摘要
为探究不同频率交流电场助燃的内在机制,利用定容燃烧试验平台比较了低频(40、60、80、100 Hz)和高频(15、20、25、30 kHz)交流电场对甲烷/空气稀薄燃烧(过量空气系数为1.2、1.4、1.6)火焰的影响,并利用机器学习方法对混合气的燃烧特征参数进行了预测研究。结果表明:低频和高频交流电场下,火焰在电场方向上均被拉伸,不同频率交流电场对火焰传播的促进效果处于同一量级,但低频交流电场下的火焰锋面更加稳定;高频交流电场对燃烧特性参数的影响更加显著,其对应的压力峰值和压力升高率峰值均比低频交流电场下变化更大;采用支持向量机方法构建的平均火焰传播速度和燃烧压力峰值的预测模型相关系数均高于0.998,平均绝对百分比误差和希尔不等系数分别小于1.093%和0.007,均具有良好的预测性能和泛化能力。研究进一步验证了低频和高频交流电场的助燃机制,证实了机器学习方法应用于基础燃烧特征参数预测的可行性,丰富了电场助燃理论。
This paper explores the internal combustion promotion mechanism of AC electric field with different frequencies.A comparison is made on the effect of low-frequency(40,60,80,100 Hz)and high-frequency(15,20,25,30 kHz)AC electric fields on methane/air lean combustion(excess air ratio of 1.2,1.4,1.6)flames with a constant volume combustion test platform.The machine learning method is applied to predict the combustion characteristic parameters of the mixture under various AC fields.The results show that,under low-frequency and high-frequency AC fields,the flame is stretched in the electric field direction;the effect in promoting flame prop agation is of the same order of magnitude,but the flame front under low-frequency AC fields is more stable;the effect of high-frequency AC fields on combustion characteristic parameters(peak pressure and peak pressure rise rate)is more significant than under low-frequency ones;the prediction models built by the support vector machine method have excellent prediction performance and generalization ability,with the correlation coefficients of higher than 0.998,and the average absolute percentage error and Hill’s coefficients of inequality of less than 1.093%and 0.007,respectively.The research further verified the combustion promotion mechanism of low-frequency and high-frequency AC fields and confirmed the feasibility of machine learning method in the prediction of fundamental combustion characteristic parameters,enriching the electric field combustion theory.
作者
段浩
尹晓军
寇海亮
张猛
曾科
DUAN Hao;YIN Xiaojun;KOU Hailiang;ZHANG Meng;ZENG Ke(School of Energy and Power Engineering,Xi’an Jiaotong University,Xi’an,710049,China)
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2023年第5期118-127,148,共11页
Journal of Xi'an Jiaotong University
基金
国家自然科学基金资助项目(52176130)。
关键词
电场辅助燃烧
双离子风效应
电化学效应
机器学习
支持向量机
electric field assisted combustion
bi-ionic wind effect
electrical-chemical effect
machine learning
support vector machine