Existence of nonnegative solutions for second order \(m\)-point boundary value problems at resonance (Q627135): Difference between revisions

Consider the \(m\)-point boundary value problem \[ \begin{gathered} (\rho(t)x')'= f(t,x(t))\quad\text{for }0< t< 1,\\ x(0)= 0,\quad x(1)= \sum^{m-2}_{i=1} \alpha_i x(\eta_i),\end{gathered}\tag{\(*\)} \] where \(0< \eta_1<\cdots< \eta_{m-2}< 1\), \(\alpha_i\in \mathbb{R}\) for \(1\leq i\leq m-2\), \(\rho\in C^1([0, 1],\mathbb{R}^+)\), \(f\in C([0, 1]\times \mathbb{R},\mathbb{R})\). the authors assume \(\sum^{m-2}_{i=1} \alpha_i= 1\), that is, the operator \(Lx:= (\rho(t)x')'\) is not invertible (the so-called resonance case). Using an extension of a fixed point theorem for an increasing operator in a normal cone, they establish a minimal and a maximal nonnegative solution of \((*)\) and construct sequences of functions converging monotone and uniformly to these extremal solutions.