How to set solver's parameters to eliminate isolated equations' effect? #365
Description
Hello,
In CFD simulation, there are solid and liquid regions. When solving pressure or velocity fields, the solid region equations are set to isolated form (A(i,i)=1, b(i)=0) when assembling the control matrix. Do these isolated solid region equations affect the convergence of AMGX (especially for pressure field )? How can the solving parameters be set to eliminate the influence of these isolated solid region equations?