Basic Concepts of Multivariable Functions | 多元函数基本概念

📚 Basic Concepts of Multivariable Functions | 多元函数基本概念

A multivariable function is a rule that assigns a single real number to each ordered tuple of independent variables. This article covers the essential definitions, properties, and computational tools you need for the IB Mathematics syllabus, with an emphasis on clarity and exam accuracy.

多元函数是将多个自变量映射到一个实数因变量的规则。本文围绕 IB 数学课程要求，系统讲解多元函数的核心定义、性质与计算方法，注重清晰性与考试准确性。

1. Definition of Multivariable Functions | 多元函数定义

Formally, a function of \(n\) variables is written as \(f : D \subseteq \mathbb{R}^n \rightarrow \mathbb{R}\). For each point \((x_1, x_2, \dots, x_n)\) in the domain \(D\), there is a unique output \(f(x_1, x_2, \dots, x_n)\). When \(n=2\), we usually write \(z = f(x,y)\); when \(n=3\), we write \(w = f(x,y,z)\).

形式化地，n 元函数记作 \(f : D \subseteq \mathbb{R}^n \rightarrow \mathbb{R}\)。定义域 \(D\) 中每一点 \((x_1, x_2, \dots, x_n)\) 都有唯一的输出值。当 \(n=2\) 时，常记作 \(z = f(x,y)\)；当 \(n=3\) 时，记作 \(w = f(x,y,z)\)。

Example: \(f(x,y) = x^2 + y^2\) is a real-valued function of two variables.
示例：\(f(x,y) = x^2 + y^2\) 是一个二元实值函数。
Example: \(f(x,y,z) = e^{x+y}\sin z\) has three independent variables.
示例：\(f(x,y,z) = e^{x+y}\sin z\) 含有三个自变量。

2. Domain and Range | 定义域与值域

The domain of a multivariable function is the set of all valid input points for which the expression is defined. Common restrictions include denominators not equal to zero, expressions under square roots non-negative, and logarithm arguments strictly positive.

多元函数的定义域是所有使表达式有意义的输入点集合。常见限制包括分母不为零、偶次根号下非负、对数真数大于零等。

For example, \(f(x,y)=\sqrt{4-x^2-y^2}\) has domain \(x^2+y^2 \le 4\), which is the disk of radius 2 centered at the origin. Its range is \([0,2]\).

例如，\(f(x,y)=\sqrt{4-x^2-y^2}\) 的定义域满足 \(x^2+y^2 \le 4\)，即中心在原点、半径为 2 的圆盘；其值域为 \([0,2]\)。

Domain of f = { (x,y) ∈ ℝ² | condition satisfied }

When finding domains, identify all restrictions simultaneously and express the result using set-builder notation or inequalities.

求定义域时，要同时考虑所有限制条件，并用集合描述法或不等式表达结果。

3. Graphs and Level Curves | 图像与等高线

The graph of a two-variable function \(z=f(x,y)\) is a surface in three-dimensional space. To visualize it, we often draw level curves: sets of points in the xy-plane where \(f(x,y)=c\) for various constants \(c\).

二元函数 \(z=f(x,y)\) 的图像是三维空间中的一张曲面。为了可视化，常引入等高线（等值线）：在 xy 平面上满足 \(f(x,y)=c\) 的点的集合，其中 \(c\) 取不同常数。

For \(f(x,y)=x^2+y^2\), level curves are circles \(x^2+y^2=c\) for \(c>0\). For three-variable functions, we use level surfaces \(f(x,y,z)=c\), e.g., \(x^2+y^2+z^2=r^2\) gives spheres.

对于 \(f(x,y)=x^2+y^2\)，等高线是圆 \(x^2+y^2=c\)（\(c>0\)）。对于三元函数，则使用等值面 \(f(x,y,z)=c\)，如 \(x^2+y^2+z^2=r^2\) 表示球面。

Level curves never cross for a single well-defined function.
同一个多元函数的任意两条等高线不会相交。
Closely spaced level curves indicate a steep surface.
等高线越密集，表示曲面越陡峭。

4. Limits of Multivariable Functions | 多元函数的极限

We say \(\lim_{(x,y)\to(a,b)} f(x,y) = L\) if \(f(x,y)\) approaches \(L\) as \((x,y)\) approaches \((a,b)\) from every possible path. Unlike single-variable limits, there are infinitely many paths to a point; if two different paths yield different limits, the limit does not exist.

若无论 \((x,y)\) 沿任何路径趋向于 \((a,b)\)，函数值都趋近于 \(L\)，则称极限存在且等于 \(L\)。与一元极限不同，多元极限需要考虑无数条趋近路径；若两条不同路径给出不同极限，则极限不存在。

Example: \(f(x,y)=\frac{xy}{x^2+y^2}\). Along \(y=mx\),

f(x,mx) = \frac{mx^2}{x^2(1+m^2)} = \frac{m}{1+m^2}

This result depends on \(m\), so the limit at \((0,0)\) does not exist.

该结果随斜率 \(m\) 变化，因此在 \((0,0)\) 处极限不存在。

To prove a limit exists, use the squeeze theorem or convert to polar coordinates and show the expression is bounded by a term tending to zero.

要证明极限存在，可用夹逼定理或转化为极坐标，证明表达式被一个趋于零的量控制。

5. Continuity | 连续性

A function \(f(x,y)\) is continuous at \((a,b)\) if three conditions hold: \(f(a,b)\) is defined, the limit exists, and \(\lim_{(x,y)\to(a,b)} f(x,y) = f(a,b)\). Sums, products, quotients (where denominator nonzero), and compositions of continuous functions are continuous.

函数 \(f(x,y)\) 在点 \((a,b)\) 连续需满足三个条件：函数在该点有定义、极限存在、极限值等于函数值。连续函数做加法、乘法、除法（分母非零）以及复合后仍连续。

Polynomials in multiple variables are continuous everywhere. Rational functions are continuous except where the denominator vanishes.

多元多项式处处连续；有理函数在其分母不为零的点处连续。

Continuity requires: \lim_{(x,y)\to(a,b)} f(x,y) = f(a,b)

6. Partial Derivatives | 偏导数

The partial derivative of \(f(x,y)\) with respect to \(x\) at \((a,b)\) is the ordinary derivative treating \(y\) as constant:

函数 \(f(x,y)\) 对 \(x\) 的偏导数在 \((a,b)\) 处的定义为将 \(y\) 看作常数后求普通导数：

f_x(a,b)= \lim_{h\to0} \frac{f(a+h,b)-f(a,b)}{h}

Similarly \(f_y(a,b)= \lim_{h\to0} \frac{f(a,b+h)-f(a,b)}{h}\). Notation: \(f_x\), \(\partial f/\partial x\), or \(z_x\).

类似地 \(f_y(a,b)= \lim_{h\to0} \frac{f(a,b+h)-f(a,b)}{h}\)。记法有 \(f_x\)、\(\partial f/\partial x\) 或 \(z_x\)。

Higher-order partial derivatives are defined by differentiating again. For example, \(f_{xy}=(f_x)_y\) and \(f_{yx}=(f_y)_x\). Clairaut’s Theorem states that if \(f_{xy}\) and \(f_{yx}\) are both continuous on an open region, then they are equal.

高阶偏导数通过再次求导得到，如 \(f_{xy}=(f_x)_y\)，\(f_{yx}=(f_y)_x\)。克拉洛定理指出：若 \(f_{xy}\) 与 \(f_{yx}\) 在开区域内连续，则二者相等。

Compute \(f_x\) by differentiating with respect to \(x\), holding \(y\) fixed.
求 \(f_x\) 时固定 \(y\)，仅对 \(x\) 求导。
Use Clairaut’s theorem to check mixed partials quickly.
利用克拉洛定理可快速验证混合偏导数。

7. Tangent Planes and Differentiability | 切平面与可微性

If \(f\) has continuous partial derivatives near \((a,b)\), then the tangent plane to \(z=f(x,y)\) at \((a,b,f(a,b))\) is given by

若 \(f\) 在 \((a,b)\) 附近有连续偏导数，则在点 \((a,b,f(a,b))\) 处的切平面方程为

z = f(a,b) + f_x(a,b)(x-a) + f_y(a,b)(y-b)

This plane provides the best linear approximation to the surface near the point. The function is called differentiable if this linearization error goes to zero faster than the distance from \((a,b)\).

该平面给出了曲面在该点附近的最佳线性近似。若线性化误差比点到 \((a,b)\) 的距离更快趋于零，则称函数在该点可微。

Differentiability is stronger than having partial derivatives; partial derivatives must be continuous for differentiability to hold (sufficient condition).

可微比偏导数存在更强；偏导数连续是可微的充分条件。

8. Gradient and Directional Derivatives | 梯度与方向导数

The gradient of \(f(x,y)\) is the vector

函数 \(f(x,y)\) 的梯度为向量

∇f = (f_x, f_y)

For a unit vector \(\mathbf{u} = (u_1, u_2)\), the directional derivative of \(f\) in the direction of \(\mathbf{u}\) is

对于单位向量 \(\mathbf{u} = (u_1, u_2)\)，函数 \(f\) 沿方向 \(\mathbf{u}\) 的方向导数为

D_\mathbf{u} f = \nabla f \cdot \mathbf{u} = f_x u_1 + f_y u_2

The gradient vector points in the direction of the steepest increase, with magnitude equal to that maximum rate of change. Perpendicular to level curves, the gradient is a powerful tool for optimization and modeling.

梯度指向函数增长最快的方向，其大小等于最大增长率。梯度垂直于等高线，是优化与建模的重要工具。

Maximum rate of change: \(\|\nabla f\|\)
最大变化率：\(\|\nabla f\|\)
Zero directional derivative occurs when \(\mathbf{u}\) is orthogonal to \(\nabla f\).
当 \(\mathbf{u}\) 与梯度正交时，方向导数为零。

9. The Chain Rule | 链式法则

If \(z=f(x,y)\), and \(x=x(t), y=y(t)\), then the total derivative with respect to \(t\) is

若 \(z=f(x,y)\)，且 \(x=x(t), y=y(t)\)，则对 \(t\) 的全导数为

\frac{dz}{dt} = \frac{\partial f}{\partial x}\frac{dx}{dt} + \frac{\partial f}{\partial y}\frac{dy}{dt}

For \(z=f(x,y)\) with \(x=x(s,t)\), \(y=y(s,t)\), the partial derivatives with respect to \(s\) or \(t\) follow the same structure:

若 \(z=f(x,y)\)，且 \(x=x(s,t)\), \(y=y(s,t)\)，则对 \(s\) 或 \(t\) 的偏导数遵循相同结构：

\frac{\partial z}{\partial s} = \frac{\partial f}{\partial x}\frac{\partial x}{\partial s} + \frac{\partial f}{\partial y}\frac{\partial y}{\partial s}

This rule extends to more variables via dependency diagrams. It is essential for implicit differentiation: if \(F(x,y)=0\), then

该法则可通过依赖关系图推广到更多变量。它也用于隐函数求导：若 \(F(x,y)=0\)，则

\frac{dy}{dx} = -\frac{F_x}{F_y}

provided \(F_y \neq 0\).

其中要求 \(F_y \neq 0\)。

10. Exam Strategies and Common Mistakes | 考试策略与常见错误

In IB exams, carefully identify the variable you are differentiating with respect to and hold other variables fixed. For limits, always test multiple paths before concluding a limit exists.

在 IB 考试中，务必看清要对哪个变量求导，同时固定其他变量。对于极限，先尝试多条路径再断言极限存在。

Common mistakes include: forgetting domain restrictions, misusing notation \(f_x\) vs. \(f_{xy}\), assuming \(f_{xy}=f_{yx}\) without continuity, and treating partial derivatives as ordinary derivatives without chain rule adjustments.

常见错误有：忽略定义域限制、混淆 \(f_x\) 与 \(f_{xy}\) 的记号、未验证连续性就认为 \(f_{xy}=f_{yx}\)、以及不按链式法则而直接把偏导数当作普通导数。

Error 错误	Correction 正确
\(f_{xy}=f_{yx}\) always assumed 默认混合偏导一定相等	Check continuity first 先验证连续性
Use \(y=mx\) only for limit nonexistence 只用一条路径判断极限	Test multiple paths; for existence use squeeze or polar 多条路径；存在性需夹逼或极坐标
Forget \(y\) depends on \(t\) in chain rule 链式法则忘记 \(y\) 依赖 \(t\)	Draw a dependency diagram and sum all paths 画出依赖树并累加所有路径

Practice with functions like \(f(x,y)=e^{xy}\ln(1+x^2y)\) to improve computational fluency and prepare for exam questions.

多练习 \(f(x,y)=e^{xy}\ln(1+x^2y)\) 等函数，以提高计算熟练度并应对考试题目。

Published by TutorHao | Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Basic Concepts of Multivariable Functions | 多元函数基本概念