论文标题
非线性耦合反应扩散系统中的时空图案形成,具有混合型模态不连续的盖素方法
Spatiotemporal pattern formation in nonlinear coupled reaction-diffusion systems with a mixed-type modal discontinuous Galerkin approach
论文作者
论文摘要
非线性耦合反应扩散(NCRD)系统在许多科学和工程领域的时空模式中很重要,包括物理和化学过程,生物学,生物学,电化学过程,分形,玻璃弹性材料,多孔培养基等。在这项研究中,为一维NCRD系统开发了混合型模态不连续的Galerkin方法,包括线性,灰色 - 斯科特,布鲁塞尔剂,等温化学和Schnakenberg模型,以产生时空模式。这些模型本质上代表了各种复杂的自然时空模式,例如斑点,斑点复制,条纹,六角形等。在这种方法中,提出了混合型公式,以解决扩散术语中出现的二阶导数。为了空间离散,使用了在正交缩放legendre多项式上的层次模态函数。此外,提出了针对NCRD系统的新反应项处理,证明了新的DG方案的内在特征,并防止由于极其非线性反应项而导致错误的溶液。所提出的方法将NCRD系统缩小到了时间的及时框架中,该框架是由明确的三阶TVD Runge-Kutta算法来解决的。使用当前方法生成的时空模式与文献中发现的时空模式非常可比。可以很容易地扩展这种方法,以处理发达的生物学和化学应用中的模型方程式出现的大型多维问题。
The nonlinear coupled reaction-diffusion (NCRD) systems are important in the formation of spatiotemporal patterns in many scientific and engineering fields, including physical and chemical processes, biology, electrochemical processes, fractals, viscoelastic materials, porous media, and many others. In this study, a mixed-type modal discontinuous Galerkin approach is developed for one- and two- dimensional NCRD systems, including linear, Gray-Scott, Brusselator, isothermal chemical, and Schnakenberg models to yield the spatiotemporal patterns. These models essentially represent a variety of complicated natural spatiotemporal patterns such as spots, spot replication, stripes, hexagons, and so on. In this approach, a mixed-type formulation is presented to address the second-order derivatives emerging in the diffusion terms. For spatial discretization, hierarchical modal basis functions premised on the orthogonal scaled Legendre polynomials are used. Moreover, a novel reaction term treatment is proposed for the NCRD systems, demonstrating an intrinsic feature of the new DG scheme and preventing erroneous solutions due to extremely nonlinear reaction terms. The proposed approach reduces the NCRD systems into a framework of ordinary differential equations in time, which are addressed by an explicit third-order TVD Runge-Kutta algorithm. The spatiotemporal patterns generated with the present approach are very comparable to those found in the literature. This approach can readily be expanded to handle large multi-dimensional problems that come up as model equations in developed biological and chemical applications.