Abstract
The gradient method for inverse heat conduction problem has been modified to improve its solution characteristics by integrating the gradient over the time domain. The integrated gradient provides a solution that overcomes lack of information and noise amplification near the final time. However, the resolving power is significantly impaired and there is no parameter to control the resolution. This study presents a gradient method that combines the original and gradient integration methods, which is a two-phase method that sequentially conducts the gradient integration method and the original gradient method. Three different parameters that control the smoothness of the solution are introduced. The effects of the parameters are examined with test examples.
Acknowledgments
This study is conducted through the Practical Application Project of Advanced Microsystems Packaging Program of Seoul Technopark (No. 10029792) and through the Development Project of Large Surface Micro-Machining System Technology, both funded by Ministry of Knowledge Economy, Korea. In addition, the Industry Academic Cooperation Foundation in Seoul National University of Science and Technology also supported this study.