An Efficient Flexible Division Algorithm for Predicting Temperature-Fields of Mechatronic System with Manufacturing Applications

This paper presents a new thermal-field modeling method referred to here as a flexible division algorithm (FDA) for predicting the temperature fields of a mechatronic system, and its real-time applications where thermal effects have a significant influence on the performance and reliability of the final products. This algorithm, which takes advantages of the flexible division in 3-D space to deal with the spatial distribution of thermal fields, is built upon physical laws to derive the governing equations in state-space representation that facilitates the reconstruction and control of the thermal field being analyzed. Three numerical models (involving both Cartesian and cylindrical coordinates) are illustrated to highlight the effectiveness and usefulness of the FDA for real-time modeling and computing. In the context of a thin-walled component machining application, the FDA is evaluated numerically and validated experimentally. Its solutions agree well with results computed using commercial finite-element analysis (FEA) software, confirming its ability to obtain accurate results, but with significantly less computation time, and its effectiveness as a complement to FEA when real-time computing of a physical field is required.