定义4.1 f ( x ) f(x) f(x)在 x 0 x_0 x0的某个邻域上有定义,如果极限 lim x → x 0 f ( x ) − f ( x 0 ) x − x 0 \lim_{x\to x_0}{\frac{f(x)-f(x_0)}{x-x_0}} x→x0limx−x0f(x)−f(x0)存在,则称 f ( x ) f(x) f(x)在 x 0 x_0 x0处可导,该极限称为 f ( x ) f(x) f(x)在 x 0 x_0 x0处的导数,记为 f ′ ( x 0 ) f^{\prime}(x_0) f′(x0),进一步地,如果 f ( x ) f(x) f(x)在区间 I I I的每一个点都可导,那么 f ′ ( x ) f^{\prime}(x) f′(x)就是区间 I I I上的函数,称为 f ( x ) f(x) f(x)在 I I I上的导函数,简称导数
定理4.1 f ( x ) f(x) f(x)在 x 0 x_0 x0处可导,那么 f ( x ) f(x) f(x)在 x 0 x_0 x0处连续
证:
f ( x ) f(x) f(x)在 x 0 x_0 x0处可导,那么极限 lim x → x 0 f ( x ) − f ( x 0 ) x − x 0 \lim_{x\to x_0}{\frac{f(x)-f(x_0)}{x-x_0}} x→x0limx−x0f(x)−f(x0)存在,而 lim x → x 0 ( x − x 0 ) = 0 \lim_{x\to x_0}{(x-x_0)}=0 x→x0lim(x−x0)=0因此,就有 lim x → x 0 ( f ( x ) − f ( x 0 ) ) = lim x → x 0 f ( x ) − f ( x 0 ) x − x 0 ( x − x 0 ) = lim x → x 0 f ( x ) − f ( x 0 ) x − x 0 . lim x → x 0 ( x − x 0 ) = 0 \lim_{x\to x_0}{(f(x)-f(x_0))}=\lim_{x\to x_0}{\frac{f(x)-f(x_0)}{x-x_0}(x-x_0)}=\lim_{x\to x_0}{\frac{f(x)-f(x_0)}{x-x_0}}.\lim_{x\to x_0}{(x-x_0)}=0 x→x0lim(f(x)−f(x0))=x→x0limx−x0f(x)−f(x0)(x−x0)=x→x0limx−x0f(x)−f(x0).x→x0lim(x−x0)=0因此, f ( x ) f(x) f(x)在 x 0 x_0 x0处连续
但连续函数不一定可导,甚至连续函数可能处处不可导。
定义4.2 f ( x ) f(x) f(x)在 x 0 x_0 x0的某个右(左)半邻域有定义,如果极限 lim x → x 0 + f ( x ) − f ( x 0 ) x − x 0 ( lim x → x 0 − f ( x ) − f ( x 0 ) x − x 0 ) \lim_{x\to {x_0}^{+}}{\frac{f(x)-f(x_0)}{x-x_0}} (\lim_{x\to {x_0}^{-}}{\frac{f(x)-f(x_0)}{x-x_0}}) x→x0+limx−x0f(x)−f(x0)(x→x0−limx−x0f(x)−f(x0))存在, f ( x ) f(x) f(x)在 x 0 x_0 x0处的右(左)导数存在,该极限值称为 f ( x ) f(x) f(x)在 x 0 x_0 x0处的右(左)导数,记为 f ′ + ( x 0 ) ( f ′ − ( x 0 ) ) f^{\prime +}(x_0)(f^{\prime -}(x_0)) f′+(x0)(f′−(x0))
根据左右极限和函数极限的关系,有
定理4.2 f ( x ) f(x) f(x)在 x 0 x_0 x0处可导的充分必要条件是 f ( x ) f(x) f(x)在 x 0 x_0 x0处的左右导数存在且相等
下面,我们来证明导数的一些运算法则:
定理4.3
(1) f ( x ) f(x) f(x)和 g ( x ) g(x) g(x)在 x 0 x_0 x0处可导,则 f ( x ) + g ( x ) f(x)+g(x) f(x)+g(x)在 x 0 x_0 x0处可导,并且 f ( x ) + g ( x ) f(x)+g(x) f(x)+g(x)在 x 0 x_0 x0处的导数为 f ′ ( x 0 ) + g ′ ( x 0 ) f^{\prime}(x_0) +g^{\prime}(x_0) f′(x0)+g′(x0)
(2) f ( x ) f(x) f(x)在 x 0 x_0 x0处可导,则对任意的实数 c c c, c f ( x ) cf(x) cf(x)在 x 0 x_0 x0处可导,并且 c f ( x ) cf(x) cf(x)在 x 0 x_0 x0处的导数为 c f ′ ( x 0 ) cf^{\prime}(x_0) cf′(x0)
(3) f ( x ) f(x) f(x)和 g ( x ) g(x) g(x)在 x 0 x_0 x0处可导,则 f ( x ) g ( x ) f(x)g(x) f(x)g(x)在 x 0 x_0 x0处可导,并且 f ( x ) g ( x ) f(x)g(x) f(x)g(x)在 x 0 x_0 x0处的导数为 f ( x 0 ) g ′ ( x 0 ) + f ′ ( x 0 ) g ( x 0 ) f(x_0)g^{\prime}(x_0)+f^{\prime}(x_0)g(x_0) f(x0)g′(x0)+f′(x0)g(x0)
(4) f ( x ) f(x) f(x)和 g ( x ) g(x) g(x)在 x 0 x_0 x0处可导,并且 g ( x 0 ) ≠ 0 g(x_0)\neq 0 g(x0)=0,则 f ( x ) g ( x ) \frac{f(x)}{g(x)} g(x)f(x)在 x 0 x_0 x0处可导, f ( x ) g ( x ) \frac{f(x)}{g(x)} g(x)f(x)在 x 0 x_0 x0处的导数为 f ′ ( x 0 ) g ( x 0 ) − g ′ ( x 0 ) f ( x 0 ) g 2 ( x 0 ) \frac{f^{\prime}(x_0)g(x_0)-g^{\prime}(x_0)f(x_0)}{g^{2}(x_0)} g2(x0)f′(x0)g(x0)−g′(x0)f(x0)
证:
(1)(2)由极限的四则运算法则是显然的,仅证(3)(4)
(3)考察变化率 f ( x ) g ( x ) − f ( x 0 ) g ( x 0 ) x − x 0 = f ( x ) g ( x ) − f ( x 0 ) g ( x ) + f ( x 0 ) g ( x ) − f ( x 0 ) g ( x 0 ) x − x 0 = g ( x ) f ( x ) − f ( x 0 ) x − x 0 + f ( x 0 ) g ( x ) − g ( x 0 ) x − x 0 \begin{aligned} \frac{ f(x)g(x) - f(x_0)g(x_0) }{ x-x_0 } =\frac{ f(x)g(x) - f(x_0)g(x) + f(x_0)g(x) - f(x_0)g(x_0) }{ x-x_0 }\\ =g(x)\frac{ f(x) - f(x_0) }{ x - x_0 } + f(x_0) \frac{ g(x)-g(x_0) }{x - x_0}\\ \end{aligned} x−x0f(x)g(x)−f(x0)g(x0)=x−x0f(x)g(x)−f(x0)g(x)+f(x0)g(x)−f(x0)g(x0)=g(x)x−x0f(x)−f(x0)+f(x0)x−x0g(x)−g(x0)再由 g ( x ) g(x) g(x)在 x 0 x_0 x0处连续,令 x → x 0 x\to x_0 x→x0,即可证得结论
(4)先证明 ( 1 g ( x ) ) ′ ( x 0 ) = − g ′ ( x 0 ) g 2 ( x 0 ) (\frac{1}{g(x)})^{\prime}(x_0) =-\frac{ g^{\prime}(x_0) }{ g^{2}(x_0) } (g(x)1)′(x0)=−g2(x0)g′(x0)考察变化率函数 1 g ( x ) − 1 g ( x 0 ) x − x 0 = − g ( x ) − g ( x 0 ) g ( x ) g ( x 0 ) ( x − x 0 ) \begin{aligned} \frac{ \frac{1}{g(x)} - \frac{1}{g(x_0)} }{x-x_0} =-\frac{g(x)-g(x_0)}{ g(x)g(x_0)(x-x_0) } \end{aligned} x−x0g(x)1−g(x0)1=−g(x)g(x0)(x−x0)g(x)−g(x0)由 g ( x ) g(x) g(x)的连续性及在 x 0 x_0 x0处的可导性,两边对 x → x 0 x\to x_0 x→x0取极限即可证得
在应用(3)的结论就可以证得(4)
复合函数也有相应的求导法则
定理4.4 f ( x ) f(x) f(x)在 x 0 x_0 x0处可导, g ( y ) g(y) g(y)在 y 0 = f ( x 0 ) y_0=f(x_0) y0=f(x0)处可导,则 g ( f ( x ) ) g(f(x)) g(f(x))在 x 0 x_0 x0处可导, g ( f ( x ) ) g(f(x)) g(f(x))处导数为 f ′ ( x 0 ) g ′ ( f ( x 0 ) ) f^{\prime}(x_0)g^{\prime}(f(x_0)) f′(x0)g′(f(x0))
证:
考察变化率函数 g ( f ( x ) ) − g ( f ( x 0 ) ) x − x 0 \frac{ g(f(x))-g(f(x_0)) }{ x-x_0 } x−x0g(f(x))−g(f(x0))由于可能在每个 x 0 x_0 x0的去心邻域上,都可能存在 x x x, f ( x ) = f ( x 0 ) f(x)=f(x_0) f(x)=f(x0),
我们不能替换成: g ( f ( x ) ) − g ( f ( x 0 ) ) x − x 0 = g ( f ( x ) ) − g ( f ( x 0 ) ) f ( x ) − f ( x 0 ) f ( x ) − f ( x 0 ) x − x 0 \frac{ g(f(x))-g(f(x_0)) }{ x-x_0 } =\frac{ g(f(x))-g(f(x_0)) }{ f(x)-f(x_0) } \frac{ f(x)-f(x_0) }{ x-x_0 } x−x0g(f(x))−g(f(x0))=f(x)−f(x0)g(f(x))−g(f(x0))x−x0f(x)−f(x0)进行证明,实际上,对 f ( x ) = f ( x 0 ) f(x)=f(x_0) f(x)=f(x0)的点,我们补充一个定义 F ( x ) = { g ′ ( y 0 ) f ( x ) = f ( x 0 ) g ( f ( x ) ) − g ( f ( x 0 ) ) f ( x ) − f ( x 0 ) f ( x ) ≠ f ( x 0 ) F(x)= \begin{cases} g^{\prime}(y_0)&f(x)=f(x_0)\\ \frac{ g(f(x))-g(f(x_0)) }{ f(x)-f(x_0) }& f(x)\neq f(x_0) \end{cases} F(x)={g′(y0)f(x)−f(x0)g(f(x))−g(f(x0))f(x)=f(x0)f(x)=f(x0)这样, g ( f ( x ) ) − g ( f ( x 0 ) ) = F ( x ) ( f ( x ) − f ( x 0 ) ) g(f(x)) - g(f(x_0)) =F(x)(f(x)-f(x_0)) g(f(x))−g(f(x0))=F(x)(f(x)−f(x0))就有 g ( f ( x ) ) − g ( f ( x 0 ) ) x − x 0 = F ( x ) f ( x ) − f ( x 0 ) x − x 0 \frac{ g(f(x)) - g(f(x_0)) }{x - x_0} =F(x) \frac{ f(x)-f(x_0) }{x-x_0} x−x0g(f(x))−g(f(x0))=F(x)x−x0f(x)−f(x0)而 F ( x ) − F ( x 0 ) = { 0 f ( x ) = f ( x 0 ) g ( f ( x ) ) − g ( y 0 ) f ( x ) − y 0 − g ′ ( y 0 ) f ( x ) ≠ f ( x 0 ) F(x)-F(x_0)= \begin{cases} 0 & f(x)=f(x_0)\\ \frac{ g(f(x))-g(y_0) }{ f(x)-y_0} -g^{\prime}(y_0) & f(x)\neq f(x_0) \end{cases} F(x)−F(x0)={0f(x)−y0g(f(x))−g(y0)−g′(y0)f(x)=f(x0)f(x)=f(x0) ∀ ε > 0 \forall \varepsilon>0 ∀ε>0, ∃ δ 1 > 0 \exists \delta_1>0 ∃δ1>0, 0 < ∣ y − y 0 ∣ < δ 1 0<|y-y_0|<\delta_1 0<∣y−y0∣<δ1时, ∣ g ( y ) − g ( y 0 ) y − y 0 − g ′ ( y 0 ) ∣ < ε |\frac{g(y)-g(y_0)}{y-y_0} - g^{\prime}(y_0)|<\varepsilon ∣y−y0g(y)−g(y0)−g′(y0)∣<ε ∃ δ 2 > 0 \exists \delta_2>0 ∃δ2>0, 0 < ∣ x − x 0 ∣ < δ 2 0<|x-x_0|<\delta_2 0<∣x−x0∣<δ2时, ∣ f ( x ) − y 0 ∣ < δ 1 |f(x)-y_0|<\delta_1 ∣f(x)−y0∣<δ1
如果 f ( x ) ≠ y 0 f(x)\neq y_0 f(x)=y0 ∣ g ( f ( x ) ) − g ( y 0 ) f ( x ) − y 0 − g ′ ( y 0 ) ∣ < ε |\frac{g(f(x))-g(y_0)}{f(x)-y_0} - g^{\prime}(y_0)|<\varepsilon ∣f(x)−y0g(f(x))−g(y0)−g′(y0)∣<ε综上, 0 < ∣ x − x 0 ∣ < δ 2 0<|x-x_0|<\delta_2 0<∣x−x0∣<δ2都有 ∣ F ( x ) − F ( x 0 ) ∣ < ε |F(x)-F(x_0)|<\varepsilon ∣F(x)−F(x0)∣<ε成立, F ( x ) F(x) F(x)在 x 0 x_0 x0处连续。对(\ref{eq6})两边取极限,就可以证得结论
定理4.5 f ( x ) f(x) f(x)在 x 0 x_0 x0附近上严格单调并在 x 0 x_0 x0可导, f ′ ( x 0 ) ≠ 0 f^{\prime}(x_0)\neq 0 f′(x0)=0,则反函数 f − 1 ( y ) f^{-1}(y) f−1(y)在 y 0 = f ( x 0 ) y_0=f(x_0) y0=f(x0)也可导,并且 f − 1 ′ ( y 0 ) = 1 f ′ ( x 0 ) f^{-1 \prime}(y_0) = \frac{1}{f^{\prime}(x_0)} f−1′(y0)=f′(x0)1
证:
考察变化率函数 f − 1 ( y ) − f − 1 ( y 0 ) y − y 0 = 1 y − y 0 f − 1 ( y ) − f − 1 ( y 0 ) = 1 f ( f − 1 ( y ) ) − f ( f − 1 ( y 0 ) ) f − 1 ( y ) − f − 1 ( y 0 ) \frac{ f^{-1}(y)-f^{-1}(y_0) }{y-y_0} =\frac{1} { \frac{y-y_0}{ f^{-1}(y)-f^{-1}(y_0) } }\\ =\frac{1} { \frac{ f(f^{-1}(y))-f(f^{-1}(y_0)) }{ f^{-1}(y)-f^{-1}(y_0) } } y−y0f−1(y)−f−1(y0)=f−1(y)−f−1(y0)y−y01=f−1(y)−f−1(y0)f(f−1(y))−f(f−1(y0))1而 lim y → y 0 f ( f − 1 ( y ) ) − f ( f − 1 ( y 0 ) ) f − 1 ( y ) − f − 1 ( y 0 ) = f ′ ( f − 1 ( y 0 ) ) \lim_{y\to y_0}{ \frac{ f(f^{-1}(y))-f(f^{-1}(y_0)) }{ f^{-1}(y)-f^{-1}(y_0) } } =f^{\prime}(f^{-1}(y_0)) y→y0limf−1(y)−f−1(y0)f(f−1(y))−f(f−1(y0))=f′(f−1(y0))两边取极限可以证得结论
例4.1 ( sin x ) ′ = cos x (\sin{x})^{\prime} = \cos{x} (sinx)′=cosx
证:
lim Δ x → 0 sin ( x + Δ x ) − sin x Δ x = lim Δ x → 0 sin Δ x cos x + cos Δ x sin x − sin x Δ x = lim Δ x → 0 sin Δ x cos x + sin x ( cos Δ x − 1 ) Δ x = cos x lim Δ x → 0 sin Δ x Δ x + sin x lim Δ x → 0 cos Δ x − 1 Δ x = cos x \begin{aligned} \lim_{\Delta x \to 0} { \frac { \sin{(x+\Delta x)} - \sin{x} } {\Delta x} } =\lim_{\Delta x \to 0}{ \frac{ \sin{\Delta x}\cos{x} + \cos{\Delta x}\sin{x} - \sin{x} } {\Delta x} }\\ =\lim_{\Delta x \to 0}{ \frac{ \sin{\Delta x}\cos{x} +\sin{x}(\cos{\Delta x}-1) } {\Delta x} } =\cos{x}\lim_{\Delta x \to 0}{ \frac{\sin{\Delta x}} {\Delta x} }+\sin{x}\lim_{\Delta x \to 0}{ \frac{\cos{\Delta x}-1} {\Delta x} } =\cos{x} \end{aligned} Δx→0limΔxsin(x+Δx)−sinx=Δx→0limΔxsinΔxcosx+cosΔxsinx−sinx=Δx→0limΔxsinΔxcosx+sinx(cosΔx−1)=cosxΔx→0limΔxsinΔx+sinxΔx→0limΔxcosΔx−1=cosx
例4.2 ( cos x ) ′ = − sin x (\cos{x})^{\prime} = -\sin{x} (cosx)′=−sinx
证:
lim Δ x → 0 cos x + Δ x − cos x Δ x = cos x lim Δ x → 0 cos Δ x − 1 Δ x − sin x lim Δ x → 0 sin Δ x Δ x = − sin x \begin{aligned} \lim_{\Delta x\to 0}{ \frac{\cos{x+\Delta x}-\cos{x}} {\Delta x} } =\cos{x}\lim_{\Delta x\to 0}{ \frac{\cos{\Delta x}-1} {\Delta x} }-\sin{x}\lim_{\Delta x\to 0}{ \frac{\sin{\Delta x}} {\Delta x} } =-\sin{x} \end{aligned} Δx→0limΔxcosx+Δx−cosx=cosxΔx→0limΔxcosΔx−1−sinxΔx→0limΔxsinΔx=−sinx
这样,其他三角函数的导数也可以求出来:
例4.3
( tan x ) ′ = ( sin x cos x ) ′ = ( sin x ) ′ cos x − ( cos x ) ′ sin x cos 2 x = 1 cos 2 x (\tan{x})^{\prime}=( \frac{\sin{x}} {\cos{x}} )^{\prime}= \frac{ (\sin{x})^{\prime}\cos{x} -(\cos{x})^{\prime}\sin{x} } {\cos^{2}x}= \frac{1} {\cos^{2}x} (tanx)′=(cosxsinx)′=cos2x(sinx)′cosx−(cosx)′sinx=cos2x1
例4.4
( arcsin x ) ′ = 1 cos arcsin x = 1 1 − sin 2 arcsin x = 1 1 − x 2 (\arcsin{x})^{\prime}=\frac{1} {\cos{\arcsin{x}}} =\frac{1} {\sqrt{1-\sin^2{\arcsin{x}}}} =\frac{1}{\sqrt{1-x^2}} (arcsinx)′=cosarcsinx1=1−sin2arcsinx1=1−x21同样地,有 ( arccos x ) ′ = − 1 1 − x 2 (\arccos{x})^{\prime} = -\frac{1} {\sqrt{1-x^2}} (arccosx)′=−1−x21
例4.5 由三角函数等式: 1 + tan 2 x = sec 2 x = 1 cos 2 x 1+\tan^2{x}=\sec^2{x}=\frac{1} {\cos^2{x}} 1+tan2x=sec2x=cos2x1 ( arctan x ) ′ = 1 tan ′ ( arctan x ) = cos 2 ( arctan x ) = 1 1 + tan 2 arctan x = 1 1 + x 2 (\arctan{x})^{\prime} = \frac{1} {\tan^{\prime}(\arctan{x})} =\cos^2(\arctan{x})= \frac{1} {1+\tan^2{\arctan{x}}} =\frac{1}{1+x^2} (arctanx)′=tan′(arctanx)1=cos2(arctanx)=1+tan2arctanx1=1+x21
这样,三角函数和反三角函数的导数都是有解析表达式的。
例4.6 指数函数的导数: ( a x ) ′ = a x ln a (a^x)^\prime = a^x\ln{a} (ax)′=axlna对数函数的导数: ( log a x ) ′ = 1 x ln a (\log_{a}{x})^{\prime} = \frac{1}{x\ln{a}} (logax)′=xlna1
证:
lim Δ x → 0 a x + Δ x − a x Δ x = a x lim Δ x → 0 a Δ x − 1 Δ x = a x lim Δ x → 0 e Δ x ln a − 1 Δ x = a x lim Δ x → 0 Δ x ln a Δ x = a x ln a \lim_{\Delta x \to 0}{ \frac{a^{x+\Delta x}-a^x} {\Delta x} }=a^x\lim_{\Delta x \to 0}{ \frac{a^{\Delta x}-1} {\Delta x} } =a^x\lim_{\Delta x \to 0}{ \frac{e^{\Delta x\ln{a}}-1} {\Delta x} }=a^x\lim_{\Delta x \to 0}{ \frac{\Delta x\ln{a}} {\Delta x} }=a^x\ln{a} Δx→0limΔxax+Δx−ax=axΔx→0limΔxaΔx−1=axΔx→0limΔxeΔxlna−1=axΔx→0limΔxΔxlna=axlna从而对数的函数的导数为 ( log a x ) ′ = 1 a log a x ln a = 1 x ln a (\log_{a}{x})^{\prime} = \frac{1} {a^{\log_{a}{x}}\ln{a} } = \frac{1}{x\ln{a}} (logax)′=alogaxlna1=xlna1特别地, ( e x ) ′ = e x (e^x)^\prime = e^x (ex)′=ex ( ln x ) ′ = 1 x (\ln{x})^{\prime} = \frac{1}{x} (lnx)′=x1
例4.6 幂函数的导数: x > 0 x>0 x>0时, ( x α ) ′ = α x α − 1 (x^\alpha)^\prime = \alpha x^{\alpha -1} (xα)′=αxα−1
证:
lim Δ x → 0 ( x + Δ x ) α − x α Δ x = x α lim Δ x → 0 ( 1 + Δ x x ) α − 1 Δ x = x α lim Δ x → 0 α Δ x x Δ x = α x α − 1 \lim_{\Delta x\to 0}{ \frac{(x+\Delta x)^\alpha - x^\alpha} {\Delta x} }= x^{\alpha}\lim_{\Delta x \to 0}{ \frac{ (1+\frac{\Delta x}{x})^\alpha -1 } {\Delta x} }=x^\alpha \lim_{\Delta x\to 0}{ \frac{\alpha \frac{\Delta x}{x}} {\Delta x} }=\alpha x^{\alpha -1} Δx→0limΔx(x+Δx)α−xα=xαΔx→0limΔx(1+xΔx)α−1=xαΔx→0limΔxαxΔx=αxα−1
导数有着明确的几何意义,有了导数,我们就可以在局部,把一个复杂的函数视为是线性函数。
定义4.2 f ( x ) f(x) f(x)在 x 0 x_0 x0附近有定义,如果 f ( x ) f(x) f(x)在 x 0 x_0 x0附近可表为 f ( x ) = f ( x 0 ) + A Δ x + o ( Δ x ) f(x)=f(x_0)+A\Delta x + o(\Delta x) f(x)=f(x0)+AΔx+o(Δx)其中, Δ x = x − x 0 \Delta x = x - x_0 Δx=x−x0, A A A是与 x x x无关的常数,则称 f ( x ) f(x) f(x)在 x 0 x_0 x0处可微,线性函数 d f ( x ) = A Δ x df(x)=A\Delta x df(x)=AΔx称为是 f ( x ) f(x) f(x)在 x 0 x_0 x0处的微分
如果 f ( x ) f(x) f(x)在 x 0 x_0 x0处可微,当 Δ x \Delta x Δx足够小时,就可以近似地认为 f ( x ) ≈ f ( x 0 ) + A Δ x f(x)\approx f(x_0)+A\Delta x f(x)≈f(x0)+AΔx
定理4.6 f ( x ) f(x) f(x)在 x 0 x_0 x0处可微的充要条件是 f ( x ) f(x) f(x)在 x 0 x_0 x0处可导,并且 d f ( x ) = f ′ ( x 0 ) Δ x df(x) = f^{\prime}(x_0)\Delta x df(x)=f′(x0)Δx
我们称 y = f ( x 0 ) + f ′ ( x 0 ) ( x − x 0 ) y=f(x_0)+f^{\prime}(x_0)(x-x_0) y=f(x0)+f′(x0)(x−x0)是曲线 y = f ( x ) y=f(x) y=f(x)在 x 0 x_0 x0处的切线,那么导数 f ′ ( x 0 ) f^{\prime}(x_0) f′(x0)就是切线的斜率。
证:
必要性,如果 f ( x ) f(x) f(x)在 x 0 x_0 x0处可微,有 f ( x ) − f ( x 0 ) = A ( x − x 0 ) + o ( x − x 0 ) f(x)-f(x_0)=A(x-x_0)+o(x-x_0) f(x)−f(x0)=A(x−x0)+o(x−x0)两边除以 x − x 0 x-x_0 x−x0,再令 x → x 0 x\to x_0 x→x0,就有 A = f ′ ( x 0 ) A=f^{\prime}(x_0) A=f′(x0)
充分性,如果 f ( x ) f(x) f(x)在 x 0 x_0 x0处可导,则 f ( x ) − f ( x 0 ) x − x 0 − f ′ ( x 0 ) \frac{f(x)-f(x_0)}{x-x_0} - f^{\prime}(x_0) x−x0f(x)−f(x0)−f′(x0)是 x → x 0 x\to x_0 x→x0过程的无穷小量,即 f ( x ) − f ( x 0 ) x − x 0 − f ′ ( x 0 ) = α \frac{f(x)-f(x_0)}{x-x_0} - f^{\prime}(x_0) = \alpha x−x0f(x)−f(x0)−f′(x0)=α其中, α \alpha α是 x → x 0 x\to x_0 x→x0过程的无穷小量,就可以得到 f ( x ) − f ( x 0 ) = f ′ ( x 0 ) ( x − x 0 ) + ( x − x 0 ) α = f ′ ( x 0 ) ( x − x 0 ) + o ( x − x 0 ) f(x)-f(x_0)=f^{\prime}(x_0)(x-x_0)+(x-x_0)\alpha =f^{\prime}(x_0)(x-x_0)+o(x-x_0) f(x)−f(x0)=f′(x0)(x−x0)+(x−x0)α=f′(x0)(x−x0)+o(x−x0)
同样地,我们可以定义导数的导数,导数的导数的导数, ⋯ \cdots ⋯。记 f ( k ) f^{(k)} f(k)为k阶导数,表示对 f ( x ) f(x) f(x)求k次导数。高阶导数的计算常常要使用数学归纳法,下面我们给出若干例子。
例4.7 f ( x ) , g ( x ) f(x),g(x) f(x),g(x)对 x 0 x_0 x0有直到 n n n阶导数,则 ( f g ) ( n ) = ∑ k = 0 n C n k f ( k ) g ( n − k ) (fg)^{(n)} = \sum_{k=0}^{n}{C_n^k f^{(k)}g^{(n-k)}} (fg)(n)=k=0∑nCnkf(k)g(n−k)
证:
用数学归纳法证明:首先 n = 1 n=1 n=1时结论是成立的。
假设 n = m n=m n=m时结论是成立的。即如果 f ( x ) , g ( x ) f(x),g(x) f(x),g(x)对 x 0 x_0 x0有直到 m m m阶导数,则 ( f g ) ( m ) = ∑ k = 0 m C m k f ( k ) g ( m − k ) (fg)^{(m)} = \sum_{k=0}^{m}{C_m^k f^{(k)}g^{(m-k)}} (fg)(m)=k=0∑mCmkf(k)g(m−k)假设 f ( x ) , g ( x ) f(x),g(x) f(x),g(x)对 x 0 x_0 x0有直到 m + 1 m+1 m+1阶导数,那么 ( f g ) ( m ) = ∑ k = 0 m C m k f ( k ) g ( m − k ) (fg)^{(m)} = \sum_{k=0}^{m}{C_m^k f^{(k)}g^{(m-k)}} (fg)(m)=k=0∑mCmkf(k)g(m−k)于是有 ( f g ) ( m + 1 ) = ( ( f g ) ( m ) ) ′ = ∑ k = 0 m C m k ( f ( k + 1 ) g ( m − k ) + f ( k ) g ( m − k + 1 ) ) = ∑ k = 1 m + 1 C m k − 1 f ( k ) g ( m − k + 1 ) + ∑ k = 0 m C m k f ( k ) g ( m − k + 1 ) = C m 0 f ( 0 ) g ( m + 1 ) + C m m f ( m + 1 ) g ( 0 ) + ∑ k = 1 m ( C m k + C m k − 1 ) f ( k ) g ( m − k + 1 ) = C m + 1 0 f ( 0 ) g ( m + 1 ) + C m + 1 m + 1 f ( m + 1 ) g ( 0 ) + ∑ k = 1 m C m + 1 k f ( k ) g ( m − k + 1 ) (fg)^{(m+1)}=((fg)^{(m)})^{\prime} =\sum_{k=0}^{m}{C_m^k (f^{(k+1)}g^{(m-k)} + f^{(k)}g^{(m-k+1)})}\\ =\sum_{k=1}^{m+1}{C_m^{k-1} f^{(k)}g^{(m-k+1)}} +\sum_{k=0}^{m}{C_m^k f^{(k)}g^{(m-k+1)}}\\ =C_m^0 f^{(0)}g^{(m+1)}+C_m^m f^{(m+1)}g^{(0)} +\sum_{k=1}^{m}{(C_m^k+C_m^{k-1}) f^{(k)}g^{(m-k+1)}}\\ =C_{m+1}^0 f^{(0)}g^{(m+1)}+C_{m+1}^{m+1} f^{(m+1)}g^{(0)} +\sum_{k=1}^{m}{C_{m+1}^k f^{(k)}g^{(m-k+1)}} (fg)(m+1)=((fg)(m))′=k=0∑mCmk(f(k+1)g(m−k)+f(k)g(m−k+1))=k=1∑m+1Cmk−1f(k)g(m−k+1)+k=0∑mCmkf(k)g(m−k+1)=Cm0f(0)g(m+1)+Cmmf(m+1)g(0)+k=1∑m(Cmk+Cmk−1)f(k)g(m−k+1)=Cm+10f(0)g(m+1)+Cm+1m+1f(m+1)g(0)+k=1∑mCm+1kf(k)g(m−k+1)由数学归纳法,对任意的 n ≥ 1 n\ge 1 n≥1等式都成立。
例4.7的公式称为莱布尼兹公式。另外,由导数的线性性质,高阶导数也是线性运算。
例4.8 由归纳法同样可以证明: ( cos x ) ( k ) = cos ( x + k π 2 ) (\cos{x})^{(k)}=\cos{(x+\frac{k\pi}{2})} (cosx)(k)=cos(x+2kπ) ( sin x ) ( k ) = sin ( x + k π 2 ) (\sin{x})^{(k)}=\sin{(x+\frac{k\pi}{2})} (sinx)(k)=sin(x+2kπ)
例4.9 求 arctan x \arctan{x} arctanx的 n n n阶导数
解:令 y ( x ) = arctan x y(x)=\arctan{x} y(x)=arctanx,由反函数的求导法则,有 y ( 1 ) ( x ) = cos 2 y ( x ) y^{(1)}(x)=\cos^2{y(x)} y(1)(x)=cos2y(x)由复合函数求导法则,再求二阶导: y ( 2 ) ( x ) = − 2 sin y ( x ) cos 3 y ( x ) = − cos 2 y ( x ) sin 2 y ( x ) y^{(2)}(x) =-2\sin{y(x)}\cos^3{y(x)}=-\cos^2{y(x)}\sin{2y(x)} y(2)(x)=−2siny(x)cos3y(x)=−cos2y(x)sin2y(x)再求三阶导 y ( 3 ) ( x ) = − y ′ ( x ) [ − 2 cos y ( x ) sin y ( x ) sin 2 y ( x ) + 2 cos 2 y ( x ) cos 2 y ( x ) ] = − 2 cos y ( x ) y ′ ( x ) cos 3 y ( x ) = − 2 cos 3 y ( x ) cos 3 y ( x ) y^{(3)}(x) =-y^{\prime}(x) [-2\cos{y(x)}\sin{y(x)}\sin{2y(x)}+2\cos^2{y(x)}\cos{2y(x)}]\\ =-2\cos{y(x)}y^{\prime}(x)\cos{3y(x)} =-2\cos^3{y(x)}\cos{3y(x)} y(3)(x)=−y′(x)[−2cosy(x)siny(x)sin2y(x)+2cos2y(x)cos2y(x)]=−2cosy(x)y′(x)cos3y(x)=−2cos3y(x)cos3y(x)再求多一阶导就可以发现规律 y ( 4 ) ( x ) = − 2 y ′ ( x ) [ − 3 cos 2 y ( x ) sin y ( x ) cos 3 y ( x ) − 3 cos 3 y ( x ) sin 3 y ( x ) ] = 6 cos 4 y ( x ) sin 4 y ( x ) y^{(4)}(x) =-2y^{\prime}(x) [-3\cos^2{y(x)}\sin{y(x)}\cos{3y(x)}-3\cos^3{y(x)}\sin{3y(x)}]\\ =6\cos^4{y(x)}\sin{4y(x)} y(4)(x)=−2y′(x)[−3cos2y(x)siny(x)cos3y(x)−3cos3y(x)sin3y(x)]=6cos4y(x)sin4y(x)
猜想: y ( k ) ( x ) = ( k − 1 ) ! cos k y ( x ) sin ( k y ( x ) + k π 2 ) y^{(k)}(x)=(k-1)!\cos^k{y(x)}\sin{(ky(x)+\frac{k\pi}{2})} y(k)(x)=(k−1)!cosky(x)sin(ky(x)+2kπ)再用数学归纳法证明即可。
接下来我们转为讨论闭区间上连续,开区间上可导的函数,首先,我们要给出一个引理。
引理4.1 f ( x ) f(x) f(x)在 x 0 x_0 x0处取得极大值(极小值),并且在 x 0 x_0 x0处导数存在,则 f ′ ( x 0 ) = 0 f^{\prime}(x_0)=0 f′(x0)=0
证:
由 f ( x ) f(x) f(x)在 x 0 x_0 x0处取得极大值,存在一个 x 0 x_0 x0的邻域 B ( x 0 , δ ) B(x_0,\delta) B(x0,δ),对任意的 x ∈ B ( x 0 , δ ) x\in B(x_0,\delta) x∈B(x0,δ),有 f ( x ) − f ( x 0 ) ≤ 0 f(x)-f(x_0)\le 0 f(x)−f(x0)≤0当 x > x 0 x> x_0 x>x0时, f ( x ) − f ( x 0 ) x − x 0 ≤ 0 \frac{f(x)-f(x_0)}{x-x_0} \le 0 x−x0f(x)−f(x0)≤0从而 lim x → x 0 + f ( x ) − f ( x 0 ) x − x 0 = f ′ ( x 0 ) ≤ 0 \lim_{x\to x_0^+}{\frac{f(x)-f(x_0)}{x-x_0}} =f^{\prime}(x_0) \le 0 x→x0+limx−x0f(x)−f(x0)=f′(x0)≤0同理,有 lim x → x 0 − f ( x ) − f ( x 0 ) x − x 0 = f ′ ( x 0 ) ≥ 0 \lim_{x\to x_0^-}{\frac{f(x)-f(x_0)}{x-x_0}} =f^{\prime}(x_0) \ge 0 x→x0−limx−x0f(x)−f(x0)=f′(x0)≥0因此, f ′ ( x 0 ) = 0 f^{\prime}(x_0)=0 f′(x0)=0
定理4.7(罗尔定理) f ( x ) f(x) f(x)在闭区间 [ a , b ] [a,b] [a,b]上连续,开区间 ( a , b ) (a,b) (a,b)上可导,如果 f ( a ) = f ( b ) f(a)=f(b) f(a)=f(b),则存在 ξ ∈ ( a , b ) \xi \in (a,b) ξ∈(a,b), f ′ ( ξ ) = 0 f^{\prime}(\xi)=0 f′(ξ)=0
证:
如果 ∀ x ∈ ( a , b ) \forall x \in (a,b) ∀x∈(a,b), f ( x ) = f ( a ) = f ( b ) f(x)=f(a)=f(b) f(x)=f(a)=f(b),那么 f ( x ) f(x) f(x)在 ( a , b ) (a,b) (a,b)上是常数函数,那么结论显然是成立
设 M M M是 f ( x ) f(x) f(x)在 [ a , b ] [a,b] [a,b]上的最大值, m m m是 f ( x ) f(x) f(x)在 [ a , b ] [a,b] [a,b]上的最小值,那么不妨设 m < M mm<M , M = f ( x 1 ) , m = f ( x 2 ) M=f(x_1),m=f(x_2) M=f(x1),m=f(x2),同时, x 1 , x 2 x_1,x_2 x1,x2至少有一个不为端点,设 x 1 x_1 x1不为端点,那么, x 1 x_1 x1是 f ( x ) f(x) f(x)的一个极值点,那么 x 1 x_1 x1就满足条件
我们可以从图中直观地感受罗尔定理:
由图示,如果 [ a , b ] [a,b] [a,b]上的连续函数在两边是相等的,那么,最值一定在中间 ( a , b ) (a,b) (a,b)取得,而最值点导数就为0,将罗尔定理“旋转”一下,就得到拉格朗日中值定理。
定理4.8 f ( x ) f(x) f(x)在 [ a , b ] [a,b] [a,b]上连续, ( a , b ) (a,b) (a,b)上可导,则存在 ξ ∈ ( a , b ) \xi \in (a,b) ξ∈(a,b), f ′ ( ξ ) = f ( b ) − f ( a ) b − a f^{\prime}(\xi) = \frac{f(b)-f(a)}{b-a} f′(ξ)=b−af(b)−f(a)
在证明之前,我们先直观地看拉格朗日中值定理的几何意义。
f ( b ) − f ( a ) b − a \frac{f(b)-f(a)}{b-a} b−af(b)−f(a)是过 A ( a , f ( a ) ) , B ( b , f ( b ) ) A(a,f(a)),B(b,f(b)) A(a,f(a)),B(b,f(b))两点直线的斜率,实际上,将坐标轴旋转,使得 x x x轴与该直线平行,在这个角度看 f ( x ) f(x) f(x), f ( a ) = f ( b ) f(a)=f(b) f(a)=f(b),因此,我们可以认为拉格朗日中值定理是在另一个坐标轴视角下的罗尔定理。
证:
令 F ( x ) = f ( x ) − f ( a ) − f ( b ) − f ( a ) b − a ( x − a ) F(x)=f(x)-f(a)-\frac{f(b)-f(a)}{b-a} (x-a) F(x)=f(x)−f(a)−b−af(b)−f(a)(x−a), F ( b ) = F ( a ) = 0 F(b)=F(a)=0 F(b)=F(a)=0,由罗尔定理,存在 ξ ∈ ( a , b ) \xi \in (a,b) ξ∈(a,b), F ′ ( ξ ) = f ′ ( ξ ) − f ( b ) − f ( a ) b − a = 0 F^{\prime}(\xi)=f^{\prime}(\xi)-\frac{f(b)-f(a)}{b-a}=0 F′(ξ)=f′(ξ)−b−af(b)−f(a)=0
为了介绍柯西中值定理,我们首先介绍函数的参数方程形式,实际上,对于一条曲线 y = f ( x ) y=f(x) y=f(x),除了以函数形式表示该曲线,还可以令 { x = t y = f ( t ) \begin{cases} x=t\\ y=f(t) \end{cases} {x=ty=f(t)来表示该条曲线。更一般地, f ( t ) , g ( t ) f(t),g(t) f(t),g(t)是 [ a , b ] [a,b] [a,b]上的连续, ( a , b ) (a,b) (a,b)上可导的函数, { x = f ( t ) y = g ( t ) \begin{cases} x=f(t)\\ y=g(t) \end{cases} {x=f(t)y=g(t)就表示平面上的一条曲线,进一步地,我们要求 f ( t ) f(t) f(t)的导数不为0,那么 f ( t ) f(t) f(t)的导数只能恒为正或恒为负。这可由达布中值定理证明。
定理4.9 f ( x ) f(x) f(x)在 [ a , b ] [a,b] [a,b]上连续, ( a , b ) (a,b) (a,b)上可导,如果存在两点 x 1 , x 2 ∈ ( a , b ) x_1,x_2\in (a,b) x1,x2∈(a,b), f ′ ( x 1 ) = A < f ′ ( x 2 ) = B f^{\prime}(x_1)=A
证:
不妨设 x 1 < x 2 x_1x1<x2 ,将 f ( x ) f(x) f(x)限制在 [ x 1 , x 2 ] [x_1,x_2] [x1,x2]之间,令
g ( x ) = f ( x ) − τ x g(x)=f(x)-\tau x g(x)=f(x)−τx则 g ′ ( x 1 ) < 0 , g ′ ( x 2 ) > 0 g^{\prime}(x_1)<0,g^{\prime}(x_2)>0 g′(x1)<0,g′(x2)>0由极限的局部保号性, ∃ δ 1 > 0 \exists \delta_1>0 ∃δ1>0, x 1 < x < x 1 + δ 1 x_1x1<x<x1+δ1 时 g ( x ) − g ( x 1 ) x − x 1 < 0 \frac{g(x)-g(x_1)} {x-x_1}<0 x−x1g(x)−g(x1)<0从而, g ( x ) < g ( x 1 ) g(x)g(x)<g(x1) , x 1 x_1 x1不是 g ( x ) g(x) g(x)在 [ x 1 , x 2 ] [x_1,x_2] [x1,x2]上的最小值点。同样 x 2 x_2 x2也不是 g ( x ) g(x) g(x)在 [ x 1 , x 2 ] [x_1,x_2] [x1,x2]上的最小值点。设 ξ \xi ξ是 g ( x ) g(x) g(x)在 [ x 1 , x 2 ] [x_1,x_2] [x1,x2]上的最小值点,则 x 1 < ξ < x 2 x_1<\xix1<ξ<x2 ,则
g ′ ( ξ ) = f ′ ( ξ ) − τ = 0 g^{\prime}(\xi)=f^{\prime}(\xi)-\tau=0 g′(ξ)=f′(ξ)−τ=0
不妨设 f ′ ( x ) > 0 ( x ∈ ( a , b ) ) f^{\prime}(x)>0(x\in(a,b)) f′(x)>0(x∈(a,b)),由拉格朗日中值定理, f ( x ) f(x) f(x)在 [ a , b ] [a,b] [a,b]上严格单调上升,这样 f ( t ) f(t) f(t)就存在反函数, t = f − 1 ( x ) t=f^{-1}(x) t=f−1(x),这样,就有 y = g ( f − 1 ( x ) ) y=g(f^{-1}(x)) y=g(f−1(x)),就把参数方程化为函数形式,由复合函数求导法则: d y d x = f − 1 ′ ( x ) g ′ ( t ) = 1 f ′ ( t ) g ′ ( t ) = g ′ ( t ) f ′ ( t ) \frac{dy}{dx}=f^{-1\prime}(x)g^{\prime}(t) =\frac{1}{f^{\prime}(t)}g^{\prime}(t) =\frac{g^{\prime}(t)}{f^{\prime}(t)} dxdy=f−1′(x)g′(t)=f′(t)1g′(t)=f′(t)g′(t)过参数方程曲线两端的直线斜率应当为 g ( b ) − g ( a ) f ( b ) − f ( a ) \frac{g(b)-g(a)}{f(b)-f(a)} f(b)−f(a)g(b)−g(a)按拉格朗日中值定理的观点,应当存在 ξ ∈ ( a , b ) \xi\in(a,b) ξ∈(a,b),满足
g ( b ) − g ( a ) f ( b ) − f ( a ) = g ′ ( ξ ) f ′ ( ξ ) \frac{g(b)-g(a)}{f(b)-f(a)} = \frac{g^\prime(\xi)}{f^\prime(\xi)} f(b)−f(a)g(b)−g(a)=f′(ξ)g′(ξ)这就是柯西中值定理。
定理4.10 f ( t ) , g ( t ) f(t),g(t) f(t),g(t)在 [ a , b ] [a,b] [a,b]上连续, ( a , b ) (a,b) (a,b)上可导,并且 f ( t ) f(t) f(t)导数不为0,则存在 ξ ∈ ( a , b ) \xi\in(a,b) ξ∈(a,b),满足 g ( b ) − g ( a ) f ( b ) − f ( a ) = g ′ ( ξ ) f ′ ( ξ ) \frac{g(b)-g(a)}{f(b)-f(a)} = \frac{g^\prime(\xi)}{f^\prime(\xi)} f(b)−f(a)g(b)−g(a)=f′(ξ)g′(ξ)
证:
令 F ( x ) = g ( x ) − g ( a ) − g ( b ) − g ( a ) f ( b ) − f ( a ) ( f ( x ) − f ( a ) ) F(x) = g(x)-g(a)-\frac{g(b)-g(a)}{f(b)-f(a)}(f(x)-f(a)) F(x)=g(x)−g(a)−f(b)−f(a)g(b)−g(a)(f(x)−f(a)), F ( b ) = F ( a ) = 0 F(b)=F(a)=0 F(b)=F(a)=0,在应用罗尔定理即可证得结论
柯西中值定理提供了一种计算极限的简化方式。
定理4.11(洛必达法则1) f ( x ) , g ( x ) f(x),g(x) f(x),g(x)在 x 0 x_0 x0的某个右半去心邻域(左半去心邻域)上可导,并且满足:
(1) lim x → x 0 + f ( x ) = lim x → x 0 + g ( x ) = 0 \lim_{x\to x_0^+}{f(x)}=\lim_{x\to x_0^+}{g(x)} = 0 limx→x0+f(x)=limx→x0+g(x)=0( lim x → x 0 − f ( x ) = lim x → x 0 − g ( x ) = 0 \lim_{x\to x_0^-}{f(x)}=\lim_{x\to x_0^-}{g(x)} = 0 limx→x0−f(x)=limx→x0−g(x)=0)
(2) g ( x ) g(x) g(x)在该邻域内导数恒不为0
(3) lim x → x 0 + f ′ ( x ) g ′ ( x ) = A \lim_{x\to x_0^+}{\frac{f^\prime(x)}{g^\prime(x)}}=A limx→x0+g′(x)f′(x)=A( lim x → x 0 − f ′ ( x ) g ′ ( x ) = A \lim_{x\to x_0^-}{\frac{f^\prime(x)}{g^\prime(x)}}=A limx→x0−g′(x)f′(x)=A)
则 lim x → x 0 + f ( x ) g ( x ) = A \lim_{x\to x_0^+}{\frac{f(x)}{g(x)}}=A limx→x0+g(x)f(x)=A( lim x → x 0 − f ( x ) g ( x ) = A \lim_{x\to x_0^-}{\frac{f(x)}{g(x)}}=A limx→x0−g(x)f(x)=A)
证:
仅证右极限情形,左极限是类似的,补充定义 f ( x 0 ) = g ( x 0 ) = 0 f(x_0)=g(x_0)=0 f(x0)=g(x0)=0,对该邻域内的一点 x x x,有 f ( x ) g ( x ) = f ( x ) − f ( x 0 ) g ( x ) − g ( x 0 ) = f ′ ( ξ ) g ′ ( ξ ) ( ξ ∈ ( x 0 , x ) ) \frac{f(x)}{g(x)} =\frac{f(x)-f(x_0)}{g(x)-g(x_0)} =\frac{f^{\prime}(\xi)}{g^{\prime}(\xi)} (\xi \in (x_0,x)) g(x)f(x)=g(x)−g(x0)f(x)−f(x0)=g′(ξ)f′(ξ)(ξ∈(x0,x))令 x 0 → x x_0\to x x0→x即可
对 x → ∞ x\to\infty x→∞,也有类似的结论。
定理4.12(洛必达法则2) f ( x ) , g ( x ) f(x),g(x) f(x),g(x)在 ( a , + ∞ ) ( ( − ∞ , a ) ) (a,+\infty)((-\infty,a)) (a,+∞)((−∞,a))可导,并且满足:
(1) lim x → + ∞ f ( x ) = lim x → + ∞ g ( x ) = 0 ( lim x → − ∞ f ( x ) = lim x → − ∞ g ( x ) = 0 ) \lim_{x\to +\infty}{f(x)}=\lim_{x\to +\infty}{g(x)}=0(\lim_{x\to -\infty}{f(x)}=\lim_{x\to -\infty}{g(x)}=0) limx→+∞f(x)=limx→+∞g(x)=0(limx→−∞f(x)=limx→−∞g(x)=0)
(2) g ( x ) g(x) g(x)在 ( a , + ∞ ) ( ( − ∞ , a ) ) (a,+\infty)((-\infty,a)) (a,+∞)((−∞,a))上导数恒不为0
(3) lim x → + ∞ f ′ ( x ) g ′ ( x ) = A \lim_{x\to +\infty}{\frac{f^\prime(x)}{g^\prime(x)}}=A limx→+∞g′(x)f′(x)=A( lim x → − ∞ f ′ ( x ) g ′ ( x ) = A \lim_{x\to -\infty}{\frac{f^\prime(x)}{g^\prime(x)}}=A limx→−∞g′(x)f′(x)=A)
则 lim x → + ∞ f ( x ) g ( x ) = A \lim_{x\to +\infty}{\frac{f(x)}{g(x)}}=A limx→+∞g(x)f(x)=A( lim x → − ∞ f ( x ) g ( x ) = A \lim_{x\to-\infty}{\frac{f(x)}{g(x)}}=A limx→−∞g(x)f(x)=A)
证:
令 F ( t ) = f ( 1 t ) , G ( t ) = g ( 1 t ) F(t)=f(\frac{1}{t}),G(t)=g(\frac{1}{t}) F(t)=f(t1),G(t)=g(t1),补充定义 F ( 0 ) = G ( 0 ) = 0 F(0)=G(0)=0 F(0)=G(0)=0,则 lim t → 0 + F ( t ) G ( t ) = lim t → 0 + F ′ ( t ) G ′ ( t ) = lim t → 0 + f ′ ( 1 t ) ( − 1 t 2 ) g ′ ( 1 t ) ( − 1 t 2 ) = lim t → 0 + f ′ ( 1 t ) g ′ ( 1 t ) = A \lim_{t\to 0^+}{F(t)}{G(t)}= \lim_{t\to 0^+}{F^{\prime}(t)}{G^{\prime}(t)} =\lim_{t\to 0^+}{ \frac{ f^\prime(\frac{1}{t}) (-\frac{1}{t^2}) } { g^\prime(\frac{1}{t}) (-\frac{1}{t^2}) } } =\lim_{t\to 0^+}{ \frac{ f^\prime(\frac{1}{t}) } { g^\prime(\frac{1}{t}) } } =A t→0+limF(t)G(t)=t→0+limF′(t)G′(t)=t→0+limg′(t1)(−t21)f′(t1)(−t21)=t→0+limg′(t1)f′(t1)=A
对 ∞ / ∞ \infty/\infty ∞/∞型极限,也有相应的洛必达法则
定理4.13(洛必达法则3) f ( x ) , g ( x ) f(x),g(x) f(x),g(x)在 x 0 x_0 x0的某个右半去心邻域(左半去心邻域)上可导,并且满足:
(1) lim x → x 0 + f ( x ) = lim x → x 0 + g ( x ) = ∞ \lim_{x\to x_0^+}{f(x)}=\lim_{x\to x_0^+}{g(x)} = \infty limx→x0+f(x)=limx→x0+g(x)=∞( lim x → x 0 − f ( x ) = lim x → x 0 − g ( x ) = ∞ \lim_{x\to x_0^-}{f(x)}=\lim_{x\to x_0^-}{g(x)} = \infty limx→x0−f(x)=limx→x0−g(x)=∞)
(2) g ( x ) g(x) g(x)在该邻域内导数恒不为0
(3) lim x → x 0 + f ′ ( x ) g ′ ( x ) = A \lim_{x\to x_0^+}{\frac{f^\prime(x)}{g^\prime(x)}}=A limx→x0+g′(x)f′(x)=A( lim x → x 0 − f ′ ( x ) g ′ ( x ) = A \lim_{x\to x_0^-}{\frac{f^\prime(x)}{g^\prime(x)}}=A limx→x0−g′(x)f′(x)=A)
则 lim x → x 0 + f ( x ) g ( x ) = A \lim_{x\to x_0^+}{\frac{f(x)}{g(x)}}=A limx→x0+g(x)f(x)=A ( lim x → x 0 − f ( x ) g ( x ) = A \lim_{x\to x_0^-}{\frac{f(x)}{g(x)}}=A limx→x0−g(x)f(x)=A)
定理4.14(洛必达法则4) f ( x ) , g ( x ) f(x),g(x) f(x),g(x)在 ( a , + ∞ ) ( ( − ∞ , a ) ) (a,+\infty)((-\infty,a)) (a,+∞)((−∞,a))可导,并且满足:
(1) lim x → + ∞ f ( x ) = lim x → + ∞ g ( x ) = ∞ ( lim x → − ∞ f ( x ) = lim x → − ∞ g ( x ) = ∞ ) \lim_{x\to +\infty}{f(x)}=\lim_{x\to +\infty}{g(x)}=\infty(\lim_{x\to -\infty}{f(x)}=\lim_{x\to -\infty}{g(x)}=\infty) limx→+∞f(x)=limx→+∞g(x)=∞(limx→−∞f(x)=limx→−∞g(x)=∞)
(2) g ( x ) g(x) g(x)在 ( a , + ∞ ) ( ( − ∞ , a ) ) (a,+\infty)((-\infty,a)) (a,+∞)((−∞,a))上导数恒不为0
(3) lim x → + ∞ f ′ ( x ) g ′ ( x ) = A \lim_{x\to +\infty}{\frac{f^\prime(x)}{g^\prime(x)}}=A limx→+∞g′(x)f′(x)=A ( lim x → − ∞ f ′ ( x ) g ′ ( x ) = A \lim_{x\to -\infty}{\frac{f^\prime(x)}{g^\prime(x)}}=A limx→−∞g′(x)f′(x)=A)
则 lim x → + ∞ f ( x ) g ( x ) = A \lim_{x\to +\infty}{\frac{f(x)}{g(x)}}=A limx→+∞g(x)f(x)=A (