📚 Mastering Differentiation: Edexcel A-Level Mathematics | 掌握微分：爱德思A-Level数学
Differentiation is a cornerstone of A-Level Mathematics, and the Edexcel specification demands a deep conceptual understanding alongside fluent computational skill. From first principles to advanced applications like connected rates of change, this topic bridges pure algebra with the geometry of curves and real-world motion. In this article, we will systematically unpack every essential technique, common pitfalls, and exam-focused strategies to help you build confidence and achieve top marks.
微分是A-Level数学的基石，爱德思考试大纲要求在掌握流畅计算技能的同时具备深刻的概念理解。从第一性原理到关联变化率等高级应用，这一主题将纯代数与曲线的几何性质及现实运动联系起来。本文将会系统性地剖析每种核心技巧、常见误区以及以考试为焦点的策略，帮助你建立信心并取得高分。
1. The Gradient Function: From Secants to Tangents | 梯度函数：从割线到切线
The derivative f' hord connecting (x, f(x)) and (x+h, f(x+h)) has gradient (f(x+h) - f(x))/h. As h tends to 0, this ratio approaches the instantaneous gradient, provided the limit exists. This is the first principle: f'(x) = lim(h→0) [f(x+h) - f(x)]/h.
导数 f'(x) 给出曲线在某一瞬间的精确梯度，但它的起源在于两点之间的平均变化率。对于函数 y = f(x)，连接点 (x, f(x)) 和 (x+h, f(x+h)) 的弦的梯度为 (f(x+h) - f(x))/h。当 h 趋于 0 时，如果极限存在，该比值便趋近于瞬间梯度。这就是第一性原理：f'(x) = lim(h→0) [f(x+h) - f(x)]/h。
xh + h², subtract f(x) giving 2xh + h², divide by h to get 2x + h, and let h→0 to obtain f' reful expansion and cancellation. Always show all steps explicitly to secure method marks.
考试中可能会要求用这个极限定义来对 x² 或 1/x 这样的简单多项式求导。对 f(x) = x²，展开 (x+h)² = x² + 2xh + h²，减去 f(x) 得到 2xh + h²，除以 h 得 2x + h，再令 h→0 即得 f'(x) = 2x。代数过程直接，但需要仔细展开和约分。务必清晰地写全所有步骤以获取方法分。
Understanding the geometric interpretation is vital: the chord's gradient approaches the tangent's gradient as the two points merge. Sketching the chord and tangent on a diagram helps visualise why the limit must exist for differentiability. A function is not differentiable where it has a corner, cusp, or discontinuity — these are common Edexcel trap questions.
理解其几何意义至关重要：随着两点不断靠近，弦的梯度便趋近于切线的梯度。画出弦与切线的示意图有助于直观理解可微性为何要求极限存在。函数在尖角、尖点或不连续处均不可导——这些都是爱德思考试中常见的陷阱题。
2. Differentiation Rules: Power, Constant Multiple, Sum and Difference | 求导法则：幂法则、常数倍法则、和与差法则
imply bring the index down as a coefficient and reduce the power by one. Combine this with the constant multiple rule d/dx [c f(x)] = c f'(x) and the sum rule d/dx [u ± v] = u' ± v', and you can differentiate any polynomial term by term.
一旦建立了第一性原理的模式，幂法则 d/dx [xⁿ] = n xⁿ⁻¹ 便成为处理多项式的主力工具。对任意实数指数 n，只需将指数作为系数带到前方，并将指数减 1。结合常数倍法则 d/dx [c f(x)] = c f'(x) 与和差法则 d/dx [u ± v] = u' ± v'，你就可以逐项对任何多项式求导。
ative is 0 because the graph is a horizontal line with zero gradient. Another mistake is mishandling negative or fractional powers. Rewrite expressions like 1/x² as x⁻² and √x as x½ before differentiating. Thus d/dx [3/x²] = 3 * (-2) x⁻³ = -6/x³.
一个常见错误是忘记常数的导数为零。对 f(x)=5，导数为 0，因为其图像是一条斜率为零的水平线。另一个错误是处理负指数或分数指数时出错。先将 1/x² 改写为 x⁻²，√x 改写为 x½，然后再求导。因此 d/dx [3/x²] = 3 * (-2) x⁻³ = -6/x³。
ms, express everything with positive or negative exponents, then differentiate term by term. For example, f(x) = (x³+2x)/x² becomes x + 2x⁻¹, giving f'(x) = 1 - 2x⁻² = 1 - 2/x².
爱德思考官常将这些法则隐藏在有理表达式中。养成先化简的习惯：将分式拆分成若干项，用正指数或负指数表示每一项，然后逐项求导。例如 f(x) = (x³+2x)/x² 可化为 x + 2x⁻¹，从而 f'(x) = 1 - 2x⁻² = 1 - 2/x²。
3. Product and Quotient Rules: When Functions Multiply or Divide | 乘积法则与商法则：函数乘积或相除时
If y = u(x)v(x), the derivative is not simply u'v'; the product rule states dy/dx = u' v + u v'. This formula is essential when two non-constant functions multiply. A mnemonic is "first times derivative of second plus second times derivative of first" identify u and v clearly before differentiating each separately, then assemble the result.
若 y = u(x)v(x)，导数并非简单的 u'v'；乘积法则为 dy/dx = u' v + u v'。当两个非常数函数相乘时，这一公式必不可少。记忆口诀是“第一乘以第二的导数再加上第二乘以第一的导数”。务必先明确 u 和 v，分别求导后再组合结果。
For the quotient y = u/v, the derivative is (u' v - u v' ngly. Practice with rational functions like y = (x²+1)/(x-2); here u = x²+1, v = x-2, so u'=2x, v'=1, yielding dy/dx = [2x(x-2) - (x²+1)]/(x-2)² = (x² - 4x -1)/(x-2)².
对于商 y = u/v，导数为 (u' v - u v') / v²。顺序很重要：分子从被除式的导数开始。许多学生会漏掉减号或错误地对分母进行平方。用有理函数练习，如 y = (x²+1)/(x-2)；此处 u = x²+1, v = x-2，从而 u'=2x, v'=1，得出 dy/dx = [2x(x-2) - (x²+1)]/(x-2)² = (x² - 4x -1)/(x-2)²。
It is often easier to use the product rule with a negative exponent rather than the quotient rule. Rewrite 1/v as v⁻¹ and apply the product rule; this can avoid the messy denominator algebra until the final step. However, know both methods, as Edexcel questions sometimes explicitly ask for the quotient rule or expect you to simplify after using either.
有时使用带负指数的乘积法则比商法则更为简便。将 1/v 写成 v⁻¹ 再应用乘积法则，可以避免在最终步骤前处理繁杂的分母代数运算。不过两种方法都应掌握，因为爱德思的题目有时明确要求使用商法则，或者期望你在使用任一种方法后进行化简。
4. Chain Rule: Differentiating Composite Functions | 链式法则：复合函数求导
if y = f(g(x)) or y = (3x+5)⁸, set u = g(x) and y = f(u), then dy/dx = dy/du × du/dx. This "outside-in" on. For y = (3x+5)⁸, u = 3x+5 so y = u⁸, dy/du = 8u⁷, du/dx = 3, giving dy/dx = 24(3x+5)⁷.
链式法则处理函数中包含函数的情况：若 y = f(g(x)) 或 y = (3x+5)⁸，令 u = g(x)，y = f(u)，则 dy/dx = dy/du × du/dx。这种“由外向内”的方法用外层函数（以内层函数为自变量）的导数乘以内层函数的导数。以 y = (3x+5)⁸ 为例，u = 3x+5 故 y = u⁸，dy/du = 8u⁷，du/dx = 3，从而 dy/dx = 24(3x+5)⁷。
ive cos(2x) multiplies by the derivative of 2x, which is 2, so dy/dx = 2cos(2x). For y = e5x, derivative is 5e5x derivative is (2x)/(x²+1). Always write the derivative of the inside function as a factor—losing this factor is the single most frequent error.
在处理三角函数、指数函数和对数函数时，链式法则至关重要。对于 y = sin(2x)，外层导数 cos(2x) 乘以内层 2x 的导数 2，因此 dy/dx = 2cos(2x)。对于 y = e5x，导数为 5e5x。对于 y = ln(x²+1)，导数为 (2x)/(x²+1)。始终要将内层函数的导数作为一个因子写出来——丢失该因子是最常见的错误。
ognise this as a product of x² and sin(3x), requiring product rule: u = x², v = sin(3x); u'=2x, v' thod marks reward clear identification of u, v, and the correct use of chain rule inside v'.
爱德思常将链式法则与乘积法则或商法则结合考查。例如，对 x² sin(3x) 求导。将其视为 x² 与 sin(3x) 的乘积，需用乘积法则：u = x²，v = sin(3x)；u'=2x，v' 需用到链式法则 → 3cos(3x)。最终答案为 2x sin(3x) + 3x² cos(3x)。清晰地标示出 u、v 以及在 v' 内正确使用链式法则，将能获得方法分。
5. Exponential and Logarithmic Differentiation | 指数函数与对数函数的微分
The natural exponential function eˣ has the remarkable property that its derivative is itself: d/dx [eˣ] = eˣ. When combined with the chain rule, d/dx [ekx] = k ekx. For bases other than e, use aˣ = ex ln a giving derivative aˣ ln a. The natural logarithm function ln x has derivative 1/x for x>0. For ln(f(x)), the derivative is f'(x)/f(x) by chain rule.
自然指数函数 eˣ 具有非凡的性质：其导数等于它自身，即 d/dx [eˣ] = eˣ。与链式法则结合时，d/dx [ekx] = k ekx。对于 e 以外的底数，可使用 aˣ = ex ln a，从而导数为 aˣ ln a。自然对数函数 ln x（x>0）的导数为 1/x。而对 ln(f(x))，根据链式法则其导数为 f'(x)/f(x)。
A classic Edexcel question: Find dy/dx for y = ln(3x²+2). Here f(x)=3x²+2, f' s, so the derivative is valid everywhere. Also practise differentiating functions like x eˣ using the product rule, and quotients like eˣ/(x+1).
一道经典的爱德思题目：求 y = ln(3x²+2) 的导数 dy/dx。此处 f(x)=3x²+2, f'(x)=6x，因此 dy/dx = 6x/(3x²+2)。原函数的定义域为所有实数（因为 3x²+2 > 0），故该导数处处有效。还应练习使用乘积法则求 x eˣ 这类函数的导数，以及求 eˣ/(x+1) 这类商的导数。
products or quotients, or when the variable appears in both base and exponent. Take y = xsin x sin x ln x. Differentiate implicitly: (1/y) dy/dx = cos x ln x + (sin x)/x. Thus dy/dx = xsin x [cos x ln x + sin x/x]. This method is not always mandatory but can simplify otherwise intractable expressions.
对数微分法是一种强大技巧，适用于复杂乘积或商，或当变量同时出现在底数和指数中的情况。以 y = xsin x 为例。两边取自然对数：ln y = sin x ln x。隐函数微分得：(1/y) dy/dx = cos x ln x + (sin x)/x。从而 dy/dx = xsin x [cos x ln x + sin x/x]。此法并非总是必需，但可以简化原本难以处理的多项表达式。
6. Trigonometric Differentiation and the Chain Rule | 三角函数的微分与链式法则
rivative of tan x can be derived using quotient rule or by writing tan x = sin x / cos x, yielding sec² x. For sec x, cosec x, and cot x, it is better to learn the standard results: d/dx [sec x] = sec x tan x, d/dx [cosec x] = -cosec x cot x, d/dx [cot x] = -cosec² x.
正弦函数和余弦函数的导数构成一个循环模式：d/dx [sin x] = cos x，d/dx [cos x] = -sin x。其他三角函数的导数均可由此推出。tan x 的导数可用商法则推导，或写成 sin x/cos x，从而得到 sec² x。对于 sec x、cosec x 和 cot x，最好记住标准结果：d/dx [sec x] = sec x tan x，d/dx [cosec x] = -cosec x cot x，d/dx [cot x] = -cosec² x。
= sin(2x), then dy/dx = 2cos(2x) and d²y/dx² = -4 sin(2x). This is particularly relevant for simple harmonic motion or verifying maxima/minima later.
当角度为倍数时，应用链式法则：d/dx [sin(ax+b)] = a cos(ax+b)。对于二阶导数，只需再求导一次。例如，若 y = sin(2x)，则 dy/dx = 2cos(2x)，d²y/dx² = -4 sin(2x)。这在简谐运动或后续验证极值时尤为重要。
includes derivatives of arcsin, arccos, and arctan. Standard results: d/dx [arcsin x] = 1/√1-x², d/dx [arccos x] = -1/√1-x² n be proved by implicit differentiation but are best memorised. For arcsin(x/a) etc., use chain rule.
一个容易出错的领域是反三角函数的微分。爱德思课程内容包括 arcsin、arccos 和 arctan 的导数。标准结果：d/dx [arcsin x] = 1/√1-x²，d/dx [arccos x] = -1/√1-x²，d/dx [arctan x] = 1/(1+x²)。这些结果可以通过隐函数微分证明，但最好加以记忆。对于 arcsin(x/a) 等，使用链式法则即可。
7. Implicit Differentiation: When y Is Not the Subject | 隐函数微分：当 y 不是主项时
h respect to x, treating y as a function of x: whenever you differentiate a y-term, multiply by dy/dx. Thus, d/dx [y²] = 2y dy/dx, and for a product like xy, use product rule: (1)(y) + x(dy/dx).
当 y 作为 x 的函数被隐式定义时，例如 x² + y² = 25 或 x³ + xy + y³ = 6，就需要使用隐函数微分。对每一项关于 x 求导，将 y 视为 x 的函数：每当对含有 y 的项求导时，要乘以 dy/dx。因此，d/dx [y²] = 2y dy/dx，而对 xy 这样的乘积则使用乘积法则：(1)(y) + x(dy/dx)。
y/dx = -x/y. Implicit differentiation can directly give gradient at a point without solving for y explicitly, which is very useful for curves that are difficult to separate.
求导后，将所有含 dy/dx 的项移至一侧，提取 dy/dx，再求解。对于圆 x² + y² = 25，我们得到 2x + 2y dy/dx = 0，因此 dy/dx = -x/y。隐函数微分可以直接给出某一点的梯度，而无需显式解出 y，这对于难以分离的曲线极为有用。
to check it satisfies the equation, then find dy/dx using implicit differentiation, evaluate the gradient at that point, then use y - y₁ = m(x - x₁) for the tangent and m perpendicular for the normal. The algebra can be messy, so work systematically and watch for sign errors.
典型的爱德思题目涉及在隐式定义的曲线上求某一点的切线或法线方程。先代入该点验证其满足方程，然后利用隐函数微分求得 dy/dx，计算该点梯度，再使用 y - y₁ = m(x - x₁) 得出切线方程，以负倒数梯度得出法线方程。代数运算可能较为繁杂，因此要系统地处理并注意符号错误。
8. Parametric Differentiation: Gradient from a Parameter | 参数方程微分：从参数求梯度
d to eliminate the parameter; just differentiate both expressions with respect to t and simplify. Second derivative d²y/dx² is found by differentiating dy/dx with respect to t and dividing by dx/dt: d²y/dx² = d/dt (dy/dx) / (dx/dt).
当曲线由参数方程 x = f(t), y = g(t) 定义时，梯度 dy/dx 由 (dy/dt) / (dx/dt) 给出，或等价地写作 dy/dt ÷ dx/dt，前提是 dx/dt ≠ 0。无需消去参数；只需将两个表达式关于 t 分别求导并化简。二阶导数 d²y/dx² 可通过先对 dy/dx 关于 t 求导，再除以 dx/dt 得到：d²y/dx² = d/dt (dy/dx) / (dx/dt)。
To find the tangent line at t = 2, compute x = 3, y = 10, then use y - 10 = (13/4)(x - 3).
例如，x = t² - 1，y = t³ + t。则 dx/dt = 2t，dy/dt = 3t² + 1，所以 dy/dx = (3t²+1)/(2t)。在 t = 2 处，梯度为 (3×4+1)/(4) = 13/4。为求 t = 2 处的切线，计算 x = 3，y = 10，然后使用 y - 10 = (13/4)(x - 3)。
). To determine the nature of a stationary point, either find d²y/dx² as above or test the sign of dy/dx either side of the parameter value. Edexcel mark schemes reward clear working; always state the condition dx/dt ≠ 0 explicitly when dividing.
注意参数形式的驻点：令 dy/dx = 0，这意味着 dy/dt = 0（前提是 dx/dt ≠ 0）。要判断驻点的性质，可以如上所述求 d²y/dx²，或者在参数值的左右两侧检查 dy/dx 的符号。爱德思的评分标准鼓励清晰的解题过程；在除法时务必明确标明条件 dx/dt ≠ 0。
9. Applications of Differentiation: Tangents, Normals, and Stationary Points | 微分的应用：切线、法线与驻点
Finding the equation of a tangent requires the gradient f'(a) and point (a, f(a)). The tangent line is y - f(a) = f'(a)(x - a). The normal is perpendicular: its gradient is -1/f'(a) provided f' substituting numbers accurately and rearranging into required forms like ax + by + c = 0.
求切线方程需要梯度 f'(a) 与切点 (a, f(a))。切线为 y - f(a) = f'(a)(x - a)。法线与之垂直：其梯度为 -1/f'(a)，前提是 f'(a) ≠ 0。你必须熟练地精确代入数值，并将方程整理成要求的形式，如 ax + by + c = 0。
Stationary points occur where f' maxima, minima, or points of inflection. To classify, use the second derivative test: if f''(a) > 0 it's a local minimum; if f''(a) < 0 it's a local maximum; if f''(a) = 0, further investigation is needed (check the sign of f' ives). For points of inflection, the gradient may not change sign, but the concavity does.
驻点出现在 f'(x) = 0 处。这些点可能是极大值、极小值或拐点。为进行分类，使用二阶导数检验：若 f''(a) > 0，为局部极小值；若 f''(a) < 0，为局部极大值；若 f''(a) = 0，则需进一步分析（检查 f'(x) 在两侧的符号，或考虑更高阶导数）。对于拐点，梯度未必改变符号，但凹凸性会改变。
Many Edexcel modelling problems ask you to find the maximum area, minimum surface area, or greatest volume under constraints. Set up an expression for the quantity to be optimised in terms of one variable, differentiate, set derivative to zero, verify the nature of the stationary point, and then compute the optimum value. Don't forget to check endpoints if the domain is closed.
许多爱德思的建模问题要求你在约束条件下求出最大面积、最小表面积或最大体积。先用单一变量表示待优化的量，对其求导，令导数为零，验证驻点性质，然后计算最优值。注意，若定义域是闭区间，别忘了检查端点。
10. Connected Rates of Change and Real-World Contexts | 关联变化率与实际问题背景
x)(dx/dt). You might be given dx/dt and need to find dy/dt, or vice versa. Set up the relationship between x and y, differentiate with respect to t, substitute known rates, and solve for the unknown rate.
关联变化率问题涉及两个或更多随时间变化的量。若 y = f(x) 且 x 和 y 都依赖于 t，那么根据链式法则有 dy/dt = (dy/dx)(dx/dt)。题目可能给出 dx/dt 并要求你求 dy/dt，或者反过来。建立 x 与 y 之间的关系，将等式两边对 t 求导，代入已知的变化率，然后解出未知变化率。
us when the radius is 10 cm. Use V = (4/3)πr³, so dV/dt = 4πr² dr/dt. Given dV/dt = 30, r = 10, solve for dr/dt = 30/(4π×100) = 3/(40π) cm/s. Always include units in the final answer.
一道典型题目：一个球形气球充气时，体积以 30 cm³/s 的恒定速率增加。求当半径为 10 cm 时半径的增加速率。使用 V = (4/3)πr³，则 dV/dt = 4πr² dr/dt。已知 dV/dt = 30，r = 10，解得 dr/dt = 30/(4π×100) = 3/(40π) cm/s。最终答案务必包含单位。
These questions often incorporate geometry (areas, volumes, Pythagoras) and require you to express one variable in terms of another before differentiating. Drawing a diagram and writing down the static relationship first is crucial. Then introduce the time derivatives. Be careful with signs: if a quantity is decreasing, its rate is negative. Edexcel expects candidates to interpret real-world rates correctly, e.g., rate of cooling, flow rates, or speed of moving parts.
这类题目往往结合几何（面积、体积、勾股定理），并要求在求导之前先用一个变量表示另一个变量。先画出简图并写下静态关系式至关重要，然后再引入对时间的导数。注意符号：如果一个量在减少，其变化率为负。爱德思期望考生能正确解释现实中的变化率，例如冷却速率、流速或运动部件的速率等。
11. Differentiation Techniques Summary and Efficient Simplification | 微分技巧总结与高效化简
Mastering differentiation demands fluency in recognising the function type and selecting the appropriate rule. A summary checklist: power rule for terms like xⁿ, constant multiples, sums/differences; product rule for uv; quotient rule for u/v or better as uv⁻¹ with product rule; chain rule for f(g(x)); implicit for equations not solved for y; parametric for x(t), y(t); logarithmic differentiation for xf(x) or messy products.
掌握微分需要熟练地识别函数类型并选择适当的法则。综合清单如下：幂法则用于 xⁿ 项，常数倍，和差；乘积法则用于 uv；商法则用于 u/v，或更好是写成 uv⁻¹ 再用乘积法则；链式法则用于 f(g(x))；隐函数用于未解出 y 的方程；参数微分用于 x(t), y(t)；对数微分法用于 xf(x) 或复杂的乘积式。
Simplifying before and after differentiation saves time and reduces errors. For instance, factor out constants early, rewrite roots and reciprocals with fractional/negative exponents, and break fractions into separate terms. After differentiating, factor common terms to express the derivative in a tidy form, which makes evaluating or finding stationary points easier.
在求导之前和之后进行化简能够节省时间并减少错误。例如，及早提取常数因子，使用分数/负指数重写根号和倒数，将分式拆分为若干项。求导之后，提取公因子以整洁的形式表达导数，这将使求值或寻找驻点更为简便。
Avoid common algebraic slip-ups: forgetting to multiply by the derivative of the inside function (chain rule oversights), mixing up the sign in the quotient rule numerator (it's u'v - uv', not uv' - u'v), and mishandling the derivative of constants. Practice until the process becomes automatic. The more you practise, the more you will spot efficient routes through complex-looking expressions.
避免常见的代数失误：忘记乘以内层函数的导数（疏忽链式法则），混淆商法则分子的符号（应为 u'v - uv'，而非 uv' - u'v），以及错误处理常数导数。反复练习直至过程变为自动反应。你练得越多，就越能发现处理看似复杂的表达式的高效路径。
12. Exam Strategy and Common Edexcel Pitfalls | 考试策略与爱德思常见陷阱
In Edexcel exams, differentiation questions appear in both the pure papers and as part of mechanics/statistics contexts. Always read the question carefully: identify what derivative is required, whether it's the first or second derivative, and the variable with respect to which you are differentiating. Time management is key; if a complicated differentiation is only worth 3 marks, do not write an essay—focus on the key steps.
在爱德思考试中，微分问题既会出现在纯数试卷中，也会作为力学/统计学背景的一部分出现。务必仔细读题：确定需要求的是哪一阶导数，是一阶还是二阶，以及对哪个变量求导。时间管理至关重要；如果一个复杂的微分题只值 3 分，就不要长篇大论——集中于关键步骤。
t d/dx [5] = 0. Pitfall 2: Product rule applied incorrectly for three functions, e.g., x eˣ sin x — ly inflection points with horizontal tangent, misclassified if only second derivative test
Published by TutorHao | A-Level 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