The temperature distributions were numerically calculated for the two-dimensional transient conduction heat transfer problem of a square plate. The obtained temperature distributions were converted into colors to create images, and they were provided as learning and test data of CNN. Classification and regression networks were constructed to predict representative wall temperatures through CNN analysis. As results, the classification networks predicted the representative wall temperatures with an accuracy of 99.91% by erroneously predicting only 1 out of 1100 images. The regression networks predicted the representative wall temperatures within errors of C. From this fact, it was confirmed that the deep learning techniques are applicable to the transient conduction heat transfer problems.

In this analysis, the analytical model was verified through the normal mode analysis of the piston for the 2.9 liter IDI (indirect injection) engine. Heat transfer analysis was carried out by selecting two cases of applied temperature using the validated model. The first case was a condition of 350℃ on the piston upper surface and 100℃ on the piston body and inner wall. In the second case, the conditions were set to give a temperature of 400℃ on the upper surface of the piston and 100℃ on the piston body and the inner wall. In addition, the temperature distribution due to heat transfer was obtained for the pistons with boundary conditions of two cases, and then the thermal stress distribution due to thermal expansion was obtained using the input. Using this analysis result, the thermal stress caused by thermal expansion due to the thermal conduction of the piston is examined and used as the basic data for design.