Abstract: Nonlinear regression based on Least square method is used to predict the equilibrium temperature with experimental data,The results are in accordance with the analysis that the measured temperature will exponentially vary with time from a certain moment after the beginning of the experiment.case,and the equilibrium teperature can be predicted efficiently by the method.The choosing of regression data is important to the reasonableness of the predicted equilibrium temperature,while choosing of the initial value of the regression parameters will not affect the results when they are convergent.