基于支持向量机的氢混天然气发动机性能预测

doi:10.12382/bgxb.2021.0199

摘要/Abstract

摘要： 为提高氢混天然气（HCNG）发动机的标定效率，精确预测发动机参数，对一台氢气体积分数为20%的HCNG燃料发动机进行试验研究和性能预测分析。基于高转速低负载工况稳态标定试验数据，采用支持向量机（SVM）方法建立发动机参数关联模型，并利用不同寻优算法为模型寻找最优参数，以提高各项参数的预测精度。结果显示：若发动机运行于最大扭矩点火正时，则等效天然气比消耗（BSFC）最小，NO_x比排放（BSNO_x）也处于较理想的水平，尤其在增加氢气比例时，这些外特性有更加显著的提升；SVM模型可以较好地描述发动机输入参数与输出参数之前的非线性关系，自变量与因变量之间的相关性较强（决定系数R²均大于0.97），模型的预测精度较高，利用遗传算法得出的最优预测模型具有较高的泛化能力，扭矩、BSFC、BSNO_x的平均绝对百分比误差分别仅为1.23%、1.98%、5.43%。

关键词: 氢混天然气, 低负载工况, 支持向量机, 发动机标定

Abstract: A hydrogen-enriched compressed natural gas (HCNG) fuel engine with hydrogen volume fraction of 20% is experimentally studied and its performance is analyzed to improve the calibration efficiency of HCNG fuel engine and accurately predict the engine parameters. An engine parameter association model is established by using support vector machine (SVM) based on the steady-state experimental calibration data under high speed and low load conditions，and the different optimization algorithms are used to improve the prediction accuracy of engine parameters. Results show that the engine has the minimum brake specific fuel consumption (BSFC) and ideal brake specific NO_x consumption level (BSNO_x) at the minimum advance for best torque.The external characteristics is obviously improved when increasing the hydrogen ratio especially. The SVM model can describe the nonlinear relationship between the input and output parameters of the engine，and has high prediction accuracy and strong correlation between independent variables and dependent variables (all coefficients of determination R² are all greater than 0.97). The optimal prediction model derived using the genetic algorithm has a high generalization capability，with mean absolute percentage errors of only 1.23%，1.98%，and 5.43% for torque，BSFC，and BSNO_x, respectively.

Key words: hydrogen-enrichedcompressednaturalgas, lowloadcondition, supportvectormachine, enginecalibration

中图分类号:

TK46⁺4

段浩，陈晖，翟兆阳，韩雨，马凡华，崔亚辉. 基于支持向量机的氢混天然气发动机性能预测[J]. 兵工学报, 2022, 43(5): 1002-1011.

DUAN Hao， CHEN Hui， ZHAI Zhaoyang， HAN Yu， MA Fanhua， CUI Yahui. Performance Prediction of Hydrogen-enriched Compressed Natural Gas Engine Based on Support Vector Machine[J]. Acta Armamentarii, 2022, 43(5): 1002-1011.

参考文献

［1］ SAGAR S M V，AGARWAL A K.Experimental investigation of varying composition of HCNG on performance and combustion characteristics of a SI engine［J］.International Journal of Hydrogen Energy，2017，42(18):13234-13244.
［2］ NGUYEN D，CHOI Y，PARK C，et al.Effect of supercharger system on power enhancement of hydrogen-fueled spark-ignition engine under low-load condition［J］. International Journal of Hydrogen Energy，2021，46(9):6928-6936.
［3］ HORA T S，AGARWAL A K. Experimental study of the composition of hydrogen enriched compressed natural gas on engine performance，combustion and emission characteristics［J］. Fuel，2015，160:470-478.
［4］ VERMA G，PRASAD R K，AGARWAL R A，et al. Experimental investigations of combustion，performance and emission characteristics of a hydrogen enriched natural gas fuelled prototype spark ignition engine［J］. Fuel，2016，178:209-217.
［5］ BHASKER J P，PORPATHAM E.Effects of compression ratio and hydrogen addition on lean combustion characteristics and emission formation in a compressed natural gas fuelled spark ignition engine［J］. Fuel，2017，208:260-270.
［6］ MEHRA R K，DUAN H，JUKNELEVICIUS R，et al. Progress in hydrogen enriched compressed natural gas (HCNG) internal combustion engines—a comprehensive review［J］.Renewable & Sustainable Energy Reviews，2017，80:1458-1498.
［7］张朝山，熊树生，任晓帅，等.天然气掺氢发动机燃气供给系统设计与试验研究［J］.兵工学报，2012，33(10):1162-1167.
ZHANG C S，XIONG S S，REN X S，et al.Design of hydrogen enriched compressed natural gas engine fuel system and test research［J］.Acta Armamentarii，2012，33(10):1162-1167. (in Chinese)
［8］张红光，白小磊，韩雪娇，等.甲烷掺混氢气的燃烧特性试验研究［J］.兵工学报，2011，32(2):230-235.
ZHANG H G，BAI X L，HAN X J，et al. Research on combustion characteristics of methane-hydrogen-air mixture［J］.Acta Armamentarii，2011，32(2):230-235. (in Chinese)
［9］ HAO D，MEHRA R K， LUO S J， et al. Experimental study of hydrogen enriched compressed natural gas (HCNG) engine and application of support vector machine (SVM) on prediction of engine performance at specific condition［J］.International Journal of Hydrogen Energy，2020，45(8):5309-5325.
［10］ MA F H，HE Y T，DENG J，et al. Idle characteristics of a hydrogen fueled SI engine［J］. International Journal of Hydrogen Energy，2011，36(7):4454-4460.
［11］ MA F H，LI S，ZHAO J B，et al. Effect of compression ratio and spark timing on the power performance and combustion characteristics of an HCNG engine［J］.International Journal of Hydrogen Energy，2012，37(23):18486-18491.
［12］冷先银，葛琪琪，何志霞，等.预燃室式天然气掺氢发动机燃烧及排放模拟［J］.内燃机学报，2021，39(1):26-33.
LENG X Y，GE Q Q，HE Z X, et al. Numerical study on the combustion and emission characteristics of a pre-chamber engine fueled with hydrogen enriched compressed natural gas［J］. Transactions of CSICE，2021，39(1):26-33.(in Chinese)
［13］李兵，徐榕，贾春宁，等.基于自适应形态提升小波与改进非负矩阵分解的发动机故障诊断方法［J］.兵工学报，2013，34(3): 353-360.
LI B，XU R，JIA C N，et al. Engine fault iagnosis utilizing adaptive morphological lifting wavelet and improved non-negative matrix factorization［J］.Acta Armamentarii，2013，34(3):353-360. (in Chinese)
［14］ LIANG X Y，ZHAO B W，WANG K， et al. Application of response surface methodology for the joint optimization of performance and emission characteristics of a diesel engine［J］.International Journal of Green Energy，2021，18(7):697-707.
［15］李星，郭晓松，郭君斌.基于HOG特征和SVM的前向车辆识别方法［J］.计算机科学，2013，40(增刊2):329-332.
LI X，GUO X S，GUO J B.HOG-Feature and SVM based method for forward vehicle recognition［J］. Computer Science，2013，40(S2): 329-332. (in Chinese)
［16］ MARIANI V C，OCH S H，COELHO L D S，et al. Pressure prediction of a spark ignition single cylinder engine using optimized extreme learning machine models［J］.Applied Energy，2019，249:204-221.
［17］ WONG P K，TAM L M，LI K，et al.Engine idle-speed system modelling and control optimization using artificial intelligence［J］.Journal of Automobile Engineering，2010，224(1):55-72.
［18］ SABOUR M H，BEHESHTI A，ESFAHANIAN V. Reduction of experimental effort in conventional engine calibration process by using reduced order model［J］.Proceedings of the Institution of Mechanical Engineers，Part D: Journal of Automobile Engineering，2021，235(2/3):639-648.
［19］ MEHRA R K，DUAN H，LUO S J，et al. Laminar burning velocity of hydrogen and carbon-monoxide enriched natural gas (HyCONG): an experimental and artificial neural network study［J］. Fuel，2019，246:476-490.
［20］杜倩颖，倪计民，陈沁青，等.HILOMOT模型在柴油机快速标定试验中的应用［J］.内燃机学报，2018， 36(4):347-352.
DU Q Y，NI J M，CHEN Q Q，et al. Application of HILOMOT in rapid calibration test of diesel engine［J］.Transactions of CSICE，2018，36(4):347-352. (in Chinese)
［21］ WANG L J，HE M L，YANG Z Z. Research on optimal calibration technology for hydrogen-fueled engine based on nonlinear programming theory［J］. International Journal of Hydrogen Energy，2010，35(7):2747-2753.
［22］ VAPNIK V，IZMAILOV R. Knowledge transfer in SVM and neural networks［J］.Annals of Mathematics and Artificial Intelligence，2017，81(1/2):3-19.
［23］ CORTES C，VAPNIK V.Support-vector networks［J］. Machine Learing，1995，20(3):273-297.
［24］ MA F H，WANG M Y，JIANG L，et al.Performance and emission characteristics of a turbocharged CNG engine fueled by hydrogen-enriched compressed natural gas with high hydrogen ratio［J］. International Journal of Hydrogen Energy，2010，35(12):6438-6447.
［25］马凡华，王宇，刘海全，等.稀燃天然气掺氢发动机的热效率与排放特性［J］.内燃机学报，2008，26(1):43-49.
MA F H，WANG Y，LIU H Q，et al. Thermal efficiency and emission characteristics of a lean burn HCNG Engine［J］.Transactions of CSICE，2008，26(1):43-49.（in Chinese）

[1]	杨建新，兰小平，冯亚东，杨一铭，郭志明. 基于改进樽海鞘群和最小二乘支持向量机算法的新型弹药质量评估方法[J]. 兵工学报, 2022, 43(5): 1012-1022.
[2]	李良钰，苏铁熊，马富康，吴小军，徐春龙. 基于集合经验模态分解-支持向量机的高压共轨系统故障诊断方法[J]. 兵工学报, 2022, 43(5): 992-1001.
[3]	马润波，董丽虹，王海斗. 基于支持向量机的热喷涂层接触疲劳寿命预测[J]. 兵工学报, 2021, 42(12): 2743-2752.
[4]	雷震烁，刘松涛，温镇铭，葛杨. 复杂电磁环境下末制导雷达工作状态实时判别方法[J]. 兵工学报, 2021, 42(1): 83-90.
[5]	王成，吴岩，杨廷飞. 利用改进单分类支持向量机提升舰船尾流目标的检测准确率[J]. 兵工学报, 2020, 41(9): 1887-1893.
[6]	吕国俊，曹建军，郑奇斌，常宸，翁年凤，彭琮. 基于多目标蚁群优化的单类支持向量机相似重复记录检测[J]. 兵工学报, 2020, 41(2): 324-331.
[7]	邓青，薛青，罗佳. 基于支持向量机的坦克驾驶模拟训练结果分析[J]. 兵工学报, 2019, 40(9): 1953-1960.
[8]	代健，李泽，郝新红，栗苹. 基于目标联合特征提取的脉冲多普勒引信抗干扰方法[J]. 兵工学报, 2019, 40(2): 225-233.
[9]	于震梁, 孙志礼, 曹汝男, 王鹏. 基于支持向量机和卡尔曼滤波的机械零件剩余寿命预测模型研究[J]. 兵工学报, 2018, 39(5): 991-997.
[10]	谢萌蕤，赵兆，李阳，许志勇. 基于多尺度子带能量集特征的膛口波识别方法[J]. 兵工学报, 2018, 39(10): 1951-1957.
[11]	帅勇，宋太亮，王建平，沈洪. 一种改进的装备保障能力并联预测模型[J]. 兵工学报, 2016, 37(6): 1089-1095.
[12]	张彪，闫晓鹏，栗苹，李志强，池庆玺. 基于支持向量机的无线电引信抗扫频式干扰研究[J]. 兵工学报, 2016, 37(4): 635-640.
[13]	赵建忠，欧阳中辉，张磊，赵建印. 基于多智能体遗传算法优化的航空电子设备状态组合预测[J]. 兵工学报, 2016, 37(4): 727-734.
[14]	丛林虎，徐廷学，荀凯. 基于D-S证据理论的导弹制导控制系统的联合最小二乘支持向量机预测模型[J]. 兵工学报, 2015, 36(8): 1466-1472.
[15]	马润波，杜建华，许世蒙. 铜基复合材料分类判别方法研究[J]. 兵工学报, 2015, 36(5): 921-927.