您的位置:站长主页 -> 繁星客栈 -> 望月殿 (数学逻辑论坛) -> (随笔)Riemann映射定理及其推广形式略谈 November 22, 2024

(随笔)Riemann映射定理及其推广形式略谈

用户登陆 | 刷新 本版嘉宾: 萍踪浪迹 季候风 星空与道德 gage

萍踪浪迹

发表文章数: 1983
武功等级: 深不可测
内力值: 645/645

(随笔)Riemann映射定理及其推广形式略谈



写篇随笔。
经典的Riemann映射定理告诉我们复平面上单连通真子集都可以和单位圆建立共形映射
因此,所有复平面上单连通真子集之间都可以建立共形映射等价
但是在多连通情况下,这个定理不再成立,即使是连通数相同的都无法保证这个性质的成立
我们研究两个同心圆环区域,这是很简单的
它们要共形等价,必须使两个圆环的内外半径成一定比例

现在我们看看多复变函数论中的单连通域情形,由于它们都是单连通,所以必定同胚,又由于它们都是单连通,我们似乎可以指望Riemann映射定理可以推广到多复变情形,因为Riemann映射定理在一维情形的单连通情况在这里被保持者。
但是实际上,这个是无法成立的。1907年,Poincare在一篇论文中指出了这一点,从而拓扑等价单连通区域与共形等价单连通区域成为两个截然不同的概念,也使得这个多复变函数论中区域的分类理论变得极为重要。
我们研究C ^2中两个最简单的单连通区域:
单位球面:|z_1|^2+|z_2|^2 <=1
双圆柱:|z_1|^2 <=1,|z_2|^2 <=1
这两个区域都是单连通区域,是单复变情形中是单位圆的两个推广,它们同胚
但是它们各自的全纯自同构群不一样,于是,它们无法建立全纯等价(共形等价)
以现代方式证明这个结论要用到H. Cartan的两个定理(Cartan定理A,Cartan定理A)
本质上是推广著名的Schwarz引理的在构造全纯自同构群方面的作用。


漫漫长夜不知晓 日落云寒苦终宵
痴心未悟拈花笑 梦魂飞度同心桥
-------------------------------------------------
红叶晚萧萧,长亭酒一瓢
残云归太华,疏雨过中条
树色随山迥,河声入海遥
帝乡明日到,犹自梦渔樵


发表时间:2005-11-23, 01:45:25  作者资料

sage

发表文章数: 1125
武功等级: 天山六阳掌
     (第六重)
内力值: 535/535

Re: (随笔)Riemann映射定理及其推广形式略谈



It is not so surprising the simple connected-ness (or vanish first homotopy group) does not implies that much. After all, it is only a very small portion of the topological information.

so what is the proper generation of that theorem to higher dimensional spaces.


发表时间:2005-11-23, 13:53:29  作者资料

季候风

发表文章数: 291
武功等级: 太极剑法
     (第四重)
内力值: 370/370

Re: (随笔)Riemann映射定理及其推广形式略谈



二维跟三维以上最显著的差距在于三维以上的“刚性”。二维流形的双曲结构构成一个模空间,三维以上流形的双曲结构则是唯一的。究其本质,就是Riemann映射定理和Liouville 定理的差别。除整个复平面以外的任何复平面的单连通开子集与单位开圆盘共形等价。但在三维以上,与单位球体共形等价的只有不同大小的球体。


书山有路勤为径
学海无涯苦作舟


发表时间:2005-11-23, 15:03:53  作者资料

sage

发表文章数: 1125
武功等级: 天山六阳掌
     (第六重)
内力值: 535/535

Re: (随笔)Riemann映射定理及其推广形式略谈



another stupid question

what is 双曲结构?


发表时间:2005-11-23, 16:15:57  作者资料

季候风

发表文章数: 291
武功等级: 太极剑法
     (第四重)
内力值: 370/370

Re: (随笔)Riemann映射定理及其推广形式略谈



截面曲率恒为 -1 的黎曼度量。


书山有路勤为径
学海无涯苦作舟


发表时间:2005-11-23, 20:56:55  作者资料

sage

发表文章数: 1125
武功等级: 天山六阳掌
     (第六重)
内力值: 535/535

Re: (随笔)Riemann映射定理及其推广形式略谈



二维流形的双曲结构构成一个模空间,三维以上流形的双曲结构则是唯一的。究其本质,就是Riemann映射定理和Liouville 定理的差别。除整个复平面以外的任何复平面的单连通开子集与单位开圆盘共形等价。
===============================================================

I understand what you want to say.

I have a question on details

Riemann theorem, applied to open simply connected set, seems to say there is no modulous (i.e., every metric is conformally equivalent). And as you said, this is not true for d>2.

On the other hand, I am a little bit confused about what you said for hyperbolic(?) structures. you said in d=2, there is modulous, and in d>2 there is no?

Or, we might be talking about different things in talking about moduli space? for me, the exitsence of modulous means that there is a class of inequivalent metric, such as a d=2 torus.


发表时间:2005-11-24, 11:18:42  作者资料

季候风

发表文章数: 291
武功等级: 太极剑法
     (第四重)
内力值: 370/370

Re: (随笔)Riemann映射定理及其推广形式略谈



well, talking about moduli requires the underlying structure, or background structure. In this case, the differential structure, or, roughly, topological structure. All the simply-connected region ( somehow borded ) are topologically the same, so that we can say the topological disk has no moduli. However, the topological ball (>=3D) has many conformal structure, because there are other regions in E^n which are topologically balls but by Liouville theorem, they are not conformally equivalent to the unit ball, or in other words, the topological ball has moduli.


书山有路勤为径
学海无涯苦作舟


发表时间:2005-11-24, 12:04:11  作者资料

季候风

发表文章数: 291
武功等级: 太极剑法
     (第四重)
内力值: 370/370

Re: (随笔)Riemann映射定理及其推广形式略谈



Sorry, I forgot what I want to say when I was typing the last article. In 2D, because of the Riemann uniformization, hyperbolic structure is a very nice representative in each conformal equivalence class of metrics. So hyperbolic moduli correspond to conformal moduli.

In higher dimensions, say, 3D, the hyperbolic space is defined as the unit ball equipped with a group action of PSL(2,C) which is embedded in the conformal group of S^3. The other topological balls in S^3 don't have a Poincare metric conformally equivalent to the metric induced from S^3, or in other words, the conformal groups of the other topologicals ball are not PSL(2,C). This is the reason why the topological ball has conformal moduli but the hyperbolic structure is unique. Only the geometric ball can serve as models of the hyperbolic space.


书山有路勤为径
学海无涯苦作舟


发表时间:2005-11-24, 13:50:50  作者资料

萍踪浪迹

发表文章数: 1983
武功等级: 深不可测
内力值: 645/645

Re: (随笔)Riemann映射定理及其推广形式略谈



::what is 双曲结构?

::截面曲率恒为 -1 的黎曼度量。
==================================================
低维情形如单位开圆盘,赋予Poincare metric以后,可以计算出其截面曲率恒为-1 。

ps:to 季侯风兄,前天给你信箱发一邮件,不知是否被系统拖到回收站(因为yahoo现在限制垃圾邮件,经常出现这种情况),请查收一下。因为我这半月没有空闲上来,所以就仰仗你和星空道德兄的管理了,先谢过:)


漫漫长夜不知晓 日落云寒苦终宵
痴心未悟拈花笑 梦魂飞度同心桥


发表时间:2005-11-25, 09:21:02  作者资料

星空与道德

发表文章数: 258
武功等级: 华山剑法
     (第九重)
内力值: 360/360

Re: (随笔)Riemann映射定理及其推广形式略谈



Sorry, I forgot what I want to say when I was typing the last article. In 2D, because of the Riemann uniformization, hyperbolic structure is a very nice representative in each conformal equivalence class of metrics. So hyperbolic moduli correspond to conformal moduli.

In higher dimensions, say, 3D, the hyperbolic space is defined as the unit ball equipped with a group action of PSL(2,C) which is embedded in the conformal group of S^3. The other topological balls in S^3 don't have a Poincare metric conformally equivalent to the metric induced from S^3, or in other words, the conformal groups of the other topologicals ball are not PSL(2,C). This is the reason why the topological ball has conformal moduli but the hyperbolic structure is unique. Only the geometric ball can serve as models of the hyperbolic space.
--------------------------
不是很明白你的意思。二维的结果是熟知的。在三维的时候,你说
This is the reason why the topological ball has conformal moduli but the hyperbolic structure is unique.

如果三维时hyperbolic structure是一个度量,他唯一确定的话难道他诱导的共形结构会有一个模空间?所以这个如果是错的。

所以你第二段第一句话是关于三维hyperbolic space的定义,这个定义是什么意思?


堕落吧,朋友!


发表时间:2005-11-25, 18:10:13  作者资料

sage

发表文章数: 1125
武功等级: 天山六阳掌
     (第六重)
内力值: 535/535

Re: (随笔)Riemann映射定理及其推广形式略谈



I forgot what I want to say when I was typing the last article. In 2D, because of the Riemann uniformization, hyperbolic structure is a very nice representative in each conformal equivalence class of metrics. So hyperbolic moduli correspond to conformal moduli.

In higher dimensions, say, 3D, the hyperbolic space is defined as the unit ball equipped with a group action of PSL(2,C) which is embedded in the conformal group of S^3. The other topological balls in S^3 don't have a Poincare metric conformally equivalent to the metric induced from S^3, or in other words, the conformal groups of the other topologicals ball are not PSL(2,C). This is the reason why the topological ball has conformal moduli but the hyperbolic structure is unique. Only the geometric ball can serve as models of the hyperbolic space.
=====================================================================

I think this, including later comments from Xing Kong yu dao De confuse me more. I think ther confusion mainly comes from the terminology. So it will be helpful if you could clarify the following definitions for me

1) moduli space to me means that a collection of metrics which are NOT equivalent to each other, such as the complex structure modulous for torus.

2) now, what is hyperbolic moduli and conformal moduli. is the conformal moduli parameterizes spaces conformally equivalent to each other, or conformal classes which are not equivalent to each other? I think it is porbably the later but I am not sure.

3) what you are saying is that in 3D, there is not a family of hyperbolic structures, right?


发表时间:2005-11-25, 23:31:26  作者资料

星空与道德

发表文章数: 258
武功等级: 华山剑法
     (第九重)
内力值: 360/360

Re: (随笔)Riemann映射定理及其推广形式略谈



这一篇应该和yinhow的"球面的自同构群"放在一起,它们之间有很强的联系。

鉴于我对复结构的兴趣,季兄谈到三维双曲流形的边界和球面的关系吸引了我,而我对三维流形一无所知,所以无法明白季兄的话。我用google搜索了一下,确定了几件事情。

1,双曲空间H^3(R3的上半平面)上的度量是
ds^ 2 = (dx^ 2 + dy^ 2 + dz^ 2 )/z^ 2

2,H^3的边界是复平面C,我们可以将H^3紧致化使得他的边界是球面,并且H^3的等距变换可以延拓为边界球面上的保角变换。

3,双曲空间H^3上的双曲度量是唯一的。也就是说给定任何两个双曲度量,我们可以找到一个等距同构。这是在2维时不具备的性质。


堕落吧,朋友!


发表时间:2005-11-26, 00:27:25  作者资料

星空与道德

发表文章数: 258
武功等级: 华山剑法
     (第九重)
内力值: 360/360

Re: (随笔)Riemann映射定理及其推广形式略谈



季兄说,
但在三维以上,与单位球体共形等价的只有不同大小的球体。
-------------------
如果单位球体是开的,也就是不包含边界。那么由于双曲度量是唯一的,如果每个共形等价类中有一个双曲度量的话,他应该只有一个共形等价类。

所以季兄说的球体应该是闭的,也就是包含边界。那么前面说到开球体的双曲度量唯一,但它们是在等距变换的意义下唯一,但这个等距变换延拓到边界上以后就不是等距的了,只是保角的了,所以可以相信
与单位"闭"球体共形等价的只有不同大小的球体。

OK,我想我基本上明白了,当然只是结论而不是明白原因。


堕落吧,朋友!


发表时间:2005-11-26, 00:41:14  作者资料

季候风

发表文章数: 291
武功等级: 太极剑法
     (第四重)
内力值: 370/370

Re: (随笔)Riemann映射定理及其推广形式略谈



如果每个共形等价类中有一个双曲度量的话
~~~~~~~~~~~~~~~~~~~~~~~~~~~~

如果我没搞错的话,这个在三维应该是不成立的。开球体还是有很多不同的保角结构。


书山有路勤为径
学海无涯苦作舟


发表时间:2005-11-26, 01:32:58  作者资料