具有2个卷积核和Hilbert核的奇异积分方程的可解性

王洪泉,  张雯雯,  类延鑫,  杨经芃,  李平润

王洪泉, 张雯雯, 类延鑫, 等. 具有2个卷积核和Hilbert核的奇异积分方程的可解性 [J]. 首都师范大学学报(自然科学版), 2026, 47(2): 32-37. DOI: 10.19789/j.1004-9398.2026.02.005
引用本文: 王洪泉, 张雯雯, 类延鑫, 等. 具有2个卷积核和Hilbert核的奇异积分方程的可解性 [J]. 首都师范大学学报(自然科学版), 2026, 47(2): 32-37. DOI: 10.19789/j.1004-9398.2026.02.005
WANG Hongquan, ZHANG Wenwen, LEI Yanxin, et al. Solvability of singular integral equations with two convolution kernels and Hilbert kernel [J]. Journal of Capital Normal University (Natural Science Edition), 2026, 47(2): 32-37. DOI: 10.19789/j.1004-9398.2026.02.005
Citation: WANG Hongquan, ZHANG Wenwen, LEI Yanxin, et al. Solvability of singular integral equations with two convolution kernels and Hilbert kernel [J]. Journal of Capital Normal University (Natural Science Edition), 2026, 47(2): 32-37. DOI: 10.19789/j.1004-9398.2026.02.005

具有2个卷积核和Hilbert核的奇异积分方程的可解性

DOI: 10.19789/j.1004-9398.2026.02.005
基金项目: 

国家自然科学基金项目 11971015;

山东省大学生创新创业训练计划项目 S202310446053.

详细信息
    通讯作者:

    王洪泉, whq211918@163.com.

  • 中图分类号: O175.5; O175.8

Solvability of singular integral equations with two convolution kernels and Hilbert kernel

  • 摘要:

    研究了函数类$ {L}^{2}[-\mathtt{π} , \mathtt{π} ] $中带有2个卷积核和Hilbert核的奇异积分方程。利用离散的Fourier变换,推广的留数定理及线性代数相关理论,将含有Hilbert核和卷积核相混合的奇异积分方程转化为线性方程组,在$ {L}^{2}[-\mathtt{π} , \mathtt{π} ] $上得到了该方程的一般解和可解条件。

     

    Abstract:

    This paper studies a class of singular integral equations with two convolution kernels and Hilbert kernels in the function class $ {L}^{2}[-\mathtt{π} , \mathtt{π} ] $. This study use discrete Fourier transform, generalized residue theorem and related theory of linear algebra to transform the singular integral equation containing the mixture of Hilbert kernel and convolution kernel into a system of linear equations. Finally, this paper obtains the general solution and solvable conditions of the equation on $ {L}^{2}[-\mathtt{π} , \mathtt{π} ] $.

     

  • Muskhelishvili[1]和Vekua[2]等数学家系统地研究和发展了奇异积分方程,使其研究转化成解析函数边值问题的相应理论。使得对卷积型和奇异型2类重要积分方程的研究形成了较完整的理论体系。目前,如何处理复杂的核函数以及在不同空间中实现高效求解是近年来国内外的热门研究。李凯雅和魏鑫[3]在不限于正则型,且积分曲线上出现非正则型单零点的情况下,讨论了Hilbert核奇异积分方程的直接解法,不同于对积分方程的数值分解方法,作者直接对方程非正则型的特点进行求解,这种直接解法在理论上更加简单明了;Abusalim等[4]借助求积方程,通过转化为非线性积分方程组研究了带有奇异核的非线性Volterra-Fredholm积分方程,并且运用Chebyshev多项式求解了该方程组,有效应对方程的非线性和奇异核这2个难点;Boikov和Pivkina[5]利用迭代投影法,提出了一种在退化情形下求解奇异积分方程的样条配置法,使得方程解的存在性、唯一性、稳定性、收敛性等方面有了更全面和精确的理论分析。另外De-Bonis和Laurita[6]、Du和Shen[7]、Giang等[8]、Li和Ren[9]等数学工作者利用再生核方法、Fourier变换卷积及配置法等,系统地研究并证明了方程解的存在性与可解性条件,提出了高效的数值算法和解析框架,为理论和应用数学提供了重要工具[10-11]。目前将经典复分析到高维的Clifford分析的推广已经成为国内外数学家研究热点[12-20]。

    含有2个卷积核和Hilbert核的奇异积分方程在多物理场耦合系统中具有重要应用,例如复合材料断裂力学和5G通信滤波器设计。此类方程通过分离不同机制的非局部相互作用与奇异性,显著提升了复杂工程问题的建模精度,解决了传统单核模型因机制耦合导致的预测误差与计算效率瓶颈。且与含有1个卷积核和Hilbert核的方程相比:一方面,前者运用离散的Fourier变换,得到的是更复杂的离散方程组,综合运用了数学多方面的知识;另一方面,前者增加的卷积核使得方程在描述物理现象和数学问题时具有更高的灵活性和准确性,为解决以前难以精确描述和分析的问题提供了可能。

    关于Hilbert核的奇异积分方程的研究为解析函数边值问题、潮汐理论、弹性理论、流体力学等领域的科学研究、工程技术以及许多数学物理问题等提供了强有力的理论基础。随着科学计算方法的发展,其重要性也越来越明显。本文将具有Cauchy核的奇异积分方程理论推广到$ {L}^{2}[-\mathtt{π} , \mathtt{π} ] $类中具有2个卷积核和Hilbert核的奇异积分方程,利用离散Fourier变换转化为方程组,利用线性代数的可解性及矩阵理论得到了方程的可解条件和一般解。此研究在数学的一些分支中具有重要的理论意义,在实际问题中也发挥着较大的应用。

    定义1  设$ F\left(\theta \right), G\left(\theta \right)\in {L}^{2}[-\mathtt{π} , \mathtt{π} ] $,其中$ {L}^{2}[-\mathtt{π} , \mathtt{π} ] $为平方可积空间。定义

    $$ F * G(\theta)=\frac{1}{2 \mathtt{π}} \displaystyle\int_{-\mathtt{π}}^\mathtt{π} F(\theta-\tau) G(\tau) \mathrm{d} \tau , $$ (1)

    称为$ F\left(\theta \right) $与$ G\left(\theta \right) $在$ [-\mathtt{π} , \mathtt{π} ] $上的卷积。

    定义2[6-11]  设$ {L}^{2}\left(\mathbb{R}\right) $为平方可积空间,$ f\left(t\right)\in {L}^{2}\left(\mathbb{R}\right) $,定义Fourier变换$ \mathcal{F} $和Fourier逆变换$ \mathcal{F}^{-1} $为

    $$ \begin{aligned} \mathcal{F}[f(s)] & =\frac{1}{2 \mathtt{π}} \displaystyle\int_{-\infty}^{+\infty} f(t) \mathrm{e}^{\mathrm{i} s t} \mathrm{~d} t, \\ \mathcal{F}^{-1}[F(t)] & =\frac{1}{2 \mathtt{π}} \displaystyle\int_{-\infty}^{+\infty} F(s) \mathrm{e}^{-\mathrm{i} s t} \mathrm{~d} s 。 \end{aligned} $$

    定义3  设$ {l}^{2}\left(\mathbb{Z}\right) $为平方可和空间,$ f={\left\{{f}_{k}\right\}}_{-\infty }^{+\infty }\in {l}^{2}\left(\mathbb{Z}\right) $为一双向无穷数列,定义算子

    $$ w f=\sum\limits_{k=-\infty}^{+\infty} f_k \mathrm{e}^{\mathrm{i} k \theta}=F(\theta)\text{,} $$ (2)
    $$ w^{-1}[F(\theta)]=\frac{1}{2 \mathtt{π}} \displaystyle\int_{-\mathtt{π}}^\mathtt{π} F(\theta) \mathrm{e}^{-\mathrm{i} k \theta} \mathrm{~d} \theta=f_k, -\mathtt{π} \leqslant \theta \leqslant \mathtt{π} , $$ (3)

    分别为离散的Fourier(逆)变换。

    对任意$ F\left(\theta \right), G\left(\theta \right)\in {L}^{2}[-\mathtt{π} , \mathtt{π} ] $,可得

    $$ \begin{gathered} w^{-1}(F * G)= \\ \frac{1}{2 \mathtt{π}} \int_{-\mathtt{π}}^\mathtt{π}\left[\frac{1}{2 \mathtt{π}} \int_{-\mathtt{π}}^\mathtt{π} F(\theta-\tau) G(\tau) \mathrm{d} \tau\right] \mathrm{e}^{-\mathrm{i} k \theta} \mathrm{~d} \theta= \\ \frac{1}{2 \mathtt{π}} \int_{-\mathtt{π}}^\mathtt{π} G(\tau)\left[\frac{1}{2 \mathtt{π}} \int_{-\mathtt{π}}^\mathtt{π} F(\theta-\tau) \mathrm{e}^{-\mathrm{i} k(\theta-\tau)} \mathrm{d}(\theta-\tau)\right] \mathrm{e}^{-\mathrm{i} k \tau} \mathrm{~d} \tau= \\ f_k g_k, \end{gathered}$$

    其中$ {w}^{-1}\left[G\right(\theta \left)\right]={g}_{k} $,因此有

    $$ {w}^{-1}\left(F*G\right)={f}_{k}{g}_{k}, k=0, \pm 1, \pm 2, \cdots 。 $$ (4)

    定义4  设$ L $为一光滑曲线,$ f\left(t\right)\in L $,若对任意的$ {t}_{1}, {t}_{2}\in L $,存在正实数$ A, \alpha (0 < \alpha \le 1) $恒满足不等式

    $$ \left|f\right({t}_{1})-f({t}_{2}\left)\right|\le A|{t}_{1}-{t}_{2}{|}^{\alpha }, $$ (5)

    则称$ f\left(t\right) $属于$ \alpha $阶Hölder函数类,记为$ f\left(t\right)\in {H}^{\alpha } $。

    定义5  若无穷数列$ f={\left\{{f}_{k}\right\}}_{-\infty }^{+\infty } $满足

    $$ \left|{f}_{k}\right|\le \frac{A}{|k{|}^{1+\mu }}, \hspace{1em}k=\pm 1, \pm 2, \cdots , $$ (6)

    其中$ 0 < \mu \le 1 $,且$ A $为正常数,那么称$ f={\left\{{f}_{k}\right\}}_{-\infty }^{+\infty }\in \left\{1\right\} $。

    若$ f\in \left\{1\right\} $,则称定义3中的$ wf=F\left(\theta \right)\in \left\{\tilde{0}\right\} $。

    引理1  若$ f\in \left\{1\right\} $,那么定义3中的$ wf=F\left(\theta \right)\in H $。

    证明  对任意的$ {\theta }_{1}, {\theta }_{2}\in [-\mathtt{π} , \mathtt{π} ] $,由Euler公式[21]

    $$ \mathrm{e}^{\mathrm{i} k \theta}=\cos \theta+\mathrm{i} \sin \theta $$

    和式(2),有

    $$ \begin{gathered} \left|F\left(\theta_1\right)-F\left(\theta_2\right)\right|=\left|\sum\limits_{k=-\infty}^{+\infty} f_k\left(\mathrm{e}^{\mathrm{i} k \theta_1}-\mathrm{e}^{\mathrm{i} k \theta_2}\right)\right|= \\ \left|\sum\limits_{k=-\infty}^{+\infty} f_k\left[\left(\cos \theta_1-\cos \theta_2\right)+\mathrm{i}\left(\sin \theta_1-\sin \theta_2\right)\right]\right| \leqslant \\ \sum\limits_{k=-\infty}^{+\infty} 2\left|f_k\right|\left|\sin \frac{\theta_1+\theta_2}{2} \sin \frac{\theta_1-\theta_2}{2}-\mathrm{i}\left(\cos \frac{\theta_1+\theta_2}{2} \sin \frac{\theta_1-\theta_2}{2}\right)\right| \leqslant \\ \sum\limits_{k=-\infty}^{+\infty} 2\left|f_k\right|\left|\sin \frac{\theta_1-\theta_2}{2}\right| \leqslant A\left|\theta_1-\theta_2\right|, \end{gathered} $$

    其中$ A=\sum \limits_{k=-\infty }^{+\infty }|{f}_{k}| < +\infty , $由Hölder连续函数类的定义可知$ wf=F\left(\theta \right)\in H $。

    引理2  定义算子$ S $,$ \tilde{S} $如下:

    $$ \begin{aligned} & S F(\theta)=\frac{1}{\mathtt{π}} \int_{-\mathtt{π}}^\mathtt{π} F(\tau) \cot \frac{\tau-\theta}{2} \mathrm{~d} \tau, \\ & \tilde{S} F(\theta)=\frac{1}{\mathtt{π}} \int_{-\mathtt{π}}^\mathtt{π} F(\tau) \cot \frac{\tau+\theta}{2} \mathrm{~d} \tau, \end{aligned} $$

    其中$ F\left(\theta \right)\in {L}^{2}[-\mathtt{π} , \mathtt{π} ] $。那么有

    (1)$ w^{-1}[S F(\theta)]=2 \mathrm{i} f_k \operatorname{sgn} k, \quad w^{-1}[\tilde{S} F(\theta)]= -2 \mathrm{i} f_{-k} \operatorname{sgn} k $;

    (2)$ \tilde{S} F(\theta) \in L^2[-\mathtt{π}, \mathtt{π}], w^{-1}[\tilde{S} F(\theta)] \in l^2(\mathbb{Z}) $。

    证明  (1)由Euler公式,有

    $$ \begin{gathered} \cot \frac{\tau+\theta}{2}=\frac{\cos \frac{\tau+\theta}{2}}{\sin \frac{\tau+\theta}{2}}=\frac{\mathrm{i}\left(\mathrm{e}^{\mathrm{i} \frac{\tau+\theta}{2}}+\mathrm{e}^{-\mathrm{i} \frac{\tau+\theta}{2}}\right)}{\mathrm{e}^{\mathrm{i} \frac{\tau+\theta}{2}}-\mathrm{e}^{-\mathrm{i} \frac{\tau+\theta}{2}}}= \\ \mathrm{i}\left(\frac{\mathrm{e}^{\mathrm{i} \tau}}{\mathrm{e}^{\mathrm{i} \tau}-\mathrm{e}^{-\mathrm{i} \theta}}-\frac{\mathrm{e}^{-\mathrm{i} \tau}}{\mathrm{e}^{-\mathrm{i} \tau}-\mathrm{e}^{\mathrm{i} \theta}}\right), \\ \tilde{S} F(\theta)=\frac{\mathrm{i}}{\mathtt{π}} \int_{-\mathtt{π}}^\mathtt{π}\left(F(\tau) \frac{\mathrm{e}^{\mathrm{i} \tau}}{\mathrm{e}^{\mathrm{i} \tau}-\mathrm{e}^{-\mathrm{i} \theta}}-F(\tau) \frac{\mathrm{e}^{-\mathrm{i} \tau}}{\mathrm{e}^{-\mathrm{i} \tau}-\mathrm{e}^{\mathrm{i} \theta}}\right) \mathrm{d} \tau, \\ w^{-1}[\tilde{S} F(\theta)]= \\ \frac{1}{2 \mathtt{π}} \int_{-\mathtt{π}}^\mathtt{π}\left[\frac{\mathrm{i}}{\mathtt{π}} \int_{-\mathtt{π}}^\mathtt{π}\left(F(\tau) \frac{\mathrm{e}^{\mathrm{i} \tau}}{\mathrm{e}^{\mathrm{i} \tau}-\mathrm{e}^{-\mathrm{i} \theta}}-F(\tau) \frac{\mathrm{e}^{-\mathrm{i} \tau}}{\mathrm{e}^{-\mathrm{i} \tau}-\mathrm{e}^{\mathrm{i} \theta}}\right) \mathrm{d} \tau\right] \mathrm{e}^{-\mathrm{i} k \theta} \mathrm{~d} \theta_0。 \end{gathered} $$

    记

    $$ \begin{aligned} & I_1=\frac{\mathrm{i}}{2 \mathtt{π}^2} \int_{-\mathtt{π}}^\mathtt{π}\left[\int_{-\mathtt{π}}^\mathtt{π} \frac{\mathrm{e}^{-\mathrm{i} k \theta}}{\mathrm{e}^{\mathrm{i} \tau}-\mathrm{e}^{-\mathrm{i} \theta}} \mathrm{~d} \theta\right] F(\tau) \mathrm{e}^{\mathrm{i} \tau} \mathrm{~d} \tau, \\ & I_2=\frac{\mathrm{i}}{2 \mathtt{π}^2} \int_{-\mathtt{π}}^\mathtt{π}\left[\int_{-\mathtt{π}}^\mathtt{π} \frac{\mathrm{e}^{-\mathrm{i} k \theta}}{\mathrm{e}^{-\mathrm{i} \tau}-\mathrm{e}^{\mathrm{i} \theta}} \mathrm{~d} \theta\right] F(\tau) \mathrm{e}^{-\mathrm{i} \tau} \mathrm{~d} \tau, \end{aligned} $$

    则

    $$ {w}^{-1}\left[\tilde{S}F\left(\theta \right)\right]={I}_{1}-{I}_{2} 。 $$ (7)

    利用指数变换$ z=\mathrm{e}^{-\mathrm{i}\theta } (0\le \theta \le 2\mathtt{π} ) $,那么

    $$ \displaystyle \int_{-\mathtt{π}}^\mathtt{π} \frac{\mathrm{e}^{-\mathrm{i} k \theta}}{\mathrm{e}^{\mathrm{i} \tau}-\mathrm{e}^{-\mathrm{i} \theta}} \mathrm{~d} \theta= \displaystyle \int_{|z|=1} \frac{z^k}{\left(\mathrm{e}^{\mathrm{i} \tau}-z\right) \mathrm{i} z} \mathrm{~d} z。 $$

    (ⅰ)当$ k\le 0 $时,由推广的留数定理[22],有

    $$ \begin{gathered} \int_{|z|=1} \frac{z^k}{\left(\mathrm{e}^{\mathrm{i} \tau}-z\right) \mathrm{i} z} d z=-2 \mathtt{π}\left\{\underset{z=0}{\operatorname{Res}} f(z)+\beta_1 \underset{z=\mathrm{e}^{\mathrm{i} \tau}}{\operatorname{Res}} f(z)\right\}= \\ -2 \mathtt{π}\left[\lim _{z \rightarrow 0} \frac{1}{(-k)!}\left(\frac{1}{z-\mathrm{e}^{\mathrm{i} \tau}}\right)^{(-k)}+\frac{1}{2} \lim _{z \rightarrow \mathrm{e}^{\mathrm{i} \tau}} \frac{1}{z^{1-k}}\right]= \\ -2 \mathtt{π}\left[-\frac{1}{\mathrm{e}^{\mathrm{i}(1-k) \tau}}+\frac{1}{2} \frac{1}{\mathrm{e}^{\mathrm{i}(1-k) \tau}}\right]=\frac{\mathtt{π}}{\mathrm{e}^{\mathrm{i}(1-k) \tau}}, \end{gathered} $$

    其中$ f(z)=\frac{1}{\left(z-\mathrm{e}^{\mathrm{i} \tau}\right) z^{1-k}} $,$ {\beta }_{1} $为$ \left|z\right|=1 $所围闭域在$z=\mathrm{e}^{\mathrm{i} \tau} $处的张度[1]。因此

    $$\begin{gathered} I_1=\frac{\mathrm{i}}{2 \mathtt{π}^2} \int_{-\mathtt{π}}^\mathtt{π}\left[\int_{-\mathtt{π}}^\mathtt{π} \frac{\mathrm{e}^{-\mathrm{i} k \theta}}{\mathrm{e}^{\mathrm{i} \tau}-\mathrm{e}^{-\mathrm{i} \theta}}\right] F(\tau) \mathrm{e}^{\mathrm{i} \tau} \mathrm{~d} \tau= \\ \frac{\mathrm{i}}{2 \mathtt{π}} \int_{-\mathtt{π}}^\mathtt{π} F(\tau) \mathrm{e}^{\mathrm{i} k \tau} \mathrm{~d} \tau=\mathrm{i} f_{-k}。 \end{gathered} $$

    (ⅱ)当$ k > 0 $时,由推广的留数定理,可以得到

    $$ \begin{gathered} \int_{|z|=1} \frac{z^k}{\left(\mathrm{e}^{\mathrm{i} \tau}-z\right) \mathrm{i} z} \mathrm{~d} z=-\frac{1}{\mathrm{i}} \int_{|z|=1} \frac{z^{k-1}}{z-\mathrm{e}^{\mathrm{i} \tau}} \mathrm{~d} z= \\ -2 \mathtt{π} \beta_2 \underset{z=\mathrm{e}^{\mathrm{i} \tau}}{\operatorname{Re}} f(z)=-\frac{\mathtt{π}}{\mathrm{e}^{\mathrm{i}(1-k) \tau}}, \end{gathered} $$

    其中$ f\left(z\right) $同上,$ {\beta }_{2} $为$ \left|z\right|=1 $所围闭域在$z=\mathrm{e}^{\mathrm{i} \tau}$处的张度。此时有

    $$ \begin{aligned} I_1= & \frac{\mathrm{i}}{2 \mathtt{π}^2} \int_{-\mathtt{π}}^\mathtt{π}\left[\int_{-\mathtt{π}}^\mathtt{π} \frac{\mathrm{e}^{-\mathrm{i} k \theta}}{\mathrm{e}^{\mathrm{i} \tau}-\mathrm{e}^{-\mathrm{i} \theta}}\right] F(\tau) \mathrm{e}^{\mathrm{i} \tau} \mathrm{~d} \tau= \\ & -\frac{\mathrm{i}}{2 \mathtt{π}} \int_{-\mathtt{π}}^\mathtt{π} F(\tau) \mathrm{e}^{\mathrm{i} k \tau} \mathrm{~d} \tau=-\mathrm{i} f_{-k }。 \end{aligned} $$

    综上所述,可以得到

    $$ I_1=-\mathrm{i} f_{-k} \operatorname{sgn} k。 $$ (8)

    同理,对于$ {I}_{2} $,此时取指数变换$ z=\mathrm{e}^{\mathrm{i}\theta } $,由推广的留数定理,有

    $$I_2=\mathrm{i} f_{-k} \operatorname{sgn} k。 $$

    因此有

    $$ I_1-I_2=-2 \mathrm{i} f_{-k} \operatorname{sgn} k, $$

    即

    $$ w^{-1}[\tilde{S} F(\theta)]=-2 \mathrm{i} f_{-k} \operatorname{sgn} k 。 $$ (9)

    类似地也有

    $$ w^{-1}[S F(\theta)]=2 \mathrm{i} f_k \operatorname{sgn} k, $$

    只需

    $$ \cot \frac{\tau-\theta}{2}=\mathrm{i}\left(\frac{\mathrm{e}^{\mathrm{i} \tau}}{\mathrm{e}^{\mathrm{i} \tau}-\mathrm{e}^{\mathrm{i} \theta}}-\frac{\mathrm{e}^{-\mathrm{i} \tau}}{\mathrm{e}^{-\mathrm{i} \tau}-\mathrm{e}^{-\mathrm{i} \theta}}\right)$$

    即可,同上面证明可得。

    (2)由Riesz定理[23]有

    $$ \left|\right|\tilde{S}F|{|}_{2}\le C|\left|F\right|{|}_{2}, C 为常数。 $$

    因为$ F\left(\theta \right)\in {L}^{2}[-\mathtt{π} , \mathtt{π} ] $,所以有$ \tilde{S}F\left(\theta \right)\in {L}^{2}[-\mathtt{π} , \mathtt{π} ] $,进而$ {w}^{-1}\left[\tilde{S}F\left(\theta \right)\right]\in {l}^{2}\left(\mathbb{Z}\right) $。

    称方程

    $$ \begin{gathered} a F(\theta)+b_1 \tilde{S} F^{+}(\theta)+b_2 \tilde{S} F^{-}(\theta)+c_1 H_1 * F^{+}(\theta)+ \\ c_2 H_2 * F^{-}(\theta)+H_3 * \tilde{S} F^{+}(\theta)+ \\ H_4 * \tilde{S} F^{-}(\theta)=G(\theta) \end{gathered} $$ (10)

    为带有2个卷积核和Hilbert核的奇异积分方程,式中:

    $$ {F}^{+}\left(\theta \right)=\left\{\begin{array}{ll}F\left(\theta \right), & \theta > 0, \\ 0, & \theta \le 0;\end{array}\right.{F}^{-}\left(\theta \right)=\left\{\begin{array}{ll}0, & \theta \ge 0, \\ -F\left(\theta \right), & \theta < 0;\end{array}\right. $$ (11)

    $ -\mathtt{π} \le \theta \le \mathtt{π} , a、{b}_{j}、{c}_{j}(j=\mathrm{1, 2}) $为常数;$ {b}_{1} $与$ {b}_{2} $,$ {c}_{1} $与$ {c}_{2} $不同时为零。已知函数$ {H}_{j}\left(\theta \right), G\left(\theta \right)\in {L}^{2}[-\mathtt{π} , \mathtt{π} ] $,不失一般性,要求未知函数$ F\left(\theta \right) $为$ {L}^{2}[-\mathtt{π} , \mathtt{π} ] $上的偶函数。对于$ F\left(\theta \right) $不是$ {L}^{2}[-\mathtt{π} , \mathtt{π} ] $上的偶函数,同理可得。

    用$ {w}^{-1} $作用于方程(10)两端,由引理2可得

    $$ \begin{gathered} \left(a+c_1 h_k^{(1)}\right) f_k^{+}-\left(2 \mathrm{i} b_1 \operatorname{sgn} k+2 \mathrm{i} h_k^{(3)} \operatorname{sgn} k\right) f_{-k}^{+}+(a+ \\ \left.c_2 h_k^{(2)}\right) f_k^{-}-\left(2 \mathrm{i} b_2 \operatorname{sgn} k+2 \mathrm{i} h_k^{(4)} \operatorname{sgn} k\right) f_{-k}^{-}=g_k, \end{gathered}$$ (12)

    式中

    $$ \begin{gathered} f_k^{ \pm}=w^{-1} F^{ \pm} ; h_k^{(j)}=w^{-1} H_j ; g_k=w^{-1} G ; \\ k=0, \pm 1, \pm 2, \cdots 。 \end{gathered}$$

    由式(11)可知

    $$ F\left(\theta \right)={F}^{+}\left(\theta \right)-{F}^{-}\left(\theta \right), $$

    又因为$ F\left(\theta \right) $为偶函数,因此

    $$ \begin{gathered} f_{-k}^{-}=\frac{1}{2 \mathtt{π}} \int_{-\mathtt{π}}^\mathtt{π} F^{-}(\theta) \mathrm{e}^{\mathrm{i} k \theta} \mathrm{~d} \theta=-\frac{1}{2 \mathtt{π}} \int_{-\mathtt{π}}^0 F(\theta) \mathrm{e}^{\mathrm{i} k \theta} \mathrm{~d} \theta= \\ -\frac{1}{2 \mathtt{π}} \int_\mathtt{π}^0 F(-\theta) \mathrm{e}^{-\mathrm{i} k \theta} \mathrm{~d}(-\theta)= \\ -\frac{1}{2 \mathtt{π}} \int_0^\mathtt{π} F(\theta) \mathrm{e}^{-\mathrm{i} k \theta} \mathrm{~d} \theta=-\frac{1}{2 \mathtt{π}} \int_{-\mathtt{π}}^\mathtt{π} F^{+}(\theta) \mathrm{e}^{-\mathrm{i} k \theta} \mathrm{~d} \theta=-f_k^{+}, \end{gathered}$$

    即

    $$ {f}_{-k}^{-}=-{f}_{k}^{+}, \hspace{1em}k=0, \pm 1, \pm 2, \cdots 。 $$ (13)

    同理也可得

    $$ {f}_{-k}^{+}=-{f}_{k}^{-}, \hspace{1em}k=0, \pm 1, \pm 2, \cdots 。 $$ (14)

    由式(13和14),方程(12)变为

    $$ \begin{aligned} & \left(a+c_2 h_k^{(2)}+2 \mathrm{i} b_1 \operatorname{sgn} k+2 \mathrm{i} h_k^{(3)} \operatorname{sgn} k\right) f_k^{-}+(-a- \\ & \left.c_1 h_k^{(1)}-2 \mathrm{i} b_2 \operatorname{sgn} k-2 \mathrm{i} h_k^{(4)} \operatorname{sgn} k\right) f_{-k}^{-}=g_{k }。 \end{aligned}$$ (15)

    在式(15)中,用$ -k $代替$ k $,则有

    $$ \begin{gathered} \left(a+c_2 h_{-k}^{(2)}-2 \mathrm{i} b_1 \operatorname{sgn} k-2 \mathrm{i} h_{-k}^{(3)} \operatorname{sgn} k\right) f_{-k}^{-}+(-a- \\ \left.c_1 h_{-k}^{(1)}+2 \mathrm{i} b_2 \operatorname{sgn} k+2 \mathrm{i} h_{-k}^{(4)} \operatorname{sgn} k\right) f_k^{-}=g_{-k}, \end{gathered}$$ (16)

    由式(15和16),有

    $$ \left\{\begin{array}{l}{\delta }_{k}^{\left(1\right)}{f}_{k}^{-}+{\delta }_{k}^{\left(2\right)}{f}_{-k}^{-}={g}_{k}, \\ {\delta }_{-k}^{\left(2\right)}{f}_{k}^{-}+{\delta }_{-k}^{\left(1\right)}{f}_{-k}^{-}={g}_{-k}, \end{array}\right. $$ (17)

    式中

    $$ \left\{\begin{array}{l} \delta_{ \pm k}^{(1)}=a+c_2 h_{ \pm k}^{(2)} \pm 2 \mathrm{i} b_1 \operatorname{sgn} k \pm 2 \mathrm{i} h_{ \pm k}^{(3)} \operatorname{sgn} k, \\ \delta_{ \pm k}^{(2)}=-a-c_1 h_{ \pm k}^{(1)} \mp 2 \mathrm{i} b_2 \operatorname{sgn} k \mp 2 \mathrm{i} h_{ \pm k}^{(4)} \operatorname{sgn} k。 \end{array}\right. $$ (18)

    因此求解方程(10)等价于求解方程组(17)。

    下面研究方程组(17)的解及可解性。记

    $$ \boldsymbol{D}_k=\left[\begin{array}{ll} \delta_k^{(1)} & \delta_k^{(2)} \\ \delta_{-k}^{(2)} & \delta_{-k}^{(1)} \end{array}\right]_{2 \times 2}, \quad \boldsymbol{B}_k=\left[\begin{array}{ccc} \delta_k^{(1)} & \delta_k^{(2)} & g_k \\ \delta_{-k}^{(2)} & \delta_{-k}^{(1)} & g_{-k} \end{array}\right]_{2 \times 3} $$ (19)

    分别为式(17)的系数矩阵与增广矩阵,其系数行列式为

    $$\begin{gathered} \left|\boldsymbol{D}_k\right|=\left(a+c_2 h_k^{(2)}+2 \mathrm{i} b_1 \operatorname{sgn} k+2 \mathrm{i} h_k^{(3)} \operatorname{sgn} k\right) \cdot \\ \left(a+c_2 h_{-k}^{(2)}-2 \mathrm{i} b_1 \operatorname{sgn} k-2 \mathrm{i} h_{-k}^{(3)} \operatorname{sgn} k\right)- \\ \left(a+c_1 h_k^{(1)}+2 \mathrm{i} b_2 \operatorname{sgn} k+2 \mathrm{i} h_k^{(4)} \operatorname{sgn} k\right)\left(a+c_1 h_{-k}^{(1)}-\right. \\ \left.2 \mathrm{i} b_2 \operatorname{sgn} k-2 \mathrm{i} h_{-k}^{(4)} \operatorname{sgn} k\right)=\epsilon+c_2^2 h_k^{(2)} h_{-k}^{(2)}-c_1^2 h_k^{(1)} h_{-k}^{(1)}, \end{gathered}$$ (20)

    式中

    $$ \begin{gathered} \mathtt{ϵ}=a\left(c_2 h_{-k}^{(2)}+c_2 h_k^{(2)}-c_1 h_{-k}^{(1)}-c_1 h_k^{(1)}\right)+ \\ 2 \mathrm{i} a \operatorname{sgn} k\left(h_{-k}^{(4)}+h_k^{(3)}-h_k^{(4)}-h_{-k}^{(3)}\right)+ \\ 2 \mathrm{i} \operatorname{sgn} k\left(b_1 c_2 h_{-k}^{(2)}-b_1 c_2 h_k^{(2)}+b_2 c_1 h_k^{(1)}-b_2 c_1 h_{-k}^{(1)}+\right. \\ \left.c_1 h_k^{(1)} h_{-k}^{(4)}-c_1 h_k^{(4)} h_{-k}^{(1)}+c_2 h_k^{(3)} h_{-k}^{(2)}-c_2 h_k^{(2)} h_{-k}^{(3)}\right)+ \\ 4\left(b_1^2-b_2^2+b_1 h_{-k}^{(3)}+b_1 h_k^{(3)}-b_2 h_{-k}^{(4)}-b_2 h_k^{(4)}+h_k^{(3)} h_{-k}^{(3)}-h_k^{(4)} h_{-k}^{(4)}\right)_{\circ} \end{gathered} $$

    因为$h_{ \pm k}^{(j)} \in l^2[-\mathtt{π}, \mathtt{π}](j=1, 2, 3, 4) $,因此有

    $$ \lim \limits_{k \rightarrow \infty} h_{ \pm k}^{(j)}=0, \quad j=1, 2, 3, 4, $$ (21)

    那么由式(21),有

    $$ \lim \limits_{k \rightarrow \infty}\left|\boldsymbol{D}_k\right|=4 b_1^2-4 b_2^2, $$ (22)

    由于要求方程在正则型的情况下求解,也就是需要满足下列正则性$ \left|{b}_{1}\right|\ne \left|{b}_{2}\right|, $所以存在正整数$ {k}_{0} $,使得当$ \left|k\right| > {k}_{0} $时,有$ \left|\boldsymbol{D}_k\right| \neq 0 $。下面分别讨论$ \left|k\right| > {k}_{0} $与$ \left|k\right|\le {k}_{0} $ 2种情况。

    (1)当$ \left|k\right| > {k}_{0} $时。

    方程组的系数行列式$\left|\boldsymbol{D}_k\right| \neq 0 $,由Cramer法则知其有唯一解

    $$ f_k^{-}=\frac{\left|\boldsymbol{D}_k^{(1)}\right|}{\left|\boldsymbol{D}_k\right|}, \quad f_{-k}^{-}=\frac{\left|\boldsymbol{D}_k^{(2)}\right|}{\left|\boldsymbol{D}_k\right|}, $$ (23)

    式中

    $$ \left|\boldsymbol{D}_k^{(1)}\right|=\left|\begin{array}{cc} g_k & \delta_k^{(2)} \\ g_{-k} & \delta_{-k}^{(1)} \end{array}\right|, \quad\left|\boldsymbol{D}_k^{(2)}\right|=\left|\begin{array}{cc} \delta_k^{(1)} & g_k \\ \delta_{-k}^{(2)} & g_{-k} \end{array}\right|。 $$

    (2)当$ \left|k\right|\le {k}_{0} $时。

    (ⅰ)$ \left|\boldsymbol{D}_k\right| \neq 0$,即$ \boldsymbol{D}_k \neq \mathbf{0}\left(|k| \leqslant k_0\right) $。此时方程组(17)有唯一解,唯一解表达形式与式(23)相同。

    (ⅱ)$ \left|\boldsymbol{D}_k\right| \neq 0\left(|k| \leqslant k_0\right), g_{ \pm k}=0 $。此时方程组(17)仅仅存在零解。

    (ⅲ)若存在$ {k}_{j} (1\le j\le {p}_{1}, -{k}_{0}\le {k}_{j}\le {k}_{0}) $,使得

    $$ \left|\boldsymbol{D}_{k_j}\right|=0, \quad R\left(\boldsymbol{D}_{k_j}\right)=R\left(\boldsymbol{B}_{k_j}\right), $$

    其中$ R\left(\boldsymbol{D}_{k_j}\right), R\left(\boldsymbol{B}_{k_j}\right) $分别为矩阵$ \boldsymbol{D}_{k_j} $与矩阵$ \boldsymbol{B}_{k_j}$的秩,此时方程组(17)有无数组解

    $$ {f}_{{k}_{j}}^{-}={c}_{{k}_{j}}^{-}, \hspace{1em}{f}_{-{k}_{j}}^{-}={c}_{-{k}_{j}}^{-} 。 $$

    (ⅳ)若存在$ {k}_{j}^{\text{'}} (1\le j\le {p}_{2}, -{k}_{0}\le {k}_{j}^{\text{'}} \le {k}_{0}) $使得$ \left|\boldsymbol{D}_{k_j^{\prime}}\right|=0, R\left(\boldsymbol{D}_{k_j^{\prime}}\right) \neq R\left(\boldsymbol{B}_{k_j^{\prime}}\right) $,此时方程组(17)无解。

    (ⅴ)若存在$ {k}_{j}^{\text{'}\text{'}} (1\le j\le {p}_{3}, -{k}_{0}\le {k}_{j}^{\text{'}\text{'}}\le {k}_{0}) $,使得$ \left|\boldsymbol{D}_{k_j^*}\right|=0, g_{ \pm k_j^*}=0 $, 则可以得到

    $$ g_{ \pm k_j^*}=w^{-1}[G(\theta)]=\frac{1}{2 \mathtt{π}} \int_{-\mathtt{π}}^\mathtt{π} G(\theta) \mathrm{e}^{ \pm \mathrm{i} k_j^* \theta} \mathrm{~d} \theta=0, $$ (24)

    即

    $$ \begin{gathered} \int_{-\mathtt{π}}^\mathtt{π} G( \pm \theta) \cos \left(k_j^{\prime \prime} \theta\right) \mathrm{d} \theta=0, \\ \int_{-\mathtt{π}}^\mathtt{π} G( \pm \theta) \sin \left(k_j^{\prime \prime} \theta\right) \mathrm{d} \theta=0, \end{gathered} $$ (25)

    因此式(25)成立时方程组(17)可解。

    下面讨论$ {g}_{k}=0 $的情况,即方程组(10)齐次问题的求解。

    此时方程组为

    $$ \left\{\begin{array}{l}{\delta }_{k}^{\left(1\right)}{f}_{k}^{-}+{\delta }_{k}^{\left(2\right)}{f}_{-k}^{-}=0, \\ {\delta }_{-k}^{\left(2\right)}{f}_{k}^{-}+{\delta }_{-k}^{\left(1\right)}{f}_{-k}^{-}=0, \end{array}\right. $$ (26)

    仍分为如下2种情况讨论。

    (1)当$ \left|k\right| > {k}_{0} $时,由上述讨论,此时方程组(26)只有零解。

    (2)当$ \left|k\right|\le {k}_{0} $时,在这种情况下,存在$ {k}_{1}, {k}_{2}, {k}_{3}, \cdots , {k}_{p} $使得$ \left|\boldsymbol{D}_{k_j}\right|=0(j=1, 2, \cdots, p) $,此时解$ {f}_{{k}_{j}}^{-}, {f}_{-{k}_{j}}^{-} $均为常数,

    $$ f_{-k_j}^{-}=\frac{1}{2 \mathtt{π}} \displaystyle \int_{-\mathtt{π}}^\mathtt{π} F^{-}(\theta) \mathrm{e}^{\mathrm{i} k_j \theta} \mathrm{~d} \theta=-\frac{1}{2 \mathtt{π}} \displaystyle \int_0^\mathtt{π} F(\theta) \mathrm{e}^{-\mathrm{i} k_j \theta} \mathrm{~d} \theta\text{,} $$ (27)

    和

    $$ f_{k_j}^{-}=\frac{1}{2 \mathtt{π}} \displaystyle \int_{-\mathtt{π}}^\mathtt{π} F^{-}(\theta) \mathrm{e}^{-\mathrm{i} k_j \theta} \mathrm{~d} \theta=-\frac{1}{2 \mathtt{π}} \displaystyle \int_{-\mathtt{π}}^0 F(\theta) \mathrm{e}^{-\mathrm{i} k_j \theta} \mathrm{~d} \theta。 $$ (28)

    因此,由式(27和28),可以得到

    $$ {f}_{{k}_{j}}=-({f}_{{k}_{j}}^{-}+{f}_{-{k}_{j}}^{-}), $$ (29)

    因而$ {f}_{{k}_{j}} $也为常数,而对于其他的$ k (k\ne {k}_{1}, {k}_{2}, \cdots , {k}_{p}) $,此时$ \left|\boldsymbol{D}_k\right| \neq 0$,因此方程组只有零解$ {f}_{k}^{-}={f}_{-k}^{-}=0, $即$ {f}_{k}=0。 $

    由式(2),可以得到

    $$ F(\theta)=f_{k_1} \mathrm{e}^{\mathrm{i} k_1 \theta}+f_{k_2} \mathrm{e}^{\mathrm{i} k_2 \theta}+\cdots+f_{k_p} \mathrm{e}^{\mathrm{i} k_p \theta} 。 $$ (30)

    事实上,因为

    $$ \begin{aligned} \left\langle\mathrm{e}^{\mathrm{i} k_s \theta}, \mathrm{e}^{\mathrm{i} k_t \theta}\right\rangle & =\int_{-\mathtt{π}}^\mathtt{π} \mathrm{e}^{\mathrm{i} k_s \theta} \overline{\mathrm{e}^{\mathrm{i} k_t \theta}} \mathrm{~d} \theta= \\ \int_{-\mathtt{π}}^\mathtt{π} \mathrm{e}^{\mathrm{i}\left(k_s-k_t\right) \theta} \mathrm{d} \theta & = \begin{cases}2 \mathtt{π}, & k_s=k_t, \\ 0, & k_s \neq k_t,\end{cases} \end{aligned} $$

    所以$ \mathrm{e}^{\mathrm{i} k_1 \theta}, \mathrm{e}^{\mathrm{i} k_2 \theta}, \cdots, \mathrm{e}^{\mathrm{i} k_p \theta}$是正交的,是线性无关的,那么由式(30)可知,方程组(26)有线性无关解$ \mathrm{e}^{\mathrm{i} k_1 \theta}, \mathrm{e}^{\mathrm{i} k_2 \theta}, \cdots, \mathrm{e}^{\mathrm{i} k_p \theta}$。

    通过上面的讨论,可得如下本文的重要结果。

    定理1  (1)当$ \left|\boldsymbol{D}_k\right| \neq 0(k=0, \pm 1, \cdots) $时,方程(10)有唯一解,且由式(29),解为

    $$ \begin{gathered} F(\theta)=-\left(F^{-}(\theta)+F^{-}(-\theta)\right)= \\ -\sum\limits_{k=-\infty}^{+\infty} f_{-k}^{-} \mathrm{e}^{\mathrm{i} k \theta}-\sum\limits_{k=-\infty}^{+\infty} f_k^{-} \mathrm{e}^{\mathrm{i} k \theta}=\sum\limits_{k=-\infty}^{+\infty} f_k \mathrm{e}^{\mathrm{i} k \theta} 。 \end{gathered} $$

    (2)当$ \left|\boldsymbol{D}_k\right| \neq 0$且$ {g}_{k}=0 (k=0, \pm 1, \cdots , \pm {k}_{0}) $时,方程(10)有解

    $$ F(\theta)=-\sum\limits_{|k|>k_0} f_{-k}^{-} \mathrm{e}^{\mathrm{i} k \theta}-\sum\limits_{|k|>k_0} f_k^{-} \mathrm{e}^{\mathrm{i} k \theta}=\sum\limits_{|k|>k_0} f_k \mathrm{e}^{\mathrm{i} k \theta} 。 $$

    (3)当$ \left|\boldsymbol{D}_{k_j}\right|=0 $且$ R\left(\boldsymbol{D}_{k_j}\right)=R\left(\boldsymbol{B}_{k_j}\right)(1 \leqslant j \leqslant p_{1}, -{k}_{0}\le {k}_{j}\le {k}_{0}) $时,方程组(17)有无数组解

    $$ {f}_{{k}_{j}}^{-}={c}_{{k}_{j}}^{-}, \hspace{1em}{f}_{-{k}_{j}}^{-}={c}_{-{k}_{j}}^{-}, $$

    其中$ {c}_{{k}_{j}}^{-}, {c}_{-{k}_{j}}^{-} $均为常数,此时方程(10)的解为

    $$ \begin{aligned} F(\theta)=- & \sum\limits_{j=1}^{p_1} c_{ \pm k_j}^{-} \mathrm{e}^{\mathrm{i} k_j \theta}-\sum\limits_{k=-\infty, k \neq \pm k_j}^{+\infty}\left(f_k^{-} \mathrm{e}^{\mathrm{i} k \theta}+f_{-k}^{-} \mathrm{e}^{\mathrm{i} k \theta}\right)= \\ & -\sum\limits_{j=1}^{p_1} c_{ \pm k_j}^{-} \mathrm{e}^{\mathrm{i} k_j \theta}+\sum\limits_{k=-\infty, k \neq \pm k_j}^{+\infty} f_k \mathrm{e}^{\mathrm{i} k \theta} 。 \end{aligned} $$

    (4)当$ \left|\boldsymbol{D}_{k_j^{\prime}}\right|=0$且$ R\left(\boldsymbol{D}_{{k}_{j}^{\text{'}}}\right)\ne R\left(\boldsymbol{B}_{{k}_{j}^{\text{'}}}\right) (1\le j\le {p}_{2}, $ $ -{k}_{0}\le {k}_{j}^{\text{'}}\le {k}_{0}) $时,此时方程(10)无解。

    (5)当$ \left|\boldsymbol{D}_{{k}_{j}^{\text{'}\text{'}} }\right|=0 $且$ {g}_{\pm {k}_{j}^{\text{'}\text{'}} }=0 (1\le j\le {p}_{3}, -{k}_{0}\le {k}_{j}^{\text{'}\text{'}} \le {k}_{0}) $时,解的情况就是上面讨论齐次的情况,当且仅当式(25)成立时,方程(10)有解,解为

    $$ F(\theta)=\sum\limits_{j=1}^{P_3} f_{k_j^*} \mathrm{e}^{\mathrm{i} k_j^* \theta}。 $$

    推论1  在方程(10)中,若$ {H}_{j}\in \left\{\tilde{0}\right\} (j=\mathrm{1, 2}, \mathrm{3, 4}) $,则式(10)在函数类$ H $的可解条件和解的情况与定理1类似。

    本文研究的带有2个卷积核和Hilbert核的奇异积分方程可推广到高维的Clifford分析中。如何将经典复分析的方法和理论引入到高维的Clifford分析中,并使其成为研究相关奇异积分方程的重要工具,也是当前研究的重点问题之一。

  • [1] MUSKHELISHVILLI N I. Singular integral equations[M]. Moscow: NauKa, 1968.
    [2] VEKUA I N. Generalized analytic functions[M]. Moscow: NauKa, 1986.
    [3] 李凯雅, 魏鑫. 一类具有Hilbert核非正则型奇异积分方程的直接解法[J]. 菏泽学院学报, 2020, 42(5): 9-14. DOI: 10.16393/j.cnki.37-1436/z.2020.05.004.

    LI K Y, WEI X. The direct method of solution for a class of singularity integral equation with Hilbert kernel[J]. Joural of Heze University, 2020, 42(5): 9-14. doi: 10.16393/j.cnki.37-1436/z.2020.05.004
    [4] ABUSALIM M S, ABDOU A M, NASR E M, et al. An algorithm for the solution of nonlinear Volterra-Fredholm integral equations with a singular kernel[J]. Fractal and Fractional, 2023, 7(10): 1-21. DOI: 10.3390/fractalfract7100730.
    [5] BOYKOV I V, PIVKINA A A. Approximate methods for solving degenerate singular integral equations[J]. Technical Physics, 2023, 68(3): 67-73. DOI: 10.1134/S1063784223020020.
    [6] DE-BONIS M C, LAURITA C. Numerical solution of systems of Cauchy singular integral equations with constant coefficients[J]. Applied Mathematics and Computation, 2012, 219(4): 1391-1410. DOI: 10.1016/j.amc.2012.08.022.
    [7] DU H, SHEN J H. Reproducing kernel method of solving singular integral equation with cosecant kernel[J]. Journal of Mathematical Analysis and Applications, 2008, 348(1): 308-314. DOI: 10.1016/j.jmaa.2008.07.037.
    [8] GIANG B T, MAU N V, TUAN N M. Convolutions for the Fourier transforms with geometric variables and applications[J]. Mathematische Nachrichten, 2010, 283(12): 1758-1770. DOI: 10.1002/mana.200710192.
    [9] LI P R, REN G B. Solvability of singular integro-differential equations via Riemann-Hilbert problem[J]. Journal of Differential Equations, 2018, 265(11): 5455-5471. DOI: 10.1016/j.jde.2018.07.056.
    [10] LI P R. Holomorphic solutions and solvability theory for a class of linear complete singular integro-differential equations with convolution by Riemann-Hilbert method[J]. Analysis and Mathematical Physics, 2022, 12(6): 1-29. DOI: 10.1007/s13324-022-00759-6.
    [11] LI P R. Existence of analytic solutions for some classes of singular integral equations of non-normal type with convolution kernel[J]. Acta Applicandae Mathematicae, 2022, 181(1): 1-5. DOI: 10.1007/s10440-022-00522-w.
    [12] QIAO Y Y, XU Y Z, YANG H J. Poincare-Bertrand transformation formula of Cauchy-type singular integrals in Clifford analysis[J]. Complex Variables and Elliptic Equations, 2012, 57(2): 197-217. DOI: 10.1080/17476933.2011.593098.
    [13] DELANGHE R, SOMMEN F. On the Schwarz reflection principle for monogenic function[J]. Archiv der Mathematik, 2007, 89(2): 170-184. DOI: 10.1007/s00013-007-2212-5.
    [14] BLOCKI Z. Suita conjecture and Ohsawa-Takegoshi extension theorem[J]. Inventiones Mathematicae, 2013, 193(1): 149-158. DOI: 10.1007/s00222-012-0423-2.
    [15] LUO W, DU J. Generalized Cauchy theorem in Clifford analysis and boundary value problems for regular functions[J]. Advances in Applied Clifford Algebras, 2017, 27(3): 2531-2583. DOI: 10.1007/s00006-017-0790-2.
    [16] KHOSRAVI H, ALLAHYARI R, HAGHIGHI A. Existence of solutions of functional integral equations of convolution type using a new construction of a measure of noncompactness on $ L^p\left(\mathrm{R}_{+}\right) $[J]. Applied Mathematics and Computation, 2015, 260(1): 140-147. DOI: 10.1016/j.amc.2015.03.035.
    [17] BRACKX F, DELANGHE F, SOMMEN F. Clifford Analysis[M]. Boston: Pitman Publishers, 1982.
    [18] CAO N, LI Z, YANG H, et al. Cauchy type integrals and a boundary value problem in a complex Clifford analysis[J]. Acta Mathematica Scientia, 2024, 44(1): 369-385. DOI: 10.1007/s10473-024-0120-4.
    [19] BRACKX F, DE SCHEPPER H, LÁVIČKA R, et al. The Cauchy integral formula in Hermitian, quaternionic and osp (4|2) Clifford analysis[J]. Computational Methods and Function Theory, 2020, 20(3): 431-464. DOI: 10.1007/s40315-020-00322-z.
    [20] LI P R. Linear BVPs and SIEs for generalized regular functions in Clifford analysis[J]. Journal of Function Spaces, 2018, 2018(1): 1-7. DOI: 10.1155/2018/6967149.
    [21] LU J K. Boundary value problems for analytic functions[M]. Singapore: World Scientific, 2004.
    [22] LI P R. Singular integral equations of convolution type with Hilbert kernel and a discrete jump problem[J]. Advances in Difference Equations, 2017, 2017(1): 1-13. DOI: 10.1186/s13662-017-1413-x.
    [23] BLAYA R A, REYES J B, BRACKX F, et al. Cauchy integral formulae in quaternionic Hermitean Clifford analysis[J]. Complex Analysis and Operator Theory, 2012, 6(5): 971-985. DOI: 10.1007/s11785-011-0168-8.
WeChat 点击查看大图
出版历程
  • 收稿日期:  2024-11-11

目录

    /

    返回文章
    返回