Pulse Tube Refrigerator Performance Impact Mechanism Analysis And Multi-objective Optimization Research Based On Surrogate Modeling

With the development of technology,low-temperature refrigeration technology is playing an increasingly extensive and irreplaceable role in many fields such as aerospace,military,medical,energy,etc.The cold end of pulse tube refrigerator has no moving components,thus having advantages such as low vibration,high stability,and long service life,and has broad application prospects in the field of low-temperature refrigeration.The multiple parameters of the key components of the pulse tube refrigerator,such as the regenerator,pulse tube,and phase modulation component,have a significant impact on the overall refrigeration performance.Collaborative optimization of these key component parameters can significantly improve the overall refrigeration performance of the pulse tube refrigerator.Due to the high computational cost of multi-dimensional Computational Fluid Dynamics(CFD)numerical simulation for pulse tube refrigerator,the current approach mainly involves optimizing key component parameters one by one to determine the optimal refrigeration performance conditions,which can easily lead to local optima and make it difficult to obtain the global optimal performance of the refrigerator.In view of the above limitations,based on the CFD numerical model of the whole pulse tube refrigerator,this paper uses the principle of Gaussian process to establish its efficient Kriging surrogate model,and further uses the efficient global optimization method combining surrogate model and multi-objective genetic algorithm to carry out efficient systematic global optimization research on multidimensional key parameters for the overall performance of the refrigerator,providing an efficient and reliable method.Based on the parameter results before and after optimization,this article further utilizes the CFD numerical model of the pulse tube refrigerator to conduct a comparative analysis of the flow and heat transfer characteristics,along the path loss characteristics,and phase relationship characteristics inside the pulse tube refrigerator,in order to reveal the internal mechanism of the influence of key components and multiple parameters on the overall performance of the refrigerator.The main research work and related results are as follows:1.In this paper,the theoretical model derivation and analysis of the basic thermodynamic process involved in the pulse tube refrigerator are first carried out,and on this basis,the modeling method of the two-dimensional axisymmetric CFD numerical model of the pulse tube refrigerator is studied.Further,the Gaussian process modeling principle of Kriging surrogate model is theoretically expounded and process derivation is carried out,and the optimization principle of NSGA-Ⅱ(Non-dominated Sorting Genetic Algorithms-Ⅱ)is analyzed.Based on the above theoretical analysis,the CFD numerical model of pulse tube refrigerator and the Kriging surrogate model of its Gaussian process are established.Multi parameter optimization is carried out based on the response surface of the surrogate model and genetic algorithm.According to the optimization results,the overall research route for comparative analysis of flow heat transfer,along the way loss characteristics and phase relationship characteristics is carried out.2.Because the structural parameters of the key components of the regenerator and pulse tube have a significant impact on the refrigeration performance of the whole refrigerator,this paper establishes a Kriging surrogate model based on the principle of Gaussian process,and realizes the correlation mapping between the refrigeration performance response and the structural parameters of the components.Through the response surface analysis of the surrogate model,the influence law of the coupling between the diameter and length of the regenerator and pulse tube on the cooling temperature and cooling capacity is explored,Further combining genetic algorithm to obtain the optimization results of key component structural parameters,a comparative analysis of the results before and after optimization was conducted based on numerical simulation.The research results indicate that the essential reasons for the differences in refrigeration performance are the increase in internal edge loss of the regenerator caused by structural differences,the full development of eddy currents inside the pulse tube,and the deterioration of the temperature field.At the same time,the comparison of the results before and after optimization shows that the cooling temperature has decreased by 31.24%compared to the original configuration,and the cooling capacity has increased by 31.7%.3.Since the phase modulation parameters are another important factor that significantly affects the refrigeration performance of the whole refrigerator,this paper explores the phase modulation parameters on the phase modulation capability of the inertance tube and gas reservoir mechanism based on the single parameter numerical simulation,compares the sensitivity of each parameter to the phase modulation capability,further carries out the mapping analysis of key parameters on the phase modulation capability based on Kriging surrogate model,and obtains the optimal combination of phase modulation parameters combined with genetic algorithm.The research results indicate that the working frequency and structural parameters of the inertance tube have the most significant impact on the phase modulation ability by changing the dominant relationship between inertia effect and resistance effect.The optimization results significantly improve the rationality of the overall phase relationship,resulting in a significant increase in cooling capacity compared to before optimization.4.Based on the conclusion of the local optimization of the regenerator,pulse tube and phase modulation mechanism of the refrigerator,and considering the coupling effect of multiple parameters on the refrigeration performance,this paper carried out the global optimization of multiple parameters of the whole pulse tube refrigerator.Through the establishment of Kriging surrogate model,the correlation mapping between the refrigeration performance response and the parameters of the whole refrigerator was realized,and further combined with genetic algorithm to obtain the optimization results of the parameters of the whole refrigerator.The research results indicate that the operating parameters,and structural parameters of key components of the pulse tube refrigerator have a significant impact on the comprehensive refrigeration performance of the pulse tube refrigerator.The comparison of results before and after optimization shows that the cooling temperature is reduced by 39.9%compared to the original configuration,and the cooling capacity is increased by 135.25%.The refrigeration performance has been significantly improved compared to before optimization.