说明

本文并不是严格的证明,只是简单的用于理解什么时候用 d x dx dx什么时候用 d s ds ds,所以采用的曲线比较特殊,为 y = x y=x y=x并且在 [ 0 , 1 ] [0,1] [0,1]区间上,实际对于复杂的曲线,取微元时 Δ y \Delta y Δy Δ x \Delta x Δx为线性关系,即 Δ y = α Δ x \Delta y=\alpha \Delta x Δy=αΔx,在本文中 Δ y = Δ x \Delta y=\Delta x Δy=Δx
将区间划为 n n n个小区间,对于区间 [ a , b ] [a,b] [a,b],每个区间 Δ x = b − a n \Delta x=\frac{b-a}{n} Δx=nba,区间的左端点为 a + i n ( b − a ) a+\frac{i}{n}(b-a) a+ni(ba),右端点为 a + i + 1 n ( b − a ) a+\frac{i+1}{n}(b-a) a+ni+1(ba) i 的 取 值 为 { 0 , 1 , 2 , . . . , n − 1 } i的取值为\{0,1,2,...,n-1\} i{0,1,2,...,n1},而对于本例中,区间范围为 [ 0 , 1 ] [0,1] [0,1],每个区间 Δ x = 1 n \Delta x=\frac{1}{n} Δx=n1,区间的左端点为 i n \frac{i}{n} ni,右端点为 i + 1 n \frac{i+1}{n} ni+1

面积与弧长的情况

取一小段区间, Δ x = 1 n \Delta x=\frac{1}{n} Δx=n1,如图所示,放大来看
在这里插入图片描述

面积

Δ A = f ( i n ) ⋅ Δ x + 1 2 ⋅ Δ x ⋅ Δ y = f ( i n ) ⋅ 1 n + 1 2 ⋅ 1 n ⋅ 1 n \Delta A=f\left( \frac{i}{n} \right) \cdot \Delta x+\frac{1}{2}\cdot \Delta x\cdot \Delta y=f\left( \frac{i}{n} \right) \cdot \frac{1}{n}+\frac{1}{2}\cdot \frac{1}{n}\cdot \frac{1}{n} ΔA=f(ni)Δx+21ΔxΔy=f(ni)n1+21n1n1
A = lim ⁡ n → + ∞ ( ∑ i = 0 n − 1 ( f ( i n ) ⋅ 1 n ) + 1 2 ⋅ 1 n ⋅ 1 n ⋅ n ) = lim ⁡ n → + ∞ ( 1 n ⋅ ∑ i = 0 n − 1 f ( i n ) + 1 2 ⋅ 1 n ) = lim ⁡ n → + ∞ ( 1 n ⋅ ∑ i = 0 n − 1 f ( i n ) ) + lim ⁡ n → + ∞ 1 2 n \begin{aligned} A&=\underset{n\rightarrow +\infty}{\lim}\left( \sum_{i=0}^{n-1}{\left( f\left( \frac{i}{n} \right) \cdot \frac{1}{n} \right)}+\frac{1}{2}\cdot \frac{1}{n}\cdot \frac{1}{n}\cdot n \right) \\ &=\underset{n\rightarrow +\infty}{\lim}\left( \frac{1}{n}\cdot \sum_{i=0}^{n-1}{f\left( \frac{i}{n} \right)}+\frac{1}{2}\cdot \frac{1}{n} \right) \\ &=\underset{n\rightarrow +\infty}{\lim}\left( \frac{1}{n}\cdot \sum_{i=0}^{n-1}{f\left( \frac{i}{n} \right)} \right) +\underset{n\rightarrow +\infty}{\lim}\frac{1}{2n} \end{aligned} A=n+lim(i=0n1(f(ni)n1)+21n1n1n)=n+lim(n1i=0n1f(ni)+21n1)=n+lim(n1i=0n1f(ni))+n+lim2n1

由于 lim ⁡ n → + ∞ 1 2 n = 0 \underset{n\rightarrow +\infty}{\lim}\frac{1}{2n}=0 n+lim2n1=0
所以
A = lim ⁡ n → + ∞ ( 1 n ⋅ ∑ i = 0 n − 1 f ( i n ) ) A=\underset{n\rightarrow +\infty}{\lim}\left( \frac{1}{n}\cdot \sum_{i=0}^{n-1}{f\left( \frac{i}{n} \right)} \right) A=n+lim(n1i=0n1f(ni))
由定积分的定义得
A = lim ⁡ n → + ∞ ( 1 n ⋅ ∑ i = 0 n − 1 f ( i n ) ) = ∫ 0 1 f ( x ) d x A=\underset{n\rightarrow +\infty}{\lim}\left( \frac{1}{n}\cdot \sum_{i=0}^{n-1}{f\left( \frac{i}{n} \right)} \right) =\int_0^1{f\left( x \right) dx} A=n+lim(n1i=0n1f(ni))=01f(x)dx

弧长

Δ S = ( Δ x ) 2 + ( Δ y ) 2 \Delta S=\sqrt{\left( \Delta x \right) ^2+\left( \Delta y \right) ^2} ΔS=(Δx)2+(Δy)2

S = lim ⁡ n → + ∞ ( ∑ i = 0 n − 1 ( Δ x ) 2 + ( Δ y ) 2 ) = lim ⁡ n → + ∞ ( ∑ i = 0 n − 1 1 + ( Δ y Δ x ) 2 ⋅ Δ x ) = ∫ 0 1 1 + [ f ′ ( x ) ] 2 d x = ∫ s d s \begin{aligned} S&=\underset{n\rightarrow +\infty}{\lim}\left( \sum_{i=0}^{n-1}{\sqrt{\left( \Delta x \right) ^2+\left( \Delta y \right) ^2}} \right)\\ &=\underset{n\rightarrow +\infty}{\lim}\left( \sum_{i=0}^{n-1}{\sqrt{1+\left( \frac{\Delta y}{\Delta x} \right) ^2}}\cdot \Delta x \right)\\ &=\int_0^1{\sqrt{1+\left[ f'\left( x \right) \right] ^2}dx}\\ &=\int_s{ds}\\ \end{aligned} S=n+lim(i=0n1(Δx)2+(Δy)2 )=n+limi=0n11+(ΔxΔy)2 Δx=011+[f(x)]2 dx=sds

旋转体体积和表面积的情况

y = x y=x y=x在区间 [ 0 , 1 ] [0,1] [0,1]上绕x轴旋转
在这里插入图片描述
则取一小段区间是一个圆台,图画的不是很好,但可以看出这个圆台的上底面半径为 f ( i n ) f\left( \frac{i}{n} \right) f(ni) ,下底面半径为 f ( i n ) + Δ y f\left( \frac{i}{n} \right)+\Delta y f(ni)+Δy,高位 Δ x \Delta x Δx
在这里插入图片描述

圆台侧面积和体积公式见
公式推导——圆台的侧面积和体积
本文直接给出
在这里插入图片描述
在这里插入图片描述

体积

Δ V = 1 3 π Δ x [ f 2 ( i n ) + f ( i n ) ⋅ ( f ( i n ) + Δ y ) + ( f ( i n ) + Δ y ) 2 ] = 1 3 π Δ x [ f 2 ( i n ) + f 2 ( i n ) + f ( i n ) ⋅ Δ y + f 2 ( i n ) + 2 ⋅ f ( i n ) ⋅ Δ y + ( Δ y ) 2 ] = 1 3 π Δ x [ 3 ⋅ f 2 ( i n ) + 3 ⋅ f ( i n ) ⋅ Δ y + ( Δ y ) 2 ] = π ⋅ f 2 ( i n ) ⋅ Δ x + π ⋅ f ( i n ) ⋅ Δ x ⋅ Δ y + 1 3 π Δ x ⋅ ( Δ y ) 2 \begin{aligned} \Delta V&=\frac{1}{3}\pi \Delta x\left[ f^2\left( \frac{i}{n} \right) +f\left( \frac{i}{n} \right) \cdot \left( f\left( \frac{i}{n} \right) +\Delta y \right) +\left( f\left( \frac{i}{n} \right) +\Delta y \right) ^2 \right] \\ &=\frac{1}{3}\pi \Delta x\left[ f^2\left( \frac{i}{n} \right) +f^2\left( \frac{i}{n} \right) +f\left( \frac{i}{n} \right) \cdot \Delta y+f^2\left( \frac{i}{n} \right) +2\cdot f\left( \frac{i}{n} \right) \cdot \Delta y+\left( \Delta y \right) ^2 \right] \\ &=\frac{1}{3}\pi \Delta x\left[ 3\cdot f^2\left( \frac{i}{n} \right) +3\cdot f\left( \frac{i}{n} \right) \cdot \Delta y+\left( \Delta y \right) ^2 \right] \\ &=\pi \cdot f^2\left( \frac{i}{n} \right) \cdot \Delta x+\pi \cdot f\left( \frac{i}{n} \right) \cdot \Delta x\cdot \Delta y+\frac{1}{3}\pi \Delta x\cdot \left( \Delta y \right) ^2 \end{aligned} ΔV=31πΔx[f2(ni)+f(ni)(f(ni)+Δy)+(f(ni)+Δy)2]=31πΔx[f2(ni)+f2(ni)+f(ni)Δy+f2(ni)+2f(ni)Δy+(Δy)2]=31πΔx[3f2(ni)+3f(ni)Δy+(Δy)2]=πf2(ni)Δx+πf(ni)ΔxΔy+31πΔx(Δy)2

V = lim ⁡ n → + ∞ [ ∑ i = 0 n − 1 ( π ⋅ f 2 ( i n ) ⋅ Δ x ) + ∑ i = 0 n − 1 ( π ⋅ f ( i n ) ⋅ Δ x ⋅ Δ y ) + ∑ i = 0 n − 1 ( 1 3 π Δ x ⋅ ( Δ y ) 2 ) ] = lim ⁡ n → + ∞ ∑ i = 0 n − 1 ( π ⋅ f 2 ( i n ) ⋅ Δ x ) + lim ⁡ n → + ∞ ∑ i = 0 n − 1 ( π ⋅ f ( i n ) ⋅ Δ x ⋅ Δ y ) + lim ⁡ n → + ∞ ∑ i = 0 n − 1 ( 1 3 π Δ x ⋅ ( Δ y ) 2 ) \begin{aligned} V&=\underset{n\rightarrow +\infty}{\lim}\left[ \sum_{i=0}^{n-1}{\left( \pi \cdot f^2\left( \frac{i}{n} \right) \cdot \Delta x \right)}+\sum_{i=0}^{n-1}{\left( \pi \cdot f\left( \frac{i}{n} \right) \cdot \Delta x\cdot \Delta y \right)}+\sum_{i=0}^{n-1}{\left( \frac{1}{3}\pi \Delta x\cdot \left( \Delta y \right) ^2 \right)} \right] \\ &=\underset{n\rightarrow +\infty}{\lim}\sum_{i=0}^{n-1}{\left( \pi \cdot f^2\left( \frac{i}{n} \right) \cdot \Delta x \right)}+\underset{n\rightarrow +\infty}{\lim}\sum_{i=0}^{n-1}{\left( \pi \cdot f\left( \frac{i}{n} \right) \cdot \Delta x\cdot \Delta y \right)}+\underset{n\rightarrow +\infty}{\lim}\sum_{i=0}^{n-1}{\left( \frac{1}{3}\pi \Delta x\cdot \left( \Delta y \right) ^2 \right)} \end{aligned} V=n+lim[i=0n1(πf2(ni)Δx)+i=0n1(πf(ni)ΔxΔy)+i=0n1(31πΔx(Δy)2)]=n+limi=0n1(πf2(ni)Δx)+n+limi=0n1(πf(ni)ΔxΔy)+n+limi=0n1(31πΔx(Δy)2)

其中
∵ lim ⁡ n → + ∞ ∑ i = 0 n − 1 ∣ ( π ⋅ f ( i n ) ⋅ Δ x ⋅ Δ y ) ∣ ≤ lim ⁡ n → + ∞ ∑ i = 0 n − 1 ∣ ( π ⋅ max ⁡ ( f ( x ) ) ⋅ Δ x ⋅ Δ y ) ∣ \because \underset{n\rightarrow +\infty}{\lim}\sum_{i=0}^{n-1}{\left| \left( \pi \cdot f\left( \frac{i}{n} \right) \cdot \Delta x\cdot \Delta y \right) \right|}\leq \underset{n\rightarrow +\infty}{\lim}\sum_{i=0}^{n-1}{\left| \left( \pi \cdot \max \left( f\left( x \right) \right) \cdot \Delta x\cdot \Delta y \right) \right|} n+limi=0n1(πf(ni)ΔxΔy)n+limi=0n1(πmax(f(x))ΔxΔy)
由于分母的数量级为 n 2 ,分子的数量级为 n ,所以当 n → + ∞ 时 \text{由于分母的数量级为}n^2\text{,分子的数量级为}n\text{,所以当}n\rightarrow +\infty \text{时} 由于分母的数量级为n2,分子的数量级为n,所以当n+
lim ⁡ n → + ∞ ∑ i = 0 n − 1 ∣ ( π ⋅ max ⁡ ( f ( x ) ) ⋅ Δ x ⋅ Δ y ) ∣ = 0 \underset{n\rightarrow +\infty}{\lim}\sum_{i=0}^{n-1}{\left| \left( \pi \cdot \max \left( f\left( x \right) \right) \cdot \Delta x\cdot \Delta y \right) \right|}=0 n+limi=0n1(πmax(f(x))ΔxΔy)=0
∴ lim ⁡ n → + ∞ ∑ i = 0 n − 1 ( π ⋅ f ( i n ) ⋅ Δ x ⋅ Δ y ) = 0 \therefore \underset{n\rightarrow +\infty}{\lim}\sum_{i=0}^{n-1}{\left( \pi \cdot f\left( \frac{i}{n} \right) \cdot \Delta x\cdot \Delta y \right)}=0 n+limi=0n1(πf(ni)ΔxΔy)=0
显然
lim ⁡ n → + ∞ ∑ i = 0 n − 1 ( 1 3 π Δ x ⋅ ( Δ y ) 2 ) = 0 \underset{n\rightarrow +\infty}{\lim}\sum_{i=0}^{n-1}{\left( \frac{1}{3}\pi \Delta x\cdot \left( \Delta y \right) ^2 \right)}=0 n+limi=0n1(31πΔx(Δy)2)=0
所以
V = lim ⁡ n → + ∞ ∑ i = 0 n − 1 ( π ⋅ f 2 ( i n ) ⋅ Δ x ) = ∫ 0 1 π x 2 d x V=\underset{n\rightarrow +\infty}{\lim}\sum_{i=0}^{n-1}{\left( \pi \cdot f^2\left( \frac{i}{n} \right) \cdot \Delta x \right)}=\int_0^1{\pi x^2dx} V=n+limi=0n1(πf2(ni)Δx)=01πx2dx

表面积

Δ A = π ( f ( i n ) + f ( i n ) + Δ y ) Δ s = π ( 2 f ( i n ) + Δ y ) Δ s \Delta A=\pi \left( f\left( \frac{i}{n} \right) +f\left( \frac{i}{n} \right) +\Delta y \right) \Delta s=\pi \left( 2f\left( \frac{i}{n} \right) +\Delta y \right) \Delta s ΔA=π(f(ni)+f(ni)+Δy)Δs=π(2f(ni)+Δy)Δs
A = lim ⁡ n → + ∞ [ ∑ i = 0 n − 1 ( π ( 2 f ( i n ) + Δ y ) Δ s ) ] = lim ⁡ n → + ∞ ∑ i = 0 n − 1 ( π ⋅ 2 f ( i n ) ⋅ Δ s ) + lim ⁡ n → + ∞ ∑ i = 0 n − 1 ( π ⋅ Δ y ⋅ Δ s ) \begin{aligned} A&=\underset{n\rightarrow +\infty}{\lim}\left[ \sum_{i=0}^{n-1}{\left( \pi \left( 2f\left( \frac{i}{n} \right) +\Delta y \right) \Delta s \right)} \right] \\ &=\underset{n\rightarrow +\infty}{\lim}\sum_{i=0}^{n-1}{\left( \pi \cdot 2f\left( \frac{i}{n} \right) \cdot \Delta s \right)}+\underset{n\rightarrow +\infty}{\lim}\sum_{i=0}^{n-1}{\left( \pi \cdot \Delta y\cdot \Delta s \right)} \end{aligned} A=n+lim[i=0n1(π(2f(ni)+Δy)Δs)]=n+limi=0n1(π2f(ni)Δs)+n+limi=0n1(πΔyΔs)
其中
lim ⁡ n → + ∞ ∑ i = 0 n − 1 ( π ⋅ Δ y ⋅ Δ s ) = 0 \underset{n\rightarrow +\infty}{\lim}\sum_{i=0}^{n-1}{\left( \pi \cdot \Delta y\cdot \Delta s \right)}=0 n+limi=0n1(πΔyΔs)=0

A = lim ⁡ n → + ∞ ∑ i = 0 n − 1 ( π ⋅ 2 f ( i n ) ⋅ Δ s ) = ∫ s 2 π f ( x ) d s A=\underset{n\rightarrow +\infty}{\lim}\sum_{i=0}^{n-1}{\left( \pi \cdot 2f\left( \frac{i}{n} \right) \cdot \Delta s \right)}=\int_s{2\pi f\left( x \right) ds} A=n+limi=0n1(π2f(ni)Δs)=s2πf(x)ds

总结

其实不管是求面积还是体积的时候不是想当然的不考虑那一小块(这是很多人误解的地方,从而导致分不清什么时候用 d x dx dx,什么时候用 d s ds ds),而是在求定积分的时候(本质是求极限)那一部分趋于0,所以在实际计算的时候可以忽略趋于0的部分。

Logo

开放原子开发者工作坊旨在鼓励更多人参与开源活动,与志同道合的开发者们相互交流开发经验、分享开发心得、获取前沿技术趋势。工作坊有多种形式的开发者活动,如meetup、训练营等,主打技术交流,干货满满,真诚地邀请各位开发者共同参与!

更多推荐