定义
微分的基本概念:
\begin{align*} \Delta y &= f(x_0 + \Delta x) - f(x_0) \quad \text{(函数增量)} \\ \Delta y &= A \Delta x + o(\Delta x) \quad \text{(线性近似+高阶无穷小)} \end{align*}微分表达式:
\begin{align*} dy = y'_x(x)dx = f'(u)g'(x)dx \quad \text{(链式法则微分形式)} \end{align*}复合函数的微分法则
乘法法则的微分形式:
\begin{align*} d(uv) &= (uv)'dx \\[5pt] \text{∵}\quad (uv)' &= u'v + v'u \quad \text{(乘积导数法则)} \\[5pt] u'dx &= du \\[5pt] v'dx &= dv \\[5pt] \text{∴}\quad d(uv) &=(u'v+v'u)dx \\ &=u'vdx+v'udx \\ &= vdu + udv \quad \text{(微分乘积法则)} \\ \end{align*}复合函数示例:
\begin{align*} y &= \sin(2x+1) \\ dy &= d(\sin u) = \cos(2x+1)d(2x+1) \\ &= \cos(2x+1) \cdot 2dx = 2\cos(2x+1)dx \text{(先外函数后内函数)} \end{align*}近似计算原理
线性近似公式:
\begin{align*} \text{∵} \Delta y=f(x_0 + \Delta x) - f(x_0)&\approx dy = f'(x_0)\Delta x \quad \text{(用切线近似函数)} \\[5pt] \text{∴} f(x_0 + \Delta x) \approx f(x_0) + f'(x_0) \Delta x \\[5pt] for \ x=(x_0 + \Delta x): \\[5pt] f(x) &\approx f(x_0) + f'(x_0)(x-x_0) \quad \text{(泰勒展开一阶项)} \end{align*}常用近似公式(当|x|很小时):
-
n次方根:
\begin{align*} \sqrt[n]{1+x} \approx f(0) + f'(0)x (x - 0) =1 +(\frac{1}{n})(x) = 1 + \frac{1}{n}x \end{align*}如 $$\sqrt{1.02} \approx 1.01$$
-
正弦函数:
\begin{align*} \sin x \approx f(0) + f'(0)(x - 0) = 0 + \cos(0)x = x \end{align*}小角度时正弦值≈弧度值
-
正切函数:
\begin{align*} \tan x &= \frac{\sin x}{\cos x} \\[5pt] \text{∵}\quad \tan'x &= \frac{\cos^2x + \sin^2x}{\cos^2x} = \frac{1}{\cos^2x} = \sec^2x \\[5pt] \text{∴}\quad \tan x &\approx f(0) + f'(0)(x - 0) = 0 + 1 \cdot x = x \end{align*}小角度时正切值≈弧度值
-
指数函数:
\begin{align*} e^x \approx f(0) + f'(0)(x - 0) = 1 + x \end{align*}如 $$e^{0.01} \approx 1.01$$
-
对数函数:
\begin{align*} \ln(1+x) \approx f(0) + f'(0)(x - 0) = 0 + 1 \cdot x = x \end{align*}如 $$\ln(1.02) \approx 0.02$$