Numerical Methods in AQA Mathematics: Key Points Summary | AQA 数学：数值方法考点精讲

📚 Numerical Methods in AQA Mathematics: Key Points Summary | AQA 数学：数值方法考点精讲

Numerical methods are essential tools in A-level Mathematics when exact analytical solutions are difficult or impossible to obtain. They provide approximate solutions to equations, integrals, and other problems through iterative processes. This revision guide covers the key numerical methods examined in AQA A-Level Mathematics, including root-finding techniques and numerical integration, with a strong focus on accuracy, error analysis, and exam technique.

数值方法是当精确解析解难以或无法求得时，A-level 数学中必不可少的工具。它们通过迭代过程提供方程、积分及其他问题的近似解。本复习指南涵盖了 AQA A-Level 数学中常考的关键数值方法，包括求根技巧和数值积分，并重点关注精度、误差分析和应试技巧。

1. Introduction to Numerical Methods | 数值方法简介

Numerical methods generate approximate solutions by repeated calculations. In the AQA specification, you will be tested on finding roots of equations f(x)=0 and estimating definite integrals. These methods are especially useful when functions are not easily solvable by algebraic manipulation.

数值方法通过重复计算生成近似解。在 AQA 考试大纲中，你将接受求解方程 f(x)=0 的根以及估算定积分方面的考核。当函数不易通过代数操作求解时，这些方法尤其有用。

The accuracy of a numerical method depends on the number of iterations, step size, and the nature of the function. Understanding convergence and error bounds is crucial for selecting the appropriate method and interpreting results.

数值方法的精度取决于迭代次数、步长和函数的性质。理解收敛性和误差界限对于选择合适的方法及解释结果至关重要。

2. Locating Roots: The Sign Change Rule | 定位根：符号变化法则

If a continuous function f(x) changes sign over an interval [a, b], i.e. f(a)×f(b) < 0, then there exists at least one root r in (a, b) such that f(r)=0. This is the Intermediate Value Theorem for continuous functions.

若连续函数 f(x) 在区间 [a, b] 上改变符号，即 f(a)×f(b) < 0，则 (a, b) 内至少存在一个根 r 使得 f(r)=0。这是连续函数的中值定理。

A change of sign is a sufficient condition for a root, but not a necessary one — a root can occur without a sign change (e.g. f(x)=(x−1)² has a root at x=1, but the sign does not change). Therefore, graphical or analytical checks are important.

符号改变是根存在的充分条件，但不是必要条件 — 根可能在没有符号改变的情况下出现（例如 f(x)=(x−1)² 在 x=1 处有根，但符号不变）。因此，图形或解析检查很重要。

When using a table of values, look for a sign change between consecutive x-values. The root is then located between those two x-values. This interval can be used as the starting point for iterative methods like bisection or linear interpolation.

使用函数值表时，寻找相邻 x 值之间的符号变化。那么根就位于这两个 x 值之间。这个区间可以作为二分法或线性插值等迭代方法的起点。

3. Bisection Method | 二分法

The bisection method systematically halves an interval [a, b] where f(a) and f(b) have opposite signs. The midpoint c = (a+b)/2 is calculated. If f(c) has the same sign as f(a), replace a with c; otherwise replace b with c. This guarantees the root remains inside the shrinking interval.

二分法系统地将区间 [a, b] 一分为二，其中 f(a) 与 f(b) 异号。计算中点 c = (a+b)/2。若 f(c) 与 f(a) 同号，则用 c 替换 a；否则用 c 替换 b。这就保证了根始终位于不断缩小的区间内。

The process repeats until the interval width (b−a) is less than the required accuracy. The midpoint of that final interval is taken as the root approximation. Bisection is robust but converges linearly, meaning it requires many iterations to achieve high precision.

重复该过程直到区间宽度 (b−a) 小于所需精度。最终区间的中点即为根的近似值。二分法稳健，但收敛速度为线性，这意味着要达到高精度需要多次迭代。

The number of iterations n needed to reduce an initial interval [a₀, b₀] to an accuracy of ε can be estimated by n ≥ log₂((b₀−a₀)/ε). In exam questions, you may be asked to perform a specific number of iterations and state the final interval or midpoint.

将初始区间 [a₀, b₀] 缩小到精度 ε 所需的迭代次数 n 可估算为 n ≥ log₂((b₀−a₀)/ε)。在考试题中，你可能需要执行指定次数的迭代，并给出最终的区间或中点。

4. Linear Interpolation (Chord Method / Regula Falsi) | 线性插值（弦截法 / 试位法）

Linear interpolation improves upon bisection by using the idea of similar triangles to estimate where the chord joining (a, f(a)) and (b, f(b)) crosses the x-axis. The formula for the new x-intercept is:

线性插值利用相似三角形的思想改进了二分法，估算连接 (a, f(a)) 和 (b, f(b)) 的弦与 x 轴的交点。新的 x 截距公式为：

c = a − f(a) × (b − a) / (f(b) − f(a))

This replaces the midpoint calculation. The interval is then reduced to [a, c] or [c, b] depending on the sign of f(c), keeping the root bracketed. This method often converges faster than bisection when the function is approximately linear.

该公式替代了中点的计算。然后根据 f(c) 的符号将区间缩小为 [a, c] 或 [c, b]，保持根包含在区间内。当函数接近于线性时，此方法通常比二分法收敛更快。

However, linear interpolation can stall if the function is highly curved; the root bracket may shrink slowly on one side. AQA questions might ask you to perform one or two iterations and compare with bisection.

然而，如果函数高度弯曲，线性插值可能会停滞不前；区间可能在某一侧缓慢缩小。AQA 的题目可能会要求你进行一到两次迭代，并与二分法进行比较。

5. Newton-Raphson Method | 牛顿-拉夫森法

The Newton-Raphson method uses tangents to approximate the root. Starting with an initial guess x₀, the iterative formula is derived from the tangent line at (xₙ, f(xₙ)):

牛顿-拉夫森法使用切线来逼近根。从初始猜测 x₀ 开始，迭代公式由点 (xₙ, f(xₙ)) 处的切线推导得出：

xₙ₊₁ = xₙ − f(xₙ) / f'(xₙ)

This formula requires the derivative f'(x). If f'(xₙ)=0 or is very small, the method fails. Newton-Raphson converges quadratically near the root, meaning the number of correct digits roughly doubles each step, provided the initial guess is close enough.

该公式需要导数 f'(x)。如果 f'(xₙ)=0 或非常小，则方法失效。牛顿-拉夫森法在根附近呈二次收敛，这意味着只要初始猜测足够接近，正确数字的位数每步大约翻倍。

In AQA exams, you will typically be given a starting value and asked to perform two or three iterations, recording your values to a specified number of decimal places. You may also need to describe a situation where the method could diverge, such as when the starting point is near a turning point.

在 AQA 考试中，通常会给出一个初始值，要求你进行两次或三次迭代，并将数值记录到指定的小数位数。你可能还需要描述方法可能发散的情况，例如当起始点靠近驻点时。

6. Fixed Point Iteration | 不动点迭代

To solve f(x)=0 by fixed point iteration, the equation is rearranged into the form x = g(x). Starting from an initial value x₀, the iteration formula is:

要用不动点迭代求解 f(x)=0，需将方程整理成 x = g(x) 的形式。从初始值 x₀ 开始，迭代公式为：

xₙ₊₁ = g(xₙ)

Convergence occurs if |g'(x)| < 1 near the root. Graphically, the method corresponds to finding the intersection of y = x and y = g(x) via staircase or cobweb diagrams. An exam question may ask you to sketch such diagrams and comment on convergence.

如果在根附近 |g'(x)| < 1，则迭代收敛。在图形上，该方法对应通过阶梯图或蛛网图找到 y = x 与 y = g(x) 的交点。考试题可能会要求你绘制此类图示并讨论收敛性。

Different rearrangements lead to different g(x). Some may converge quickly, others diverge or converge slowly. You may be asked to suggest a suitable rearrangement and test it with a given starting value.

不同的变形会得到不同的 g(x)。有些可能收敛得很快，有些则可能发散或收敛得很慢。你可能会被要求提出一个合适的变形并用给定的初始值进行检验。

7. Numerical Integration: Trapezium Rule | 数值积分：梯形法则

The trapezium rule estimates the definite integral ∫ₐᵇ f(x) dx by dividing the area under the curve into n trapezoids of equal width h = (b−a)/n. The approximation is:

梯形法则通过将曲线下的面积分成 n 个等宽 h = (b−a)/n 的梯形来估算定积分 ∫ₐᵇ f(x) dx。其近似公式为：

∫ₐᵇ f(x) dx ≈ ½ h [y₀ + 2(y₁ + y₂ + … + yₙ₋₁) + yₙ]

where yᵢ = f(a + ih). You will often be given a table of heights and asked to apply the formula. The accuracy increases as n increases (or h decreases), because the trapezoids better fit the curve.

其中 yᵢ = f(a + ih)。通常题目会给出高度表，要求你应用该公式。随着 n 增加（或 h 减小），精度会提高，因为梯形能更好地拟合曲线。

The trapezium rule overestimates when the curve is concave upwards and underestimates when concave downwards. You may be asked to state whether the approximation is an over- or under-estimate by examining the graph or the second derivative.

当曲线向上凹时，梯形法则会高估；向下凹时会低估。你可能需要根据图形或二阶导数来判断该近似值是高估还是低估。

8. Simpson’s Rule | 辛普森法则

Simpson’s rule provides a more accurate estimate by fitting parabolic arcs through groups of three adjacent points. It requires an even number of sub-intervals (n must be even). With h = (b−a)/n, the formula is:

辛普森法则通过在三组相邻点之间拟合抛物线弧来提供更精确的估计。它要求子区间数为偶数（n 必须为偶数）。设 h = (b−a)/n，公式为：

∫ₐᵇ f(x) dx ≈ ⅓ h [y₀ + 4(y₁ + y₃ + … + yₙ₋₁) + 2(y₂ + y₄ + … + yₙ₋₂) + yₙ]

Simpson’s rule is exact for polynomials of degree up to 3. In AQA papers, you will usually be given the ordinates and asked to calculate an approximation. Always check that the number of strips is even.

辛普森法则对于次数不超过 3 的多项式是完全精确的。在 AQA 试卷中，通常会给出纵坐标，要求你计算近似值。务必检查分段数为偶数。

When the number of strips is odd, Simpson’s rule cannot be applied directly; you may need to apply the trapezium rule to the last strip or use a composite rule. However, standard AQA questions provide an even number of intervals.

当分段数为奇数时，辛普森法则不能直接使用；你可能需要对最后一段应用梯形法则或使用复合法则。不过，标准的 AQA 题目会提供偶数个区间。

9. Error Analysis and Accuracy | 误差分析与精度

Understanding error in numerical methods is essential. For root-finding, the number of iterations determines the accuracy. In bisection, the maximum error after n iterations is (b₀−a₀)/2ⁿ. For linear interpolation and Newton-Raphson, error estimation is more complex but can be bounded by comparing successive approximations.

理解数值方法中的误差至关重要。对于求根，迭代次数决定着精度。在二分法中，经过 n 次迭代后的最大误差为 (b₀−a₀)/2ⁿ。对于线性插值和牛顿-拉夫森法，误差估计更为复杂，但可以通过比较连续近似值来界定。

For integration, the error in the trapezium rule is roughly proportional to 1/n², while Simpson’s rule error is proportional to 1/n⁴ for sufficiently smooth functions. In exam answers, you can comment on the effect of doubling n: for trapezium, error reduces by a factor of about 4; for Simpson, by a factor of about 16.

对于积分，梯形法则的误差大致与 1/n² 成正比，而对于足够光滑的函数，辛普森法则的误差与 1/n⁴ 成正比。在答题时，你可以评述将 n 加倍的效果：对于梯形法则，误差约缩小为原来的 1/4；对于辛普森法则，约缩小为原有的 1/16。

Always quote your answers to the degree of accuracy asked for in the question, typically to a given number of decimal places or significant figures. Carry forward unrounded values in intermediate steps to avoid rounding error buildup.

务必按照题目要求的精度给出答案，通常是保留指定的小数位数或有效数字。在中间步骤中保留未经舍入的值，以避免舍入误差累积。

10. Common Pitfalls and Exam Tips | 常见陷阱与应试技巧

Misidentifying sign changes: Check consecutive values carefully. A sign change indicates a root, but a zero value means you have found an exact root — still acceptable.

误判符号变化：仔细检查连续的函数值。符号改变表明有根，而函数值为零说明你找到了一个精确的根 — 同样可以接受。
Forgetting to check the domain: Ensure the function is continuous over the interval. Discontinuities invalidate the sign change argument.

忘记检查定义域：确保函数在区间上连续。间断点会使符号变化论据失效。
Newton-Raphson without derivative: You must differentiate correctly. A common mistake is to misplace brackets or use an incorrect derivative.

牛顿-拉夫森法缺少导数：必须正确求导。一个常见错误是括号位置错误或使用了不正确的导数。
Using the wrong number of ordinates in trapezium/Simpson: Note that n strips correspond to n+1 ordinates. Miscounting leads to an incorrect formula.

梯形/辛普森法则中纵坐标数量有误：注意 n 个分段对应 n+1 个纵坐标。数错会导致公式错误。
Rounding in intermediate steps: Keep full calculator accuracy during iterations; round only the final answer. Premature rounding can cause significant error.

中间步骤中的舍入：迭代过程中保持完整的计算器精度；仅对最终答案进行舍入。过早舍入可能造成显著误差。
Not justifying convergence or divergence: If asked to state whether an iterative formula converges, refer to |g'(x)| < 1 near the root. Use a sketch or value to support your argument.

未论证收敛性或发散性：如果要求说明迭代公式是否收敛，需参考根附近 |g'(x)| < 1。使用草图或数值来支撑你的论点。
Simpson’s rule with odd n: AQA will not give this scenario unless a composite approach is intended. If you suspect it, re-read the question; usually n is a multiple of 2.

奇数 n 使用辛普森法则：除非意在考察复合方法，否则 AQA 不会给出这种情形。如果你有疑问，重读题目；通常 n 是 2 的倍数。

Published by TutorHao | Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Numerical Methods in AQA Mathematics: Key Points Summary | AQA 数学：数值方法 考点精讲