Aiming at the problems of large temperature error and poor accuracy in infrared thermal imaging nondestructive testing system, a detection method based on nondestructive testing simulation model is proposed. Several surface temperatures at the excitation point and defect center in the same experimental environment are selected as reference values to modify and correct the simulation model. First, the temperature variation trend of both sides of the excitation point is analyzed by changing the laser excitation time to reveal the influence of defects on the temperature field of the specimen. Then, the temperature resistance effect of the defect is analyzed by comparing the temperature values of the regions on both sides of the excitation point. Finally, the fitting curve and function relationship between the excitation time and the maximum temperature difference of the temperature measuring point are obtained by using the method of data fitting. The quantitative relationship between the excitation time and the maximum temperature difference of the temperature measuring point is analyzed quantitatively. The results show that the detection method based on the simulation model of nondestructive testing has a remarkable effect on the detection of parts defects, it can solve the problems of large temperature measurement error and poor accuracy, and has a good ability to detect parts defects.