The precision of temperature field control in the hypersonic wind tunnel directly affects the accuracy of wind tunnel test data. In view of the control problems of hypersonic wind tunnel temperature field control, such as large delay, nonlinear and multivariable coupling, phase space reconstruction of data affecting temperature was carried out and support vector regression was applied to the hypersonic wind tunnel temperature field predictive control to improve the accuracy and efficiency of hypersonic wind tunnel temperature field control. At the same time, considering that the selection of kernel function in support vector regression machine and the optimization of kernel function parameters affect the accuracy of prediction results, the support vector machine model was established based on different kernel functions, and the optimal kernel function was selected through comparative verification and analysis, and the corresponding PSR-SVR model was established to predict the temperature field of the hypersonic wind tunnel, so as to improve the temperature prediction accuracy. The analysis of actual temperature field data shows the effectiveness of the proposed method.