A linearly implicit and local energy-preserving scheme for the sine-Gordon equation based on the invariant energy quadratization approach

TL;DR

This paper introduces a novel linearly implicit, energy-preserving numerical scheme for the sine-Gordon equation, utilizing the invariant energy quadratization approach to ensure local energy conservation and optimal error bounds.

Contribution

The paper develops a new energy-preserving scheme based on invariant energy quadratization, specifically designed for the sine-Gordon equation, with proven energy conservation and error estimates.

Findings

The scheme exactly preserves the discrete local energy conservation law.

The scheme achieves unconditional and optimal error estimates.

Numerical examples confirm theoretical results and advantages over existing methods.

Abstract

In this paper, we develop a novel, linearly implicit and local energy-preserving scheme for the sine-Gordon equation. The basic idea is from the invariant energy quadratization approach to construct energy stable schemes for gradient systems, which are energy dispassion. We here take the sine-Gordon equation as an example to show that the invariant energy quadratization approach is also an efficient way to construct linearly implicit and local energy-conserving schemes for energy-conserving systems. Utilizing the invariant energy quadratization approach, the sine-Gordon equation is first reformulated into an equivalent system, which inherits a modified local energy conservation law. The new system are then discretized by the conventional finite difference method and a semi-discretized system is obtained, which can conserve the semi-discretized local energy conservation law. Subsequently, the linearly implicit structure-preserving method is applied for the resulting semi-discrete system to arrive at a fully discretized scheme. We prove that the resulting scheme can exactly preserve the discrete local energy conservation law. Moveover, with the aid of the classical energy method, an unconditional and optimal error estimate for the scheme is established in discrete $H_h^1$-norm. Finally, various numerical examples are addressed to confirm our theoretical analysis and demonstrate the advantage of the new scheme over some existing local structure-preserving schemes.