Adaptive thermal model for real-time peak temperature and cooling rate prediction in laser material processing

Laser materials processing (LMP) is highly sensitive to small disturbances in the process inputs, resulting in changes in the thermal dynamics, particularly, in the peak temperature and cooling rate. These thermal dynamic changes in turn vary the consistency of the material and the mechanical properties of the processed material. There are numerous models to estimate and control these thermal dynamics in LMP processes. However, none of these existing models can deal with thermal dynamic changes in real-time. In this research, an adaptive thermal model is developed that can deal with disturbances in real-time. A two-dimensional thermal model of the laser heat treatment (LHT) process is developed for real-time prediction of the peak temperature and cooling rate. In addition, an optimization algorithm is developed to make the thermal model adaptive to thermal dynamic changes in real-time. The adaptive thermal model has been implemented on an LMP setup, and multiple LHT experiments have been performed to study and validate the performance of the adaptive thermal model. Results show that the adaptive thermal model can effectively deal with the thermal dynamic changes in real-time, resulting in accurate peak temperature and cooling rate predictions during LHT. This work can be considered as an important step towards the development of accurate and computationally efficient models for real-time prediction and control of LMP.