← 返回资讯屏轩教育 · 行业资讯

Differentiation of Functions of Two Variables | 二元函数的微分

2026-08-03 02:46 · 本文部分内容由AI辅助创作

📚 Differentiation of Functions of Two Variables | 二元函数的微分

In single-variable calculus, differentiation gives the rate of change of a function with respect to one independent variable. When a function depends on two independent variables, such as temperature across a metal plate or gravitational potential on a surface, we need to analyse how it changes with each variable individually and how these changes combine. This leads to the concept of partial derivatives, the total differential, the chain rule, directional derivatives, gradients, and optimisation of functions of two variables. These tools form the foundation of multivariable calculus and are essential in IB Mathematics Analysis and Approaches at Higher Level.

在单变量微积分中,微分描述函数关于一个自变量的变化率。当函数依赖于两个自变量时——例如金属板上的温度分布或曲面上的引力势——我们需要分析它如何随每个变量单独变化,以及这些变化如何结合。由此引出了偏导数、全微分、链式法则、方向导数、梯度以及二元函数的优化等概念。这些工具构成了多元微积分的基础,也是IB数学分析与方法高水平课程中的关键内容。

1. Partial Derivatives – Definition | 偏导数的定义

by differentiating f with respect to x while treating y as constant. Similarly, ∂f/∂y or fy found by differentiating with respect to y, treating x as constant. Formally, ∂f/∂x = limh→0 usly for ∂f/∂y. For example, if f(x, y) = x²y + y³, then ∂f/∂x = 2xy and ∂f/∂y = x² + 3y².

考虑函数 z = f(x, y)。关于 x 的偏导数,记作 ∂f/∂x 或 fₓ,是将 y 视为常数时对 f 关于 x 的微分得到。同理,∂f/∂y 或 fy 是将 x 视为常数时对 f 关于 y 的微分。正式定义为 ∂f/∂x = limh→0 [f(x+h, y) - f(x, y)] / h,∂f/∂y 类似。例如,若 f(x, y) = x²y + y³,则 ∂f/∂x = 2xy,∂f/∂y = x² + 3y²。


2. Geometric Interpretation | 几何解释

e x = a. Thus, partial derivatives measure the instantaneous rate of change along directions parallel to the coordinate axes.

z = f(x, y) 的图像是三维空间中的一个曲面。在点 (a, b) 处的偏导数 ∂f/∂x 表示用平面 y = b 截割曲面所得曲线在该点的切线斜率。类似地,∂f/∂y 表示用平面 x = a 截割曲面所得曲线的切线斜率。因此,偏导数衡量了沿平行于坐标轴方向的瞬时变化率。


3. Higher-Order Partial Derivatives | 高阶偏导数

n y). For f(x, y) = x²y + y³, we have ∂²f/∂x² = 2y, ∂²f/∂y² = 6y, ∂²f/∂x∂y = 2x, and ∂²f/∂y∂x = 2x.

我们可以对偏导数再次微分,得到二阶偏导数。其中包括 ∂²f/∂x²(对 x 微分两次)、∂²f/∂y²(对 y 微分两次)以及混合偏导数 ∂²f/∂x∂y(先对 y 再对 x)和 ∂²f/∂y∂x(先对 x 再对 y)。对于 f(x, y) = x²y + y³,∂²f/∂x² = 2y,∂²f/∂y² = 6y,∂²f/∂x∂y = 2x,∂²f/∂y∂x = 2x。


4. Symmetry of Mixed Partials | 混合偏导数的对称性

es computations and appears in many physical laws, such as in thermodynamics and elasticity.

在二阶混合偏导数在某个区域上连续的条件下,克莱罗定理指出求导顺序无关紧要:∂²f/∂x∂y = ∂²f/∂y∂x。这种对称性简化了计算,并在许多物理定律中出现,如热力学和弹性力学。


5. The Total Differential | 全微分

and dy. If the function is differentiable, the linear approximation Δz ≈ fₓ(x₀,y₀) Δx + fy from 1 to 1.01 and y from 2 to 2.02, we use dz = 2xy dx + x² dy = 4·0.01 + 1·0.02 = 0.06.

z = f(x, y) 的全微分定义为 dz = (∂f/∂x) dx + (∂f/∂y) dy。它提供了当 x 和 y 有微小增量 dx 和 dy 时 z 的近似变化量。若函数可微,则线性近似 Δz ≈ fₓ(x₀,y₀) Δx + fy(x₀,y₀) Δy 在 (x₀,y₀) 附近成立。例如,估计 f(x,y) = x²y 在 x 从1变至1.01、y从2变至2.02时的变化,可用 dz = 2xy dx + x² dy = 4·0.01 + 1·0.02 = 0.06。


广告 / Sponsored

6. The Chain Rule | 链式法则

(∂x/∂u) + (∂f/∂y)(∂y/∂u) and similarly for v. This rule is essential for handling related rates and implicit differentiation in multiple variables.

当 x 和 y 本身是单变量 t 的函数时,f(x(t), y(t)) 对 t 的导数由链式法则给出:df/dt = (∂f/∂x)(dx/dt) + (∂f/∂y)(dy/dt)。若 x 和 y 依赖于两个变量 u 和 v,则 f 对 u 和 v 的偏导数为 ∂f/∂u = (∂f/∂x)(∂x/∂u) + (∂f/∂y)(∂y/∂u),对 v 类似。该法则在处理相关变化率与多元隐函数求导时至关重要。


7. Directional Derivatives | 方向导数

oint in the direction of a unit vector u = (a, b), we compute the directional derivative Du f = ∂f/∂x a + ∂f/∂y b = ∇ ection on the surface. If f is differentiable at the point, this linear combination holds.

偏导数仅给出沿坐标轴方向的变化率。要得到 f 在某点沿单位向量 u = (a, b) 方向的变化率,我们计算方向导数 Du f = ∂f/∂x a + ∂f/∂y b = ∇f · u。它表示曲面在该方向上的切线斜率。如果函数在该点可微,这一线性组合便成立。


8. The Gradient Vector | 梯度向量

The gradient of f is the vector ∇ ∂f/∂y). It points in the direction of the maximum rate of increase of f, and its magnitude |∇f| h always points radially outward from the origin and is orthogonal to circles x² + y² = c.

函数 f 的梯度是向量 ∇f = (∂f/∂x, ∂f/∂y)。它指向 f 增加最快的方向,其大小 |∇f| 给出该最大变化率。梯度与等高线 f(x, y) = c 垂直。例如,f(x, y) = x² + y² 的梯度为 (2x, 2y),总是从原点径向指向外,并与圆 x² + y² = c 正交。


9. Extrema of Functions of Two Variables | 二元函数的极值

is determined by the second derivative test using the discriminant D = (∂²f/∂x²)(∂²f/∂y²) – (∂²f/∂x∂y)².

要寻找 f(x, y) 的局部极大值与极小值,首先找出所有偏导数同时为零的临界点:∂f/∂x = 0 且 ∂f/∂y = 0。这些点可能是极值点或鞍点。在局部极大值处曲面出现峰,在局部极小值处出现谷。临界点的性质由二阶判别式 D = (∂²f/∂x²)(∂²f/∂y²) – (∂²f/∂x∂y)² 来判定。


10. Saddle Points and the Second Derivative Test | 鞍点与二阶判别式

um. If D > 0 and ∂²f/∂x² < 0, it is a local maximum. If D < 0, the point is a saddle point – t (0,0) is 0, requiring further analysis; however, a slight shift reveals saddle behaviour.

在临界点 (x₀, y₀) 处计算 D。若 D > 0 且 ∂²f/∂x² > 0,该点为局部极小值点。若 D > 0 且 ∂²f/∂x² < 0,则为局部极大值点。若 D < 0,则为鞍点——曲面在一个方向向上弯曲,在另一方向向下弯曲。若 D = 0,判别法无效。例如 f(x, y) = x³ - 3xy²,在 (0,0) 处 ∂f/∂x = 3x² - 3y²,∂f/∂y = -6xy,二者为零;D = (6x)(-6x) - (-6y)² = -36x² -36y²,在 (0,0) 处为0,需进一步分析;不过轻微移动即可揭示鞍点特征。


Published by TutorHao | Mathematics Revision Series | aleveler.com

广告 / Sponsored
IBDP国际文凭计算机科学学生用书 高级及标准水平 Computer Science for the IB Diplo…
计算机 备考教辅 · 正版现货 · 与本文内容相关
查看详情 / View on Taobao参考价 ¥524.0

更多咨询请联系16621398022(同微信)

备考资讯每日更新,关注不错过
考纲变化 · 科目分析 · 学习方法
扫码关注公众号 tutorhao
在微信里长按二维码识别关注
或截图后用「扫一扫 → 相册」选择本图
公众号:tutorhao