Effect Of External Sound Field On Soot Suppression Of Acetylene Diffusion Flame

In recent years,people’s demand for vehicles is increasing day by day.And the production of vehicles in China is also expanding day by day.While bringing convenience to people’s,the problem of automobile exhaust emission has also caused a great burden on the ecological environment.Soot particles play an important role in environmental pollution caused by motor vehicles exhaust emissions.How to improve soot suppression efficiency is of great significance to protect the environment.As a new combustion technology,compared with steady-state combustion,pulsating combustion has compact device,lower cost,higher combustion efficiency and combustion intensity,and can effectively reduce the emission of soot particles.In this study,the relation of acoustic oscillation angles(the angle between the acoustic oscillation direction and horizontal axis)and positions(the acoustic oscillation applied to the top,average,and bottom regions of the flame)with the soot suppression efficiency was investigated for acetylene laminar diffusion flames.The soot suppression efficiency depends on acoustic oscillation angles.With an increase in the flow rate,the angle required for optimum soot suppression increased.Moreover,acoustic oscillation positions affect the soot suppression efficiency.When the acoustic oscillation acts on the bottom region of the flame,the effect of soot suppression is substantial.And the LSTM model is obtained through Tensor Flow framework.The soot suppression efficiency is predicted combined with the relevant data of the experimental part.The suppression of soot particles is closely related to the vehicle environmental protection.Based on the data measured in the actual experiment,the data set is constructed and the model is trained to predict the data of soot suppression.The predicted value obtained by LSTM neural network can provide reference data for the environmental protection function of vehicles.The prediction results show that the soot suppression efficiency is the best when the acoustic oscillation acts on the top area of the flame.The predicted results are consistent with the above pulsating combustion experiments.The LSTM long-term and short-term memory neural network can use its own structural advantages to go deeper into the characteristics of long-time series data.So as to more accurately predict the development trend of pulsating combustion for soot suppression efficiency.