Ma316-001-F10 further explanation, §3.2
further explanation
existence and uniqueness
Consider two functions p(t) and q(t):
In[122]:=
Out[122]=
We know that the differential equation
y '' + p(t) y ' + q(t) y = 0,
with initial conditions
,
In[145]:=
Out[145]=
Has a unique solution:
In[125]:=
Out[125]=
...as long as p(t) and q(t) are continuous
In[150]:=
Out[150]=
superposition of two (linearly independent) solutions can satisfy any initial condition
Now, suppose that we have two solutions,
and
, of
y '' + p(t) y ' + q(t) y = 0
In[177]:=
Out[177]=
Then, if their Wronskian is non-zero at
, given an initial condition there:
In[179]:=
Out[179]=
we can find a linear combination of
and
,
that solves the initial value problem
y '' + p(t) y ' + q(t) y = 0,
,
:
In[202]:=
Out[202]=
superposition of two (linearly independent) solutions gives the general solution
Now suppose that we are given any solution φ(t) of the differential equation
y '' + p(t) y ' + q(t) y = 0.
We want to show that we can write φ(t) as a linear combination of
and
, that is, that we can find a
and
so that
![]()
So, suppose that we have such a φ and that we know that at some point
the Wronskian of
and
is non-zero.
In[210]:=
Out[210]=
Then, at
, we know that φ(t) satisfies the initial value problem
![]()
That is, a solution to the differential equation with the initial conditions:
In[249]:=
Out[249]=
That is, we're solving the differential equation with these initial conditions:
In[251]:=
Out[251]=
But, by the work in the previous section, we know that we can find a
and a
such that
![]()
satisfies the initial value problem
.
Thus:
In[246]:=
Out[246]=
And the uniqueness theorem (the first section above) says that this solution is unique—that is, it must be the same function as the φ(t) that we started with:
In[248]:=
Out[248]=
Thus, because we started off with the assumption that φ(t) was any solution to the differential equation, we've shown that the linear combination
![]()
is the general solution to the differential equation!
Super cool!