The role of interplay between coefficients in the \(G\)-convergence of some elliptic equations (Q1687602)

The paper studies solutions of \[ \-\nabla \cdot (M_n(x)\nabla u_n)+a_n u_n=f_n\text{ in }\Omega,\quad u_n=0\text{ on }\partial \Omega, \] in \(W^{1,2}_0(\Omega )\cap L^\infty (\Omega ),\) where \(\Omega \) is a bounded open subset of \(\mathbb R^N\) and \(M_n\) are matrices. It is shown that if \(M_n \to M_0\), \(a_n \to a_0\), \(f_n \to f_0\) in some sense, then \(u_n\) converges weakly in \(W^{1,2}_0(\Omega )\) and weakly\(^*\) in \(L^\infty (\Omega )\) to the solution \(u_0\) of \[ -\nabla \cdot (M_0(x)\nabla u_0)+a_0 u_0=f_0\text{ in }\Omega , \quad u_0=0\text{ on }\partial \Omega. \]