4. 微分
4.4 复合函数求导法则及其应用
【例4.4.3】 y = e 1 + cos x y=e^{\sqrt{1+\cos x}} y=e1+cosx,求 y ′ y' y′
【解】 y ′ = e 1 + cos x ⋅ 1 2 1 + cos x ⋅ ( − sin x ) = − sin x 2 1 + cos x e 1 + cos x y'=e^{\sqrt{1+\cos x}}\cdot\frac{1}{2\sqrt{1+\cos x}}\cdot(-\sin x)=-\frac{\sin x}{2\sqrt{1+\cos x}}e^{\sqrt{1+\cos x}} y′=e1+cosx⋅21+cosx1⋅(−sinx)=−21+cosxsinxe1+cosx
4.4.2 幂指函数求导法则
y = f ( x ) = u ( x ) v ( x ) y=f(x)=u(x)^{v(x)} y=f(x)=u(x)v(x)
两边取对数得 ln y = ln f ( x ) = v ( x ) ln u ( x ) \ln y=\ln f(x)=v(x)\ln u(x) lny=lnf(x)=v(x)lnu(x)
等式两边同时对 x x x求导得 1 y y ′ = 1 u ( x ) v ( x ) y ′ = v ′ ( x ) ln u ( x ) + v ( x ) u ( x ) u ′ ( x ) \frac{1}{y}y'=\frac{1}{u(x)^{v(x)}}y'=v'(x)\ln u(x)+\frac{v(x)}{u(x)}u'(x) y1y′=u(x)v(x)1y′=v′(x)lnu(x)+u(x)v(x)u′(x)
y ′ ( x ) = u ( x ) v ( x ) ( v ′ ( x ) ln u ( x ) + v ( x ) u ′ ( x ) u ( x ) ) y'(x)=u(x)^{v(x)}(v'(x)\ln u(x)+\frac{v(x)u'(x)}{u(x)}) y′(x)=u(x)v(x)(v′(x)lnu(x)+u(x)v(x)u′(x))
【例】 y = ( sin x ) cos x y=(\sin x)^{\cos x} y=(sinx)cosx,求 y ′ y' y′
【解】 ln y = cos x ln ( sin x ) \ln y=\cos x\ln(\sin x) lny=cosxln(sinx)
y ′ y = − sin x ln ( sin x ) + cos x 1 sin x ⋅ cos x \frac{y'}{y}=-\sin x\ln(\sin x)+\cos x\frac{1}{\sin x}\cdot \cos x yy′=−sinxln(sinx)+cosxsinx1⋅cosx
y ′ = ( sin x ) cos x ( cos 2 x sin x − sin x ln ( sin x ) ) y'=(\sin x)^{\cos x}(\frac{\cos ^2 x}{\sin x}-\sin x\ln(\sin x)) y′=(sinx)cosx(sinxcos2x−sinxln(sinx))
4.4.3 导数运算法则和微分运算法则
表要记住
4.4.4 一阶微分形式不变性
只有一阶微分有形式不变性
- y = f ( u ) , y ′ ( u ) = f ′ ( u ) , d y = f ′ ( u ) d u , u y=f(u),y'(u)=f'(u),dy=f'(u)du,u y=f(u),y′(u)=f′(u),dy=f′(u)du,u是自变量;
- y = f ( u ) , u = g ( x ) , y = f ( g ( x ) ) , y ′ ( x ) = f ′ ( u ) g ′ ( x ) = f ′ ( g ( x ) ) g ′ ( x ) , d y = f ′ ( g ( x ) ) g ′ ( x ) d x = f ′ ( g ( x ) ) d g ( x ) = f ′ ( u ) d u , u y=f(u),u=g(x),y=f(g(x)),y'(x)=f'(u)g'(x)=f'(g(x))g'(x),dy = f'(g(x))g'(x)dx=f'(g(x))dg(x)=f'(u)du,u y=f(u),u=g(x),y=f(g(x)),y′(x)=f′(u)g′(x)=f′(g(x))g′(x),dy=f′(g(x))g′(x)dx=f′(g(x))dg(x)=f′(u)du,u是中间变量;
不管 u u u是自变量还是中间变量,都有一个式子成立即 d y = f ′ ( u ) d u dy=f'(u)du dy=f′(u)du,这就叫做一阶微分的形式不变性。
4.4.5 隐函数得求导与求微分
隐函数的表达式: F ( x , y ) = 0 F(x,y)=0 F(x,y)=0
【例】 x 2 a 2 + y 2 b 2 = 1 \frac{x^2}{a^2}+\frac{y^2}{b^2}=1 a2x2+b2y2=1,限定 y > 0 y>0 y>0就是上半椭圆, y < 0 y<0 y<0就是下半椭圆,求微分。
d ( x 2 a 2 + y 2 b 2 ) = 0 d(\frac{x^2}{a^2}+\frac{y^2}{b^2})=0 d(a2x2+b2y2)=0
即 1 a 2 2 x d x + 1 b 2 2 y d y = 0 \frac{1}{a^2}2xdx+\frac{1}{b^2}2ydy=0 a212xdx+b212ydy=0
亦即 d y = − b 2 a 2 ⋅ x y d x dy=-\frac{b^2}{a^2}\cdot\frac{x}{y}dx dy=−a2b2⋅yxdx
所以 d y d x = − b 2 a 2 ⋅ x y \frac{dy}{dx}=-\frac{b^2}{a^2}\cdot\frac{x}{y} dxdy=−a2b2⋅yx
【例4.4.5】 e x y + x 2 y − 1 = 0 e^{xy}+x^2y-1=0 exy+x2y−1=0,求 y ′ . y'. y′.
【解】它的显函数写不出来,用隐函数求导法则
左右两边对 x x x求导
d d x ( e x y + x 2 y − 1 ) = 0 \frac{d}{dx}(e^{xy}+x^2y-1)=0 dxd(exy+x2y−1)=0
则 e x y ( y + x y ′ ) + 2 x y + x 2 y ′ = 0 e^{xy}(y+xy')+2xy+x^2y'=0 exy(y+xy′)+2xy+x2y′=0
y ′ ( x ) = − 2 x y + y e x y x e x y + x 2 = − ( e x y + 2 x ) y ( e x y + x ) x y'(x)=-\frac{2xy+ye^{xy}}{xe^{xy}+x^2}=-\frac{\left(\mathrm{e}^{x y}+2 x\right) y}{\left(\mathrm{e}^{x y}+x\right) x} y′(x)=−xexy+x22xy+yexy=−(exy+x)x(exy+2x)y
【例4.4.6】 sin y 2 = cos x \sin y^2=\cos \sqrt{x} siny2=cosx,求 y ′ y' y′
【解】等式两边同时求微分得
2 y cos y 2 d y = − sin x ⋅ 1 2 x d x 2y\cos y^2 dy=-\sin\sqrt{x}\cdot\frac{1}{2\sqrt{x}}dx 2ycosy2dy=−sinx⋅2x1dx
由一阶微分的形式不变性
y ′ = d y d x = − sin x 4 y ( cos y 2 ) x y' =\frac{dy}{dx}=-\frac{\sin\sqrt{x}}{4y(\cos y^2) \sqrt{x}} y′=dxdy=−4y(cosy2)xsinx
【例4.4.7】 e x + y − x y − e = 0 e^{x+y}-xy-e=0 ex+y−xy−e=0几何上表示平面上一条曲线, ( 0 , 1 ) (0,1) (0,1)在曲线上,求过 ( 0 , 1 ) (0,1) (0,1)的切线方程。
【解】等式两边求导得 e x + y ( 1 + y ′ ) − y − x y ′ = 0 e^{x+y}(1+y')-y-xy'=0 ex+y(1+y′)−y−xy′=0
即 y ′ ( x ) = y − e x + y e x + y − x y'(x)=\frac{y-e^{x+y}}{e^{x+y}-x} y′(x)=ex+y−xy−ex+y
将 ( 0 , 1 ) (0,1) (0,1)代入 y ′ ( x ) y'(x) y′(x),则 y ′ ( 0 ) = 1 − e 1 e 1 − 0 = 1 − e e y'(0)=\frac{1-e^{1}}{e^{1}-0}=\frac{1-e}{e} y′(0)=e1−01−e1=e1−e
则切线方程为 y − 1 = 1 − e e x y-1=\frac{1-e}{e}x y−1=e1−ex
4.4.6 归纳
(1) y = 1 g ( x ) , y ′ = − g ′ ( x ) g 2 ( x ) y=\frac{1}{g(x)},y'=-\frac{g'(x)}{g^2(x)} y=g(x)1,y′=−g2(x)g′(x),也可以看成 y = 1 u , u = g ( x ) , y ′ = − 1 u 2 g ′ ( x ) = − g ′ ( x ) g 2 ( x ) y=\frac{1}{u},u=g(x),y'=-\frac{1}{u^2}g'(x)=-\frac{g'(x)}{g^2(x)} y=u1,u=g(x),y′=−u21g′(x)=−g2(x)g′(x)
(2) y = f ( x ) , x = f − 1 ( y ) , f − 1 ( f ( x ) ) = x y=f(x),x=f^{-1}(y),f^{-1}(f(x))=x y=f(x),x=f−1(y),f−1(f(x))=x
等式 f − 1 ( f ( x ) ) = x f^{-1}(f(x))=x f−1(f(x))=x两边对 x x x求导得 1 = ( f − 1 ( y ) ) ′ f ′ ( x ) 1=(f^{-1}(y))'f'(x) 1=(f−1(y))′f′(x),所以 f − 1 ( y ) ) ′ = 1 f ′ ( x ) f^{-1}(y))'=\frac{1}{f'(x)} f−1(y))′=f′(x)1
4.4.7 函数的参数表示(参数方程)求导
{ x = φ ( t ) , y = ψ ( t ) , α ⩽ t ⩽ β \left\{\begin{array}{l} x=\varphi(t), \\ y=\psi(t), \end{array} \quad \alpha \leqslant t \leqslant \beta\right. {x=φ(t),y=ψ(t),α⩽t⩽β且 φ , ψ \varphi,\psi φ,ψ都可导, φ \varphi φ严格单调且 φ ′ ( t ) ≠ 0 \varphi'(t)\ne 0 φ′(t)=0
由反函数的可导定理, t = φ − 1 ( x ) , y = ψ ( φ − 1 ( x ) ) t=\varphi^{-1}(x),y=\psi(\varphi^{-1}(x)) t=φ−1(x),y=ψ(φ−1(x))
d y d x = ψ ′ ( φ − 1 ( x ) ) ( φ − 1 ( x ) ) ′ = ψ ′ ( φ − 1 ( x ) ) φ ′ ( t ) = ψ ′ ( t ) φ ′ ( t ) \frac{dy}{dx}=\psi'(\varphi^{-1}(x))(\varphi^{-1}(x))'=\frac{\psi'(\varphi^{-1}(x))}{\varphi'(t)}=\frac{\psi'(t)}{\varphi'(t)} dxdy=ψ′(φ−1(x))(φ−1(x))′=φ′(t)ψ′(φ−1(x))=φ′(t)ψ′(t)
实际上也可以从微分公式(一阶微分形式的不变性)出发推出来
{ d x = φ ′ ( t ) d t , d y = ψ ′ ( t ) d t , \left\{\begin{array}{l} dx=\varphi'(t)dt, \\ dy=\psi'(t)dt, \end{array} \right. {dx=φ′(t)dt,dy=ψ′(t)dt,,即 d y d x = ψ ′ ( t ) φ ′ ( t ) \frac{dy}{dx}=\frac{\psi'(t)}{\varphi'(t)} dxdy=φ′(t)ψ′(t)
【例】【旋轮线(摆线)】 { x = t − sin t , y = 1 − cos t , 0 ⩽ t ⩽ π \left\{\begin{array}{l} x=t-\sin t, \\ y=1-\cos t, \end{array} \quad 0 \leqslant t \leqslant \pi\right. {x=t−sint,y=1−cost,0⩽t⩽π,求 d y d x \frac{dy}{dx} dxdy
【解】 d y d x = sin t 1 − cos t \frac{dy}{dx}=\frac{\sin t}{1-\cos t} dxdy=1−costsint