Numerical Methods for IB CCEA Mathematics | IB CCEA 数学：数值方法考点精讲

📚 Numerical Methods for IB CCEA Mathematics | IB CCEA 数学：数值方法考点精讲

Numerical methods provide powerful techniques for obtaining approximate solutions to mathematical problems when exact analytical methods are impractical or impossible. In IB and CCEA mathematics, these methods are essential for solving equations, evaluating integrals, and modelling real-world phenomena. This article covers key concepts such as root-finding algorithms, iterative processes, numerical integration, and error analysis, equipping you with the knowledge to tackle exam questions confidently.

当精确的解析方法难以或无法使用时，数值方法为解决数学问题提供了强大的近似求解技术。在 IB 和 CCEA 数学课程中，数值方法对于求解方程、计算积分以及模拟真实世界现象至关重要。本文涵盖求根算法、迭代过程、数值积分和误差分析等核心概念，帮助你充满信心地应对考试题目。

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

Numerical methods approximate solutions by performing a finite sequence of arithmetic operations. They are particularly useful for transcendental equations, complex integrals, and differential equations that lack closed-form solutions. In your exam, you will be expected to apply specific algorithms and understand their convergence behaviour and limitations.

数值方法通过执行有限次算术运算来逼近解。它们对于超越方程、复杂积分以及没有封闭形式解的微分方程特别有用。在考试中，你需要能够应用特定算法，并理解其收敛行为和局限性。

A key feature of numerical methods is that they yield approximate results; hence, the analysis of errors becomes a central theme. Distinguishing between truncation error and rounding error is vital for interpreting computed values.

数值方法的一个关键特征是它们产生近似结果，因此误差分析成为核心主题。区分截断误差和舍入误差对于解读计算值至关重要。

2. Error Analysis and Approximations | 误差分析与近似

Every numerical computation introduces some degree of error. Understanding the difference between absolute error and relative error allows you to assess the quality of an approximation. Absolute error is defined as |approximate value − true value|, while relative error is the absolute error divided by the true value, often expressed as a percentage.

每次数值计算都会引入一定程度的误差。理解绝对误差和相对误差的区别能让你评估近似解的质量。绝对误差定义为 |近似值 − 真值|，而相对误差是绝对误差除以真值，通常以百分比表示。

Truncation error arises when an infinite process is replaced by a finite one, such as truncating a Taylor series. Rounding error occurs because computers and calculators store numbers with finite precision. In the context of iterative root-finding, you may be asked to estimate the error in the n-th iterate or to determine how many iterations are needed to achieve a desired tolerance.

当无限过程被有限过程替代时，例如截断泰勒级数，就会产生截断误差。由于计算机和计算器以有限精度存储数字，舍入误差也会出现。在迭代求根法的背景下，你可能会被要求估计第 n 次迭代的误差，或确定需要多少次迭代才能达到给定的容差。

For IB and CCEA problem-solving, it is common to use the stopping criterion |xₙ₊₁ − xₙ| < ε or |f(xₙ)| < ε, where ε is a pre-set tolerance such as 10⁻⁴. Always check which criterion is specified in the question.

在 IB 和 CCEA 的解题中，通常使用停止准则 |xₙ₊₁ − xₙ| < ε 或 |f(xₙ)| < ε，其中 ε 是预设的容差，例如 10⁻⁴。务必检查题目中指定了哪种准则。

3. The Bisection Method | 二分法

The bisection method is a bracketing method that repeatedly halves an interval [a, b] where f(a) and f(b) have opposite signs, guaranteeing a root exists by the Intermediate Value Theorem. It is robust and conceptually simple but converges relatively slowly compared to other methods.

二分法是一种括根法，它通过重复将区间 [a, b] 对半分来逼近根，其中 f(a) 和 f(b) 异号，由介值定理保证根的存在。该方法稳健且概念简单，但与其他方法相比收敛较慢。

At each step, compute the midpoint c = (a+b)/2 and evaluate f(c). If f(c) = 0, c is the exact root. Otherwise, replace the endpoint whose function value has the same sign as f(c) with c, thus maintaining the sign change. The width of the interval after n iterations is (b−a)/2ⁿ, so the maximum absolute error is bounded by half that width.

在每一步中，计算中点 c = (a+b)/2 并求 f(c)。如果 f(c) = 0，则 c 为精确根。否则，将函数值与 f(c) 同号的端点替换为 c，从而保持符号变化。n 次迭代后区间宽度为 (b−a)/2ⁿ，因此最大绝对误差不超过该宽度的一半。

Students often need to perform a given number of bisection steps manually in an exam. Always tabulate values clearly, showing a, b, c, f(a), f(b), f(c), and the sign of f(c) to justify the choice of the new interval.

学生在考试中常需手动执行指定次数的二分法步骤。务必清晰地列表，给出 a、b、c、f(a)、f(b)、f(c) 及 f(c) 的符号，以说明新区间的选择依据。

4. Linear Interpolation (False Position) | 线性插值法（试位法）

The method of false position, also known as linear interpolation, improves on bisection by taking a weighted average of the endpoints using their function values. The new estimate c is calculated as:

试位法，又称线性插值法，通过使用端点函数值取加权平均来改进二分法。新的估计值 c 计算如下：

c = (a·f(b) − b·f(a)) / (f(b) − f(a))

This formula comes from the intersection of the secant line with the x-axis. The method retains the sign-change property, so convergence is guaranteed, but the speed depends on the shape of f near the root.

该公式来源于割线与 x 轴的交点。该方法保留异号性质，因此收敛是有保证的，但速度取决于 f 在根附近的形状。

In practice, you replace either a or b by c in the same way as bisection, ensuring the root remains bracketed. An exam question might ask you to compare the efficiency of false position with bisection or Newton-Raphson for a particular function.

在实践中，像二分法那样将 a 或 b 替换为 c，确保根依然被括住。考试题可能会让你针对某个特定函数比较试位法与二分法或牛顿-拉夫森法的效率。

5. Newton-Raphson Method | 牛顿-拉夫森法

The Newton-Raphson method is one of the most widely used iterative techniques due to its rapid convergence near a simple root. Starting from an initial guess x₀, each iteration uses the formula:

牛顿-拉夫森法因其在单根附近的快速收敛性而成为使用最广泛的迭代技术之一。从初始猜测值 x₀ 开始，每次迭代使用公式：

xₙ₊₁ = xₙ − f(xₙ)/f′(xₙ)

Geometrically, this corresponds to following the tangent line at xₙ to its intersection with the x-axis. You must be able to derive this formula from the first-order Taylor expansion or the equation of the tangent.

几何上，这对应于沿着 xₙ 处的切线求其与 x 轴的交点。你必须能够从一阶泰勒展开或切线方程推导出这个公式。

Newton-Raphson converges quadratically if f′(α) ≠ 0, meaning the number of correct digits roughly doubles with each step near the root. However, it can fail spectacularly if f′(xₙ) is close to zero, leading to division by a very small number, or if the initial guess is far from the root.

如果 f′(α) ≠ 0，牛顿-拉夫森法具有二次收敛性，这意味着在根附近每一步的正确位数大约翻倍。然而，如果 f′(xₙ) 接近零，导致除以一个非常小的数，或者初始猜测值离根太远，该方法会严重失效。

Examiners often test your ability to perform two or three iterations and to critique the method’s limitations. Always state the derivative clearly before substituting into the iteration formula.

考官经常测试你执行两到三次迭代的能力以及评价该方法的局限性。在代入迭代公式前，务必先写出明确的导数表达式。

6. Secant Method | 割线法

The secant method avoids the need to compute the derivative by approximating f′(xₙ) with a finite difference based on two previous iterates. Its iteration formula is:

割线法通过用基于前两次迭代值的有限差分来逼近 f′(xₙ)，从而避免计算导数。其迭代公式为：

xₙ₊₁ = xₙ − f(xₙ)·(xₙ − xₙ₋₁) / (f(xₙ) − f(xₙ₋₁))

Because it uses two starting values, it is not a single-point method like Newton-Raphson. Convergence is superlinear, with order approximately 1.618, making it faster than bisection but slightly slower than Newton-Raphson near the root.

由于它使用两个初始值，因此它不像牛顿-拉夫森法那样是单点迭代法。收敛是超线性的，收敛阶约为 1.618，这使得它比二分法快，但在根附近略慢于牛顿-拉夫森法。

In exam problems, you may be asked to generate the next iterate given x₀ and x₁, or to discuss the advantages of the secant method when the derivative is difficult to obtain analytically. Be mindful of stagnation if the denominator becomes very small.

在考试问题中，你可能会被要求根据给定的 x₀ 和 x₁ 生成下一个迭代值，或者讨论当导数难以解析求得时割线法的优势。注意如果分母变得非常小，迭代可能停滞。

7. Iterative Formulas and Convergence | 迭代公式与收敛性

Apart from Newton-Raphson and secant, iteration can be based on rearranging f(x)=0 into the form x = g(x). Starting from an initial value, successive approximations are generated by xₙ₊₁ = g(xₙ). Convergence occurs if |g′(x)| < 1 in a neighbourhood of the root.

除了牛顿-拉夫森法和割线法，迭代还可以基于将 f(x)=0 重排成 x = g(x) 的形式。从初始值开始，通过 xₙ₊₁ = g(xₙ) 生成逐次逼近值。如果在根的一个邻域内 |g′(x)| < 1，迭代就会收敛。

A common exam task is to show that a given rearrangement will converge to a specific root. You may be asked to calculate a few iterates using a given g(x) and to illustrate convergence using a staircase or cobweb diagram. Be ready to explain how the derivative condition guarantees contraction.

常见的考题是证明给定的重排形式会收敛到某一特定根。你可能会被要求使用给定的 g(x) 计算出几个迭代值，并用阶梯图或蛛网图说明收敛性。要准备好解释导数条件如何保证压缩性。

8. Numerical Integration: Trapezium Rule | 梯形法则

The trapezium rule approximates 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 approximate area is:

梯形法则通过将曲线下的区域划分为 n 个等宽 h = (b−a)/n 的梯形来近似计算定积分 ∫ₐᵇ f(x) dx。近似面积为：

∫ₐᵇ f(x) dx ≈ (h/2)[f(a) + 2∑ f(xᵢ) + f(b)]

where the sum runs over the interior points x₁, x₂, …, xₙ₋₁. The trapezium rule is a Newton-Cotes formula of degree 1, and its error is proportional to f″(ξ) for some ξ in (a, b).

其中求和涵盖内部点 x₁, x₂, …, xₙ₋₁。梯形法则是一个 1 阶牛顿-科茨公式，其误差与 (a,b) 内某点 ξ 处的 f″(ξ) 成正比。

In IB and CCEA questions, you will typically be given a table of values or asked to calculate ordinates yourself. Remember that increasing n reduces the width h and therefore improves accuracy, but also increases computational effort. The overestimate/underestimate property depends on the concavity of the function.

在 IB 和 CCEA 的考题中，通常会给你一个数值表，或要求你自己计算纵坐标。请记住，增加 n 会减少宽度 h，从而提高精度，但也会增加计算量。具体是高估还是低估则取决于函数的凹性。

9. Simpson’s Rule | 辛普森法则

Simpson’s rule provides a more accurate approximation by fitting quadratic polynomials through three consecutive points. It requires an even number of strips n, and the formula is:

辛普森法则通过用二次多项式拟合三个连续点来提供更精确的近似。它要求 strip 数 n 为偶数，公式为：

∫ₐᵇ f(x) dx ≈ (h/3)[f(a) + 4∑ f(x_odd) + 2∑ f(x_even) + f(b)]

where h = (b−a)/n, odd indices refer to the first, third, fifth ordinate, etc., and even indices refer to the intermediate ordinates. It is especially effective when f(x) is well-approximated by quadratics.

其中 h = (b−a)/n，奇数项指第 1, 3, 5 等纵坐标，偶数项指中间的纵坐标。当 f(x) 可被二次函数很好地近似时，此法特别有效。

You must be careful to apply the 4, 2, 4, 2,…, 4 pattern correctly. A typical exam question gives a data set and asks you to use Simpson’s rule with a specific number of strips. Comparing trapezium and Simpson’s estimates for the same number of points often highlights the superior accuracy of Simpson’s rule.

你必须小心地正确应用 4, 2, 4, 2,…, 4 的模式。典型的考题会给出数据并要求你用指定数量的 strip 应用辛普森法则。比较相同点数下梯形法则与辛普森法则的估计值，通常能突出辛普森法则的优越精度。

10. Applications and Exam Tips | 应用与考试技巧

Numerical methods appear in a variety of contexts: solving transcendental equations like eˣ = 2 − x, calculating the area under an irregular curve obtained from experimental data, and modelling population growth where an explicit formula is unavailable. You must be able to choose an appropriate method based on the information given.

数值方法出现在多种情境中：求解超越方程如 eˣ = 2 − x，计算由实验数据获得的不规则曲线下方面积，以及在无法获得显式公式时对人口增长进行建模。你必须能根据给定信息选择适当的方法。

When presenting iterative solutions, always round to the required number of decimal places or significant figures only at the end. Show all substitutions explicitly. In tables, maintain a consistent level of precision. For error questions, state clearly whether you are using absolute error, relative error, or an error bound.

在给出迭代解时，只在最后才按要求的小数位数或有效数字进行舍入。明确展示所有代入步骤。在表格中保持一致的精度水平。对于误差问题，要清楚地说明你使用的是绝对误差、相对误差还是误差界。

Time management in exams is critical. Familiarity with the steps of each method will allow you to perform manual calculations quickly. Practice with past-paper questions from both IB and CCEA to recognise common patterns, such as verifying convergence or applying Simpson’s rule with a supplied table.

考试中的时间管理至关重要。熟悉每种方法的步骤将使你能够快速进行手动计算。通过练习 IB 和 CCEA 的历年真题，识别常见模式，例如验证收敛性或对给定的表格应用辛普森法则。

11. Common Pitfalls | 常见陷阱

One frequent mistake in Newton-Raphson is forgetting to differentiate correctly or omitting the negative sign in the formula. Always double-check f′(x) independently. For the secant method, students sometimes mix up the order of xₙ and xₙ₋₁ in the numerator; remember the formula is symmetric in a specific way.

牛顿-拉夫森法中一个常见的错误是求导错误或在公式中遗漏负号。务必独立地检查 f′(x)。对于割线法，学生有时会混淆分子中 xₙ 和 xₙ₋₁ 的顺序；要记住该公式在特定方式下是对称的。

In trapezium and Simpson’s rule, forgetting to use the correct h or miscounting the number of ordinates can lead to an entirely wrong answer. With Simpson’s rule, ensure n is even; if the question gives an odd number of ordinates, use the trapezium rule for the last strip or combine methods as indicated.

在梯形法则和辛普森法则中，忘记使用正确的 h 或者数错纵坐标个数，会导致完全错误的答案。使用辛普森法则时，要确保 n 为偶数；如果题目给出的纵坐标个数为奇数，则需对最后一个 strip 使用梯形法则或根据指示组合方法。

Finally, do not misinterpret convergence criteria. Just because |xₙ₊₁ − xₙ| is small does not guarantee that you are close to the true root, especially if the derivative is large. Always link the stopping test back to the context of the problem.

最后，不要误解收敛准则。仅仅因为 |xₙ₊₁ − xₙ| 很小，并不能保证你已经接近真根，特别是在导数值很大的情况下。始终将停止测试与问题背景联系起来。

12. Summary and Revision Checklist | 总结与复习清单

To excel in numerical methods for IB and CCEA mathematics, ensure you can: state and apply the bisection, false position, Newton-Raphson and secant methods; rearrange equations into convergent iteration forms; calculate approximations using the trapezium rule and Simpson’s rule; estimate errors and determine rates of convergence; and interpret results critically, understanding the limitations of each algorithm.

要想在 IB 和 CCEA 数学的数值方法中取得优异成绩，请确保你能：陈述并应用二分法、试位法、牛顿-拉夫森法和割线法；将方程重排为收敛的迭代形式；使用梯形法则和辛普森法则计算近似值；估计误差并确定收敛速度；以及批判性地解释结果，理解每种算法的局限性。

Create a concise formula sheet referencing the key iteration and integration formulas. Practice writing clear solutions that an examiner can follow effortlessly. With consistent practice, numerical methods will become one of the most reliable and rewarding topics on your exam paper.

制作一份简洁的公式表，列出关键的迭代和积分公式。练习书写清晰、易于考官理解的解答。通过持之以恒的练习，数值方法将成为你考卷上最可靠、最易得分的主题之一。

Published by TutorHao | Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Numerical Methods for IB CCEA Mathematics | IB CCEA 数学：数值方法 考点精讲