Exponential Modelling | 指数模型

📚 Exponential Modelling | 指数模型

rrent value. In A-Level Mathematics, you will learn to work with models of the form y = ae^kx and y = ab^x, apply logarithms to linearise data, estimate parameters, and interpret constants such as growth rate and half‑life. Mastering these techniques is essential for tackling modelling questions in the Edexcel exam.

指数模型在描述许多现实世界中量与其自身成正比增长或衰减的现象时非常关键。在 A‑Level 数学中，你将学习使用形如 y = ae^kx 和 y = ab^x 的模型，运用对数把数据线性化，估计参数，并解释如增长率和半衰期等常数。掌握这些方法对于应对 Edexcel 考试中的建模题至关重要。

1. Introduction to Exponential Models | 指数模型简介

lf. This leads to the differential equation dy/dx = k y, whose general solution is y = A e^kx, where A and k are constants. If k > 0, we have exponential growth; if k < 0, we have exponential decay.

当某个量的变化率与这个量本身成正比时，就可建立指数模型。由此得到微分方程 dy/dx = k y，其通解为 y = A e^kx，其中 A 和 k 是常数。若 k > 0，则为指数增长；若 k < 0，则为指数衰减。

In many practical contexts the independent variable is time t, so we write y = a e^kt is structure forms the backbone of the modelling questions you will encounter in Edexcel A‑Level Mathematics.

在许多实际情境中自变量是时间 t，因此常写作 y = a e^kt。初值为 y(0) = a。这一结构正是你在 Edexcel A‑Level 数学考试中会遇到的各种建模题的核心。

---

2. General Form: y = ae^kx | 一般形式 y = ae^kx

The most commonly used exponential model in the Edexcel specification is y = a e^kx xponential growth, k < 0 produces exponential decay, and k = 0 gives a constant function y = a.

Edexcel 考纲中最常用的指数模型为 y = a e^kx，其中 a ≠ 0 与 k 为实数常数。参数 a 给出了 x = 0 时 y 的值。k 的符号决定过程的性质：k > 0 产生指数增长，k < 0 产生指数衰减，而 k = 0 则退化为常值函数 y = a。

It is also common to see the model written as y = a b^x especially when dealing with discrete growth factors. These two forms are linked by b = e^k ould be comfortable converting between them, as exam questions may present either version.

模型也常写作 y = a b^x，尤其在处理离散增长因子时。这两种形式可通过 b = e^k 和 k = ln b 相互转换。你应当熟练在两者之间切换，因为考试题可能以任何一种形式出现。

y = a e^kx ⇔ y = a b^x, b = e^k, k = ln b

---

3. Understanding Growth and Decay | 理解增长与衰减

th rate: the quantity multiplies by the same factor over equal time intervals. For y = a e^kt with k > 0, the doubling time is constant. In contrast, exponential decay means the quantity reduces by the same proportion in each period, giving a constant half‑life when k < 0.

指数增长的特征是保持相对增长率恒定：在相等的时间间隔内，量总是乘以相同的倍数。对于 y = a e^kt 且 k > 0 的情况，翻倍时间是常数。相反，指数衰减意味着每个周期内量均以相同比例减少，当 k < 0 时具有恒定的半衰期。

Property	k > 0 (Growth)	k < 0 (Decay)
Behaviour as x → ∞	y → ∞	y → 0
Graph shape	Rising curve, increasing gradient	Falling curve, gradient approaching zero
Key constant	Doubling time Td = (ln 2)/k	Half‑life T1/2 = (ln 2)/|k|

中文：性质对比：k > 0 时，随着 x → ∞，y → ∞；图形上升且斜率增大；关键常数为翻倍时间 T_d = (ln 2)/k。k < 0 时，y → 0；图形下降且斜率趋于零；关键常数为半衰期 T_1/2 = (ln 2)/|k|。

---

4. The Role of e and Natural Logarithms | 自然常数 e 与自然对数的角色

Exponential models are intimately tied to the number e because the derivative of e^kx is k e^kx ural logarithms convert the model into a straight line: taking ln of both sides of y = a e^kx gives ln y = ln a + kx. The gradient is k and the vertical intercept is ln a.

指数模型与 e 紧密相关，因为 e^kx 的导数恰为其自身的 k 倍，这使得 e 成为微积分中最自然的底数。当我们对指数关系进行线性化时，自然对数将模型转化为直线：对 y = a e^kx 两边取自然对数得 ln y = ln a + kx。此时斜率为 k，纵截距为 ln a。

This is why Edexcel questions almost exclusively use natural logarithms (ln) rather than log base 10. You may be asked to plot ln y against x, draw a line of best fit, and then deduce a and k from the gradient and intercept.

正因如此，Edexcel 的考题几乎都使用自然对数 ln，而非以 10 为底的对数。你可能会被要求画出 ln y 对 x 的散点图，添加最佳拟合直线，然后由斜率和截距求出 a 和 k。

---

5. Linearising Exponential Data | 将指数数据线性化

rd technique is to take natural logarithms of the dependent variable. If the model y = a e^kx ient of this line estimates k, and the vertical intercept estimates ln a, from which a = e^intercept.

对于疑似遵循指数规律的数据，标准方法是取因变量的自然对数。若模型 y = a e^kx 成立，则 ln y (纵轴) 对 x (横轴) 的图应大致为一条直线。该直线的斜率即为 k 的估计值，纵截距为 ln a 的估计值，从而 a = e^截距。

Edexcel exam papers frequently provide a table of values and ask you to complete a table of ln y. You then plot the transformed data, fit a straight line, and use it to find the constants. Always label axes correctly, for example 'ln y' on the vertical axis and 'x' on the horizontal axis.

Edexcel 试卷常给出一个数值表，并要求你完成 ln y 的表格。然后你需绘制转换后的数据，拟合一条直线，并用它求出常数。务必正确标记坐标轴，例如纵轴标为 ln y，横轴标为 x。

---

广告 / Sponsored

6. Using Logarithms to Find Parameters | 利用对数求参数

ly on an exponential curve, you can find a and k without plotting. Substitute into y = a e^kx ing natural logarithms yields k = (ln y₂ − ln y₁) / (x₂ − x₁). Then a can be found by back‑substitution.

当给出精确位于指数曲线上的两个数据点 (x₁, y₁) 和 (x₂, y₂) 时，你可以不必作图便求出 a 和 k。将它们代入 y = a e^kx 得到两个方程，两式相除消去 a，再取自然对数得到 k = (ln y₂ − ln y₁) / (x₂ − x₁)。再回代即可求得 a。

If the data contain experimental errors, a better approach is to take natural logarithms of all y‑values, plot ln y against x, and use the gradient and intercept of a least‑squares regression line. This method is often tested in the statistics element of the course.

若数据含有实验误差，更可靠的方法是对所有 y 值取自然对数，作 ln y 对 x 的图，并使用最小二乘回归直线的斜率和截距。这一方法常在课程的统计部分被考查。

---

7. Half‑Life and Doubling Time | 半衰期与翻倍时间

Half‑ uired for a decaying exponential quantity to reduce to half its initial value. For y = a e^kt with k < 0, set y = a/2. Solving a/2 = a e^kt gives t = (ln 2) / |k|. The half‑life is independent of the initial amount a, which is a hallmark of exponential decay.

半衰期定义为衰变的指数量降到其初值一半所需的时间。对于 y = a e^kt 且 k < 0，令 y = a/2，解 a/2 = a e^kt 得 t = (ln 2) / |k|。半衰期与初始量 a 无关，这是指数衰减的标志性特点。

> 0), the doubling time is the time taken for the quantity to double. Solving 2a = a e^kt be able to derive these formulas and use them in context, for example with radioactive decay or bacterial growth.

对于指数增长 (k > 0)，翻倍时间是量翻一倍所需的时间。解 2a = a e^kt 得 t = (ln 2) / k。你应该能推导这些公式，并结合具体情境（例如放射性衰变或细菌繁殖）加以应用。

T_1/2 = ln 2 / |k| for decay; T_d = ln 2 / k for growth

---

8. Modelling Population Growth | 人口增长建模

A simple exponential model for population P at time t is P = P₀ e^rt, where P₀ is the initial population and r is the continuous relative growth rate. This model assumes unlimited resources, so it is most accurate over short time periods. In Edexcel questions you might be asked to predict a future population or estimate the time needed to reach a certain size.

人口数量 P 关于时间 t 的简单指数模型为 P = P₀ e^rt，其中 P₀ 为初始人口，r 为连续相对增长率。该模型假设资源无限，因此在较短时间内最为准确。在 Edexcel 题目中，你可能需要预测未来人口，或估计达到某个规模所需的时间。

For example, if a population grows at 2% 7 years. Always interpret r correctly: a value like 0.02 means a continuous growth rate of 2% per unit time.

例如，若某人口以每年连续增长 2%，则 r = 0.02。翻倍时间为 (ln 2)/0.02 ≈ 34.7 年。务必正确理解 r：形如 0.02 意味着单位时间内连续增长率为 2%。

---

9. Radioactive Decay Modelling | 放射性衰减建模

mass m of a radioactive substance that decays exponentially is often modelled as m = m₀ e^−λt, where λ > 0 is the decay constant. The half‑life is then T = ln 2 / λ. Students should be able to find λ given the half‑life, and vice versa, and use the model to compute residual mass after a given time.

放射性物质的指数衰减质量 m 通常建模为 m = m₀ e^−λt，其中 λ > 0 为衰减常数。半衰期为 T = ln 2 / λ。学生应能根据半衰期求 λ，或反之，并利用该模型计算经过一定时间后的残余质量。

Exam questions often provide a table of count rates or masses and ask you to confirm an exponential relationship. By plotting ln m against t, you should obtain a straight line with gradient −λ. This linear relationship is a powerful check on the model's validity.

考试题常给出计数率或质量的数据表，并要求你验证指数关系。通过绘制 ln m 对 t 的图，你应该得到一条斜率为 −λ 的直线。该线性关系是检验模型有效性的有力手段。

---

10. Compound Interest and Continuous Growth | 复利与连续增长

In finance, the value of an investment with annual compounding is A = P (1 + r/n)^nt. As n → ∞, this becomes A = P e^rt tly an exponential model. The continuous growth rate r here plays the role of k in y = a e^kx. Edexcel may link financial contexts to exponential modelling.

金融领域中，按年复利计算的投资价值为 A = P (1 + r/n)^nt。当 n → ∞ 时，它就变成 A = P e^rt，这正是一个指数模型。这里的连续增长率 r 就扮演了 y = a e^kx 中 k 的角色。Edexcel 可能将金融情境与指数建模相结合。

You should be able to move between discrete and continuous compounding. For a given annual interest rate, the equivalent continuous rate r satisfies e^r rate. This is a useful skill for questions that blur the lines between pure and applied mathematics.

你应能在离散复利和连续复利之间转换。对于给定的年利率，等效连续复利 r 满足 e^r = 1 + i，其中 i 为有效年利率。对于跨越纯数学和应用数学界限的题目，这是一项有用的技能。

---

11. Practical Example: Data Fitting | 实例：数据拟合

with corresponding y = 4.8, 7.9, 13.0, 21.4, 35.2. Suspecting an exponential model y = a e^kx gradient m ≈ (3.561 − 1.569)/4 = 0.498, so k ≈ 0.498. The intercept c ≈ 1.569, hence a ≈ e¹.569 ≈ 4.80. The model is y ≈ 4.80 e^0.498x.

考虑一组实验数据：x = 0, 1, 2, 3, 4，对应的 y = 4.8, 7.9, 13.0, 21.4, 35.2。猜测模型为 y = a e^kx，我们先计算 ln y：1.569, 2.067, 2.565, 3.063, 3.561。绘制 ln y 对 x 的图，点几乎落在一条直线上。斜率 m ≈ (3.561 − 1.569)/4 = 0.498，故 k ≈ 0.498。截距 c ≈ 1.569，因此 a ≈ e¹.569 ≈ 4.80。模型为 y ≈ 4.80 e^0.498x。

This model can then be used to predict y at x = 5: y ≈ 4.80 e^0.498×5 ≈ 4.80 × e² ng an exam question, always show the linearisation steps, give the equation of the line, and clearly state the final exponential model. Also comment on the reliability of extrapolation.

该模型随后可用于预测 x = 5 时的 y：y ≈ 4.80 e^0.498×5 ≈ 4.80 × e².49 ≈ 4.80 × 12.06 ≈ 57.9。在考试作答时，务必展示线性化的步骤，给出直线方程，并清晰陈述最终的指数模型。同时

Published by TutorHao | A-Level Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）