📚 Newton-Raphson Method | 牛顿-拉弗森方法
-Level Edexcel Mathematics, it is a core part of the Pure Mathematics syllabus, often introduced in Year 2 under numerical methods. Many students playfully mispronounce its name as “The New Right,” thod is serious business, especially for calculator-based iteration questions in the exam.
牛顿-拉弗森方法是求解方程 f(x) = 0 近似根的一种强大数值技术。在 Edexcel A-Level 数学体系中，它是纯数学部分的核心内容，通常在第二年数值方法章节引入。不少学生开玩笑把它的名字谐音念成“新右派”，成了一个轻松的课堂梗。尽管昵称有趣，掌握这个方法对考试来说可是件严肃的事，尤其在涉及计算器迭代的题目中。
1. What is the Newton-Raphson Method? | 什么是牛顿-拉弗森方法？
The Newton-Raphson method uses an iterative formula to progressively find better approximations to the roots of a real-valued function. Starting from an initial guess x₀, a sequence of values x₁, x₂, x₃, ... is generated that converges to a root α, provided certain conditions are satisfied. It is essentially an application of linear approximation, using the tangent line at a current approximation to point towards the root.
牛顿-拉弗森方法通过一个迭代公式，逐步逼近实函数的根。从一个初始猜测值 x₀ 开始，生成序列 x₁, x₂, x₃, ...，在满足某些条件时收敛到根 α。它本质上是线性逼近的应用，利用当前近似点处的切线来指示根的方向。
2. Deriving the Iterative Formula | 迭代公式的推导
tangent to the curve y = f(x) at the point (xₙ, f(xₙ)). The equation of this tangent is y – f(xₙ) = f'(xₙ)(x – xₙ). This line crosses the x‑axis where y = 0, so 0 – f(xₙ) = f'(xₙ)(x – xₙ). Solving for x gives the improved estimate xₙ₊₁ = xₙ – f(xₙ) / f'(xₙ). This is the Newton-Raphson formula.
假设我们有方程 f(x) = 0 的一个根的估计值 xₙ。在点 (xₙ, f(xₙ)) 处作曲线 y = f(x) 的切线。此切线的方程为 y – f(xₙ) = f'(xₙ)(x – xₙ)。该直线与 x 轴相交时 y = 0，因此 0 – f(xₙ) = f'(xₙ)(x – xₙ)。从中解出 x，便得到改进的估计值 xₙ₊₁ = xₙ – f(xₙ) / f'(xₙ)。这便是牛顿-拉弗森公式。
xₙ₊₁ = xₙ – f(xₙ) / f'(xₙ)
In this formula, f'(xₙ) must not be zero, otherwise the tangent is horizontal and never meets the x‑axis, causing the method to fail. The formula is remarkably simple and can be programmed into most scientific calculators for repeated iteration.
在这个公式中，f'(xₙ) 不能为零，否则切线水平，永远不会与 x 轴相交，导致方法失效。公式出奇地简单，可以编入大多数科学计算器，进行反复迭代。
3. Geometric Interpretation | 几何解释
Geometrically, each step of the Newton-Raphson method consists of following the tangent line at the current point until it hits the x‑axis, and then using that intercept as the next approximation. Imagine a curve crossing the axis at α. Choose a starting x₀ close to α, draw the tangent, and find its x‑intercept x₁. Then repeat using x₁. The sequence of intercepts should zig‑zag towards the true root under favourable conditions.
几何上，牛顿-拉弗森方法的每一步就是沿着当前点处的切线前行，直到它与 x 轴相交，然后把那个交点作为下一个近似值。想象一条曲线在 α 处穿过 x 轴。选择一个靠近 α 的起点 x₀，画出切线，找到它的 x 截距 x₁。然后在 x₁ 处重复此过程。在有利条件下，截距序列会以之字形趋向真实的根。
4. Step-by-Step Procedure | 逐步执行过程
ion, follow these steps: (i) Rearrange the equation into the form f(x) = 0. (ii) Compute f' se a suitable starting value x₀, often given in the question. (iv) Substitute into x₁ = x₀ – f(x₀)/f' tions agree to the required degree of accuracy, or after a specified number of iterations.
在考题中应用该方法时，按以下步骤进行：(i) 把方程变形为 f(x) = 0 的形式。(ii) 求导得到 f'(x)。(iii) 选取合适的起始值 x₀，通常题目会给出。(iv) 代入 x₁ = x₀ – f(x₀)/f'(x₀)。(v) 用 x₁ 作为新值求 x₂，以此类推。当连续两次近似值在所需精度下一致时停止，或按要求完成指定次数的迭代。
5. Worked Example – Solving x³ – 2x – 5 = 0 | 例题 – 求解 x³ – 2x – 5 = 0
Let f(x) = x³ – 2x – 5. Then f'(x) = 3x² – 2. Starting with x₀ = 2 (f(2) = –1, f(2) is negative while f(3) = 16, so a root lies between 2 and 3).
令 f(x) = x³ – 2x – 5，则 f'(x) = 3x² – 2。取初始值 x₀ = 2（f(2) = –1 是负的，而 f(3) = 16 为正，因此有一个根在 2 和 3 之间）。
First iteration: x₁ = 2 – (–1) / (3×2² – 2) = 2 – (–1)/10 = 2.1.
第一次迭代：x₁ = 2 – (–1) / (3×2² – 2) = 2 – (–1)/10 = 2.1。
Second iteration: f(2.1) = 2.1³ – 4.2 – 5 = 9.261 – 9.2 = 0.061; f'(2.1) = 3×2.1² – 2 = 13.23 – 2 = 11.23. So x₂ = 2.1 – 0.061/11.23 ≈ 2.09457.
第二次迭代： f(2.1) = 2.1³ – 4.2 – 5 = 9.261 – 9.2 = 0.061； f'(2.1) = 3×2.1² – 2 = 13.23 – 2 = 11.23。因此 x₂ = 2.1 – 0.061/11.23 ≈ 2.09457。
Third iteration: f(2.09457) ≈ 0.00092, f' o 5 decimal places is 2.09455. The rapid convergence demonstrates the power of the method.
第三次迭代：f(2.09457) ≈ 0.00092，f'(2.09457) ≈ 11.161，得到 x₃ ≈ 2.09455。精确到 5 位小数的根为 2.09455。快速的收敛性展示了该方法的强大。
6. Conditions for Convergence | 收敛条件
Newton-Raphson does not always converge. For successful convergence, the starting guess x₀ should be sufficiently close to the true root α. Moreover, f'(x) must not be zero near the root, and f''(x) should not be too large relative to f'(x). In practice, if |f(xₙ)| is small and f'(xₙ) is not too small, the method usually works well.
牛顿-拉弗森法并非总能收敛。要成功收敛，初始猜测值 x₀ 必须足够接近真实根 α。此外，在根附近 f'(x) 不能为零，且 f''(x) 相对于 f'(x) 不应太大。实际中，若 |f(xₙ)| 很小而 f'(xₙ) 不太小，该方法通常表现良好。
7. When Newton-Raphson Fails – Pitfalls | 失效情形与陷阱
Common failure modes include: the tangent being horizontal (f' the sequence oscillating around a local extremum instead of moving towards the root; or the approximations diverging to infinity if the initial guess is on the wrong side of a turning point. Students must be able to identify such behaviour from a diagram or iteration table.
常见的失效模式包括：切线水平 (f'(xₙ)=0) 导致分母为零；序列围绕局部极值振荡而不趋向根；若初始猜测在转折点的错误一侧，近似值可能发散到无穷。学生必须能从图形或迭代表中识别这类行为。
8. Choosing a Good Starting Value | 选择良好的初始值
A good x₀ can often be found by evaluating f(x) at integer points and looking for a sign change, confirming the root lies between two values. Alternatively, exam questions typically provide a starting value. If you need to choose one yourself, pick a value where |f(x₀)| is relatively small and f'(x₀) is not close to zero. A rough sketch of the function helps immensely.
好的初始值 x₀ 通常可以通过计算整数点处的 f(x) 值并寻找符号变化来确定，证实根位于两个数值之间。此外，考题一般会直接给出起始值。若需自行选择，就选一个 |f(x₀)| 较小且 f'(x₀) 不接近零的值。粗略画出函数图形有很大帮助。
9. Comparison with Other Numerical Methods | 与其他数值方法的比较
ared with bisection, Newton-Raphson is much faster (quadratic convergence) but less robust— tion as x = g(x) and may converge more slowly. Newton-Raphson essentially chooses g(x) = x – f(x)/f'(x) and is highly efficient near simple roots.
在 Edexcel 大纲中，学生还会接触到变号法（区间二分法），有时还有割线法。与二分法相比，牛顿-拉弗森收敛速度更快（二次收敛），但稳定性较差——可能发散。不动点迭代将方程改写为 x = g(x)，收敛可能较慢。牛顿-拉弗森实质上选择了 g(x) = x – f(x)/f'(x)，在单根附近效率极高。
| Method | Speed | Reliability | Requirements |
|---|---|---|---|
| Bisection | Linear (slow) | Very reliable | Sign change over interval |
| Newton-Raphson | Quadratic (fast) | Can diverge | Derivative available, good x₀ |
| Fixed-point iteration | Linear (usually) | Conditional | |g'(α)| < 1 near root |
上表简要对比了三种常见数值方法。考试中可能要求解释为何在特定情境下选择牛顿法而非二分法。
10. Exam Tips and Common Mistakes | 考试技巧与常见错误
In an exam, always write down the formula explicitly before substituting values. Show each step clearly to gain method marks. A common mistake is misdifferentiating f(x) or forgetting to compute f'(x) correctly. Another error is rounding off intermediate values too early, which can propagate and cause the final answer to be inaccurate. Use your calculator's memory or ANS button to store values precisely. Also, when asked to 'explain why the method fails', refer to the geometry: horizontal tangent, division by zero, or oscillation.
考试中，务必在代入数值前先写出公式。每一步都清晰展示，以便获得方法分。常见的错误是求导 f(x) 出错，或忘记正确计算 f'(x)。另一个错误是过早舍入中间值，会导致误差传递，使最终答案不准。利用计算器的记忆功能或 ANS 键精确存储数值。此外，若要求“解释方法为何失效”，要联系几何意义：切线水平、除以零或发生振荡。
11. Using a Calculator for Iteration | 使用计算器进行迭代
culator with iterative capability. For Newton-Raphson, you can enter the formula xₙ₊₁ = xₙ – f(xₙ)/f') using the ANS variable. For example, type the initial guess, press equals, then type ANS – f(ANS)/f' to write down the displayed numbers to the required degree of accuracy, not the calculator's full floating-point output.
大多数 A-Level 考试允许使用具备迭代功能的现代科学计算器。针对牛顿-拉弗森方法，你可以利用 ANS 变量输入公式 xₙ₊₁ = xₙ – f(xₙ)/f'(xₙ)。例如，输入初始猜测值，按等号，然后键入 ANS – f(ANS)/f'(ANS) 并反复按等号。记录屏幕上出现的数值，达到所需精度后停止。注意按题目要求的准确度写下显示的数字，而不是计算器上的全部浮点数。
12. Summary and Key Takeaways | 总结与关键要点
t-based iterative technique for solving f(x)=0. It requires both f(x) and its derivative f' ugely valuable in mathematics and engineering. A-Level examiners test not only the mechanical use of the formula but also the understanding of convergence, failure, and geometric interpretation. Revise the worked example, remember the formula, and you will tackle any Newton-Raphson question with confidence.
牛顿-拉弗森方法是一种基于切线的快速迭代技术，用于求解 f(x)=0。它需要 f(x) 及其导数 f'(x)，以及一个良好的初始近似值。尽管它在驻点附近或初始选择较差时可能失效，但其二次收敛特性使它在数学和工程中极具价值。A-Level 考官不仅测试公式的机械运用，还考查对收敛性、失效情形和几何解释的理解。复习好例题，牢记公式，你就能信心十足地应对所有牛顿-拉弗森考题。
Published by TutorHao | 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