📚 Differentiation from First Principles | 基于第一原理的微分
Differentiation from first principles lies at the very heart of calculus. It moves beyond memorising derivative rules and asks the fundamental question: what happens to the average rate of change as the interval shrinks to zero? In IB Mathematics, this topic not only appears in exam questions but also builds a deep conceptual bridge between limits, gradients, and the behaviour of functions. Mastering first principles gives you a robust understanding that supports everything from optimisation to differential equations.
基于第一原理的微分是微积分最核心的内容。它不仅让你摆脱死记求导公式，更追问一个本质问题：当间隔趋近于零时，平均变化率会怎样变化？在IB数学中，这个主题既会直接出现在考题里，也为理解极限、斜率和函数整体性质搭建起一座深刻的概念桥梁。掌握好第一原理，你对最优化和微分方程等后续内容的理解都会变得格外扎实。
1. The Concept of a Derivative | 导数的概念
A derivative measures how a function changes as its input changes. Imagine walking along a curved path – your instantaneous speed at any moment is a derivative. In mathematical language, we begin with an average rate of change over an interval [x, x+h], then shrink that interval until the change happens at a single point. This process gives the gradient of the tangent, rather than the slope of a secant line connecting two points. The derivative is therefore a limit of average rates of change.
导数衡量的是函数随自变量变化而变化的快慢程度。设想沿着一条弯曲的小路行走，你在任意瞬间的速度其实就是一种导数。用数学语言来说，我们首先考察区间 [x, x+h] 上的平均变化率，然后让这个区间无限缩小，直到变化是在一个点附近发生。这个过程最终给出的是切线的斜率，而非两点间割线的坡度。因此，导数本质上是一系列平均变化率的极限。
2. Definition via Limits | 通过极限的定义
The crucial idea is to let h approach zero without ever letting h equal zero. We write limh→0 to capture this intention. A limit describes the value a function approaches, not necessarily the value it reaches. In the context of differentiation, we need the limit of the difference quotient (f(x+h) – f(x)) / h as h tends to 0. Without the limit concept, we would be forced to divide by zero, which is undefined. Limits give rigour to our intuition of “infinitesimally small” changes.
关键的想法是让 h 趋近于零，却从不允许 h 恰好等于零。我们用 limh→0 来表示这一意图。极限描述的是函数趋近的值，并不一定是它真正抵达的值。在微分情境下，我们需要的正是差商 (f(x+h)–f(x))/h 在 h 趋于 0 时的极限。如果没有极限这个工具，我们就会被迫面对除以零的困难，而那样毫无定义。极限赋予了我们“无穷小变化”这一直觉以严谨的数学基础。
3. First Principles Formula | 第一原理公式
The formal definition of the derivative from first principles is written as:
f'(x) = limh→0 [f(x+h) – f(x)] / h
Here f'(x) denotes the derivative of f with respect to x. The expression inside the limit is called the difference quotient. Notice the structure: we take the function's value at a slightly shifted input x+h, subtract the value at x, and then divide by the tiny horizontal step h. The limit compresses this average slope into the instantaneous slope. This definition works for any function f, provided the limit exists.
基于第一原理的导数正式定义为：
f'(x) = limh→0 [f(x+h) – f(x)] / h
其中 f'(x) 表示 f 关于 x 的导数。极限内部的分式称作差商。请注意它的结构：我们先取函数在稍稍偏移后的自变量 x+h 处的值，减去其在 x 处的值，再除以微小的水平步长 h。极限过程把这条平均坡线压缩成瞬时坡线。只要极限存在，这一定义对所有函数 f 都适用。
4. Worked Example: f(x) = x² | 计算实例：f(x) = x²
Let us apply the definition to the simple quadratic function f(x) = x2. First, compute f(x+h) = (x+h)2 = x2 + 2xh + h2. The difference f(x+h) – f(x) becomes (x2 + 2xh + h2) – x2 = 2xh + h2. Dividing by h gives (2xh + h2) / h = 2x + h, for h ≠ 0. Finally, we take the limit as h→0:
f'(x) = limh→0 (2x + h) = 2x
The term containing h vanishes, leaving only 2x. Thus, the derivative of x2 is 2x, which matches the power rule you already know.
我们把定义应用到简单的二次函数 f(x)=x2 上。首先求出 f(x+h) = (x+h)2 = x2 + 2xh + h2。差 f(x+h)–f(x) 变为 (x2 + 2xh + h2) – x2 = 2xh + h2。除以 h 得到 (2xh + h2)/h = 2x + h，其中 h ≠ 0。最后，令 h→0 取极限：
f'(x) = limh→0 (2x + h) = 2x
含有 h 的项趋于零，只留下 2x。因此，x2 的导数是 2x，与你早已熟悉的幂法则完全吻合。
5. Worked Example: f(x) = √x | 计算实例：f(x) = √x
Consider f(x) = √x, defined for x > 0. The difference quotient is (√x+h – √x) / h. To handle the square roots, we multiply numerator and denominator by the conjugate √x+h + √x. This rationalisation yields:
[(x+h) – x] / [h (√x+h + √x)] = h / [h (√x+h + √x)] = 1 / (√x+h + √x)
Now we safely cancel h and take the limit as h→0. Since √x+h → √x, the denominator approaches 2√x. Hence,
f'(x) = limh→0 1 / (√x+h + √x) = 1 / (2√x)
This result reveals the derivative of the square root function and illustrates why algebraic tricks like rationalisation are essential when radicals appear.
考虑 f(x) = √x，其定义域为 x > 0。差商为 (√x+h – √x) / h。为了处理根号，我们将分子分母同乘共轭表达式 √x+h + √x。通过有理化，式子化为：
[(x+h) – x] / [h (√x+h + √x)] = h / [h (√x+h + √x)] = 1 / (√x+h + √x)
此时我们可以安全地约去 h，并令 h→0 取极限。注意到 √x+h → √x，分母趋近于 2√x。因此，
f'(x) = limh→0 1 / (√x+h + √x) = 1 / (2√x)
这一结果给出了平方根函数的导数，也说明当遇到根式时，有理化这类代数技巧不可或缺。
6. Generalising to Other Functions | 推广到其他函数
similar limit process, we can find derivatives of many elementary functions. For f(x) = xn with integer n, the binomial theorem expands (x+h)n and the derivative emerges as n xn-1. For trigonometric functions like sin x, the first-principles limit requires the identity for sin(x+h) and the fundamental limits of (sin h)/h and (1-cos h)/h. The table below summarises some common derivatives obtained from first principles:
应用类似的极限过程，我们可以求出许多基本函数的导数。对 f(x) = xn（n 为整数），二项式定理展开 (x+h)n 后导数就会以 n xn-1 的形式出现。对于 sin x 等三角函数，基于第一原理的极限需要用到 sin(x+h) 的恒等式以及 (sin h)/h 与 (1-cos h)/h 的基本极限。下表总结了一些由第一原理推得的常见导数：
| f(x) | f'(x) from first principles |
|---|---|
| x² | 2x |
| x³ | 3x² |
| 1/x (x ≠ 0) | -1/x² |
| √x | 1/(2√x) |
| sin x | cos x |
Each entry confirms the power of the limit definition as the source of all derivative rules.
表格中的每一项都印证了极限定义作为所有求导法则源泉的强大威力。
7. The Relationship to Tangent Slope | 与切线斜率的关系
Graphically, the difference quotient (f(x+h)-f(x))/h gives the slope of the secant line joining the points (x, f(x)) and (x+h, f(x+h)). As h shrinks, the secant rotates and draws closer to the position of the tangent line at x. In the limit, the secant slope becomes the tangent slope f'(x). This geometric interpretation is why the derivative is often called the gradient function. Visualising this motion helps many students move from algebra to geometry in calculus.
从图形上看，差商 (f(x+h)-f(x))/h 给出的是连接点 (x, f(x)) 和 (x+h, f(x+h)) 的割线斜率。随着 h 缩小，割线发生旋转并不断逼近在 x 处的切线位置。在极限中，割线斜率就化作了切线斜率 f'(x)。正是基于这一几何解释，导数常被称为梯度函数。在心中想象这个动态过程，能帮助许多学生顺利地从微积分的代数层面过渡到几何直观。
8. Differentiability and Continuity | 可微性与连续性
continuity alone does not guarantee differentiability. The classic counterexample is f(x)=|x| ded limits differ, the two-sided limit does not exist, and the derivative is undefined at the cusp. First principles therefore provides a direct test for differentiability.
若函数在某点可导，则该点必然连续。但连续性本身并不能保证可导。最经典的反例是 f(x)=|x| 在 x=0 处的行为。该函数是连续的，但差商的左极限为 -1，右极限为 +1。由于两侧极限不相等，双侧极限不存在，导数在该尖点也就无定义。因此，第一原理本身就为判断可微性提供了一种直接的测试手段。
9. Common Mistakes | 常见错误
es, students often forget to apply the limit and stop at the simplified algebra, writing f' r is to treat f(x+h) as f(x)+f(h), which is false for almost all functions. With radicals, a failure to rationalise the numerator can trap the algebra in an indeterminant form. Additionally, handle the h only as a non-zero quantity while simplifying, but remember to take the limit at the end. Keeping these pitfalls in mind will improve both accuracy and confidence.
在使用第一原理推导时，学生常忘记取极限，在代数化简后就停下，写出 f'(x)=2x+h 而不是 2x。另一个错误是把 f(x+h) 视为 f(x)+f(h)，这对绝大多数函数都不成立。处理根式时，若不对分子进行有理化，代数过程就会陷入未定式而无法继续。此外，化简过程中要把 h 当作非零量来操作，最后再取极限。将这些易错点牢记在心，能显著提升准确率与自信心。
10. Exam-style Practice | 考试风格练习
irst principles, find the derivative of g(x) = 1/x, where x ≠ 0. Show all steps carefully.
Solution outline: Write the difference quotient (1/(x+h) – 1/x) / h. Combine the fractions in the numerator to obtain ((x – . Cancel h (since h≠0) to get -1 / (x(x+h)). Finally, take the limit as h→0 to arrive at g'(x) = -1 / x2. This matches the power rule for negative exponents and shows the consistency of the first-principles method.
你能试着完成下面这道典型的IB风格题目吗？使用第一原理求 g(x)=1/x (x≠0) 的导数，并仔细写出所有步骤。
解答要点：写出差商 (1/(x+h) – 1/x) / h。将分子中的分式合并，得到 ((x – (x+h)) / (x(x+h))) / h = (-h) / (h x(x+h))。约去 h（h≠0）得 -1 / (x(x+h))。最后令 h→0 取极限，得到 g'(x) = -1 / x2。这与负指数的幂法则一致，同时也体现了第一原理方法的内在统一性。
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