📚 The Gradient Vector in Three Dimensions | 三维空间中的梯度向量
In multivariable calculus, extending the concept of the derivative to functions of three variables leads to the gradient vector. This powerful tool captures both the direction and rate of steepest ascent of a scalar field at any point, and it serves as a foundation for tangent planes, directional derivatives, and optimization in three-dimensional space. Understanding the gradient in 3D is essential for studies in physics, engineering, and advanced mathematics, where quantities vary smoothly through space.
在多变量微积分中，将导数的概念推广到三元函数就产生了梯度向量。这个强大的工具能够捕捉标量场在任意点上最陡上升的方向和变化率，也是研究三维空间中切平面、方向导数和最优化问题的基础。理解三维空间中的梯度，对于在物理、工程和高等数学中处理随空间平滑变化的物理量而言不可或缺。
1. Introduction to the Gradient in 3D | 三维空间中梯度的简介
For a function of two variables, the gradient was a vector in the xy-plane indicating the direction of maximum increase. In three dimensions, we consider a scalar field f(x, y, z) that assigns a real number to each point in space. The gradient of f becomes a three-dimensional vector field, pointing to where f grows most rapidly and with its magnitude equal to that maximum rate of change. This extension is vital for modeling temperature distributions, potential energy fields, and other spatial phenomena.
对于二元函数，梯度是位于 xy 平面内的一个向量，指示函数值增加最快的方向。在三维情形下，我们考虑一个标量场 f(x, y, z)，它为空间中的每一个点赋予一个实数。f 的梯度就成为一个三维向量场，指向 f 增长最迅速的方向，其大小等于该最大变化率。这一推广对于模拟温度分布、势能场以及其他空间现象至关重要。
2. Partial Derivatives in Three Variables | 三个变量的偏导数
re computed by treating the other two variables as constants. These derivatives measure the instantaneous rate of change of f along the coordinate axes. For a differentiable function, all three partial derivatives exist at a point and allow us to construct the gradient.
为了定义梯度，我们必须先计算 f 对 x、y 和 z 的一阶偏导数。偏导数 ∂f/∂x 在给定点处是通过固定 y 和 z，再对 x 求导得到的。类似地，∂f/∂y 和 ∂f/∂z 则是将另外两个变量视为常数后求导。这些偏导数衡量了 f 沿坐标轴方向的瞬时变化率。对于一个可微函数，在一点处这三个偏导数都存在，从而允许我们构造梯度。
3. Definition of the Gradient Vector | 梯度向量的定义
The gradient of a scalar field f(x, y, z) is denoted by ∇f (pronounced "del f") and is defined as the vector whose components are the partial derivatives:
∇f = (∂f/∂x, ∂f/∂y, ∂f/∂z) = (∂f/∂x) i + (∂f/∂y) j + (∂f/∂z) k
标量场 f(x, y, z) 的梯度记作 ∇f（读作“del f”），定义为以各偏导数为分量的向量：
∇f = (∂f/∂x, ∂f/∂y, ∂f/∂z) = (∂f/∂x) i + (∂f/∂y) j + (∂f/∂z) k
Here i, j, and k are the standard unit vectors along the x-, y-, and z-axes. The symbol ∇ ("nabla") acts as a vector differential operator: ∇ ient is effectively the operator applied to the scalar function f. It produces a vector field from a scalar field.
这里 i、j、k 分别为沿 x、y、z 轴的标准单位向量。符号 ∇（"nabla"）充当一个向量微分算子：∇ = i ∂/∂x + j ∂/∂y + k ∂/∂z。因此，梯度实际上就是该算子作用于标量函数 f 的结果。它将一个标量场变成一个向量场。
4. Geometric Interpretation: Direction of Steepest Ascent | 几何意义：最速上升方向
direction of a unit vector u, the rate of change of f is the directional derivative Dᵤf = ∇f · u. This dot product is maximised when u is parallel to ∇f, giving the maximum rate of change equal to |∇f|. Therefore, ∇f itself indicates the path of steepest ascent, while the opposite vector −∇f shows the direction of steepest descent.
函数在点 P 处的梯度向量指向 f 增加最迅速的方向。更确切地说，如果我们沿单位向量 u 的方向移动一小段距离 ds，f 的变化率就是方向导数 Dᵤf = ∇f · u。当 u 平行于 ∇f 时，这个点积取最大值，该最大变化率等于 |∇f|。因此，∇f 本身指出了最速上升的路径，而相反的向量 −∇f 则表示最速下降的方向。
In contour maps of a hill, the gradient would be perpendicular to contour lines and point uphill. In three dimensions, level surfaces replace contour lines, and the gradient maintains its perpendicularity to these surfaces.
在小山的等高线图中，梯度将垂直于等高线并指向上坡。在三维空间中，等值面取代了等高线，而梯度仍与这些曲面保持垂直关系。
5. Gradient and Level Surfaces | 梯度与等值面
oints where f(x, y, z) = c for a constant c. For a differentiable function f, the gradient ∇ ted at a point on the level surface is orthogonal to the surface at that point. This means ∇ his geometric property is central to finding equations of tangent planes and normal lines.
函数 f(x, y, z) 的等值面是满足 f(x, y, z) = c（c 为常数）的所有点的集合。对于可微函数 f，在等值面上任意一点处的梯度 ∇f 都与该点处的曲面正交。这意味着 ∇f 充当了等值面切平面的法向量。这一几何性质对于求切平面和法线方程至关重要。
z(t)) lying on the level surface, so f(r(t)) = c. Differentiating with respect to t gives ∇f · r'(t) = 0, showing ∇ ar to any tangent vector to the surface. This orthogonality is a hallmark of the gradient.
要理解其原因，考虑一条位于等值面上的曲线 r(t) = (x(t), y(t), z(t))，则有 f(r(t)) = c。对 t 求导得出 ∇f · r'(t) = 0，这表明 ∇f 垂直于曲面的任意切向量。这种正交性是梯度的一大特征。
6. Directional Derivative in 3D | 三维中的方向导数
point (x₀, y₀, z₀) in the direction of a unit vector u = (u₁, u₂, u₃) is defined by Dᵤf = ∇ vatives along the coordinate axes. If u is not a unit vector, it must first be normalised.
函数在点 (x₀, y₀, z₀) 处沿单位向量 u = (u₁, u₂, u₃) 的方向导数 Dᵤf 定义为 Dᵤf = ∇f(x₀, y₀, z₀) · u。它给出了沿 u 方向通过该点时 f 的瞬时变化率。这推广了偏导数——它们仅仅是沿坐标轴方向的方向导数。如果 u 不是单位向量，则必须先将其归一化。
The maximum value of Dᵤf is |∇f|, attained when u = ∇f / |∇f|; the minimum value is −|∇f|, when u = −∇f / |∇f| Thus, the gradient not only gives the direction of steepest ascent but also its magnitude.
Dᵤf 的最大值为 |∇f|，当 u = ∇f / |∇f| 时取得；最小值为 −|∇f|，当 u = −∇f / |∇f| 时取得。因此，梯度不仅给出了最速上升的方向，还给出了其大小。
7. Gradient and Tangent Planes | 梯度与切平面
F(x, y, z) = c, the tangent plane at a point (x₀, y₀, z₀) on the surface has normal vector ∇F(x₀, y₀, z₀). The equation of the tangent plane is:
Fₓ(x₀, y₀, z₀)(x − x₀) + Fᵧ(x₀, y₀, z₀)(y − y₀) + Fz(x₀, y₀, z₀)(z − z₀) = 0
对于由隐函数 F(x, y, z) = c 定义的等值面，曲面上一点 (x₀, y₀, z₀) 处的切平面具有法向量 ∇F(x₀, y₀, z₀)。切平面方程为：
Fₓ(x₀, y₀, z₀)(x − x₀) + Fᵧ(x₀, y₀, z₀)(y − y₀) + Fz(x₀, y₀, z₀)(z − z₀) = 0
e it as F(x, y, z) = g(x, y) − z = 0, and the gradient gives a normal vector (gₓ, gᵧ, −1).
这里 Fₓ 表示 ∂F/∂x，对 y 和 z 类似。这种平面近似可用于局部线性化函数。如果曲面以显函数形式 z = g(x, y) 给出，我们可以将其改写为 F(x, y, z) = g(x, y) − z = 0，此时梯度给出的法向量为 (gₓ, gᵧ, −1)。
8. Properties of the Gradient | 梯度的性质
The gradient shares several algebraic properties analogous to those of ordinary derivatives, making it invaluable for computation and theoretical work.
梯度具有若干类似于普通导数的代数性质，这使其在计算和理论研究中极为有用。
Linearity: ∇(af + bg) = a ∇f + b ∇g for constants a, b.
线性性：∇(af + bg) = a ∇f + b ∇g，其中 a、b 为常数。
Product Rule: ∇(fg) = f ∇g + g ∇f.
乘法法则：∇(fg) = f ∇g + g ∇f。
Quotient Rule: ∇(f/g) = (g ∇f − f ∇g)/g² (g ≠ 0).
除法法则：∇(f/g) = (g ∇f − f ∇g)/g²（g ≠ 0）。
Chain Rule: If h(t) = f(r(t)), then dh/dt = ∇f(r(t)) · r'(t). This is a special case of the total derivative.
链式法则：如果 h(t) = f(r(t))，则 dh/dt = ∇f(r(t)) · r'(t)。这是全导数的一种特殊情况。
These rules can be verified directly from the definition of ∇ and the properties of partial derivatives. They greatly simplify the differentiation of composite expressions.
这些法则可以通过 ∇ 的定义和偏导数的性质直接验证。它们极大地简化了复合表达式的求导。
9. Example Calculations | 实例计算
Let f(x, y, z) = x²y + yz³ + xz. Compute its gradient at the point (1, −1, 2).
设 f(x, y, z) = x²y + yz³ + xz。计算它在点 (1, −1, 2) 处的梯度。
First find the partial derivatives: ∂f/∂x = 2xy + z, ∂f/∂y = x² + z³, ∂f/∂z = 3yz² + x.
首先求出偏导数：∂f/∂x = 2xy + z，∂f/∂y = x² + z³，∂f/∂z = 3yz² + x。
2 = 0; ∂f/∂y = 1² + 2³ = 1 + 8 = 9; ∂f/∂z = 3(−1)(2²) + 1 = 3(−1)(4) + 1 = −12 + 1 = −11.
在 (1, −1, 2) 处求值：∂f/∂x = 2(1)(−1) + 2 = −2 + 2 = 0；∂f/∂y = 1² + 2³ = 1 + 8 = 9；∂f/∂z = 3(−1)(2²) + 1 = 3(−1)(4) + 1 = −12 + 1 = −11。
Thus, ∇f(1, −1, 2) = (0, 9, −11). The maximum rate of increase at this point is |∇f| = √(0² + 9² + (−11)²) = √81 + 121 of steepest ascent is along the vector (0, 9, −11), or normalised: (0, 9/√202, −11/√202).
因此，∇f(1, −1, 2) = (0, 9, −11)。该点的最大增长率为 |∇f| = √(0² + 9² + (−11)²) = √81 + 121 = √202 ≈ 14.21。最速上升方向沿向量 (0, 9, −11)，归一化后为 (0, 9/√202, −11/√202)。
The table below shows gradients of some common three-variable functions.
下表展示了一些常见三元函数的梯度。
| Function f(x, y, z) | Gradient ∇f |
|---|---|
| ax + by + cz | (a, b, c) |
| x² + y² + z² | (2x, 2y, 2z) |
| xyz | (yz, xz, xy) |
| ex+2y−z | (ex+2y−z, 2ex+2y−z, −ex+2y−z) |
| sin(x)cos(y)z | (cos(x)cos(y)z, −sin(x)sin(y)z, sin(x)cos(y)) |
10. Applications: Optimization and Normal Vectors | 应用：最优化与法向量
multipliers. To find extrema of f(x, y, z) subject to a constraint g(x, y, z) = c, we set ∇f = λ ∇ re parallel. This yields a system of equations that can be solved for the critical points.
在带约束的最优化问题中，梯度通过拉格朗日乘数法发挥了关键作用。为了在约束 g(x, y, z) = c 下求 f(x, y, z) 的极值，我们令 ∇f = λ ∇g，其中 λ 是一个标量乘数。该条件表明，在约束曲面上的极值点处，f 和 g 的梯度平行。由此得到方程组，可解出临界点。
way to compute unit normal vectors. For the surface F(x, y, z) = c, the unit normal is n = ∇F / |∇F|. This is heavily used in surface integrals and flux calculations in vector calculus.
此外，由于梯度垂直于等值面，它提供了一种直接计算单位法向量的方法。对于曲面 F(x, y, z) = c，单位法向量为 n = ∇F / |∇F|。这在向量积分中的面积分和通量计算中被大量使用。
11. The Gradient in Physical Contexts | 物理背景中的梯度
radient appears in many contexts. If f represents the temperature at each point in a room, ∇ tter regions, and heat flows against this gradient (from hot to cold) according to Fourier' n. In electrostatics, the electric field E is related to the electric potential V by E = −∇ icular to equipotential surfaces. In fluid dynamics, the gradient of pressure drives flow.
在物理学中，梯度出现在许多情境里。如果 f 表示房间内每一点的温度，∇f 就是一个从低温区指向高温区的向量场，而热量则根据傅里叶热传导定律逆着这个梯度（从高温流向低温）传递。在静电学中，电场 E 与电势 V 的关系为 E = −∇V；电场线垂直于等势面。在流体动力学中，压力梯度驱动流动。
rvative vector field is also tied to the gradient: a vector field F is conservative if F = ∇ s. This is a direct extension of the fundamental theorem of calculus to higher dimensions.
保守向量场的概念也与梯度紧密相连：一个向量场 F 是保守的，如果存在某个标量势 φ 使得 F = ∇φ。此时，F 的线积分与路径无关，做功仅取决于端点。这是微积分基本定理在高维空间的直接推广。
12. Summary and Key Points | 总结与关键点
ons is a central concept in multivariable calculus. It is defined as the vector of first-order partial derivatives, ∇ ion indicates the path of steepest ascent; its magnitude equals the maximum directional derivative. The gradient is orthogonal to level surfaces, providing a normal vector for tangent planes. Its algebraic properties mirror those of single-variable derivatives, and it is fundamental to optimisation, physics, and vector analysis.
三维空间中的梯度向量是多变量微积分的核心概念。它定义为一阶偏导数组成的向量，∇f = (∂f/∂x, ∂f/∂y, ∂f/∂z)。其方向指示最速上升的路径，其大小等于最大的方向导数。梯度与等值面正交，从而为切平面提供法向量。它的代数性质与单变量导数类似，并且是最优化、物理学和向量分析的基础。
Mastery of the gradient in 3D requires practice in computing partial derivatives, interpreting geometric relationships, and applying the concept to tangent planes and directional derivatives. Always remember that ∇f transforms a scalar field into a vector field that encapsulates local change information at every point in space.
掌握三维空间中的梯度需要多加练习偏导数的计算，理解其几何关系，并将其应用于切平面和方向导数。请始终牢记，∇f 将标量场转化为向量场，后者囊括了空间每一点处的局部变化信息。
Published by TutorHao | Mathematics Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
或截图后用「扫一扫 → 相册」选择本图