Exponential Models | 指数模型

📚 Exponential Models | 指数模型

Exponential models describe situations where a quantity changes at a rate proportional to its current value. They appear in population growth, radioactive decay, compound interest, and the cooling of objects. In Edexcel A-Level Mathematics, you learn to formulate, solve and interpret these models using the exponential function eˣ and the associated differential equations dy/dx = ky. Understanding the properties of exponentials and logarithms is essential for linking data to the underlying growth or decay constant.

指数模型描述的是某个量以其当前值的一定比例发生变化的情形。这些模型出现在人口增长、放射性衰变、复利和物体冷却过程中。在 Edexcel A-Level 数学中，你将学习利用指数函数 eˣ 以及相关微分方程 dy/dx = k y 来建立、求解并解释这些模型。理解指数函数和对数的性质，是将数据与潜在的生长或衰减常数联系起来的关键。

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

An exponential model describes a quantity that grows or decays by a fixed percentage per unit time. The key feature is that the rate of change is proportional to the current amount. This leads to a function of the form y = a · bˣ or, more commonly in calculus-based work, y = a eᵏˣ. Exponential models are distinct from linear models, where the absolute change is constant, and from polynomial models, where the rate of change itself changes in a non-proportional way.

指数模型描述的是一个量在单位时间内按固定百分比增长或衰减的情况。其关键特征是变化率与当前量成正比。这就导出了形如 y = a · bˣ 的函数，或者在微积分中更常用的 y = a eᵏˣ。指数模型不同于绝对变化为常数的线性模型，也不同于变化率本身以非比例方式变化的多项式模型。

Examples include the growth of a bacterial colony where each cell divides at a constant rate, the decay of a radioactive isotope where each nucleus has a fixed probability of decay per second, and the temperature change of a hot object placed in a cooler environment. Recognising an exponential pattern in a table of data often involves noticing that the ratio of successive y‑values is constant for equal increments in x.

例子包括细菌菌落以恒定速率分裂的生长、放射性同位素中每个原子核每秒以固定概率发生的衰变，以及热物体在较冷环境中的温度变化。在一组数据表中识别指数模式，通常需要注意到在 x 等量增加时，相邻 y 值的比值是恒定的。

2. The Exponential Function eˣ | 指数函数 eˣ

The natural exponential function, denoted eˣ or exp(x), is the unique function that equals its own derivative. That is, if f(x) = eˣ, then f'(x) = eˣ. The number e is approximately 2.71828 and it appears throughout mathematics whenever continuous growth or decay is modelled. Unlike exponential expressions with an arbitrary base b, eˣ simplifies differentiation and integration, making it the preferred base in A‑Level modelling.

自然指数函数记作 eˣ 或 exp(x)，它是唯一一个导数等于自身的函数。也就是说，若 f(x) = eˣ，则 f'(x) = eˣ。数字 e 约等于 2.71828，在数学中只要涉及连续增长或衰减，就会出现 e。与任意底数 b 的指数式不同，eˣ 能简化微分和积分，因此成为 A‑Level 建模中的首选底数。

Using e allows us to express any exponential relationship y = a bˣ in the form y = a eˡⁿ⁽ᵇ⁾ˣ, since b = eˡⁿ ᵇ. This conversion is fundamental when solving differential equations or when linearising data through logarithmic transformations. The derivative of y = a eᵏˣ is dy/dx = a k eᵏˣ = k y, directly linking to the proportionality constant k.

利用 e，我们可以将任何指数关系 y = a bˣ 表示成 y = a eˡⁿ⁽ᵇ⁾ˣ 的形式，因为 b = eˡⁿ ᵇ。在求解微分方程或通过对数变换将数据线性化时，这一转换至关重要。y = a eᵏˣ 的导数为 dy/dx = a k eᵏˣ = k y，直接与比例常数 k 相关联。

3. Exponential Growth and Decay | 指数增长与衰减

Exponential growth occurs when the proportionality constant k is positive. The quantity increases over time and the gradient of the tangent at any point is proportional to the y‑value. Exponential decay corresponds to k < 0, where the quantity decreases and approaches zero asymptotically. In both cases, the graph either rises from or falls towards the x‑axis, never touching it.

当比例常数 k 为正时，出现指数增长。量随时间增加，且任一点处切线的斜率与 y 值成正比。指数衰减对应于 k < 0，此时量逐渐减少并渐近于零。无论是增长还是衰减，图像都是从 x 轴上升起或向 x 轴下降，但永远不会触及 x 轴。

In many real‑world processes, growth cannot continue indefinitely; the exponential model is often valid only for a limited time range. Decay, however, may accurately describe radioactive substances over many half‑lives. The table below summarises the sign of k and the behaviour of the model.

在许多真实过程中，增长并不会无限持续；指数模型通常只在有限的时间范围有效。然而，衰减可以在很多半衰期内精确地描述放射性物质。下表总结了 k 的符号以及模型的行为。

k > 0	Growth, y increases	增长，y 增大
k < 0	Decay, y decreases	衰减，y 减小
k = 0	Constant, y unchanged	恒定，y 不变

4. The General Form y = a eᵏˣ | 一般形式 y = a eᵏˣ

In Edexcel examinations, the standard exponential model is written as y = a eᵏˣ (or y = a eᵏᵗ with time t). The parameter a represents the initial value when x = 0, because e⁰ = 1. The constant k governs the speed of growth or decay: a larger |k| means faster change. If the model has a non‑zero starting point at x = x₀, the form becomes y = a eᵏ⁽ˣ⁻ˣ⁰⁾, but the zero‑shifted form is rarely tested in pure contexts.

在 Edexcel 考试中，标准指数模型写为 y = a eᵏˣ（或以时间 t 表示为 y = a eᵏᵗ）。参数 a 代表 x = 0 时的初始值，因为 e⁰ = 1。常数 k 支配增长或衰减的速度：|k| 越大，变化越快。如果模型在 x = x₀ 处有非零起点，则形式变为 y = a eᵏ⁽ˣ⁻ˣ⁰⁾，但在纯数情境中很少考查这种平移形式。

When analysing data, it is common to take natural logarithms of both sides: ln y = ln a + k x. This transforms the exponential model into a linear relationship where ln y is plotted against x. The slope of the resulting line is k, and the vertical intercept is ln a. This linearisation is crucial for estimating k and a from experimental data.

在分析数据时，通常对两边取自然对数：ln y = ln a + k x。这将指数模型转化为线性关系，其中以 ln y 对 x 作图。所得直线的斜率为 k，纵截距为 ln a。这种线性化对于从实验数据中估计 k 和 a 至关重要。

5. From Differential Equations: dy/dx = k y | 来自微分方程：dy/dx = k y

Many exponential models arise from the simple differential equation dy/dx = k y, which states that the rate of change of y is proportional to y itself. This is solved by separation of variables: ∫ (1/y) dy = ∫ k dx, giving ln |y| = k x + C, and hence y = A eᵏˣ where A = ±eᶜ. If an initial condition y(0) = y₀ is given, the solution becomes y = y₀ eᵏˣ.

许多指数模型源于简单的微分方程 dy/dx = k y，该方程表明 y 的变化率与 y 本身成正比。可通过分离变量法求解：∫ (1/y) dy = ∫ k dx，得到 ln |y| = k x + C，进而有 y = A eᵏˣ，其中 A = ±eᶜ。若给定初始条件 y(0) = y₀，解则为 y = y₀ eᵏˣ。

In modelling, k can be determined from additional data. For example, if a population doubles from 500 to 1000 in 3 hours, we set up 1000 = 500 e³ᵏ, leading to k = (ln 2)/3. This hands‑on link between differential equations and real behaviour is a key skill in A‑Level Mathematics, especially in the applied context of mechanics or population dynamics.

在建模中，k 可以由附加数据确定。例如，如果一个种群在 3 小时内从 500 增加一倍到 1000，我们建立方程 1000 = 500 e³ᵏ，从而解得 k = (ln 2)/3。这种微分方程与实际行为之间的直接联系是 A‑Level 数学中的一项关键技能，尤其是在力学或种群动力学的应用背景中。

6. Determining Parameters from Data | 从数据确定参数

Given a table of experimental observations, the parameters a and k in the model y = a eᵏˣ can be found by taking natural logarithms and applying linear regression. First, add a column for ln y. Then plot ln y against x; the points should lie approximately on a straight line if the exponential model is appropriate. The slope gives k, and the intercept gives ln a, so a = eⁱⁿᵗᵉʳᶜᵉᵖᵗ.

给定实验观测数据表，模型 y = a eᵏˣ 中的参数 a 和 k 可以通过取自然对数并应用线性回归求得。首先，添加一列 ln y。然后以 ln y 对 x 作图；如果指数模型合适，这些点应近似落在一条直线上。斜率给出 k，截距给出 ln a，因此 a = eⁱⁿᵗᵉʳᶜᵉᵖᵗ。

It is important to check that the transformed data exhibits a linear trend; systematic curvature suggests that a simple exponential may not be the best model. In examination questions, you may be given a graph of ln y against x and asked to find k and a from the gradient and intercept, or to write down the full exponential equation.

务必检验变换后的数据是否呈现线性趋势；若存在系统性弯曲，则表明简单指数模型可能不是最佳选择。在考试题中，可能会给出 ln y 对 x 的图像，要求你根据斜率和截距求出 k 和 a，或者写出完整的指数方程。

Worked example: For data points (0, 5.0), (1, 6.1), (2, 7.4), (3, 9.0), calculate ln y values. The slope between first and last points is (ln9.0 – ln5.0)/3 ≈ (2.197 – 1.609)/3 = 0.196. So k ≈ 0.196, and a = 5.0 since y = a at x = 0. The model is y = 5.0 e⁰·¹⁹⁶ˣ.

例题：对于数据点 (0, 5.0)、(1, 6.1)、(2, 7.4)、(3, 9.0)，计算 ln y 值。首末点之间的斜率为 (ln9.0 – ln5.0)/3 ≈ (2.197 – 1.609)/3 = 0.196。因此 k ≈ 0.196，且由于 x=0 时 y=a，故 a=5.0。模型为 y = 5.0 e⁰·¹⁹⁶ˣ。

7. Half-Life and Doubling Time | 半衰期与倍增时间

For exponential decay, the half‑life t₁/₂ is the time required for the quantity to reduce to half its initial value. Setting y = ½ a in y = a eᵏˣ (with k negative) gives eᵏᵗ = ½. Taking natural logs yields k t₁/₂ = ln(½) = –ln2, so t₁/₂ = ln2 / |k|. This half‑life is constant, independent of the initial amount, which is a defining property of exponential decay.

对于指数衰减，半衰期 t₁/₂ 是量减少到初始值一半所需的时间。在 y = a eᵏˣ（k 为负）中令 y = ½ a，得 eᵏᵗ = ½。取自然对数得 k t₁/₂ = ln(½) = –ln2，因此 t₁/₂ = ln2 / |k|。该半衰期为常数，与初始量无关，这是指数衰减的一个定义性特征。

For exponential growth, the doubling time T₂ is the time for the quantity to double. Setting y = 2a gives eᵏᵀ² = 2, so T₂ = ln2 / k. Thus the same constant ln2 appears in both contexts, with k positive for doubling and negative for half‑life calculations.

对于指数增长，倍增时间 T₂ 是量增加一倍所需的时间。令 y = 2a 得 eᵏᵀ² = 2，因此 T₂ = ln2 / k。同一常数 ln2 在两种情形下均会出现，只是增长时 k 为正，半衰期计算时 k 为负。

These concepts are regularly examined in A‑Level Maths, often by asking you to find k from a known half‑life or doubling time and then predict future values. For instance, if a radioactive substance has a half‑life of 8 days, then |k| = ln2/8 ≈ 0.0866 day⁻¹, and the decay model is y = y₀ e⁻⁰·⁰⁸⁶⁶ᵗ.

这些概念在 A‑Level 数学中经常考查，通常要求根据已知的半衰期或倍增时间求出 k，然后预测未来值。例如，若某放射性物质的半衰期为 8 天，则 |k| = ln2/8 ≈ 0.0866 天⁻¹，衰减模型为 y = y₀ e⁻⁰·⁰⁸⁶⁶ᵗ。

8. Modelling Population Growth | 人口增长建模

Simple population models assume that the rate of population increase is proportional to the current population, leading to dP/dt = k P. The solution is P(t) = P₀ eᵏᵗ. This model works well for short‑term growth of bacteria or yeast under unlimited resources. The constant k is often called the intrinsic growth rate and can be estimated from census data over regular intervals.

简单的人口模型假设种群增长速率与当前人口数成正比，从而得到 dP/dt = k P。解为 P(t) = P₀ eᵏᵗ。该模型适用于资源无限条件下细菌或酵母的短期增长。常数 k 通常被称为内禀增长率，可通过定期人口普查数据估算。

In practice, exponential growth cannot be sustained due to limited food, space or other factors. Nevertheless, the model provides a baseline and is often the first step before introducing more realistic logistic models. In A‑Level questions, you may be told that a population doubles every T years and asked to find the population after a certain time, or to determine the growth rate k.

实际上，由于食物、空间或其他因素的限制，指数增长不可能持续。尽管如此，该模型提供了一个基准，通常是引入更实际的逻辑模型前的第一步。在 A‑Level 试题中，可能会告知某人口每 T 年翻一番，要求计算某一时间后的人口数，或求出增长率 k。

Example: A bacteria culture starts with 200 cells and triples every 4 hours. Find the population after 10 hours. First, triple means k = (ln3)/4. Then P(10) = 200 e^(10·ln3/4) = 200 · 3^(10/4) = 200 · 3²·⁵ ≈ 200 × 15.59 ≈ 3118 cells. Always use exact expressions first, then approximate.

例题：一个细菌培养物初始有 200 个细胞，每 4 小时增至三倍。求 10 小时后的种群数量。首先，三倍意味着 k = (ln3)/4。然后 P(10) = 200 e^(10·ln3/4) = 200 · 3^(10/4) = 200 · 3²·⁵ ≈ 200 × 15.59 ≈ 3118 个细胞。务必先使用精确表达式，再取近似值。

9. Radioactive Decay | 放射性衰变

Radioactive substances decay according to the exponential law N = N₀ e⁻ᵏᵗ, where N is the number of undecayed nuclei at time t, and λ (or k) is the decay constant. The activity, which is the rate of decay –dN/dt, is also exponential: A = λ N₀ e⁻ᵏᵗ. The half‑life t₁/₂ = ln2 / λ is a characteristic property of each isotope.

放射性物质按照指数规律 N = N₀ e⁻ᵏᵗ 衰变，其中 N 为时刻 t 未衰变的原子核数，λ（或 k）是衰变常数。活度即衰变速率 –dN/dt，也是指数形式：A = λ N₀ e⁻ᵏᵗ。半衰期 t₁/₂ = ln2 / λ 是每种同位素的特征性质。

In Edexcel questions, you may encounter graphs of ln N against t, which should yield a straight line with negative slope –k. You could be asked to determine the age of a sample using the ratio N/N₀, often in the context of carbon‑14 dating. If 60% of the original carbon‑14 remains, then 0.60 = e⁻ᵏᵗ, and using the known half‑life of 5730 years, t can be found.

在 Edexcel 考题中，可能会遇到 ln N 对 t 的图像，应得到一条斜率为 –k 的直线。你可能需要利用比值 N/N₀ 确定样本的年代，常出现在碳‑14 定年的情境中。若原始碳‑14 还剩 60%，则 0.60 = e⁻ᵏᵗ，利用已知半衰期 5730 年，便可求出 t。

10. Newton’s Law of Cooling | 牛顿冷却定律

Newton’s law of cooling states that the rate of change of the temperature of an object is proportional to the difference between its temperature and the ambient temperature. This gives the differential equation dT/dt = –k (T – Tₐ), where Tₐ is the constant surrounding temperature. The solution is T(t) = Tₐ + (T₀ – Tₐ) e⁻ᵏᵗ, which shows exponential decay of the temperature difference.

牛顿冷却定律指出，物体温度的变化率与其自身温度和环境温度之差成正比。由此得到微分方程 dT/dt = –k (T – Tₐ)，其中 Tₐ 为恒定的环境温度。其解为 T(t) = Tₐ + (T₀ – Tₐ) e⁻ᵏᵗ，体现了温差呈指数衰减。

This model is widely used in forensic science to estimate time of death, in engineering for predicting cooling of components, and in everyday situations such as a cup of tea cooling. Graphically, the temperature approaches Tₐ asymptotically. The constant k depends on the object’s surface area, insulation and other physical properties.

该模型广泛用于法医学中推断死亡时间、工程中预测部件冷却，以及像一杯茶冷却这类日常情景。从图像上看，温度渐近于 Tₐ。常数 k 取决于物体的表面积、隔热情况和其他物理性质。

11. Solving Exponential Equations | 解指数方程

In modelling contexts, you often need to solve an equation of the form a eᵏˣ = b for the unknown x. Taking natural logarithms both sides gives ln a + k x = ln b, so x = (ln b – ln a)/k. This is essential for predicting when a population reaches a certain size or when a decaying substance falls to a safe level.

在建模情境中，常需就未知数 x 求解形如 a eᵏˣ = b 的方程。两边取自然对数得 ln a + k x = ln b，因此 x = (ln b – ln a)/k。这对于预测种群何时达到特定规模或衰减物质何时降至安全水平至关重要。

When dealing with exponential expressions with different bases, such as 3ˣ = 5(2ˣ), you can take logarithms of both sides. Using natural logs, x ln3 = ln5 + x ln2, then x (ln3 – ln2) = ln5, yielding x = ln5 / (ln3 – ln2). Mastery of logarithmic properties is assumed and tested regularly.

当处理底数不同的指数式时，如 3ˣ = 5(2ˣ)，可对两边取对数。利用自然对数，得 x ln3 = ln5 + x ln2，进而 x (ln3 – ln2) = ln5，于是 x = ln5 / (ln3 – ln2)。熟练掌握对数性质是必备技能，并会经常被考查。

In some questions, a quadratic in eˣ may appear, for instance e²ˣ – 5 eˣ + 6 = 0. Let u = eˣ, then u² – 5u + 6 = 0, so u = 2 or 3, giving x = ln2 or x = ln3. This technique links exponential models with algebraic manipulation.

在某些问题中，可能出现关于 eˣ 的二次方程，例如 e²ˣ – 5 eˣ + 6 = 0。令 u = eˣ，则 u² – 5u + 6 = 0，故 u = 2 或 3，得到 x = ln2 或 x = ln3。这种技巧将指数模型与代数操作联系起来。

12. Graphical Transformations of Exponential Functions | 指数函数的图像变换

Understanding transformations of y = eˣ helps interpret models with parameters. The graph of y = a eˣ is a vertical stretch by factor a. The graph of y = eᵏˣ is a horizontal stretch: if k > 1 it compresses horizontally, if 0 < k < 1 it stretches horizontally. A negative k reflects the graph in the y‑axis, turning growth into decay.

理解 y = eˣ 的变换有助于解释带有参数的模型。y = a eˣ 的图像是竖直方向拉伸 a 倍。y = eᵏˣ 的图像则是水平方向拉伸：若 k > 1，则水平压缩；若 0 < k < 1，则水平伸展。负的 k 会使图像关于 y 轴反射，将增长变为衰减。

Vertical shifts are not appropriate for pure exponential models because y = eˣ + c would no longer satisfy the proportionality dy/dx = k y. However, in cases like Newton’s cooling, the model is y = Tₐ + (T₀ – Tₐ) e⁻ᵏᵗ, which is a vertical shift of a decaying exponential. This combination of transformations is important for sketching and interpreting real‑world models.

竖直方向的平移不适用于纯指数模型，因为 y = eˣ + c 不再满足比例关系 dy/dx = k y。不过，在像牛顿冷却定律这样的情况中，模型为 y = Tₐ + (T₀ – Tₐ) e⁻ᵏᵗ，它就是一个衰减指数函数的竖直平移。这种变换组合对于绘制草图以及解释现实模型非常重要。

When using exponential graphs to solve equations, remember that eˣ is always positive, and as x → –∞, eˣ → 0. This asymptotic behaviour sets bounds for growth or decay models and helps avoid unrealistic predictions.

当利用指数图像求解方程时，要记住 eˣ 恒为正，且当 x → –∞ 时，eˣ → 0。这种渐近行为为增长或衰减模型设定了界限，有助于避免不切实际的预测。

Published by TutorHao | Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Exponential Models | 指数模型