Abstract

A meshfree postprocessing for the computation of field lines suggests itself in the case of fast boundary element methods. Field values are only calculated and stored in points, which are absolutely necessary for field line computations. Hence, the total amount of processed data is dramatically reduced in comparison to a mesh-based approach along with precomputed field values in all mesh nodes. An automatic and robust domain detection method enables reliable identifications of intersections of field lines with domain boundaries even in the case of complex-shaped surfaces. A combination of automatic domain detection and adaptive step size control of the underlying Runge–Kutta–Fehlberg method results in correctly and efficiently computed field lines, including sharp bends at domain boundaries.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call