Two-scale convergence of first-order operators (Q2454948)

The notion of two-scale convergence and some of its main properties are recalled. The weak two-scale limit of the gradient of bounded sequences in \(W^{1,p}\) is studied: if \(u_\varepsilon \rightarrow u\) in \(W^{1,p}\) a sequence \(u_{\varepsilon}\) is constructed such that \(u_{\varepsilon}(x) \rightarrow u_1(x,y)\) and \(\nabla u_\varepsilon(x) \rightarrow \nabla u(x) + \nabla_y u_1(x,y)\) weakly two-scale. Analogous constructions are performed for \((W^{1,p})^N\), \(L^2_{\text{rot}}(\mathbb R^3)^3\), \(L^2_{\text{div}}(\mathbb R^N)^N\), \(L^2_{\text{div}}(\mathbb R^N)^{N^2}\). The applications to some classical equations of electromagnetism and continuum mechanics are given and the method is applied to quasilinear elliptic equations \(\nabla \times [ A(u_\varepsilon(x),x, \frac{x}{\varepsilon})\nabla \times u_\varepsilon]= f\).