📚 13C Exponential Models | 13C 指数模型
ey describe processes where a quantity changes at a rate proportional to its current value – ers, and real-world meaning. This article covers everything from the basic structure y = a ⋅ bˣ to natural exponential functions, half‑life, doubling time, logarithmic transformation, and regression on a GDC.
指数模型是应用数学中最强大的工具之一。它们描述的是量的变化率与其当前值成正比的许多过程——从人口增长、病毒传播到放射性衰变和冷却过程。在 IB 数学中，掌握指数模型意味着理解代数形式、图像、参数与现实意义之间的联系。本文涵盖从基本结构 y = a ⋅ bˣ 到自然指数函数、半衰期、倍增时间、对数变换以及用图形计算器做指数回归的所有内容。
1. What is an Exponential Model? | 什么是指数模型？
An exponential model is a function of the form f(x) = a ⋅ bˣ (or equivalently f(x) = a ⋅ ekx onstant, and a is the initial value when x = 0. This structure leads to a characteristic J‑ for decay. Unlike linear models, the rate of change itself grows or shrinks exponentially.
指数模型是指自变量出现在指数位置的函数，形式为 f(x) = a ⋅ bˣ（或等价地 f(x) = a ⋅ ekx）。底数 b 是一个正常数，a 是 x = 0 时的初始值。这种结构导致增长时呈 J 形曲线，衰减时则迅速下降。与线性模型不同，其变化率本身也呈指数增长或缩减。
In the IB syllabus, you encounter exponential models in contexts such as compound interest (AI), population ecology (AI/AA), radioactive decay, and Newton's law of cooling. Recognising when a situation calls for an exponential model is a key skill: if a quantity increases by a fixed percentage per unit time, the process is exponential.
在 IB 课程中，你会在复利（AI 课程）、种群生态学（AI/AA 通用）、放射性衰变和牛顿冷却定律等情境中遇到指数模型。识别何时需要使用指数模型是一项关键技能：如果某个量每单位时间按固定百分比增加，那么该过程就是指数的。
2. The Basic Form y = a ⋅ bˣ | 基本形式 y = a ⋅ bˣ
The simplest exponential model is y = a ⋅ bˣ, where a is the initial value (y‑ meter b is never negative because that would cause the function to oscillate and lose real‑world meaning for most applications.
最简单的指数模型是 y = a ⋅ bˣ，其中 a 是初始值（y 轴截距），b 是增长因子。若 b > 1，函数代表指数增长；若 0 < b < 1，则代表指数衰减。参数 b 永远不会为负，否则函数会发生振荡，并在大多数应用中失去实际意义。
acterial colony starting with 200 cells and doubling every hour can be modelled by y = 200 ⋅ 2ˣ, with x in hours. When x = 3, y = 200 ⋅ 2³ = 1600 cells. The base b directly encodes the multiplier per unit change in x: a 3% increase per year corresponds to b = 1.03.
例如，一个从 200 个细胞开始、每小时翻倍的细菌菌落可用 y = 200 ⋅ 2ˣ 建模，x 以小时为单位。当 x = 3 时，y = 200 ⋅ 2³ = 1600 个细胞。底数 b 直接编码了每单位 x 变化的倍数：每年增长 3% 对应 b = 1.03。
3. Natural Exponential Function y = a ⋅ ekx | 自然指数函数 y = a ⋅ ekx
≈ 2.71828 has unique calculus properties, exponential models are often expressed as y = a ⋅ ekx ositive k gives growth; negative k gives decay. The relationship between b and k is: b = ek, or equivalently k = ln b.
由于自然指数底数 e ≈ 2.71828 具有独特的微积分性质，指数模型常写为 y = a ⋅ ekx。这里 k 是连续增长率。k 为正表示增长，k 为负表示衰减。b 与 k 的关系为：b = ek，或等价地 k = ln b。
ly useful when solving differential equations, because dy/dx = k y directly yields y = a ekx B, you will use this form in contexts like continuously compounded interest, where A = P ert, and in physics for decay where N = N₀ e−λt.
在求解微分方程时这种形式尤其有用，因为 dy/dx = k y 直接给出 y = a ekx。在 IB 中，你会在连续复利（A = P ert）以及物理衰变（N = N₀ e−λt）等情境中使用此形式。
4. Growth vs Decay: Interpreting Parameters | 增长与衰减：参数解释
uantity never actually reaches zero in finite time, though it may become negligibly small.
区分增长和衰减很简单：观察底数 b 或 k 的符号。若 b > 1（或 k > 0），模型增长；若 0 < b < 1（或 k < 0），模型衰减。水平渐近线为 y = 0，意味着在有限时间内该量永远不会真正达到零，尽管它可能变得微乎其微。
The parameter a always represents the initial value at x = 0. In real‑ ond to the starting population, initial investment, original mass, etc. The graph of y = a ⋅ bˣ ugh (0, a) and, depending on b, either rises rapidly or falls towards zero as x increases.
参数 a 始终表示 x = 0 时的初始值。在现实问题中，a 可对应初始种群、初始投资、原始质量等。y = a ⋅ bˣ 的图像经过 (0, a)，并且随着 x 增大，依据 b 的不同，要么迅速上升，要么下降趋向于零。
5. Half-Life and Doubling Time | 半衰期与倍增时间
Two key intervals characterise exponential change: half‑life (T₁⁄₂) for decay and doubling time for growth. Half‑life is the time taken for a quantity to reduce to half its original value; doubling time is the time required for a quantity to double. These periods are constant, independent of the starting value – a hallmark of exponential behaviour.
两个关键时间间隔刻画了指数变化：衰减的半衰期（T₁⁄₂）和增长的倍增时间。半衰期是某个量减少到原始值一半所需的时间；倍增时间是某个量翻倍所需的时间。这些周期是恒定的，与起始值无关——这是指数行为的一个标志。
Using the model N = N₀ e−λt, the half‑life is found by setting N = N₀/2: e−λ T = 1/2, so T = (ln 2)/λ. For the doubling time in y = a ⋅ 2t/d, d is the doubling time. Converting between half‑life/doubling time and the decay/growth constant is a routine IB assessment skill.
使用模型 N = N₀ e−λt，令 N = N₀/2 可得半衰期：e−λ T = 1/2，故 T = (ln 2)/λ。对于 y = a ⋅ 2t/d 中的倍增时间，d 即为倍增时间。在半衰期/倍增时间与衰减/增长常数之间进行换算是 IB 考试中的常规技能。
6. Modelling Real-World Data | 真实世界数据建模
Given a data set, the first step is to recognise an exponential trend. If successive y‑values change by roughly the same ratio when x increases by a fixed step, an exponential model is appropriate. This can be spotted in a table or by plotting log y against x and observing a straight line.
给定一组数据，第一步是识别指数趋势。如果当 x 按固定步长增加时，相邻 y 值的变化比例大致相同，那么指数模型就是合适的。这可以从表格中发现，也可以通过绘制 log y 对 x 的图像并观察是否得到一条直线来判断。
onential behaviour is confirmed, you can find the parameters a and b by selecting two well‑spaced data points. For example, with points (0, 50) and (4, 405), solving 50 ⋅ b⁴ = 405 gives b = (8.1)¼ ≈ 1.8. Alternatively, use your GDC's exponential regression function (ExpReg) to obtain the best‑fit model in the form y = a ⋅ bˣ or y = a ⋅ ekx.
一旦确认指数行为，你可以通过选取两个间隔较远的数据点来求参数 a 和 b。例如，有数据点 (0, 50) 和 (4, 405)，解 50 ⋅ b⁴ = 405 可得 b = (8.1)¼ ≈ 1.8。或者，使用图形计算器（GDC）的指数回归功能（ExpReg）得到形如 y = a ⋅ bˣ 或 y = a ⋅ ekx 的最优拟合模型。
7. Using Logarithms to Linearise | 利用对数线性化
Exponential equations are inherently non‑ ithm (or log base 10) of both sides transforms them into a linear relationship. If y = a ekx then ln y = ln a + k x. Plotting ln y against x yields a straight line with slope k and y‑ amental in experimental data analysis and is explicitly tested in IB Internal Assessments.
指数方程本是非线性的，但对两边取自然对数（或以 10 为底的对数）可将其转化为线性关系。若 y = a ekx，则 ln y = ln a + k x。以 ln y 对 x 作图可得一条斜率为 k、截距为 ln a 的直线。这一技巧在实验数据分析中至关重要，也是 IB 内部评估明确考查的内容。
Similarly, for y = a ⋅ bˣ, taking log base 10 gives log y = log a + x log b. The slope of the log‑ nd then convert back to the exponential form. It also provides a visual check of the model's appropriateness.
类似地，对 y = a ⋅ bˣ 取以 10 为底的对数得 log y = log a + x log b。对数图像的斜率为 log b。这种线性化方法可让你借助简单线性回归提取 a 和 b，再回代到指数形式。它同时提供了模型适用性的视觉检验手段。
8. Exponential Regression on a GDC | 图形计算器上的指数回归
via STAT > CALC > ExpReg or similar). The calculator outputs either coefficients for y = a ⋅ bˣ or for y = a ⋅ ekx, depending on the command. Always note which form is given and be ready to convert if the question demands a specific base.
现代 IB 考试要求熟练使用图形计算器。将双变量数据输入列表后，你可以执行指数回归（通常通过 STAT > CALC > ExpReg 或类似路径访问）。计算器输出的是 y = a ⋅ bˣ 或 y = a ⋅ ekx 的系数，具体取决于所用指令。务必留意给出的形式，并做好在题目要求特定底数时进行转换的准备。
Pay attention to the correlation coefficient r or the coefficient of determination R². Values close to 1 indicate a strong exponential fit. If the GDC returns a negative base error, it means the data does not follow a pure exponential trend, and you may need to consider a shifted exponential model or transform the data.
注意相关系数 r 或决定系数 R²。接近 1 的值表明指数拟合程度很高。如果计算器返回负底数错误，意味着数据并不遵循纯指数趋势，你可能需要考虑平移指数模型或对数据进行变换。
9. Applications: Population Growth | 应用：人口增长
Unrestricted population growth is often modelled by the exponential equation P = P₀ ert, where P₀ is the initial population and r is the intrinsic growth rate. For short time scales, this model works well, such as bacteria in a Petri dish or yeast in fermentation. It is also the basis for modelling human population trends before resource limitations become significant.
不受限制的人口增长常用指数方程 P = P₀ ert 建模，其中 P₀ 为初始人口数，r 为内禀增长率。在较短时间尺度上该模型效果良好，例如培养皿中的细菌或发酵过程中的酵母。它也是在资源限制变得显著之前模拟人类人口趋势的基础。
n size, or compute the growth rate from two data points. Setting up the equation P₂ = P₁ erΔt forward. Many IB questions also link growth to doubling time via the relation r = ln 2 / Tdouble.
解题时，你可能需要预测未来人口、求出达到特定规模所需的时间，或根据两个数据点计算增长率。建立方程 P₂ = P₁ erΔt 并求解 r 或 t 是直接了当的。许多 IB 考题还会通过 r = ln 2 / Tdouble 将增长与倍增时间联系起来。
10. Applications: Radioactive Decay | 应用：放射性衰变
xponential decay process. The mass M of a radioactive substance at time t follows M = M₀ e−kt, where k is the decay constant. The half‑life T₁⁄₂ is related by k = ln 2 / T₁⁄₂. Carbon‑14 dating, uranium decay series, and medical tracer studies all rely on this model.
放射性衰变是典型的指数衰减过程。放射性物质在时刻 t 的质量 M 遵循 M = M₀ e−kt，其中 k 为衰变常数。半衰期 T₁⁄₂ 的关系为 k = ln 2 / T₁⁄₂。碳‑14 年代测定、铀衰变链以及医学示踪剂研究均依赖此模型。
ning after a given time or the time elapsed given the remaining fraction. Because the half‑life is constant, you can also use the fraction form M = M₀ (½)t/T. Both approaches are acceptable; choose whichever simplifies the arithmetic, but be consistent.
典型的 IB 题目会要求计算给定时间后的剩余百分比，或给定剩余比例求所经过的时间。由于半衰期恒定，你也可以使用分数形式 M = M₀ (½)t/T。两种方法均可接受；选择能使算术简化的一种，但需保持一致。
11. Applications: Compound Interest | 应用：复利
l model. For n compounding periods per year, the amount A after t years is A = P (1 + r/n)nt. As n → ∞, this becomes continuous compounding: A = P ert. IB questions often compare different compounding frequencies and ask which yields a higher return.
复利或许是最为人熟知的指数模型。对于每年复利 n 次的情况，t 年后的金额 A = P (1 + r/n)nt。当 n → ∞ 时，它变为连续复利：A = P ert。IB 题目常会比较不同的复利频率，并询问哪种能带来更高的收益。
The exponential nature lies in the fact that interest earns interest on interest. Understanding the time value of money through these models is essential for financial applications. You may also need to solve for the nominal interest rate r, the time t, or the principal P using logarithms and algebraic manipulation.
指数的特性在于利息会孳生利息。通过这些模型理解货币的时间价值对金融应用至关重要。你也可能需要运用对数与代数运算求解名义利率 r、时间 t 或本金 P。
12. Limitations and Assumptions | 局限与假设
Pure exponential models assume constant growth or decay rate and unrestricted resources. In reality, exponential growth cannot continue indefinitely due to carrying capacity, competition, or resource depletion. This leads to logistic models (sigmoid curves) in more advanced topics. Nevertheless, exponential models provide excellent local approximations and are foundational.
纯指数模型假设增长或衰减速率恒定且资源不受限制。现实中，由于环境容纳量、竞争或资源枯竭，指数增长不可能无限持续下去。这便引向了更高级主题中的逻辑斯蒂模型（S 形曲线）。然而，指数模型仍提供出色的局部近似，且是奠基性的。
Similarly, exponential decay models assume a perfectly random process with constant probability, which holds well for radioactivity but may fail for complex chemical reactions with intermediate steps. Always state the assumptions and domain of validity when applying an exponential model in an IA or exam solution.
类似地，指数衰减模型假设一个完全随机且概率恒定的过程，这对放射性很适用，但对于涉及中间步骤的复杂化学反应可能失效。在内部评估或考试解答中应用指数模型时，务必陈述其假设和适用范围。
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