T, Kouchi T 2005 Combust. Flame. 142 187 Google Scholar
[2]	Micka D J, Driscoll J F 2009 Proc. Combust. Inst. 32 2397 Google Scholar
[3]	Cao Z B, Li J, Song W Y, Li J P, Zhou Q, Wang C Z, Xu Z Y, Meng G, Hou K Y, Ding P J 2025 Energy 322 135458 Google Scholar
[4]	Li B, Zhang D Y, Liu J X, Tian Y F, Gao Q, Li Z S 2019 Appl. Sci. 9 1906 Google Scholar
[5]	Liu C, Xu L J 2019 Appl. Spectrosc. Rev. 54 1 Google Scholar
[6]	Alden M, Bood J, Li Z, Richter M 2011 Proc. Combust. Inst. 33 69 Google Scholar
[7]	Hosseinnia A, Raveesh M, Dominguez A, Ruchkina M, Linne M, Bood J 2022 Opt. Express 30 32204 Google Scholar
[8]	Bohlin A, Nordstrom E, Carlsson H, Bai X S, Bengtsson P E 2013 Proc. Combust. Inst. 34 3629 Google Scholar
[9]	Zhang Y, Zhang T L, Li H 2021 Spectrochim. Acta Part B At. Spectrosc. 181 106218 Google Scholar
[10]	Kiefer J, Troeger J W, Li Z, Seeger Z, Alden M, Leipertz A 2012 Combust. Flame. 159 3576 Google Scholar
[11]	Tripathi M M, Srinivasan K K, Krishnan S R 2013 Fuel 106 318 Google Scholar
[12]	Fu J, Tang C L, Jin W, Thi L D, Huang Z H 2013 Int. J. Hydrogen Energy 38 1636 Google Scholar
[13]	Yamammoto K, Isii S, Ohnishi M 2011 Proc. Combust. Inst. 33 1285 Google Scholar
[14]	Jeon M G, Deguchi Y, Kamimoto T, Doh D H, Cho G R 2017 Appl. Therm. Eng. 115 1148 Google Scholar
[15]	Zhou W B, Cao Z, Zhao K, Wang Z C, Xu L J 2024 IEEE Trans. Instrum. Meas 73 1 Google Scholar
[16]	Xia H H, Kan R F, Liu J G, Xu Z Y, He Y B 2016 Chin. Phys. B 25 064205 Google Scholar
[17]	Ma L, Cai W W 2008 Appl. Opt. 47 3751 Google Scholar
[18]	Wang Z Z, Deguchi Y, Kamimoto T, Tainaka K, Tanno K, 2020 Fuel 268 117370 Google Scholar
[19]	Gao X, Cao Z, Tian Y, Xu L J 2021 IEEE International Conference on Imaging Systems and Techniques (IST) Kaohsiung, China, August 24–26, 2021 p1
[20]	Yang H Q, Wang Z H, Song K L 2022 Eng. Comput. 38 2469 Google Scholar
[21]	黄安 2023 博士学位论文 (合肥: 中国科学技术大学) Huang A 2023 Ph. D. Dissertation (Hefei: University of Science and Technology of China
[22]	Shaddix C R 2017 10^th U. S. National Combustion Meeting College Park, Maryland, USA, April 23–26, 2017 SAND2017-4406C

施引文献

图 1 自适应算法流程图

Fig. 1. Adaptive algorithm flow chart.

下载: 全尺寸图片幻灯片

图 2 三种模拟分布

Fig. 2. Three simulated distributions.

下载: 全尺寸图片幻灯片

图 3 模型真值拟合误差

Fig. 3. Fitting errors of three models.

下载: 全尺寸图片幻灯片

图 4 光路布局

Fig. 4. Light path diagram.

下载: 全尺寸图片幻灯片

图 5 三种分布不同方法重建结果　(a) 四种算法的单高斯峰温度浓度重建结果; (b) 四种算法的双高斯峰温度浓度重建结果; (c) 四种算法的Top-Hat分布温度浓度重建结果

Fig. 5. Reconstruct results of three distribution: (a) Reconstruction results of single peak distribution; (b) reconstruction results of double peak distribution; (c) reconstruction results of Top-Hat distribution.

下载: 全尺寸图片幻灯片

图 6 算法温度、浓度重建误差　(a) 温度分布重建误差; (b) 浓度分布重建误差; (c) 中心区域温度重建误差; (d) 中心区域浓度重建误差

Fig. 6. Temperature and concentration reconstruction errors: (a) Temperature distribution reconstruction error; (b) concentration distribution reconstruction error; (c) central region temperature reconstruction error; (d) central region concentration reconstruction error.

下载: 全尺寸图片幻灯片

图 7 中心温度对比结果

Fig. 7. Comparison results of central temperature.

下载: 全尺寸图片幻灯片

图 8 噪声模拟结果　(a) 单高斯峰分布; (b) 双高斯峰分布; (c) Top-Hat分布

Fig. 8. Noise simulation results: (a) Single peak; (b) double peak; (c) Top-Hat.

下载: 全尺寸图片幻灯片

图 9 实验示意图

Fig. 9. Experimental schematic diagram.

下载: 全尺寸图片幻灯片

图 10 实验装置图

Fig. 10. Experimental setup diagram.

下载: 全尺寸图片幻灯片

图 11 0°第15路光束原始拟合信号

Fig. 11. The 15 th original fitting signal.

下载: 全尺寸图片幻灯片

图 12 实验重建结果

Fig. 12. Experimental reconstruction results.

下载: 全尺寸图片幻灯片

图 13 中心区域沿轴向温度分布结果

Fig. 13. Temperature distribution results along the axial direction in the central area.

下载: 全尺寸图片幻灯片

表 1 所选吸收线光谱参数

Table 1. Spectral parameters of absorption line.

v₀/cm^–1 S(T₀)/(cm^–2·atm^–1) E''

7467.7695 1.2174×10^–5 2551.4835

7444.3961 0.0011 1790.7113

7185.5962 0.0195 1045.058

6807.8350 6.1737×10^–7 3319.4485

下载: 导出CSV

map

[1]	Mitani T, Kouchi T 2005 Combust. Flame. 142 187 Google Scholar
[2]	Micka D J, Driscoll J F 2009 Proc. Combust. Inst. 32 2397 Google Scholar
[3]	Cao Z B, Li J, Song W Y, Li J P, Zhou Q, Wang C Z, Xu Z Y, Meng G, Hou K Y, Ding P J 2025 Energy 322 135458 Google Scholar
[4]	Li B, Zhang D Y, Liu J X, Tian Y F, Gao Q, Li Z S 2019 Appl. Sci. 9 1906 Google Scholar
[5]	Liu C, Xu L J 2019 Appl. Spectrosc. Rev. 54 1 Google Scholar
[6]	Alden M, Bood J, Li Z, Richter M 2011 Proc. Combust. Inst. 33 69 Google Scholar
[7]	Hosseinnia A, Raveesh M, Dominguez A, Ruchkina M, Linne M, Bood J 2022 Opt. Express 30 32204 Google Scholar
[8]	Bohlin A, Nordstrom E, Carlsson H, Bai X S, Bengtsson P E 2013 Proc. Combust. Inst. 34 3629 Google Scholar
[9]	Zhang Y, Zhang T L, Li H 2021 Spectrochim. Acta Part B At. Spectrosc. 181 106218 Google Scholar
[10]	Kiefer J, Troeger J W, Li Z, Seeger Z, Alden M, Leipertz A 2012 Combust. Flame. 159 3576 Google Scholar
[11]	Tripathi M M, Srinivasan K K, Krishnan S R 2013 Fuel 106 318 Google Scholar
[12]	Fu J, Tang C L, Jin W, Thi L D, Huang Z H 2013 Int. J. Hydrogen Energy 38 1636 Google Scholar
[13]	Yamammoto K, Isii S, Ohnishi M 2011 Proc. Combust. Inst. 33 1285 Google Scholar
[14]	Jeon M G, Deguchi Y, Kamimoto T, Doh D H, Cho G R 2017 Appl. Therm. Eng. 115 1148 Google Scholar
[15]	Zhou W B, Cao Z, Zhao K, Wang Z C, Xu L J 2024 IEEE Trans. Instrum. Meas 73 1 Google Scholar
[16]	Xia H H, Kan R F, Liu J G, Xu Z Y, He Y B 2016 Chin. Phys. B 25 064205 Google Scholar
[17]	Ma L, Cai W W 2008 Appl. Opt. 47 3751 Google Scholar
[18]	Wang Z Z, Deguchi Y, Kamimoto T, Tainaka K, Tanno K, 2020 Fuel 268 117370 Google Scholar
[19]	Gao X, Cao Z, Tian Y, Xu L J 2021 IEEE International Conference on Imaging Systems and Techniques (IST) Kaohsiung, China, August 24–26, 2021 p1
[20]	Yang H Q, Wang Z H, Song K L 2022 Eng. Comput. 38 2469 Google Scholar
[21]	黄安 2023 博士学位论文 (合肥: 中国科学技术大学) Huang A 2023 Ph. D. Dissertation (Hefei: University of Science and Technology of China
[22]	Shaddix C R 2017 10^th U. S. National Combustion Meeting College Park, Maryland, USA, April 23–26, 2017 SAND2017-4406C

[1]	邓锐, 黄渝峰, 张逸凡, 张莹莹, 宋远红. 射频容性耦合等离子体中下极板凹槽对尘埃颗粒集体行为的影响研究. , 2025, 74(24): . doi: 10.7498/aps.74.20251047
[2]	佟磊, 赵明亮, 张钰如, 宋远红, 王友年. 带有射频偏压源的感性耦合Ar/O₂/Cl₂等离子体放电的混合模拟研究. , 2024, 73(4): 045201. doi: 10.7498/aps.73.20231369
[3]	庞维煦, 李宁, 黄孝龙, 康杨, 李灿, 范旭东, 翁春生. 基于分数阶Tikhonov正则化的激光吸收光谱燃烧场二维重建光路优化研究. , 2023, 72(3): 037801. doi: 10.7498/aps.72.20221731
[4]	张钰如, 高飞, 王友年. 低气压感性耦合等离子体源模拟研究进展. , 2021, 70(9): 095206. doi: 10.7498/aps.70.20202247
[5]	王振兴, 曹志远, 李瑞, 陈峰, 孙丽琼, 耿英三, 王建华. 纵磁作用下真空电弧单阴极斑点等离子体射流三维混合模拟. , 2021, 70(5): 055201. doi: 10.7498/aps.70.20201701
[6]	臧益鹏, 许振宇, 黄安, 艾苏曼, 夏晖晖, 阚瑞峰. 基于改进模拟退火算法的非均匀燃烧场分布重建. , 2021, 70(13): 134205. doi: 10.7498/aps.70.20202124
[7]	李宁, TuXin, 黄孝龙, 翁春生. 基于Tikhonov正则化参数矩阵的激光吸收光谱燃烧场二维重建光路设计方法. , 2020, 69(22): 227801. doi: 10.7498/aps.69.20201144
[8]	周建印, 项杰, 黄思训. 确定大气边界层顶高度的新方法及数值实验. , 2020, 69(9): 090201. doi: 10.7498/aps.69.20191992
[9]	杨成, 周昕. 液态水中的多种局域结构. , 2016, 65(17): 176501. doi: 10.7498/aps.65.176501
[10]	谢正超, 王飞, 严建华, 岑可法. 炉膛三维温度场重建中Tikhonov正则化和截断奇异值分解算法比较. , 2015, 64(24): 240201. doi: 10.7498/aps.64.240201
[11]	伍飞飞, 廖瑞金, 杨丽君, 刘兴华, 汪可, 周之. 棒-板电极直流负电晕放电特里切尔脉冲的微观过程分析. , 2013, 62(11): 115201. doi: 10.7498/aps.62.115201
[12]	王晶, 马瑞玲, 王龙, 孟俊敏. 采用混合模型数值模拟从深海到浅海内波的传播. , 2012, 61(6): 064701. doi: 10.7498/aps.61.064701
[13]	廖瑞金, 伍飞飞, 刘兴华, 杨帆, 杨丽君, 周之, 翟蕾. 大气压直流正电晕放电暂态空间电荷分布仿真研究. , 2012, 61(24): 245201. doi: 10.7498/aps.61.245201
[14]	姜祝辉, 黄思训, 何然, 周晨腾. 合成孔径雷达资料反演海面风场的正则化方法研究. , 2011, 60(6): 068401. doi: 10.7498/aps.60.068401
[15]	周玉淑, 曹洁. 有限区域风场的分解和重建. , 2010, 59(4): 2898-2906. doi: 10.7498/aps.59.2898
[16]	姜祝辉, 黄思训, 杜华栋, 刘博. 利用变分结合正则化方法对高度计风速资料调整海面风场的研究. , 2010, 59(12): 8968-8977. doi: 10.7498/aps.59.8968
[17]	张亮, 黄思训, 刘宇迪, 钟剑. 变分同化结合广义变分最佳分析对微波散射计资料进行海面风场反演. , 2010, 59(4): 2889-2897. doi: 10.7498/aps.59.2889
[18]	李加庆, 陈进, 杨超, 贾文强. 波叠加声场重构精度的影响因素分析. , 2008, 57(7): 4258-4264. doi: 10.7498/aps.57.4258
[19]	于飞, 陈心昭, 李卫兵, 陈剑. 空间声场全息重建的波叠加方法研究. , 2004, 53(8): 2607-2613. doi: 10.7498/aps.53.2607
[20]	唐坤发, 胡嘉桢. 无序区域上推广混合自旋模型的严格解. , 1988, 37(9): 1564-1568. doi: 10.7498/aps.37.1564

计量

文章访问数: 311
PDF下载量: 4
被引次数: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

搜索

留言板

基于自适应区域权重混合模型的燃烧场温度和气体浓度二维重建方法