基于切比雪夫小波基与年龄相关种群模型的数值解

基于切比雪夫小波基给出与年龄相关种群模型的数值解.利用切比雪夫小波基的性质使得所求偏微分方程转化为矩阵方程,从而简化了数值解的求解过程.最后通过数值例子验证其理论结果.     

基于组合神经网络的时间分数阶扩散方程计算方法

该文首次采用一种组合神经网络的方法，求解了一维时间分数阶扩散方程.组合神经网络是由径向基函数（RBF）神经网络与幂激励前向神经网络相结合所构造出的一种新型网络结构.首先，利用该网络结构构造出符合时间分数阶扩散方程条件的数值求解格式，同时设置误差函数，使原问题转化为求解误差函数极小值问题；然后，结合神经网络模型中的梯度下降学习算法进行循环迭代，从而获得神经网络的最优权值以及各项最优参数，最终得到问题的数值解.数值算例验证了该方法的可行性、有效性和数值精度.该文工作为时间分数阶扩散方程的求解开辟了一条新的途径.     

The Cubic B-Spline Method for a Class of Caputo-Fabrizio Fractional Differential Equations北大核心CSCD

基于分数阶微积分基本定理和三次B样条理论,构造了求解线性Caputo-Fabrizio型分数阶微分方程数值解的三次B样条方法,利用分数阶微积分基本定理将初值问题转化为关于解函数的表达式,再使用三次B样条函数逼近表达式中积分项的被积函数,进而计算了一类Caputo-Fabrizio型分数阶微分方程的数值解.给出了所构造的三次B样条方法的误差估计、收敛性和稳定性的理论证明.数值实验表明,该文数值方法在求解一类Caputo-Fabrizio型分数阶微分方程数值解时具有一定的可行性和有效性,且计算精度和计算效率优于现有的两种数值方法.     

一类分数阶Langevin方程block-by-block算法的数值分析

分数阶Langevin方程有重要的科学意义和工程应用价值,基于经典block-by-block算法,求解了一类含有Caputo导数的分数阶Langevin方程的数值解.Block-by-block算法通过引入二次Lagrange基函数插值,构造出逐块收敛的非线性方程组,通过在每一块耦合求得分数阶Langevin方程的数值解.在0<α<1条件下,应用随机Taylor展开证明block-by-block算法是3+α阶收敛的,数值试验表明在不同α和时间步长h取值下,block-by-block算法具有稳定性和收敛性,克服了现有方法求解分数阶Langevin方程速度慢精度低的缺点,表明block-by-block算法求解分数阶Langevin方程是高效的.     

一个新的分数阶比较定理

研究了Caputo和Riemann-Liouville两型分数阶微分方程的比较定理.首先,讨论了一类线性分数阶微分不等式解得非负性.其次,引入单边Lipschitz条件,将微分方程解的比较问题化为线性微分不等式非负解问题,通过线性分数阶微分方程的求解,得到分数阶比较定理.最后,为进一步说明结论,给出了两个数值仿真例子.     

基于切比雪夫多项式逼近的6级6阶隐式Runge-Kutta方法

以切比雪夫偏差点为插值点,利用切比雪夫多项式逼近理论和高斯-洛巴托-切比雪夫求积公式,构造了一个6级6阶的隐式Runge-Kutta方法.理论分析发现,该算法具有良好的稳定性——A_0稳定,较大α值的A(α)稳定,较小D值的刚性稳定和几乎L稳定.数值算例显示了该算法的有效性.     

分数阶模糊时滞神经网络模型解的存在唯一性和有限时间稳定性

本文介绍了一类分数阶模糊时滞神经网络模型.利用压缩映射原理,讨论了带时滞的分数阶神经网络模型解的存在性和唯一性,并根据Gronwall不等式结合分数阶微分方程的性质,证明了分数阶神经网络模型平衡点的有限时间稳定性,给出了有限时间稳定性的判断准则.最后,给出数值仿真说明了理论结果的正确性.     

一类分数阶非线性波方程的精确解及应用

基于分离变量的思想构造了分数阶非线性波方程含常系数的解的形式.在用待定系数法求解时,根据原方程确定假设解中的待定参数,得到具体解的表达式.利用该方法求解了3个非线性波方程,即分数阶CH(Camassa-Holm)方程、时间分数阶空间五阶Kdv-like方程、分数阶广义Ostrovsky方程.比较简便地得到了这些方程的精确解.文献中关于整数阶非线性波方程的结果成为本文结果的特例.通过数值模拟给出了部分解的图像.对能够通过待定系数法求出精确解的分数阶微分方程所应满足的条件进行了阐述.     

空间分数阶偏微分方程非标准有限差分方法的稳定性和收敛性（英文）

本文研究空间分数阶偏微分方程非标准有限差分方法数值解的相关问题.采用Grünwald-Letnikov公式和平移Grünwald-Letnikov公式分别对两个空间分数阶导数进行离散.再运用带有时间和空间步长的分母函数构造非标准有限差分方法.进而利用von Neumann分析方法对差分格式的稳定性和收敛性进行研究,获得了一些新的结果.数值例子验证了非标准有限差分方法用于求解空间分数阶偏微分方程的有效性.     

利用Block-Pulse函数求解分数阶泊松方程数值解

借助于二维Block-Pulse函数求解分数阶泊松方程的数值解,并讨论了Dirichlet边界条件,方法是基于Block-Pulse函数的定义及性质,并结合相应的分数阶微分算子矩阵将原问题转化为含有未知变量的代数方程组,进而离散未知变量,求得原问题的数值解.而且还对所提方法进行了误差分析,最后给出的数值算例也验证了所提算法的有效性及可行性.     

Numerical treatment for solving fractional Riccati differential equation

This paper presents an accurate numerical method for solving fractional Riccati differential equation (FRDE). The proposed method so called fractional Chebyshev finite difference method (FCheb-FDM). In this technique, we approximate FRDE with a finite dimensional problem. The method is based on the combination of the useful properties of Chebyshev polynomials approximation and finite difference method. The Caputo fractional derivative is replaced by a difference quotient and the integral by a finite sum. By this method the given problem is reduced to a problem for solving a system of algebraic equations, and by solving this system, we obtain the solution of FRDE. Special attention is given to study the convergence analysis and estimate an error upper bound of the obtained approximate formula. Illustrative examples are included to demonstrate the validity and applicability of the proposed technique.     

Numerical solution of multi-order fractional differential equations with multiple delays via spectral collocation methods

This paper discusses a general framework for the numerical solution of multi-order fractional delay differential equations (FDDEs) in noncanonical forms with irrational/rational multiple delays by the use of a spectral collocation method. In contrast to the current numerical methods for solving fractional differential equations, the proposed framework can solve multi-order FDDEs in a noncanonical form with incommensurate orders. The framework can also solve multi-order FDDEs with irrational multiple delays. Next, the framework is enhanced by the fractional Chebyshev collocation method in which a Chebyshev operation matrix is constructed for the fractional differentiation. Spectral convergence and small computational time are two other advantages of the proposed framework enhanced by the fractional Chebyshev collocation method. In addition, the convergence, error estimates, and numerical stability of the proposed framework for solving FDDEs are studied. The advantages and computational implications of the proposed framework are discussed and verified in several numerical examples.     

A numerical technique for solving fractional variational problems

This paper presents an accurate numerical method for solving a class of fractional variational problems (FVPs). The fractional derivative in these problems is in the Caputo sense. The proposed method is called fractional Chebyshev finite difference method. In this technique, we approximate FVPs and end up with a finite‐dimensional problem. The method is based on the combination of the useful properties of Chebyshev polynomials approximation and finite difference method. The Caputo fractional derivative is replaced by a difference quotient and the integral by a finite sum. The fractional derivative approximation using Clenshaw and Curtis formula introduced here, along with Clenshaw and Curtis procedure for the numerical integration of a non‐singular functions and the Rayleigh–Ritz method for the constrained extremum, is considered. By this method, the given problem is reduced to the problem for solving a system of algebraic equations, and by solving this system, we obtain the solution of FVPs. Special attention is given to study the convergence analysis and evaluate an error upper bound of the obtained approximate formula. Illustrative examples are included to demonstrate the validity and applicability of the proposed technique. A comparison with another method is given. Copyright © 2012 John Wiley & Sons, Ltd.     

Numerical Solution of the Two-Sided Space–Time Fractional Telegraph Equation Via Chebyshev Tau Approximation

The operational matrices of left Caputo fractional derivative, right Caputo fractional derivative, and Riemann–Liouville fractional integral, for shifted Chebyshev polynomials, are presented and derived. We propose an accurate and efficient spectral algorithm for the numerical solution of the two-sided space–time Caputo fractional-order telegraph equation with three types of non-homogeneous boundary conditions, namely, Dirichlet, Robin, and non-local conditions. The proposed algorithm is based on shifted Chebyshev tau technique combined with the derived shifted Chebyshev operational matrices. We focus primarily on implementing the novel algorithm both in temporal and spatial discretizations. This algorithm reduces the problem to a system of algebraic equations greatly simplifying the problem. This system can be solved by any standard iteration method. For confirming the efficiency and accuracy of the proposed scheme, we introduce some numerical examples with their approximate solutions and compare our results with those achieved using other methods.     

一类变系数分数阶微分方程的数值解法

本文研究了一类变系数分数阶微分方程的数值解法问题. 利用Cheyshev小波推导出的分数阶微分方程的算子矩阵把分数阶微分方程转换为代数方程组. 同时给出了Cheyshev小波基的收敛性和误差估计表达式, 并给出数值算例说明所提方法的精确性和有效性     

Simulation of nonlinear fractional dynamics arising in the modeling of cognitive decision making using a new fractional neural network

By the rapid growth of available data, providing data-driven solutions for nonlinear (fractional) dynamical systems becomes more important than before. In this paper, a new fractional neural network model that uses fractional order of Jacobi functions as its activation functions for one of the hidden layers is proposed to approximate the solution of fractional differential equations and fractional partial differential equations arising from mathematical modeling of cognitive-decision-making processes and several other scientific subjects. This neural network uses roots of Jacobi polynomials as the training dataset, and the Levenberg-Marquardt algorithm is chosen as the optimizer. The linear and nonlinear fractional dynamics are considered as test examples showing the effectiveness and applicability of the proposed neural network. The numerical results are compared with the obtained results of some other networks and numerical approaches such as meshless methods. Numerical experiments are presented confirming that the proposed model is accurate, fast, and feasible.     

Numerical technique based on generalized Laguerre and shifted Chebyshev polynomials

In this study, we present a numerical scheme for solving a class of fractional partial differential equations. First, we introduce psi -Laguerre polynomials like psi-shifted Chebyshev polynomials and employ these newly introduced polynomials for the solution of space-time fractional differential equations. In our approach, we project these polynomials to develop operational matrices of fractional integration. The use of these orthogonal polynomials converts the problem under consideration into a system of algebraic equations. The solution of this system provide us the desired results. The convergence of the proposed method is analyzed. Finally, some illustrative examples are included to observe the validity and applicability of the proposed method.     

Numerical treatment of a well-posed Chebyshev Tau method for Bagley-Torvik equation with high-order of accuracy

The main purpose of this study is to develop and analyze a new high-order operational Tau method based on the Chebyshev polynomials as basis functions for obtaining the numerical solution of Bagley-Torvik equation which has a important role in the fractional calculus. It is shown that some derivatives of the solutions of these equations have a singularity at origin. To overcome this drawback we first change the original equation into a new equation with a better regularity properties by applying a regularization process and thereby the operational Chebyshev Tau method can be applied conveniently. Our proposed method has two main advantages. First, the algebraic form of the Tau discretization of the problem has an upper triangular structure which can be solved by forward substitution method. Second, Tau approximation of the problem converges to the exact ones with a highly rate of convergence under a more general regularity assumptions on the input data in spite of the singularity behavior of the exact solution. Numerical results are presented which confirm the theoretical results obtained and efficiency of the proposed method.     

Wavelets method for solving systems of nonlinear singular fractional Volterra integro-differential equations

This paper presents a computational method for solving a class of system of nonlinear singular fractional Volterra integro-differential equations. First, existences of a unique solution for under studying problem is proved. Then, shifted Chebyshev polynomials and their properties are employed to derive a general procedure for forming the operational matrix of fractional derivative for Chebyshev wavelets. The application of this operational matrix for solving mentioned problem is explained. In the next step, the error analysis of the proposed method is investigated. Finally, some examples are included for demonstrating the efficiency of the proposed method.