Edexcel Pure Maths 微分 积分 向量 证明

Edexcel Pure Maths 微分 积分 向量 证明

刚拿到 Edexcel A-Level 数学（9MA0）大纲的学生，常把纯数当成一张长长的知识点清单，然后从第一页开始背。真正拖分的往往不是清单太长，而是不清楚两卷纯数怎么分工。现行大纲是 Issue 4（2020 年 2 月），Pearson 在大纲里写得很直白：Paper 1 与 Paper 2 都叫 Pure Mathematics 1 与 Pure Mathematics 2，时长、分值、权重三项完全一致，任何纯数内容都可能出现在任意一卷。换句话说，没有哪一章可以押在第二卷再学。把三卷的分值和题量先记住，比从第一页开始背公式省时间得多。

This introduction is for students who have just started the Pearson Edexcel A-Level in Mathematics, and for the parents paying for it. Two of the three papers are pure mathematics. They share the same time allowance, the same mark total and the same weight, and the specification says outright that pure content can appear on either paper. There is no safe chapter to postpone.

两卷都考纯数：P1 与 P2 的分值与题量

官方结构是这样的：9MA0 一共三卷，各 2 小时、各 100 分、各占 33.33%。Paper 1 和 Paper 2 是纯数，Paper 3 是统计与力学，卷内统计 50 分、力学 50 分。AS 的 8MA0 换了一套比例，Paper 1 纯数 2 小时 100 分占 62.5%，Paper 2 统计与力学 1 小时 15 分钟 60 分占 37.5%。同样叫纯数，AS 一年级的纯数权重高出一大截，这也是很多一年制学生误判复习重心的地方。

The A-Level structure is easy to remember once you see it as three equal papers: two hours, 100 marks and 33.33% each, with only Paper 3 splitting into 50 marks of statistics and 50 of mechanics. The AS qualification uses a different split. Its pure paper alone is worth 62.5% of the whole award, which is why a one-year student who plans revision like an A-Level student tends to run out of time on pure topics.

证明题：反证法与穷举法在 P1 的固定位置

大纲 Topic 1 是 Proof，分三种写法：演绎证明、穷举法、反证法。反证法的经典例子是证明根号 2 是无理数，标准过程假设根号 2 等于最简分数 p 除以 q，推出 p、q 同为偶数，与最简矛盾。常见的错误是学生在写矛盾结论时只说「矛盾了」，没有点明与哪一条前提冲突，阅卷时这一句就是丢分点。穷举法则要写清枚举范围，例如证明某个整数命题只需分奇偶两类，范围不写等于没证。

Proof is the first topic on the specification and it rewards structure rather than cleverness. Deduction, exhaustion and contradiction each have a template. The irrationality of the square root of two is the standard contradiction question, and markers look for an explicit statement of which assumption has been violated, not just a line saying the result is impossible. For exhaustion, the range of cases has to be stated before the cases are worked through.

代数与函数：因式定理、判别式与部分分式

因式定理和余数定理是这一章的骨架：多项式除以一次式时，代入求值就能判断是否为因式。二次方程的判别式决定交点个数，遇到「求 k 的取值范围使曲线与直线有两个交点」这类题，把它转成判别式大于零的不等式，比画图可靠。部分分式在 P2 出现，分母有重复因式时要写全每一项，漏掉中间项是最隐蔽的错误。很多学生以为部分分式只考分解，实际常与二项展开连考。

Factor theorem and remainder theorem do most of the work in this topic, and both are really statements about substitution. Discriminant questions look like geometry but behave like algebra: convert the intersection condition into an inequality and the range of the unknown falls out. Partial fractions belong to Paper 2, and repeated factors in the denominator require every power to be written out, since a missing middle term is almost invisible in the final answer.

二项式展开：负指数与分数指数的收敛条件

正整指数的二项展开是纯计算，系数用组合数写，注意二项式系数与项的系数不是同一个东西。真正拉开差距的是负指数和分数指数：展开式只对绝对值小于 1 的 x 成立，题目要求说明展开有效的 x 范围时必须写出这个条件。有效范围写错，后面的近似值计算全部连带失分。用展开式估算根号数值时，取前两三项就够，再往后项的大小反而会扩大误差。

Positive integer powers of a binomial are routine arithmetic, and the usual slip is confusing the binomial coefficient with the coefficient of the term. Negative and fractional indices are where the marks hide. The expansion is only valid when the absolute value of x is below one, and the validity statement is a marked part of the question, not a formality. Estimation questions usually need only two or three terms before the series stops converging usefully.

坐标几何：圆方程的配方与切线斜率

圆的方程要以配方法化成标准形式，圆心和半径才能直接读出，直接背公式在系数不为 1 时容易出错。直线与圆的位置关系用圆心到直线的距离与半径比较，比联立方程省时间。切线斜率靠半径与切线垂直，法线再取负倒数。常见的错误是圆与直线相切题里学生求出切点却忘了回答切线方程，题干问什么就答什么。

Circle equations are best handled by completing the square, because the centre and radius then read off directly and the leading coefficient stops causing trouble. Position between a line and a circle is a distance comparison against the radius, which is quicker and less error prone than solving a quadratic system. Tangents follow from the radius being perpendicular, and normals from the negative reciprocal of that gradient.

数列与级数：等比求和的收敛区间

等差数列的求和公式要能自己推一遍，考场上临时记错首项与项数的对应关系是高频失分点。等比级数的无穷求和只在公比绝对值小于 1 时收敛，和为 a 除以 1 减 r，这个条件同样要写出来。实际应用题里每年增长百分之几是等比模型，别当等差处理。建模题的第二问常要求解释为什么长期值不能无限增长，答案落在收敛条件上。

Arithmetic series formulas are worth deriving once by hand, since most errors come from misreading which term is first and how many there are. An infinite geometric series converges only when the common ratio sits between minus one and one, and the sum formula is meaningless without that condition stated. Growth questions are geometric, not arithmetic, and the follow-up part usually asks why a long-run total stops growing, which is a question about convergence.

三角恒等式：谐波形式与弧度解方程

恒等式这一章有三件常用工具：正余弦平方和为一、正切的定义式、以及把 a sinx 加 b cosx 合成 R sin(x 加 a) 的形式，其中 R 等于 a 平方加 b 平方再开方。合成形式用来求最大值最小值，也用来解方程。解三角方程务必把定义域内的所有解找全，弧度制下别把角度制的答案混进来。很多学生在求完一个解就停手，漏掉同周期内的第二个解。

Three tools cover most of this topic: the Pythagorean identity, the definition of tangent, and the harmonic form that rewrites a combination of sine and cosine as a single shifted sine with amplitude R. The harmonic form answers both maximum value questions and equation questions. When solving, every solution inside the stated interval has to be listed, and in radians the answers must stay in radians throughout.

指数与对数：ln 定律与建模转化

对数定律要熟到不用想：乘积变加法、幂次变系数、换底公式处理不同底数。指数方程的标准解法是两边取对数，把未知数从指数上拿下来。自然指数与自然对数互为反函数，求导与积分里会反复用到。建模题中按比例衰减写成 e 的负 kt 次方，先由初始条件定系数，再由另一个数据点定 k，顺序反了会得到两个未知数解不出的方程。

Logarithm laws need to be automatic, since turning a product into a sum and a power into a multiplier is the whole technique. Exponential equations are solved by taking logs of both sides, which brings the unknown down to ground level. Decay models are written as an exponential with a negative constant, and the constants are fixed in order: initial condition first, then a second data point to pin the rate.

微分：链式、积、商与二阶导判别

链式法则处理复合函数，外层导数乘内层导数，漏乘内层是错得最多的一类。积法则与商法则要能写准符号，商法则分子上是前导后不导减后导前不导。驻点用一阶导为零求出，判断极大极小时用二阶导符号，二阶导为零时得回到一阶导的符号变化。切线方程题别忘了把 x 值代回原函数求 y 坐标，只写斜率不算答完。

The chain rule carries most of the marks here, and forgetting the inner derivative is the single most common error in the whole pure syllabus. Product and quotient rules are about signs and order, and quotient rule errors usually come from reversing the numerator. Stationary points need the first derivative set to zero, then the second derivative to classify, with a fallback to sign change when the second derivative vanishes.

积分：换元、分部与梯形法则的误差

定积分先看能不能直接积，多项式与简单三角函数可以直接做。换元法的关键是把 dx 一起换掉，上下限在换元后必须同步改，忘了改限是定积分最常见的扣分理由。分部积分用于多项式乘指数或对数的情形，选谁作 u 决定整题难度。梯形法则给的是近似值，题目问是偏大还是偏小时要根据曲线凹凸性回答，不是随便猜。

Definite integrals start with the question of whether the integrand can be handled directly, and polynomials usually can. Substitution only works if the differential is replaced as well, and the limits have to be converted at the same time, which is where most lost marks come from. Integration by parts suits products of a polynomial with an exponential or a logarithm, and the choice of u decides the difficulty. The trapezium rule gives an approximation, and whether it overestimates or underestimates follows from the concavity of the curve.

向量：点积求夹角与位置向量

向量在纯数里以两种形式出现：i、j 分量形式，以及两点之间的位置向量。点积用来求夹角，公式是模长乘积乘夹角余弦，等于零就垂直。求直线方程要有位置向量和方向向量，两者不能混。学生最常把方向向量当成点坐标代入，写出的直线看起来对，代入检验就不成立。三维向量只考基本运算，不涉及叉积。

Vectors appear in two costumes, as components along i and j, and as position vectors between points. The scalar product answers angle questions, and a zero result means perpendicularity. A line needs a position vector plus a direction vector, and swapping those two is the mistake that produces a plausible but wrong line. Three dimensional vectors stay within basic operations, and the cross product is not required.

参数方程：消参与 dy/dx 的链式写法

参数方程属于 P2 的内容。求导时用 dy 比 dx 等于 dy 比 dt 乘以 dt 比 dx，中间那一层不能省。求曲线围成的面积要把 y 与 dx 都换成参数 t 的表达式，积分限也换成 t 的取值。消参题优先用三角恒等式，遇到 sin 与 cos 就直接用平方和为一。很多学生以为参数方程是独立章节，其实它只是把前面的微分与积分换了变量。

Parametric equations belong to Paper 2 and are really the chain rule wearing a different hat: the derivative of y with respect to x is built from two derivatives with respect to the parameter. Areas under parametric curves need both y and dx converted, along with the limits. Eliminating the parameter is usually a trigonometric identity question in disguise, and the Pythagorean identity does most of the work.

考场上的评分点：方法分与公式册

Edexcel 的评分方案按方法分、准确分、独立分给分，方法分只看用了正确的方法，即使算错也能拿到。这意味着中间步骤必须写出来，直接写答案在方法分上等于放弃。考试提供绿色的数学公式与统计表册，常用公式不必背，但要知道哪一页有什么。计算器可以用，Pearson 的规定是不允许带符号代数运算、符号求导与积分功能的机型，进考场前确认自己的型号合规。

Marking on these papers is method, accuracy and independent marks, and method marks survive an arithmetic slip as long as the working is visible. That is the practical reason to show steps rather than jump to an answer. A green booklet of mathematical formulae and statistical tables is provided in the exam, so the skill is knowing what is in it and where. Calculators are allowed, subject to Pearson rules that exclude symbolic algebra, symbolic differentiation and symbolic integration.

复习安排上，把纯数两卷当成一份 200 分的大卷来排，比按章节顺序推进更接近真实考试。每周挑一个往年真题的整卷限时做，做完只统计失分属于哪一类：概念不清、方法选错、还是计算失误。三类问题的补法完全不同，混在一起刷题只会重复同样的错误。

A revision plan that treats the two pure papers as one 200-mark body of work matches the exam better than a chapter-by-chapter march. One timed past paper a week, followed by sorting every lost mark into concept, method or arithmetic, tells you what to fix next. The three categories need different remedies, and mixing them under the label of more practice tends to reproduce the same errors.

海外直订Higher Mathematics for AQA GCSE 高等数学AQA GCSE¥645.0📱 去京东购买京东精选

更多咨询请联系16621398022（同微信）

需要一对一辅导？Need One-on-One Tutoring?

A-Level / GCSE / IB / KS2-KS3 全科辅导

16621398022 同微信

Visit aleveler.com