权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Self-validated computation of singular integral and integral equations

奇异积分和积分方程的自验证计算

基本信息

批准号：
15540111
负责人：
YAMAMOTO Nobito
金额：
$ 1.66万
依托单位：
The University of Electro-Ccmmunications
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (C)
财政年份：
2003
资助国家：
日本
起止时间：
2003 至 2004
项目状态：
已结题

项目摘要

The present research has two purposes.1. Establishment of self-validated computation for singular integral using DE transformation, which is one of most effective methods for calculating approximate values of singular integrals.2. Development of numerical verification methods for the existence and the local uniqueness of solutions to integral equations.On the self-validated computation of DE transformation, we constructed a set of basic techniques for a computer program library of computation with guaranteed accuracy of singular integrals. For the arguments of the programs, we suppose integrands composed of elementary functions. The class of the DE transformation is chosen corresponding to the singularity of the integrand at the ends of the interval of integral. In order to estimate the error bounds, we need to verify the regularity of the integrand on an expanded area in a complex region. For this purpose, the method by Sugiura et al. is adopted. These results are described in detail in the report of the research results.On the numerical verification of integral equations, we carried out our study as follows.1. Establish a numerical verification for the local uniqueness of solutions to function equations including integral equations.2. Develop a numerical verification methods for the systems of ordinal differential equations with initial values which are derived from integral equations3. Implementation of 1 to 2.For 1, we have got an important result which will be useful for a wide range of self-validation. For 2 and 3, we have developed a new method and are now improving it for practical use.

本研究有两个目的。利用DE变换建立奇异积分的自验证计算，这是计算奇异积分近似值的最有效方法之一。积分方程解的存在性和局部唯一性数值验证方法的发展。在DE变换的自验证计算上，构造了一套保证奇异积分计算精度的计算机程序库的基本技术。对于程序的参数，我们假设由初等函数组成的被积函数。根据被积函数在积分区间端点处的奇异性来选择DE变换的类别。为了估计误差范围，我们需要在一个复杂区域的扩展区域上验证被积函数的正则性。为此，采用Sugiura等人的方法。这些结果在研究成果报告中有详细的描述。在积分方程的数值验证上，我们进行了如下的研究。建立了包括积分方程在内的函数方程解的局部唯一性的数值验证。研究了由积分方程导出的初值有序微分方程组的数值验证方法。实现1到2。对于1，我们已经得到了一个重要的结果，它将对广泛的自我验证有用。对于2和3，我们已经开发了一种新方法，现在正在改进它以供实际使用。