📚 11E Using Maclaurin Series to Solve Differential Equations | 11E 使用麦克劳林级数解微分方程
presentation of the solution, valid near the origin. By assuming the solution can be written as an infinite polynomial, substituting into the differential equation and equating coefficients, we can determine the series coefficients step by step.
许多微分方程的解无法用初等函数表示，或者标准求解过程过于复杂。利用麦克劳林级数（以 x = 0 为中心的泰勒级数）的方法，我们能够求出解在原点附近的幂级数表达式。假设解可以写成一个无限多项式，代入微分方程并比较系数，就可以逐步确定级数的各项系数。
1. Maclaurin Series Recap | 麦克劳林级数回顾
A Maclaurin series is a Taylor series expansion about x = 0, given by f(x) = f(0) + f'(0)x + f''(0)x²/2! + f'''(0)x³/3! + ... = Σ f⁽ⁿ⁾(0) xⁿ / n! o the function in some interval, we can use this series to represent the function locally.
麦克劳林级数是函数在 x = 0 处的泰勒展开，形式为 f(x) = f(0) + f'(0)x + f''(0)x²/2! + f'''(0)x³/3! + ... = Σ f⁽ⁿ⁾(0) xⁿ / n!（n 从 0 到 ∞）。当函数在 0 处无限可导且级数在某个区间内收敛到该函数时，我们可以局部地用这个级数表示该函数。
2. The Idea: Assuming a Power Series Solution | 思路：假设一个幂级数解
aₙ xⁿ, where aₙ are constants to be determined. The derivatives are obtained by differentiating term by term: y' = Σ n aₙ xⁿ⁻¹, y'' = Σ n(n-1) aₙ xⁿ⁻², and so on.
为了用级数求解微分方程，我们假设解 y(x) 可以表示为关于 x = 0 的幂级数：y = a₀ + a₁x + a₂x² + a₃x³ + ... = Σ aₙ xⁿ，其中 aₙ 是待定常数。导数可以通过逐项求导得到：y' = Σ n aₙ xⁿ⁻¹, y'' = Σ n(n-1) aₙ xⁿ⁻²，等等。
3. Substituting the Series into the Differential Equation | 将级数代入微分方程
We substitute the power series expressions for y, y', y'' etc. into the given differential equation. To compare coefficients, it is usually necessary to rewrite all series so that they share the same power of x, for instance xⁿ. This often involves shifting indices, such as replacing Σ n aₙ xⁿ⁻¹ with Σ (n+1) aₙ₊₁ xⁿ. After substitution, the equation becomes a single sum over xⁿ set equal to the right-hand side (often zero or a polynomial).
我们将 y, y', y'' 等的幂级数表达式代入给定的微分方程。为了比较系数，通常需要将所有级数改写成具有相同 x 幂次的形式，例如 xⁿ。这往往涉及指标平移，例如把 Σ n aₙ xⁿ⁻¹ 替换为 Σ (n+1) aₙ₊₁ xⁿ。代入后，方程变成关于 xⁿ 的单个求和，并等于右侧（通常为零或某个多项式）。
4. Equating Coefficients and Recurrence Relations | 比较系数与递推关系
Since the power series representation is unique (within the radius of convergence), we equate the coefficients of like powers of x on both sides of the equation. This yields a recurrence relation linking the coefficients aₙ. Typically, a recurrence relation expresses aₙ₊₁ or aₙ₊₂ in terms of earlier coefficients. The initial coefficients a₀, a₁, etc. are obtained from the initial conditions of the differential equation.
由于幂级数表示（在收敛半径内）是唯一的，我们令方程两边同次幂 x 的系数相等。这样就得到联系系数 aₙ 的递推关系。通常，递推关系用前面的系数表示 aₙ₊₁ 或 aₙ₊₂。初始系数 a₀, a₁ 等由微分方程的初始条件决定。
5. Role of Initial Conditions | 初始条件的作用
Initial conditions such as y(0) = y₀ and y' ries coefficients. Because the Maclaurin series is centred at 0, we have a₀ = y(0), a₁ = y'(0), a₂ = y''(0)/2!, a₃ = y'''(0)/3! efficients. This ensures the series solution satisfies the specific initial value problem.
初始条件如 y(0) = y₀ 和 y'(0) = y₁ 直接给出前几个级数系数。因为麦克劳林级数以 0 为中心，我们有 a₀ = y(0), a₁ = y'(0), a₂ = y''(0)/2!, a₃ = y'''(0)/3! 等。在我们的级数方法中，我们只需令 a₀ 和 a₁ 为给定值，然后利用递推关系求出更高阶系数。这保证了级数解满足特定的初值问题。
6. Example 1: Solving y' = y | 例 1：解 y' = y
Consider the equation y' = y with y(0) = 1. Let y = Σ aₙ xⁿ, then y' = Σ n aₙ xⁿ⁻¹. Re-index the derivative: y' ives (n+1) aₙ₊₁ = aₙ, so aₙ₊₁ = aₙ/(n+1). With a₀ = 1, we find a₁ = 1, a₂ = 1/2, a₃ = 1/6, … aₙ = 1/n!. Thus y = Σ xⁿ/n! = eˣ.
考虑方程 y' = y，y(0) = 1。设 y = Σ aₙ xⁿ，则 y' = Σ n aₙ xⁿ⁻¹。重新设置导数的求和指标：y' = Σ (n+1) aₙ₊₁ xⁿ（从 n=0 开始）。方程变为 Σ (n+1) aₙ₊₁ xⁿ = Σ aₙ xⁿ。比较系数得 (n+1) aₙ₊₁ = aₙ，因此 aₙ₊₁ = aₙ/(n+1)。由 a₀ = 1，得 a₁ = 1, a₂ = 1/2, a₃ = 1/6, … aₙ = 1/n!。因此 y = Σ xⁿ/n! = eˣ。
7. Example 2: y' = x + y | 例 2：y' = x + y
Solve y' = x + y, with y(0) = 0. Assume y = Σ aₙ xⁿ, then y' a₃ = a₂/3 = 1/6, a₄ = a₃/4 = 1/24, ... . The solution is y = x²/2 + x³/6 + x⁴/24 + ... = eˣ - x - 1.
解 y' = x + y，y(0) = 0。假设 y = Σ aₙ xⁿ，则 y' = Σ (n+1) aₙ₊₁ xⁿ。右侧为 x + Σ aₙ xⁿ = a₀ + (a₁+1)x + a₂x² + a₃x³ + ... 。由 y(0)=0 有 a₀ = 0。现在比较同次幂系数：n=0: a₁ = a₀ → a₁ = 0。n=1: 2a₂ = a₁ + 1 → 2a₂ = 1 → a₂ = 1/2。n≥2: (n+1)aₙ₊₁ = aₙ。于是 a₃ = a₂/3 = 1/6，a₄ = a₃/4 = 1/24，...。解为 y = x²/2 + x³/6 + x⁴/24 + ... = eˣ - x - 1。
8. Example 3: Second-Order Equation y'' + y = 0 | 例 3：二阶方程 y'' + y = 0
Take y'' + y = 0, with y(0) = 0 and y'(0) = 1. Substitute y = Σ aₙ xⁿ, then y'' = Σ n(n-1) aₙ xⁿ⁻². Re-index: y'' ₂ = 0, a₃ = -1/6, a₄ = 0, a₅ = 1/120, ... This yields y = x - x³/6 + x⁵/120 - ... = sin x.
考虑 y'' + y = 0，y(0) = 0，y'(0) = 1。代入 y = Σ aₙ xⁿ，则 y'' = Σ n(n-1) aₙ xⁿ⁻²。重新指标：y'' = Σ (n+2)(n+1) aₙ₊₂ xⁿ。方程 Σ (n+2)(n+1) aₙ₊₂ xⁿ + Σ aₙ xⁿ = 0 给出 (n+2)(n+1) aₙ₊₂ + aₙ = 0，因此 aₙ₊₂ = -aₙ / [(n+2)(n+1)]。利用 a₀ = 0, a₁ = 1，计算得 a₂ = 0, a₃ = -1/6, a₄ = 0, a₅ = 1/120, ... 于是得到 y = x - x³/6 + x⁵/120 - ... = sin x。
9. Managing the Recurrence: Finding the General Term | 管理递推：求出通项
n be unrolled to express aₙ in closed form, often involving factorials. For instance, in y' = y we got aₙ = 1/n! Even without a closed form, the series provides an approximate solution valid near x = 0.
在简单情形下，递推关系可以展开，写出 aₙ 的通式，其中常包含阶乘。例如在 y' = y 中我们得到 aₙ = 1/n!。在更复杂的方程中，也许只能求出最初几项。在 IB 问题中，通常只要求得到级数到某一次幂，如 x⁴ 或 x⁵。即使没有封闭形式，级数也提供了一个在 x = 0 附近有效的近似解。
10. Radius of Convergence and Validity | 收敛半径与有效性
The series solution is only guaranteed to converge within some radius R. Usually, for linear differential equations with ordinary points, the series converges at least up to the nearest singular point. In our examples, eˣ and sin x have infinite radii of convergence, so the series represent the functions for all x. However, for other equations, the series might diverge beyond a finite interval. It is good practice to state any known convergence restrictions when presenting the solution.
级数解只在某个收敛半径 R 内保证收敛。通常，对于具有正常点的线性微分方程，级数至少收敛到最近的奇点。在我们的例子中，eˣ 和 sin x 的收敛半径无穷大，因此级数对所有 x 都表示该函数。但对其他方程，级数可能在有限区间外发散。在给出解时，最好说明已知的收敛限制。
11. Comparison with Standard Solution Techniques | 与标准求解方法的比较
l when the differential equation has variable coefficients that are not easily handled by integrating factors or characteristic equations. For instance, equations like y'' lved in terms of elementary functions, but a Maclaurin series solution can be found systematically. In IB exams, you may be asked to apply the series method to verify a known solution or to find an approximation for an otherwise unsolvable equation.
级数方法在微分方程具有变系数且不易用积分因子或特征方程处理时特别有用。例如 y'' - x y = 0（艾里方程）就无法用初等函数求解，但可以通过麦克劳林级数系统地获得解。在 IB 考试中，你可能会被要求用级数方法验证已知解，或为一个原本不可解的方程寻找近似解。
12. Tips and Common Pitfalls | 提示与常见错误
Always write down the assumed series clearly and differentiate carefully, paying attention to index shifts. When shifting indices, ensure the starting index and exponent match. A common mistake is forgetting to adjust the starting index when changing n, leading to a missing first term. Always use initial conditions to fix a₀, a₁, etc., before applying the recurrence. Finally, presenting the solution as a sum of known series (like eˣ or sin x) when recognised can earn full marks and demonstrate deeper understanding.
始终清晰地写下假设的级数并小心求导，注意指标的平移。在平移指标时，确保起始下标和指数匹配。一个常见错误是改变 n 时忘记调整起始下标，导致遗漏首项。在应用递推前，始终用初始条件确定 a₀, a₁ 等。最后，若能将解识别为已知级数的和（如 eˣ 或 sin x）并写出，通常能获得满分并展示更深的理解。
Published by TutorHao | 数学 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