摘要:反應(yīng)-擴(kuò)散方程在科學(xué)和工程的許多分支中有著重要的應(yīng)用,對(duì)此類方程數(shù)值解的研究具有重要意義. 鑒于計(jì)算域的復(fù)雜形狀、大量的自由度等導(dǎo)致計(jì)算非常困難,提出張量積型二元三次 B 樣條法求解一 類分?jǐn)?shù)階反應(yīng)-擴(kuò)散方程和交叉反應(yīng)擴(kuò)散系統(tǒng),首先計(jì)算得出二元三次 B 樣條擬插值的矩陣表達(dá)式,然后利用 Matlab 進(jìn)行數(shù)值模擬,最后將數(shù)值模擬解與精確解進(jìn)行對(duì)比. 研究表明,當(dāng)變量 t 的迭代次數(shù)較低時(shí),所提方法行之有效.
關(guān)鍵詞:反應(yīng)-擴(kuò)散方程;B 樣條擬插值;張量積型;數(shù)值模擬
中圖分類號(hào):TP391. 41 文獻(xiàn)標(biāo)志碼:A 文章編號(hào):1001-8395(2024)03-0411-11
doi:10. 3969 / j. issn. 1001-8395. 2024. 03. 014
對(duì)流-擴(kuò)散-反應(yīng)系統(tǒng),包括對(duì)流-擴(kuò)散和反應(yīng)-擴(kuò)散方程,在科學(xué)和工程的許多分支中有重要的應(yīng)用 [1],這些應(yīng)用包括方程模型運(yùn)輸動(dòng)力學(xué) [2-4]、人口動(dòng)力學(xué) [5]、化學(xué)反應(yīng)和燃燒 [6]、傳熱和傳質(zhì) [7-9]、 環(huán)境問(wèn)題中濃度或污染的演變 [10]以及趨化性、模式形成和細(xì)胞生長(zhǎng)過(guò)程 [11-15]. 因此,此類方程的準(zhǔn)確解對(duì)于準(zhǔn)確描述問(wèn)題中的動(dòng)力學(xué)和傳遞過(guò)程具有重要意義.
由于計(jì)算域的復(fù)雜形狀、相互作用的強(qiáng)度、大量的自由度以及所使用的數(shù)值求解器的穩(wěn)定性,反應(yīng)-擴(kuò)散方程所涉及的相關(guān)復(fù)雜性通常會(huì)非??量蹋?Henry 等 [16]描述了用于模擬包含反常擴(kuò)散的活化劑-抑制劑動(dòng)態(tài)系統(tǒng)的雙組分反應(yīng)-擴(kuò)散方程;文獻(xiàn)[9]研究了一類反應(yīng)-擴(kuò)散方程的前沿動(dòng)態(tài)系統(tǒng).目前已經(jīng)開(kāi)發(fā)了許多數(shù)值方法來(lái)解決分?jǐn)?shù)反應(yīng)擴(kuò)散問(wèn)題,例如:有限差分法 [5]、有限元法 [17-18]和光譜法 [19]. 然而,由于分?jǐn)?shù)階差分算子的非局部特性 [20-21],數(shù)值方案的穩(wěn)定性通常變得非常敏感 [21].
此外,反應(yīng)-擴(kuò)散方程數(shù)值離散化通常會(huì)生成完整且密集的系數(shù)矩陣,這會(huì)導(dǎo)致嚴(yán)重的計(jì)算困難.然而,與反應(yīng)-擴(kuò)散系統(tǒng)不同,交叉擴(kuò)散系統(tǒng)允許交叉擴(kuò)散系數(shù)為負(fù). 顯著的特點(diǎn)是反應(yīng)擴(kuò)散的均勻穩(wěn)態(tài)是穩(wěn)定的,但對(duì)于交叉擴(kuò)散系統(tǒng)則不穩(wěn)定 [22]. 例如具有自擴(kuò)散和交叉擴(kuò)散系統(tǒng) [22-23]、非均勻反應(yīng)系統(tǒng) [24-25]、流行病模型 [26-27]和植被模式多樣性 [28]的捕食者-獵物模型. 交叉反應(yīng)-擴(kuò)散過(guò)程在自然界中質(zhì)量和化學(xué)成分的傳輸中起著重要作用.這些系統(tǒng)包含豐富多樣的行為. 由于它們的特性, 人們對(duì)交叉反應(yīng)-擴(kuò)散方程的研究越來(lái)越感興趣.
通過(guò)使用有限體積方法 [29-31]、有限差分方法 [32],已經(jīng)開(kāi)發(fā)了一些交叉擴(kuò)散模型. 然而,這些方法在處理復(fù)雜幾何形狀的傳輸時(shí)被證明具有挑戰(zhàn)性.本文提出二元三次 B 樣條擬插值方法解決反應(yīng)-擴(kuò)散方程的數(shù)值逼近問(wèn)題,是一元三次 B 樣條擬插值的推廣形式,可以直接構(gòu)造,不需要求解線性方程組,具有良好的保形性、計(jì)算量小等優(yōu)點(diǎn),它在計(jì)算幾何和數(shù)值逼近方面有著廣泛的應(yīng)用 [33-34].
文獻(xiàn)[35-37]已證實(shí)用一維三次 B 樣條擬插值法求解 Burgers-Huxley 等偏微分方程得到的數(shù)值解與解析解非常吻合. 因此本文用二元 B 樣條擬插值法求解更為復(fù)雜的偏微分方程并考慮其精度.
本文的安排如下:第一節(jié)基于三次 B 樣條基函數(shù)的表達(dá)式,構(gòu)造二元三次 B 樣條擬插值算子,并給出其張量積型的矩陣表達(dá)式;第二節(jié)分別對(duì)二元分?jǐn)?shù)階反應(yīng)-擴(kuò)散方程和交叉反應(yīng)-擴(kuò)散方程的 2 個(gè)例子進(jìn)行數(shù)值實(shí)驗(yàn),得到相應(yīng)的數(shù)值模擬圖和數(shù)值模擬解并與其精確解進(jìn)行對(duì)比分析;最后在第三節(jié)得出相應(yīng)的結(jié)論.
4 結(jié)論
本文主要探討了基于二維三次 B 樣條擬插值法求解分?jǐn)?shù)階反應(yīng)-擴(kuò)散方程和交叉反應(yīng)-擴(kuò)散方程的誤差,利用這種方法不需要求解線性方程組,具有計(jì)算量小、保多項(xiàng)式性等優(yōu)點(diǎn),從數(shù)值模擬的結(jié)果和相應(yīng)的誤差分析可知,在迭代次數(shù)較低時(shí),用這種方案所得到的數(shù)值解和解析解非常吻合. 數(shù)值模擬解逼近最優(yōu)解,如果迭代次數(shù)較高,也即 t→∞時(shí),運(yùn)用二維 B 樣條擬插值法誤差偏大,需要進(jìn)一 步探究該方法的收斂性,但是誤差在可控的范圍內(nèi),因此對(duì)于一些復(fù)雜的整數(shù)階偏微分方程可以利用此方案得到相應(yīng)的數(shù)值解和數(shù)值模擬圖.
參考文獻(xiàn)
[1]AHMED S,LIU X F. High order integration factor methods for systems with inhomogeneous boundary conditions[J]. Journal ofComputational and Applied Mathematics,2019,348:89-102.
[2]ALZAHRANI S S,KHALIQ A Q M. Fourier spectral exponential time differencing methods for multi-dimensional space-fractionalreaction-diffusion equations[J]. Journal of Computational and Applied Mathematics,2019,361:157-175.
[3]BRATSOS A G,KHALIQ A Q M. An exponential time differencing method of lines for Burgers-Fisher and coupled Burgers equa-tions[J]. Journal of Computational and Applied Mathematics,2019,356:182-197.
[4]BURRAGE K,HALE N,KAY D. An efficient implicit FEM scheme for fractional-in-space reaction-diffusion equations[J].SIAM Journal on Scientific Computing,2012,34(4):A2145-A2172.
[5]CHEN H,L S J,CHEN W P. Finite difference / spectral approximations for the distributed order time fractional reaction diffu-sion equation on an unbounded domain[J]. Journal of Computational Physics,2016,315:84-97.
[6]CHEN S Q,ZHANG Y T. Krylov implicit integration factor methods for spatial discretization on high dimensional unstructuredmeshes:application to discontinuous Galerkin methods[J]. Journal of Computational Physics,2011,230(11):4336-4352.
[7]COLIZZA V,PASTOR-SATORRAS R,VESPIGNANI A. Reaction-diffusion processes and metapopulation models in heterogene-ous networks [J]. Nature Physics,2007,3(4):276-282.
[8]COX S M,MATTHEWS P C. Exponential time differencing for stiff systems[J]. Journal of Computational Physics,2002, 176(2):430-455.
[9]DEL-CASTILLO-NEGRETE D,CARRERAS B A ,LYNCH V E. Front dynamics in reaction-diffusion systems with Levy flights: a fractional diffusion approach[J]. Physical Review Letters,2003,91:018302.
[10]FERREIRA S C,MARTINS M L,VILELA M J. Reaction-diffusion model for the growth of avascular tumor[J]. Physical Re-view E,2002,65(2):021907.
[11]GARVIE M R. Finite-difference schemes for reaction-diffusion equations modeling predator-prey interactions in Matlab[J]. Bul-letin of Mathematical Biology,2007,69(3):931-956.
[12]GATENBY R A,GAWLINSKI E T. A reaction-diffusion model of cancer invasion[J]. Cancer Res,1996,56(24):5745-5753.
[13]GU X M,HUANG T Z,ZHAO X L,et al. Strang-type preconditioners for solving fractional diffusion equations by boundary val-ue methods[J]. Journal of Computational and Applied Mathematics,2015,277:73-86.
[14]HAJIPOUR M,JAJARMI A,BALEANU D,et al. On an accurate discretization of a variable-order fractional reaction-diffusionequation[J]. Communications in Nonlinear Science and Numerical Simulation,2019,69:119-133.
[15]HENRY B I,WEARNE S L. Fractional reaction-diffusion[J]. Physica A:Statistical Mechanics and Its Applications,2000,276(3 / 4):448-455.
[16]HENRY B I,WEARNE S L. Existence of Turing instabilities in a two-species fractional reaction-diffusion system[J]. SIAMJournal on Applied Mathematics,2002,62(3):870-887.
[17]LI M,GU X M,HUANG C M,et al. A fast linearized conservative finite element method for the strongly coupled nonlinear frac-tional Schrdinger equations[J]. Journal of Computational Physics,2018,358:256-282.
[18]LIU Y,DU Y W,LI H,et al. An H1 -Galerkin mixed finite element method for time fractional reaction-diffusion equation[J].Journal of Applied Mathematics and Computing,2015,47(1 / 2):103-117.
[19]PINDZA E,OWOLABI K M. Fourier spectral method for higher order space fractional reaction-diffusion equations[J]. Commu-nications in Nonlinear Science and Numerical Simulation,2016,40:112-128.
[20]MEERSCHAERT M M,TADJERAN C. Finite difference approximations for two-sided space-fractional partial differential equa-tions[J]. Applied Numerical Mathematics,2006,56(1):80-90.
[21]ZHANG L,SUN H W,PANG H K. Fast numerical solution for fractional diffusion equations by exponential quadrature rule[J].Comput Phys,2015,299:130-143.
[22]GUIN L N. Existence of spatial patterns in a predator-prey model with self- and cross-diffusion[J]. Applied Mathematics andComputation,2014,226:320-335.
[23]PETROVSKII S V,MALCHOW H. A minimal model of pattern formation in a prey-predator system[J]. Mathematical and Com-puter Modelling,1999,29(8):49-63.
[24]GOLOVIN A A,MATKOWSKY B J,VOLPERT V A. Turing pattern formation in the Brusselator model with superdiffusion [J].SIAM Journal on Applied Mathematics,2008,69(1):251-272.
[25]GAMBINO G,LOMBARDO M C,SAMMARTINO M. Pattern formation driven by cross-diffusion in a 2D domain[J]. NonlinearAnalysis:Real World Applications,2013,14(3):1755-1779.
[26]DIEKMANN O,KRETZSCHMAR M. Patterns in the effects of infectious diseases on population growth[J]. Journal of Mathe-matical Biology,1991,29(6):539-570.
[27]FAN Y. Pattern formation of an epidemic model with cross diffusion [J]. Applied Mathematics and Computation,2014, 228:311-319.
[28]VON HARDENBERG J,MERON E,SHACHAK M,et al. Diversity of vegetation patterns and desertification[J]. Physical Re-view Letters,2001,87(19):198101.
[29]ANDREIANOV B,BENDAHMANE M,RUIZ-BAIER R. Analysis of a finite volume method for a cross-diffusion model in popu-lation dynamics[J]. Mathematical Models and Methods in Applied Sciences,2011,21(2):307-344.
[30]RUIZ-BAIER R,TIAN C R. Mathematical analysis and numerical simulation of pattern formation under cross-diffusion[J].Nonlinear Analysis:Real World Applications,2013,14(1):601-612.
[31]LIN Z G,RUIZ-BAIER R,TIAN C R. Finite volume element approximation of an inhomogeneous Brusselator model with cross-diffusion[J]. Journal of Computational Physics,2014,256:806-823.
[32]WANG W M,LIN Y Z,ZHANG L,et al. Complex patterns in a predator-prey model with self and cross-diffusion[J]. Commu-nications in Nonlinear Science and Numerical Simulation,2011,16(4):2006-2015.
[33]王仁宏,李崇君,朱春鋼. 計(jì)算幾何教程[M]. 北京:科學(xué)出版社,2008.
[34]JIAN H Y,HUANG T Z,GU X M. Fast compact implicit integration factor method with non-uniform meshes for the two-dimen-sional nonlinear Riesz space-fractional reaction-diffusion equation[J]. Applied Numerical Mathematics,2020,156:346-363
[35]ZHU C G,WANG R H . Numerical solution of Burgers’equation by cubic B-spline quasi-Inter-polation[J]. Applied Mathemat-ics and Computation,2009,208(1):260-272.
[36]ZHU C G,KANG W S. Numerical solution of Burgers-Fisher equation by cubic B-spline quasi-interpolation[J]. Applied Mathe-matics and Computation,2010,216(9):2679-2686.
[37]SUN L Y,ZHU C G. Cubic B-spline quasi-interpolation and an application to numerical solution of generalized Burgers-Huxleyequation[J]. Advances in Mechanical Engineering,2020,12(11):1-8.
[38]齊梓萱,錢江. 對(duì)流-擴(kuò)散方程數(shù)值解的四次 B 樣條方法[J]. 圖學(xué)學(xué)報(bào),2020,41(5):716 - 724.
[39]DE BOOR C,HOLLIG K,RIEMENSCHNEIDER S. Box splines[M]. New York:Springer-Verlag,1993.
[40]SABLONNIERE P. Univariate spline quasi-interpolants and applications to numerical analysis[J]. Rendiconti del SeminarioMatematico,2005,63(3):211-222.
[41]錢江. 遞推算法與多元插值[M]. 北京:科學(xué)出版社,2020.
[42]錢江,劉雯星. 基于數(shù)值積分的最佳平方逼近樣條函數(shù)[J]. 安徽師范大學(xué)學(xué)報(bào)(自然科學(xué)版),2022,45(2):107-116.
[43]ZHANG J Y,YAN G W. Lattice Boltzmann simulation of pattern formation under cross-diffusion[J]. Computers & Mathematicswith Applications,2015,69(3):157-169
(編輯 陶志寧)
四川師范大學(xué)學(xué)報(bào)(自然科學(xué)版)2024年3期