A two-step SOR-Newton method for nonsmooth equations
From MaRDI portal
A two step successive overrelaxation Newton method is proposed for solving a system of nonlinear equations \(F(x)=0\). Here \(F: {\mathbb R}^n\to {\mathbb R}^n\) is strongly monotone, locally Lipschitz continuous but not differentiable. Convergence of the method is established and approaches of choosing the relaxation parameters are discussed. Numerical examples are presented to illustrate fast convergence of the method.
Recommendations
Cites work
- A generalized Jacobian based Newton method for semismooth block-triangular system of equations
- A modified secant method for semismooth equations
- A nonsmooth version of Newton's method
- Approximate Newton methods for nonsmooth equations
- Efficient continuation Newton-like method for solving systems of non-linear equations
- Iterative Solution of Nonlinear Equations in Several Variables
- On a new class parametrized Newton-like method for semismooth equations
- On convergence of SOR methods for nonsmooth equations
- On the convergence of Newton's method for a class of nonsmooth operators
- On the secant method for solving nonsmooth equations
Cited in
(6)- Numerical solutions to nonsmooth Dirichlet problems based on lumped mass finite element discretization
- Two-step modified Newton method for nonlinear Lavrentiev regularization
- Convergence of the modified SOR-Newton method for non-smooth equations
- On convergence of AOR-Newton methods for nonsmooth equations
- On convergence of SOR methods for nonsmooth equations
- Some quadrature-based versions of the generalized Newton method for solving nonsmooth equations
This page was built for publication: A two-step SOR-Newton method for nonsmooth equations
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q838030)