📚 Estimating Small Changes | 估算小变化量
In calculus, estimating small changes in a function' change in the function f(x), denoted by Δy, can be approximated by the differential dy = f' called linear approximation, is widely applied in science, economics, and error analysis.
在微积分中，利用导数快速估算函数值的小变化是一种非常实用的技巧。当自变量 x 发生微小变化 Δx 时，函数的相应变化量 Δy 可以近似地用微分 dy = f'(x) dx 表示。这种方法通常称为线性近似，在科学、经济学和误差分析中应用广泛。
1. Understanding Differentials | 理解微分
Δx) − f(x) is the exact change along the curve. For a differentiable function, the difference between Δy and dy becomes negligible as Δx approaches zero, making dy an excellent approximation for small dx.
微分 dy 表示在某点处沿切线的变化量，而实际变化量 Δy = f(x+Δx) − f(x) 是沿曲线的精确变化量。对于可微函数，当 Δx 趋近于零时，Δy 与 dy 之间的差异可以忽略不计，因此对于微小的 dx，dy 是一个极好的近似值。
Geometrically, the approximation replaces the curve by its tangent line near the point of interest. This is why it is also called tangent line approximation. The smaller the Δx, the closer the true function lies to its tangent, and the more accurate the estimate becomes.
从几何角度看，这种近似是在目标点附近用切线代替曲线。这也是它被称为切线近似的原因。Δx 越小，真实函数与其切线就越接近，估算也就越精确。
2. The Core Approximation Formula | 核心近似公式
The fundamental relationship for estimating small changes is Δy ≈ f'(x) Δx, where Δx = dx is the small increment in x, and f'(x) is the derivative evaluated at the starting point. We often rewrite it as f(x+Δx) ≈ f(x) + f'(x) Δx to directly obtain an approximate function value.
估算小变化的基本关系是 Δy ≈ f'(x) Δx，其中 Δx = dx 是 x 的微小增量，f'(x) 是在起始点计算的导数。我们经常将其改写为 f(x+Δx) ≈ f(x) + f'(x) Δx，以便直接得到函数的近似值。
This formula follows directly from the definition of the derivative: f' e limit and assuming a small but finite Δx, we obtain a simple linear model for the change. The approximation is most reliable when the function is locally nearly linear, i.e., when its second derivative is not excessively large.
该公式直接由导数的定义得出：f'(x) = lim (Δx→0) Δy / Δx。去掉极限并假设一个微小但有限的 Δx，我们就得到了一个简单的线性变化模型。当函数局部近乎线性，即二阶导数不过大时，这种近似最为可靠。
3. Step-by-Step Estimation Process | 逐步估算过程
To estimate Δy for a given function, follow these steps: (1) Identify the function f(x) and the starting x-value a. (2) Determine the small change Δx (or dx). (3) Compute the derivative f'(x). (4) Evaluate f'(a) at the starting point. (5) Multiply f'(a) by Δx to obtain the approximate change dy. (6) Approximate the new value as f(a) + dy. Always use the exact known base value f(a).
要估算给定函数的 Δy，请遵循以下步骤：(1) 确定函数 f(x) 和起始 x 值 a。(2) 确定微小变化量 Δx（或 dx）。(3) 计算导数 f'(x)。(4) 在起始点求出 f'(a)。(5) 用 f'(a) 乘以 Δx 得到近似变化量 dy。(6) 新值的近似值为 f(a) + dy。务必使用精确已知的基值 f(a)。
- Choose a base point a where f(a) and f'(a) are easy to compute.
- 选择基值点 a，使得 f(a) 和 f'(a) 易于计算。
- Keep Δx small relative to a; otherwise, the approximation degrades.
- 保持 Δx 相对于 a 较小；否则近似效果会变差。
- State the approximation clearly: f(a+Δx) ≈ f(a) + f'(a) Δx.
- 明确写出近似式：f(a+Δx) ≈ f(a) + f'(a) Δx。
4. Example 1: Estimating a Power Function | 例题1：估算幂函数
Estimate the change in y = x3 when x increases from 2 to 2.01. Here f(x)=x3, a=2, Δx=0.01. f'(x)=3x2, so f' 1 = 0.12. The approximate new value is f(2.01) ≈ 8 + 0.12 = 8.12. The exact change is 2.013 − 8 = 8.120601 − 8 = 0.120601, so the error is only 0.000601.
估计当 x 从 2 增加到 2.01 时 y = x3 的变化量。此处 f(x)=x3，a=2，Δx=0.01。导数 f'(x)=3x2，因此 f'(2)=3×4=12。则 Δy ≈ 12 × 0.01 = 0.12。近似新值为 f(2.01) ≈ 8 + 0.12 = 8.12。精确变化量为 2.013 − 8 = 8.120601 − 8 = 0.120601，误差仅为 0.000601。
This demonstrates the strength of linear approximation: for Δx as large as 0.01, the relative error is less than 0.5%. Even larger Δx values like 0.1 (estimating 2.13) would give Δy ≈ 12×0.1 = 1.2, while true is 9.261−8=1.261, with error about 5%, still reasonable for quick estimates.
这展示了线性近似的优势：即使 Δx 达到 0.01，相对误差也小于 0.5%。若用更大的 Δx 如 0.1（估算 2.13），Δy ≈ 12×0.1 = 1.2，而真实值为 9.261−8=1.261，误差约 5%，对于快速估算仍可接受。
5. Example 2: Square Root Approximation | 例题2：平方根近似
Find an approximate value for √16.5. Let f(x)=√x, a=16, Δx=0.5. f'(x)=1/(2√x), so f'(16)=1/(2×4)=0.125. Δy ≈ 0.125 × 0.5 = 0.0625. Then √16.5 ≈ 4 + 0.0625 = 4.0625. The true value is about 4.062019, giving a miniscule error.
求 √16.5 的近似值。设 f(x)=√x，a=16，Δx=0.5。导数 f'(x)=1/(2√x)，则 f'(16)=1/(2×4)=0.125。Δy ≈ 0.125 × 0.5 = 0.0625。因此 √16.5 ≈ 4 + 0.0625 = 4.0625。真实值约为 4.062019，误差极小。
Choosing a perfect square like 16 as the base point makes the arithmetic simple. The key is to pick an a where f and f' are easy to evaluate. Other common bases include multiples of π for trigonometric functions or integer powers for exponentials.
选择完全平方数 16 作为基值点能让计算变得简单。关键在于选取容易计算 f 和 f' 的 a 值。其他常见的基值点包括三角函数的 π 的倍数，或指数函数的整数次幂。
6. Percentage Change and Relative Error | 百分比变化与相对误差
e in y for a given percentage change in x. The approximate proportional change is Δy/y ≈ f'(x) Δx / f(x). As a percentage, multiply by 100% idely used in economics for price elasticity or in physics for estimating relative errors.
我们常常需要知道在 x 发生一定百分比变化时，y 的百分比变化。近似的比例变化为 Δy/y ≈ f'(x) Δx / f(x)。要换算为百分比，乘以 100% 即可。这广泛用于经济学中的价格弹性，或物理学中的相对误差估算。
For example, if the radius of a circle increases by 2%, what is the approximate percentage increase in area? A = πr2, dA/dr = 2πr, so ΔA/A ≈ (2πr Δr)/(πr2) = 2 Δr/r. A 2% increase in r gives approximately a 4% increase in area.
例如，圆的半径增加 2%，面积大约增加百分之几？A = πr2，dA/dr = 2πr，因此 ΔA/A ≈ (2πr Δr)/(πr2) = 2 Δr/r。半径增加 2% 会导致面积约增加 4%。
The relative error of the approximation |Δy − dy|/|Δy| ero as Δx → 0, but for larger Δx it can become significant, especially near points where f''(x) is large. Always check the size of the second derivative to gauge accuracy.
当 Δx → 0 时，近似的相对误差 |Δy − dy|/|Δy| 趋于零，但对于较大的 Δx，误差可能变得显著，尤其是在 f''(x) 较大的点附近。始终检查二阶导数的大小以判断精度。
7. Error Propagation in Measurements | 测量中的误差传播
uncertainty ±Δx. The resulting uncertainty in a calculated quantity f(x) is approximately |f'(x)| ropagation formula is essential for estimating maximum possible errors in derived results.
在实验科学中，测量值 x 可能具有不确定性 ±Δx。计算量 f(x) 中由此产生的不确定性约为 |f'(x)| Δx。这个线性误差传播公式对于估计导出结果的最大可能误差至关重要。
e side length of a cube is measured as 10 cm with a possible error of 0.1 cm. Volume V = s3, dV/ds = 3s2. So the maximum error in volume is approximately 3×102×0.1 = 30 cm3. If you need a more conservative bound, use the absolute value.
假设立方体的边长测量为 10 cm，可能误差为 0.1 cm。体积 V = s3，dV/ds = 3s2。因此体积的最大误差约为 3×102×0.1 = 30 cm3。如需更保守的界限，使用绝对值即可。
8. Chain Rule for Related Small Changes | 相关小变化的链式法则
bine their small changes. If y = f(u) and u = g(x), then a small change Δx produces Δy ≈ f'(u) g'(x) Δx. This mirrors the chain rule dy/dx = f'(u) g'(x) and helps in layered estimation problems.
当变量通过中间函数关联时，我们可以合并它们的小变化。若 y = f(u) 且 u = g(x)，则小变化 Δx 产生的 Δy ≈ f'(u) g'(x) Δx。这与链式法则 dy/dx = f'(u) g'(x) 相似，有助于解决多层次的估算问题。
elting ice cube shrinks at 0.2 cm/min. How fast does volume decrease when the edge is 5 cm? Let s=5, ds/dt=−0.2, V=s3. dV/dt = 3s2 ds/dt = 3×25×(−0.2) = −15 cm3/min. Over a small time Δt, the volume change is ΔV ≈ −15 Δt.
例题：融化的冰块边长以 0.2 cm/min 的速度缩小。当边长为 5 cm 时，体积减少的速度是多少？设 s=5，ds/dt=−0.2，V=s3。dV/dt = 3s2 ds/dt = 3×25
Published by TutorHao | IB 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