Hi Zhaoyu, I am not sure what to suggest in this case. The techniques should also work in this case but I doubt they have been full tested in this scenario. Have you tried using both the localtoglobal approach and the standard mass matrix inverse and both do not work? Cheers, Spencer. Spencer Sherwin FREng, FRAeS Head of Aerodynamics Section, Director of Research Computing Service, Professor of Computational Fluid Mechanics, Department of Aeronautics, South Kensington Campus, Imperial College London, SW7 2AZ, UK Phone: +44 (0)20 7594 5052 http://www.imperial.ac.uk/people/s.sherwin/ On 12 Oct 2020, at 06:41, 王昭予 <clemden@sjtu.edu.cn<mailto:clemden@sjtu.edu.cn>> wrote: Dear Prof. Sherwin Thanks for your reply. There is a problem is when I use variable polynomial order in the field, the results near the boundary between 2 domains with different polynomial order cant be smoothed by C0projection module. So how can I handle the discontinuous results in this case? Many thanks, Zhaoyu Wang Shanghai Jiao Tong University