ENGR 232 Recitation 2

Initial Value Problems Recall that a general form of a differential equation can be expressed as , , , , = 0

In order to solve this initial value problem, we need to find an interval I that satisfies x0, the initial value of independent variable, over the n initial conditions. In other words, an nth order differential equation must have n restrictions (constraints) to solve it. If we want to solve the differential equation by looking at all of the initial values of the n-1 derivatives and the one value of the dependent value at x0, we solve the initial value problem. This means (0 ) = 0

Where the values a0 to an-1 are constants, or the initial conditions of the differential equation.

1 (0 ) = 1 1

(0 ) = 1 . . .

Example Verify that () = 15 3 12 5

Is a solution to the initial value problem

first, we calculate all of the derivatives for ():

2 (0) (0) + 8 + 15 = 0; = 3; = 15 2

Next, check to see if the proposed solution is a good candidate by first applying the solution and its derivatives into the differential equation: 135 3 300 5 + 8(45 3 + 60 5 ) + 15(15 3 12 5 ) = 0 (0) = 15 3(0) 12 5(0) = 3 (0) (0) = = 45 3(0) + 60 5(0) = 15

2 2 = = 135 3 300 5 2 2

= = 45 3 + 60 5

So the proposed equation has passed its first hurdle by justifying the differential equation (here, 0=0). Now for the initial value analysis:

So the proposed solution is a solution to the initial value problem

Summary To determine if a proposed solution is a solution to a differential equation: 1. Calculate the derivatives of the proposed solution with respect to the independent variable 2. Substitute the proposed solution and its derivatives into the differential equation. If the equality is met, the solution is a potential solution; otherwise it is not a solution. 3. Evaluate the initial conditions of the proposed solution and its derivatives at the initial independent value. If those conditions/equalities are met, then the proposed solution is a solution to the initial value problem.

Existence and Uniqueness Consider the first order initial value problem = (, ), (0 ) = 0

If both f and f/y are continuous over a rectangle

That contains the point (x0,y0), then the initial value problem has a unique solution in some interval x0-< x <x0+, where is a positive number. Example Consider the initial value problem = 8 2 2 ; (3) = 10 = 8 2 2 both the function f and its partial derivative with respect to the dependent variable are continuous around in any rectangle containing the point (3,10), so there is a unique solution in the interval x=3 about (3-,3+), where is a positive number. = 2 1

= {(, ): < < , < < }

Is there a unique solution to this initial value problem around the point (3,10)?

Example Consider the initial value problem = 8 2 2 ; (3) = 0 = 8 2 2 both the function f and its partial derivative with respect to the dependent variable are not defined about y=0. This means we cannot use the uniqueness theorem to determine if there is a unique solution. = 2 3 1

Is there a unique solution to this initial value problem around the point (3,10)?

Direction Fields

For first order differential equations, it is often easier to find the solution by looking at the direction field of the equation. With a first order differential equation, we can put the derivative term on the left hand side of the equality, and the other terms on the other side of the equality. Because the derivative is the slope of the tangent line, we can plot the slopes of various points over the region of interest. Consider the figure below that is the direction field of the equation = 2 + 3 8

The direction field tells us the direction of travel for the solution. Consider the case where we put a marble on a contorted plane. The marble will move in the direction of the slope.

Now consider the case where our initial value, y(t=0) has a value of -4, shown in the figure below. We place our marble at that location, and ride in the direction of the arrows. Thus, we can graphically solve for the solution very easily!

As a matter of fact, it is possible that even though the initial value can be different for many different cases, the solution can converge onto the same solution "track", as noted in the figure below:

First Order Separable Differential Equations

A first order differential equation is separable when the equation is of the form () = ()

In other words, the numerator h(x) is only a function of the independent variable, and g(y) is a function only of the dependent variable. This allows us to simply cross-multiply so that the left hand side of the equation is a function of one variable, while the right hand side of the equation is a function of the other variable: () = ()

Now that we have the desired form, we can perform indefinite integration: () = ()

Because the indefinite integrals will have constants on both sides of the equation, one of the constants can be subtracted from the other constant to yield another constant.

Example In the case of radioactivity, the rate of change of a radioactive particle behaves as the first order differential equation =

where N is the number of particles in question, and is the decay constant. Performing separation of variables, we have = =

So we integrate both sides

where we can pull out the - from the integral because it is a constant. This leads to the equation || + 1 = + 2 || = +

The next step is to combine both constants. We can subtract C1 from C2 to make an effective constant C

Now, because the number of particles is always non-negative, we can remove the absolute value brackets. Note the removal is due to physicality, and not mathematics, so be careful when you have a problem like this that can have a negative value: () = + = +

Next, we exponentiate each side, leaving us

However, the right hand side can be broken up, and a new constant K can be formed: = + = = (0) = (0) = = 0 () = 0

The final part is to find the value of K. This requires the initial value of the number of particles, at time t=0, which we can call N0. Thus,

So now we have the final equation for N(t):

Example Solve the initial value problem 3 2 + 4 + 2 = , (0) = 1 2( 1)

First, is the differential equation separable? Yes, because the numerator is a function of only the independent variable, and the denominator is a function of only the dependent variable. Therefore, we commence with separating the variables: 2( 1) = 3 2 + 4 + 2 2( 1) = 3 2 + 4 + 2 Combining constants to form a new constant K, 2 2 + 1 = 3 + 2 2 + 2 + 2 2 2 = 3 + 2 2 + 2 +

Next, we can integrate:

Now we apply the initial condition, y(0)=-1:

This leaves us with K=3. Thus, the equation is now

(1)2 2(1) = (0)3 + 2(0)2 + 2(0) + 2 2 = 3 + 2 2 + 2 + 3 = 1 3 + 2 2 + 2 + 4

This is a quadratic equation, which has the solution

But this has two solutions. How do we figure out which solution is correct? Simple. Realizing the initial value y(0)=-1, the solution that meets the initial value requirement is = 1 3 + 2 2 + 2 + 4

Where the restriction to this equation is the limitation of x, such that the radical is nonnegative. The only real solution that satisfies a nonnegative radical is x=-2. Thus, the region which can meet this solution is when x>=-2.