Imagine a physical system—a rising loan balance, a falling body, or a population of endangered species. The anatomy of a first-order differential equation (ODE) is the mathematical bridge that allows us to predict the future state of these systems. It formalizes the relationship between an independent variable $t$, a dependent variable $y$, and its instantaneous rate of change.
1. The Structural Taxonomy
At its core, a first-order ODE relates the derivative to the variables: $$\frac{dy}{dt} = f(t, y) \quad (1)$$ or in its implicit form $F(t, y) = 0$. Equations are classified by their "skeleton":
- Linear Anatomy: Equations like $\frac{dy}{dt} = -ay + b$ (2), where the function is linear in $y$. Note: Thus we will use the term 'general solution' only when discussing linear equations.
- Autonomous Anatomy: When the rate depends solely on the state, $dy/dt = f(y)$. These often feature a Threshold level (T): a critical population level below which a species cannot propagate and becomes extinct.
- Exact Anatomy: Verified by the condition $M_y(x, y) = N_x(x, y)$. If this fails, as in Example 3, there is no $\psi(x, y)$ satisfying the system.
Step 1: Construction of the Model
Physical situations, like EXAMPLE 4 | Escape Velocity (a body of mass $m$ projected from Earth), must be translated into mathematical terms. We must account for gravity and initial velocity $v_0$.
Step 2: Stability and Existence
We rely on the Lipschitz condition: $|f(t, y_1) - f(t, y_2)| \le K|y_1 - y_2|$ to guarantee that a solution exists and is unique. Without this, the "anatomy" of the problem might be broken or multi-valued.
2. Solutions and Visualization
Any differentiable function $y = \phi(t)$ that satisfies the equation for all $t$ in some interval is called a solution. Geometrically, we plot this as an integral curve. For Bernoulli Equations, we use the Substitution $v = y^{1-n}$ to linearize the anatomy.
🎯 Critical Observation: Euler's Method
In EXAMPLE 1 (Loan balance $S(t)$ with 12% interest), discrete approximations using Euler’s method $y_{n+1} = y_n + f_n \cdot (t_{n+1} - t_n)$ are often greater than actual continuous values. This is because the solution graph is concave down, causing tangent line approximations to lie above the graph.
$\frac{dS}{dt} = rS - k \implies y_n = \rho^n y_0$