Forgot password?
 Register account
View 6031|Reply 11

[几何] 一直不知怎么解释椭圆内接矩形是“方方正正”

[Copy link]

770

Threads

4692

Posts

310K

Credits

Credits
35048

Show all posts

isee Posted 2018-9-5 20:27 |Read mode
就是必有一组对边垂直于$x$轴,个人基本每次都是在选修4-5中坐标伸缩变换中“提”一下。

但是,奇怪的是为什么大家(无论是试卷里,还是做题目者)都默认就是“方方正正”的叱?

770

Threads

4692

Posts

310K

Credits

Credits
35048

Show all posts

 Author| isee Posted 2018-9-5 20:38
“提”一下是指:
将椭圆伸缩为圆,则“椭圆内接矩形”变成“圆内接平行四边形”,再由对称互补且相等,此平行边形亦是矩形。

圆内接矩形两对角线的中点即为坐标原点,从而椭圆内接矩形的对角线交点(即矩形中心)亦是坐标原点,再(在椭圆中)计算对角线长,由每一个象限内的“弧”到原点距离是单调的~

总之,很麻烦。。。。。


我印象里大学解析几何里好像有解析证明,,但真的忘记了。

686

Threads

110K

Posts

910K

Credits

Credits
91229
QQ

Show all posts

kuing Posted 2018-9-5 23:19
回复 2# isee

解几很容易证啊,就是常规的点差法。

设曲线为 `Ax^2+By^2=1`,四顶点顺次为 `P_i(x_i,y_i)`, `i=1`, `2`, `3`, `4`,则
\begin{align*}
A(x_1-x_2)(x_1+x_2)+B(y_1-y_2)(y_1+y_2)&=0,\\
A(x_3-x_4)(x_3+x_4)+B(y_3-y_4)(y_3+y_4)&=0,
\end{align*}
由矩形知 `x_1-x_2=x_4-x_3`, `y_1-y_2=y_4-y_3`,可见直线\[A(x_1-x_2)x+B(y_1-y_2)y=0\]经过边 `P_1P_2` 及 `P_3P_4` 的中点,也就是说对边中点的连线过原点,另一对边同理,从而矩形中心在原点上,然后再利用对角线相等即容易得出结论。

PS、由此证法可知结论在椭圆和双曲线里都成立(双曲线可无法伸缩变圆喔

1

Threads

55

Posts

329

Credits

Credits
329

Show all posts

huing Posted 2018-9-5 23:50
椭圆上的点从一个顶点滑动到一个相邻顶点,到中心的距离是单调的,这个事实不用繁琐的计算证明。
假定不是单调的,将至少有一个与椭圆同心的圆与椭圆有至少8个交点,这不可能。

1

Threads

55

Posts

329

Credits

Credits
329

Show all posts

huing Posted 2018-9-6 08:01
椭圆的内接矩形必定与椭圆同轴亦可由5点决定一个椭圆来理解。一个给定矩形的外接椭圆已经有了4个点,还剩一个自由度,再给定一点就能唯一确定。不妨把剩下的这个自由点取在矩形的一个主轴上,由此5点决定的椭圆已知有一个与矩形同轴,由五点决定论知它是唯一的。余略。

7

Threads

578

Posts

3956

Credits

Credits
3956

Show all posts

游客 Posted 2018-9-6 13:42
以矩形中心为圆心,对角线长为直径,画个圆,对称性显然。

3

Threads

41

Posts

228

Credits

Credits
228

Show all posts

wwdwwd117 Posted 2018-9-6 15:54
回复 6# 游客


    首先你要证明矩形中心和椭圆中心重合

770

Threads

4692

Posts

310K

Credits

Credits
35048

Show all posts

 Author| isee Posted 2018-9-6 16:52
综合楼上看来,这个问题并不是特别显而易见的。

5#这个高观下解释厉害了。

3#这个直接证明了7#从而6楼成立了。


就3#而言,后半段我用斜率积,当$P_1P_3$过椭圆$x^2/a^2+y^2/b^2=1(a>b>0)$中心,则$$k_{P_2P_1}\cdot k_{P_2P_3}=-\frac {b^2}{a^2},$$这与$$\angle P_1P_2P_3=90^\circ$$矛盾。

770

Threads

4692

Posts

310K

Credits

Credits
35048

Show all posts

 Author| isee Posted 2018-9-6 16:57
感谢楼主各位,学习了。

另外:网上查了下,另有两种证明一个是反证,一个是将5#具体化了。

7

Threads

578

Posts

3956

Credits

Credits
3956

Show all posts

游客 Posted 2018-9-7 09:06
回复 7# wwdwwd117


    圆心从原点移开,四边形就不是平行四边形:
椭圆平行弦的中点轨迹是过椭圆中心的。

3159

Threads

7941

Posts

610K

Credits

Credits
63770
QQ

Show all posts

hbghlyj Posted 2023-3-14 02:14

互链

椭圆内接矩形的边,为什么垂直于坐标轴?
isee的回答

3159

Threads

7941

Posts

610K

Credits

Credits
63770
QQ

Show all posts

hbghlyj Posted 2023-3-14 05:56

任何内接于球体的平行六面体都必须是直角平行六面体

caltech physics set4 solution
First, we show that any parallelepiped inscribed in a sphere must be a right parallelepiped. We use a sphere of unit radius without lack of generality. We first choose $\vec{r}=(x, y, z)$, the position of one of the vertices. It must hold that
\begin{equation}
x^2+y^2+z^2=1\label1
\end{equation}
Now, let the vectors $\vec{A}, \vec{B}$, and $\vec{C}$ define the directions to the nearest three vertices from $\vec{R}$, placing those vertices at $\vec{R}+\vec{A}, \vec{R}+\vec{B}$, and $\vec{R}+\vec{C}$. Then we also have the equations
\begin{align}
& \left(x+x_A\right)^2+\left(y+y_A\right)^2+\left(z+z_A\right)^2=1 \label{RA}\\
& \left(x+x_B\right)^2+\left(y+y_B\right)^2+\left(z+z_B\right)^2=1 \label{RB}\\
& \left(x+x_C\right)^2+\left(y+y_C\right)^2+\left(z+z_C\right)^2=1
\end{align}
Since the sphere defines a full 2-dimensional surface, there is an infinite number of possible choices of $\vec{A}, \vec{B}$, and $\vec{C}$ that stay on the sphere. But now the remainder of the vertices are fully determined: they are at $\vec{R}+\vec{A}+\vec{B}, \vec{R}+\vec{A}+\vec{C}, \vec{R}+\vec{B}+\vec{C}$, and $\vec{R}+\vec{A}+\vec{B}+\vec{C}$. We thus have four more equations that must be satisfied:
\begin{align}
\left(x+x_A+x_B\right)^2+\left(y+y_A+y_B\right)^2+\left(z+z_A+z_B\right)^2=1\label5 \\
\left(x+x_A+x_C\right)^2+\left(y+y_A+y_C\right)^2+\left(z+z_A+z_C\right)^2=1\label6\\
\left(x+x_B+x_C\right)^2+\left(y+y_B+y_C\right)^2+\left(z+z_B+z_C\right)^2=1\label7\\
\left(x+x_A+x_B+x_C\right)^2+\left(y+y_A+y_B+y_C\right)^2+\left(z+z_A+z_B+z_C\right)^2=1\label8
\end{align}
Now, if one subtracts \eqref{RA}\eqref{RB} from \eqref{5} and adds \eqref{1}, and does similarly for the equations \eqref{6}\eqref{7}, one obtains\begin{array}{l}2 x_{A} x_{B}+2 y_{A} y_{B}+2 z_{A} z_{B}=0 \\ 2 x_{A} x_{C}+2 y_{A} y_{C}+2 z_{A} z_{C}=0 \\ 2 x_{B} x_{C}+2 y_{B} y_{C}+2 z_{B} z_{C}=0\end{array}These are just dot-product equations, indicating that $\vec A, \vec B,$ and $\vec C$ must be mutually perpendicular. Hence, the parallelepiped must be a right parallelepiped.

Mobile version|Discuz Math Forum

2025-5-31 11:20 GMT+8

Powered by Discuz!

× Quick Reply To Top Edit