Dynamical Systems Method (DSM) for solving nonlinear operator equations in Banach spaces

TL;DR

This paper introduces a dynamical systems method (DSM) for solving nonlinear operator equations in Banach spaces, proving convergence under certain conditions on the operator and its derivatives.

Contribution

The paper develops and analyzes a new DSM approach for nonlinear equations in Banach spaces, establishing convergence with specific decay rates of the parameter a(t).

Findings

Proves convergence of DSM to the solution y as t approaches infinity.

Establishes conditions on the operator's derivatives for convergence.

Demonstrates the method's applicability under certain solvability assumptions.

Abstract

Let $F(u)=h$ be an operator equation in a Banach space $X$, $\|F'(u)-F'(v)\|\leq \omega(\|u-v\|)$, where $\omega\in C([0,\infty))$, $\omega(0)=0$, $\omega(r)>0$ if $r>0$, $\omega(r)$ is strictly growing on $[0,\infty)$. Denote $A(u):=F'(u)$, where $F'(u)$ is the Fr\'{e}chet derivative of $F$, and $A_a:=A+aI.$ Assume that (*) $\|A^{-1}_a(u)\|\leq \frac{c_1}{|a|^b}$, $|a|>0$, $b>0$, $a\in L$. Here $a$ may be a complex number, and $L$ is a smooth path on the complex $a$-plane, joining the origin and some point on the complex $a-$plane, $0<|a|<\epsilon_0$, where $\epsilon_0>0$ is a small fixed number, such that for any $a\in L$ estimate (*) holds. It is proved that the DSM (Dynamical Systems Method) \bee \dot{u}(t)=-A^{-1}_{a(t)}(u(t))[F(u(t))+a(t)u(t)-f],\quad u(0)=u_0,\ \dot{u}=\frac{d u}{dt}, \eee converges to $y$ as $t\to +\infty$, where $a(t)\in L,$ $F(y)=f$, $r(t):=|a(t)|$, and $r(t)=c_4(t+c_2)^{-c_3}$, where $c_j>0$ are some suitably chosen constants, $j=2,3,4.$ Existence of a solution $y$ to the equation $F(u)=f$ is assumed. It is also assumed that the equation $F(w_a)+aw_a-f=0$ is uniquely solvable for any $f\in X$, $a\in L$, and $\lim_{|a|\to 0,a\in L}\|w_a-y\|=0.$