Research on effects of flame tube hole diameters on combustor combustion performance based on neural network

doi:10.3969/j.issn.1674-8484.2026.01.009

Abstract

Abstract:

The combustion performance of combustors in micro turbine engine is affected by flame tube hole diameters, lacking analytical methods for evaluating the combined effects of multiple flame tube hole diameters. It is difficult to clearly determine the diameter distribution ranges of primary hole, middle hole, and dilution hole of flame tube that can achieve high combustion efficiency and low overall temperature distribution factor (OTDF). To address this issue, a combustion performance prediction model was established by learning from the results obtained from the calibrated model of a combustor with the help of neural work. It has been utilized to study the combined effects of multiple hole diameters on combustion performance. The results show that the influence of dilution hole, primary hole and middle hole diameters on combustion efficiency gradually decreases, and the influence of dilution hole diameter on the overall temperature distribution factor (OTDF) is significantly higher than that of primary hole and middle hole diameters. Combustion efficiency is low and OTDF is high at different primary hole and middle hole diameters in case of small dilution hole diameter. With the increase of dilution hole diameter, high temperature combustion zone enlarges obviously in middle zone, which is benefit to increase combustion efficiency. In the meantime, the jet depth of the air flowing from dilution holes increases, which strengthens the mixing of air with high temperature gas and reduces OTDF.

Key words: combustor, flame tube hole diameters, neural network, combustion efficiency, overall temperature distribution factor (OTDF)

CLC Number:

TK473.2

FU Xueqing, CHEN Chuang, ZHOU Daoqing, ZHANG Yan, CAO Xiaolin, LI Xin, JI Jianbo, CHEN Peng, XUE Xingxu, LI Yaozong. Research on effects of flame tube hole diameters on combustor combustion performance based on neural network[J]. Journal of Automotive Safety and Energy, 2026, 17(1): 88-95.

Figures/Tables 12

References 16

[1]	张扬军, 钱煜平, 诸葛伟林, 等. 飞行汽车的研究与关键技术[J]. 汽车安全与节能学报, 2020, 11(1): 1-16.
	ZHANG Yangjun, QIAN Yuping, ZHUGE Weilin, et al. Progress and key technologies of flying cars[J]. *J Autom Safe Energ*, 2020, 11(1): 1-16. (in Chinese)
[2]	朱炳杰, 杨希祥, 宗件安, 等. 分布式混合电推进飞行器技术[J]. 航空学报, 2022, 43(7): 025556.
	ZHU Bingjie, YANG Xixiang, ZONG Jianan, et al. Review of distributed hybrid electric propulsion aircraft technology[J]. *Acta Aeron Sin*, 2022, 43(7): 025556. (in Chinese)
[3]	LI Jingqi, LI Yulong. Micro gas turbine: Developments, applications, and technologies on components[J]. *Prop Powe Res*, 2023, 12(1): 1-43.
[4]	袁春, 陈彬兵, 陈兆海, 等. 微型燃气轮机发电技术[M]. 北京: 中国机械工业出版社, 2012: 39-50.
	YUAN Chun, CHEN Binbing, CHEN Zhaohai, et al. Micro Gas Turbine Power Generation Technology[M]. Beijing: China Machine Press, 2012: 39-50. (in Chinese)
[5]	Rodgers C. Low cost microturbines via the turbo-charger route[C]// Proc ASME Turbo Expo 2011, British Columbia, Canada. 2011.
[6]	Lefebvre A H, Ballal D R. Gas Turbine Combustion Alter-ative Fuels and Emissions[M]. New York: CRC Press, 2010: 15-17.
[7]	Gogineni S, Shouse D, Frayne C, et al. Combustion air jet influence on primary zone characteristics for gas-turbine combustors[J]. *J Propul Power*, 2015, 18(2): 407-416. doi: 10.2514/2.5949 URL
[8]	Elkady A M, Jeng S M, Mongia H, et al. The influence of primary air jets on flow and pollutant emission characteristics within a model gas turbine combustor[C]// 44th AIAA Aeros Sci Meet Exhib, 2006, Reno, Nevada.
[9]	Povey T, Chana K S, Jones T V, et al. The effect of hot-streaks on HP vane surface and end wall heat transfer: An experimental and numerical study[J]. *J Turbomach, 2007, 129*(1): 32-43. doi: 10.1115/1.2370748 URL
[10]	陈闯, 沈世成, 付雪青, 等. 主燃孔参数对双级轴向旋流燃烧室流动与燃烧特性的影响[J]. 内燃机工程, 2025, 46(3): 39-47.
	CHEN Chuang, SHEN Shicheng, FU Xueqing, et al. Effects of primary inlet hole parameters on the flow and combustion characteristics of a two-stage axial swirl combustor[J]. *Chin Inte Comb Enge Engi*, 2025, 46(3): 39-47. (in Chinese)
[11]	郑顺, 王成军. 掺混孔位置对中心分级燃烧室性能影响的数值模拟[J]. 邵阳学院学报(自然科学版), 2021, 18(1): 51-59.
	ZHEN Shun, WANG Chengjun. Numerica simulation of effect of mixing hole location on performance of a central staged combustor[J]. *J Shaoyang Univ (Nat Sci Edit)*, 2021, 18(1): 51-59. (in Chinese)
[12]	林宏军, 程明. 喷嘴匹配方案及火焰筒开孔对燃烧室性能影响的试验研究[J]. 航空发动机, 2012, 38(5): 13-17.
	LIN Hongjun, CHEN Ming. Effect of nozzle matching and flame tube holes on combustion performance[J]. *Aeroengine*, 2012, 38(5): 13-17. (in Chinese)
[13]	Fayyad U, Uthurusamy R. Evolving data mining into solutions for insights[J]. *Commun ACM*, 2002, 45(8): 28-31.
[14]	陶焰明, 冯剑寒, 江立军, 等. 基于神经网络的双轴向旋流器结构参数影响分析[J]. 推进技术, 2024, 45(10): 2306031.
	TAO Yanming, FENG Jianhan, JIANG Lijun, et al. Effects of structural parameters and dual axial swirlers based on neural network[J]. *J Propul Tech*, 2024, 45(10): 2306031. (in Chinese)
[15]	赵刚, 朱华昕, 李苏辉, 等. 基于数据和神经网络的燃气轮机NO_x排放预测与优化[J]. 动力工程学报, 2021, 41(1): 22-27. doi: 10.19805/j.cnkij.cspe.2021.01.004
	ZHAO Gang, ZHU Huaxin, LI Suhui, et al. NO_x emission prediction and optimization for gas turbines based on data and neural network[J]. *J Chin Soc Power Engi*, 2021, 41(1): 22-27. (in Chinese)
[16]	尹林子, 李乐, 蒋朝辉. 基于粗糙集理论与神经网络的铁水硅含量预测[J]. 钢铁研究学报, 2019, 31(8): 689-698. doi: 10.13228/j.boyuan.issn1001-0963.20190010
	YIN Linzi, LI Le, JIANG Chaohui. Prediction of silicon content in hot metal using neural network and rough set theory[J]. *J Iron Steel Res*, 2019, 31(8): 689-698. (in Chinese) doi: 10.13228/j.boyuan.issn1001-0963.20190010

工况	测量值		油气比	测量值		计算值
工况	喷油流量 (g·s^-1)	空气流量 (kg·s^-1)	油气比	压力 kPa	温度 K	压力 kPa	温度 K
1	4.21	0.29	0.014 5	132.1	953.2	130.9	922.9
2	5.63	0.41	0.013 7	168.9	1 010.5	167.2	995.6
3	6.63	0.54	0.012 3	240.1	979.3	238.6	962.7

工况	测量值		油气比	测量值		计算值
工况	喷油流量 (g·s^-1)	空气流量 (kg·s^-1)	油气比	压力 kPa	温度 K	压力 kPa	温度 K
1	4.21	0.29	0.014 5	132.1	953.2	130.9	922.9
2	5.63	0.41	0.013 7	168.9	1 010.5	167.2	995.6
3	6.63	0.54	0.012 3	240.1	979.3	238.6	962.7

组合	d_p /mm	d_m /mm	d_d /mm	η_c	OTDF
A	16.8	10.8	16.0	0.92	0.41
B	16.8	10.5	20.0	0.96	0.32
C	16.5	11.0	24.0	0.98	0.24

组合	d_p /mm	d_m /mm	d_d /mm	η_c	OTDF
A	16.8	10.8	16.0	0.92	0.41
B	16.8	10.5	20.0	0.96	0.32
C	16.5	11.0	24.0	0.98	0.24