📚 Definition of Linear Independence
In the context of homogeneous differential equations, a set of solutions $y_1(x), y_2(x), ..., y_n(x)$ is said to be linearly independent on an interval $I$ if the equation
$c_1y_1(x) + c_2y_2(x) + ... + c_ny_n(x) = 0$
holds for all $x$ in $I$ only when $c_1 = c_2 = ... = c_n = 0$. Otherwise, the solutions are linearly dependent.
📜 Historical Context
The concept of linear independence arose from the study of linear algebra and its applications to differential equations. The use of determinants, particularly the Wronskian, to test for linear independence was developed in the 19th century by mathematicians like Josef Hoene-Wroński.
🔑 Key Principles for Proving Linear Independence
- 🔍 Wronskian Determinant: Calculate the Wronskian $W(y_1, y_2, ..., y_n)(x)$, which is the determinant of the matrix formed by the solutions and their derivatives:
$W(y_1, y_2, ..., y_n)(x) = \begin{vmatrix}
y_1 & y_2 & ... & y_n \\
y_1' & y_2' & ... & y_n' \\
... & ... & ... & ... \\
y_1^{(n-1)} & y_2^{(n-1)} & ... & y_n^{(n-1)}
\end{vmatrix}$
- 💡 Non-Zero Wronskian: If $W(y_1, y_2, ..., y_n)(x) \neq 0$ for at least one $x$ in the interval $I$, then the solutions are linearly independent.
- 📝 Zero Wronskian: If the solutions are known to be solutions of an $n$-th order linear homogeneous differential equation, and $W(y_1, y_2, ..., y_n)(x) = 0$ for all $x$ in $I$, then the solutions are linearly dependent. However, if the solutions are not known to be solutions to the same differential equation, a zero Wronskian does not necessarily imply linear dependence.
- 🌱 Direct Verification: Sometimes, linear independence can be shown directly by assuming $c_1y_1(x) + c_2y_2(x) + ... + c_ny_n(x) = 0$ and proving that $c_1 = c_2 = ... = c_n = 0$.
➗ Example 1: Second-Order Linear Homogeneous DE
Consider the differential equation $y'' - y = 0$. Two solutions are $y_1(x) = e^x$ and $y_2(x) = e^{-x}$.
Calculate the Wronskian:
$W(e^x, e^{-x})(x) = \begin{vmatrix}
e^x & e^{-x} \\
e^x & -e^{-x}
\end{vmatrix} = -e^x e^{-x} - e^x e^{-x} = -2$
Since $W(e^x, e^{-x})(x) = -2 \neq 0$, the solutions $e^x$ and $e^{-x}$ are linearly independent.
➕ Example 2: Using Trigonometric Functions
Consider the differential equation $y'' + y = 0$. Two solutions are $y_1(x) = \sin(x)$ and $y_2(x) = \cos(x)$.
Calculate the Wronskian:
$W(\sin(x), \cos(x))(x) = \begin{vmatrix}
\sin(x) & \cos(x) \\
\cos(x) & -\sin(x)
\end{vmatrix} = -\sin^2(x) - \cos^2(x) = -1$
Since $W(\sin(x), \cos(x))(x) = -1 \neq 0$, the solutions $\sin(x)$ and $\cos(x)$ are linearly independent.
💡 Conclusion
Proving linear independence of solutions to homogeneous differential equations is crucial for finding the general solution. The Wronskian determinant provides a powerful tool for this purpose. Remember to check if the Wronskian is non-zero to confirm linear independence, and always consider the context of the differential equation when interpreting the results.
