I9 - Solving coupled models by elimination
A coupled model needs two functions of the same independent variable, with each function satisfying its own rate equation. I9 extends elimination to oscillatory and forced systems, reconstructs the second variable and interprets a predator–prey approximation. You will check the full pair and distinguish deviations from actual populations.
Eliminating one dependent variable
Can a rise in prey numbers be followed by a later rise in predators, even when both populations repeatedly return to their starting levels? Coupled rates can create that delay. First see how two first-order equations turn into one oscillation equation.
Let share the independent variable , and use primes for time derivatives. From , , differentiation gives . The resulting equation has complex auxiliary roots, so its solutions oscillate. This is different from the real exponential modes in the earlier tank model.
Elimination with constant coefficients
For
with constant and differentiable , use
Differentiating the first equation and substituting gives
Not in the formula booklet: derive it by substitution rather than relying on memory.
Indeed, . Replace by , collect the and terms, and the displayed equation follows. Notice the derivative : a time-dependent input must also be differentiated.
If , do not divide by it. The first equation is already independent of ; solve it first and then solve the second first-order equation, or eliminate instead when . The method must respect the actual coefficients.
Worked example 1: elimination produces an oscillation
Question. For , , obtain a second-order equation for and find its general solution.
Plan. Differentiate and substitute the other rate equation.
Answer. , with .
Check. Differentiating the proposed twice gives . We have checked the eliminated equation; the original pair still needs , which we reconstruct next.
Add only the time-dependent input to the first rate equation. Its derivative will supply a constant forcing term.
Worked example 2: an input becomes a forcing term
Question. For , , obtain a second-order equation for and find its general solution.
Plan. Differentiate the entire first equation, including , then use a constant particular integral.
Answer. .
Check. Here , so the second-order equation holds. Omitting the derivative of would wrongly remove the constant response.
An eliminated equation is a tool for finding one component, not the finished coupled solution. The relation used to eliminate the other component must be retained.
Watch out
Wrong: replacing by , or differentiating as . Every term is a function of the same time; differentiate the input as well.
Change the input rate and feedback coefficient in a fresh system.
Reconstructing the second solution
The first original equation tells us how to reconstruct from and . It also turns starting values for into a starting slope for . Neither operation introduces an extra arbitrary constant.
For example, if , then . At , . At a different starting time , the input is not zero: .
Worked example 3: reconstructing an unforced pair
Question. Solve , , given , .
Plan. Use the eliminated solution, obtain from the first equation, then reconstruct .
Answer. , .
Check. and . At zero, . Both equations and both conditions hold.
Keep the initial values and add only the input from Worked example 2. The reconstruction now includes its subtraction.
Worked example 4: reconstructing a forced pair
Question. Solve , , given , .
Plan. Fix the constants in , then use .
Answer. , .
Check. The first right side is . The second derivative is . The initial values are . Both original equations work, including the input.
The two constants in control the whole pair. Solving an independent second-order equation for and assigning it two more unrelated constants would usually produce pairs that fail the original system.
The elimination formula is reversible when : reconstructing enforces the first equation, and the eliminated equation then enforces the second. Checking both directly is still a quick way to catch a sign or input error.
Watch out
Wrong: taking in a system with . The correct relation is . Also derive the starting slope from the original rate equation, rather than confusing with .
Solve a new system with changed input and initial values.
Continue this lesson.
You’re previewing NovaLearn without an account. Create a free account to read the rest.
Create free account