Research On The Mechanism Of Active Phase Shift And Optimization Method Of Pulse Tube Refrigerator

With the continuous development of modern science and industry,low-temperature refrigeration technology has been increasingly widely applied in fields such as energy,transportation,medical treatment,electronic technology,major scientific installations,national defense,aerospace,etc.,as many instruments and equipment need to perform at low temperatures to achieve excellent performance.Pulse tube refrigerators(PTRs)are more suitable for special fields with high requirements due to their advantages such as small size,light weight,high efficiency,low vibration and long service life,which therefore stand out from many small low-temperature refrigerators and have become a mainstream choice at present.In recent years,with the continuous improvement of refrigeration requirements,pulse tube refrigerators are developing towards miniaturization.However,when the compression volume of the compressor is extremely small,traditional phase shifters(such as inertance tube and reservoir)gradually cannot meet the phase shift of refrigerators,active phase shifters are therefore receiving attention.At present,research on active phase shift pulse tube refrigerators(APSPTRs)mainly focuses on one-dimensional simulation,lacking analysis of multidimensional flow and heat transfer characteristics within the cold finger,which poses a challenge to revealing the essence of the active phase shift mechanism of pulse tube refrigerators.At the same time,relying on the traditional thermodynamic model to optimize the parameters of the refrigerator is more complex,which is not conducive to the realization of global optimization of multiple parameters.In view of the above research limitations,this paper establishes a two-dimensional CFD model to compare and analyze the phase shift mechanism of the APSPTR and the traditional inertance tube pulse tube refrigerator(ITPTR),further establishes its Kriging surrogate model and completes the multi parameter optimization of APSPTR with genetic algorithm.The specific research work is as follows:1.A multi-dimensional flow and heat transfer simulation was conducted on APSPTR and the ITPTR using CFD model,and the differences in cooling performance between the two at different temperatures were compared.Simulation results show that the former outperforms the latter in terms of phase relationship within the regenerator,overall cooling capacity and coefficient of performance(COP)at different cooling temperatures.Due to the power recovery effect and excellent phase shift ability of the displacer,the hot end loss and enthalpy flow loss of the regenerator in the former are both smaller than those in the latter.Due to the stronger jet effect caused by the displacer,the non-ideal expansion loss in the pulse tube is greater than that in the latter,and the conclusion is also confirmed through comparative analysis of temperature and flow contours.2.Based on the cooling advantages of APSPTR,multi-dimensional simulation was used to conduct in-depth comparative studies on the impact mechanisms of pulse tube radius,pulse tube length,frequency and displacer parameters on various losses and cooling performance of the refrigerator.The results show that increasing the radius or length of the pulse tube can to some extent improve the figure of merit(FOM)of pulse tube to improve the cooling performance.An increase in frequency will reduce the FOM inside the pulse tube,while affecting the phase relationship of the regenerator to change the enthalpy flow loss.There is an optimal value for FOM of the pulse tube regarding the parameters of the displacer,and the enthalpy flow loss of the regenerator mainly depends on the influence of the displacer on the phase relationship.3.Based on the complex influence mechanism of each parameter of APSPTR on the cooling performance,the Kriging surrogate model,which is more efficient than the original CFD model,was established to establish the correlation mapping between the parameters and the cooling capacity,COP,total input power and cold end phase angle,and genetic algorithm was used to achieve single objective and multi-objective optimization of cooling performance.The results show that the root mean square errors(RMSE)of the established surrogate model for accurate prediction of cooling capacity,COP,total input power and cold end phase angle are 0.08 W,3.46 W,0.24%and 1.12° respectively,which proves its reliability.Kriging response surface analysis shows that the more parameters involved in optimization,the more advantageous it is to achieve better cooling performance.Global single objective optimization can achieve a maximum cooling capacity of 7.29 W and a maximum COP of 7.38%.A multi-objective optimization was achieved using Pareto optimal solution,resulting in a comprehensive optimization of cooling capacity and COP,and the values of each parameter were determined.The corresponding cooling capacity was 6.62 W and COP was 7.02%through simulation,which significantly improved the cooling performance compared to the 2.65 W and 3.99%before optimization.