Introduction to DEs
Mathematical problems are “puzzles” that can be formulated precisely in terms of
mathematical language.
The goal often is to find some “mathematical objects” that satisfy a list of conditions
given. We call these “mathematical objects” a solution, i.e., any key to the “puzzle”. A
solution can be numbers, coordinates (i.e., vectors), or functions.
Differential equations (DE) are simply one class of mathematical problems.
A differential equation is an equation that contains derivatives or differentials.
Classification of DEs: Ordinary and Partial
■ One important classification is based on whether the unknown function depends on a
single independent variable or on several independent variables.
■ In the first case, only ordinary derivatives appear in the differential equation, and it is
said to be an ordinary differential equation. In the second case, the derivatives are
partial derivatives, and the equation is called a partial differential equation.
■ Examples
2 𝜕𝜕2 𝑢𝑢 𝑥𝑥,𝑡𝑡
2 𝜕𝜕 𝑢𝑢 𝑥𝑥,𝑡𝑡
a) 𝑎𝑎
𝜕𝜕𝑥𝑥 2
=
𝜕𝜕𝑡𝑡 2
wave equation
𝑑𝑑𝑑𝑑 𝑃𝑃
b) 𝑑𝑑𝑑𝑑
= 𝑘𝑘𝑘𝑘 1 −
𝐾𝐾
logistic equation
Introduction to ODEs
■ Ordinary differential equations (ODEs) are a class of mathematical problems involving
derivatives of functions of one variable.
Notation. Let 𝑦𝑦: ℝ → ℝ 𝑜𝑜𝑜𝑜 (𝑦𝑦: (𝑎𝑎, 𝑏𝑏) → ℝ)) be a “nice” function of real variable 𝑡𝑡. Then
𝑦𝑦 ′ = 𝑦𝑦𝑦(𝑡𝑡): first derivative of 𝑦𝑦 ; 𝑦𝑦 ′′ = 𝑦𝑦𝑦𝑦(𝑡𝑡): second derivative of 𝑦𝑦;
𝑦𝑦 ′′′ = 𝑦𝑦𝑦𝑦𝑦(𝑡𝑡): third derivative of 𝑦𝑦; 𝑦𝑦 (𝑘𝑘) = 𝑦𝑦 𝑘𝑘
𝑡𝑡 , 𝑘𝑘 = 0,1,2, … the 𝑘𝑘 𝑡𝑡𝑡 order derivative of 𝑦𝑦,
with the understanding that 𝑦𝑦 (0) = 𝑦𝑦.
Definition. An ODE is an expression of the form
𝐹𝐹(𝑡𝑡, 𝑦𝑦(𝑡𝑡), 𝑦𝑦′(𝑡𝑡), 𝑦𝑦′′(𝑡𝑡), … , 𝑦𝑦 (𝑛𝑛) (𝑡𝑡)) = 0,
or simply
𝐹𝐹(𝑡𝑡, 𝑦𝑦, 𝑦𝑦 ′ , 𝑦𝑦 ′ , … , 𝑦𝑦 (𝑛𝑛) ) = 0,
where 𝑡𝑡 ∈ ℝ is the independent variable, and 𝑛𝑛 ∈ ℕ is a fixed natural number called the order of
the ODE.
A solution to an ODE is a differentiable function 𝑦𝑦: ℝ → ℝ 𝑜𝑜𝑜𝑜 (𝑦𝑦: (𝑎𝑎, 𝑏𝑏) → ℝ), where (𝑎𝑎, 𝑏𝑏) is an
open interval, that satisfies the equation above, evaluated at 𝑡𝑡 ∈ (𝑎𝑎, 𝑏𝑏).
Introduction to ODEs
Examples:
■ Show that 𝑦𝑦 = 𝑥𝑥 2 – 5𝑥𝑥 is a solution of the differential equation
𝑥𝑥𝑦𝑦 ′ − 𝑦𝑦 = 𝑥𝑥 2 .
The first derivative of 𝑦𝑦 is 𝑦𝑦 ′ = 2𝑥𝑥 − 5. Substituting, we have
𝑥𝑥𝑦𝑦 ′ − 𝑦𝑦 = 𝑥𝑥 2𝑥𝑥 − 5 − 𝑥𝑥 2 − 5𝑥𝑥 = 2𝑥𝑥 2 − 5𝑥𝑥 − 𝑥𝑥 2 + 5𝑥𝑥 = 𝑥𝑥 2 .
Hence, the function 𝑦𝑦 = 𝑥𝑥 2 – 5𝑥𝑥 is indeed a solution to the ODE 𝑥𝑥𝑦𝑦 ′ − 𝑦𝑦 = 𝑥𝑥 2 .
■ Show that 𝑦𝑦 = 𝑥𝑥 2 − 𝑥𝑥 −1 is a solution of the differential equation
′′ −
2
𝑦𝑦 𝑦𝑦 = 0.
𝑥𝑥 2
The first derivative of 𝑦𝑦 is 𝑦𝑦 ′ = 2𝑥𝑥 + 𝑥𝑥 −2 and the second derivative is 𝑦𝑦 ′′ = 2 − 2𝑥𝑥 −3 .
Substituting, we have
′′
2 −3
2 2 −1 −3
2 2 2 −1
𝑦𝑦 − 2 𝑦𝑦 = 2 − 2𝑥𝑥 − 2 𝑥𝑥 − 𝑥𝑥 = 2 − 2𝑥𝑥 − 2 𝑥𝑥 + 2 𝑥𝑥 =
𝑥𝑥 𝑥𝑥 𝑥𝑥 𝑥𝑥
−3 −3
2 − 2𝑥𝑥 − 2 + 2𝑥𝑥 = 0.
2
Hence, the function 𝑦𝑦 = 𝑥𝑥 2 – 𝑥𝑥 −1 is indeed a solution to the ODE 𝑦𝑦 ′′ − 𝑦𝑦 = 0.
𝑥𝑥 2
Remark. In the definition above, the ODE is an example of one-dimensional ODEs.
Examples of ODEs
𝑦𝑦 ′ = 𝑒𝑒 𝑡𝑡 ; 𝑦𝑦′ + 𝑝𝑝(𝑡𝑡)𝑦𝑦 = 𝑞𝑞 𝑡𝑡 𝑦𝑦 𝑛𝑛 ;
𝑦𝑦′ + 𝑡𝑡𝑡𝑡 = 𝑒𝑒 𝑡𝑡 ; 𝑦𝑦𝑦𝑦′′ + 𝑡𝑡𝑡𝑡′ = 0;
𝑦𝑦 ′′ + 2𝑦𝑦 ′ + 𝑦𝑦 = 0; sin(𝑡𝑡𝑡𝑡′′) = 0;
𝑡𝑡 2 𝑦𝑦𝑦𝑦 + 3𝑡𝑡𝑡𝑡𝑡 + 𝑦𝑦 = 0; 𝑦𝑦𝑦𝑦′𝑦𝑦′′𝑦𝑦′′′ = tan 𝑡𝑡 .
Classification of ODEs:
Systems of Differential Equations
■ Another classification of differential equations depends on the number of unknown
functions that are involved. If there is a single function to be determined, then one
differential equation is sufficient. However, if there are two or more unknown
functions, then a system of differential equations is required.
■ Example
𝑑𝑑𝑑𝑑
= 𝑎𝑎𝑎𝑎 − 𝛼𝛼𝑥𝑥𝑥𝑥
𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑 predator prey model
= −𝑐𝑐𝑐𝑐 + 𝛾𝛾𝑥𝑥𝑥𝑥.
𝑑𝑑𝑑𝑑
Classification of ODEs:
Order-Degree
The order of a differential equation is the highest order of derivatives that show up in
the equation. While, the degree of a differential equation is defined as the power of
the highest order derivative, provided the differential equation is polynomial in the
highest order derivative.
Example
Find the order and degree of each equation below:
𝑦𝑦 ′′′ + 3𝑦𝑦 ′′ + 3 𝑦𝑦 ′ 4 + 𝑦𝑦 = 0; 3rd order, 1st degree 𝑦𝑦𝑦𝑦 ′ 𝑦𝑦 ′′ 𝑦𝑦 ′′′ 2 = tan 𝑡𝑡 ; 3rd order, 2nd degree
𝑦𝑦 𝑦𝑦 ′′ 3 + 𝑡𝑡𝑦𝑦 ′ = 0; 2nd order, 3rd degree 𝑡𝑡 5 𝑦𝑦 (4) + 𝑡𝑡 2 𝑦𝑦 ′′′ + 𝑦𝑦 6 = 0; 4th order, 1st degree
sin 𝑡𝑡𝑦𝑦 ′′ = 0; 2nd order 𝑦𝑦𝑦𝑦𝑦 + sin(𝑦𝑦 (12) ) = 0. 12th order
Remark
Generally speaking, higher order equations are harder to solve.
Classification of ODEs:
Linearity and Homogeneity
An important class of differential equations is linear equations. It is the class of all equations
taking the form
𝑎𝑎𝑛𝑛 𝑡𝑡 𝑦𝑦 (𝑛𝑛) + 𝑎𝑎𝑛𝑛−1 (𝑡𝑡)𝑦𝑦 (𝑛𝑛−1) + · · · + 𝑎𝑎1 (𝑡𝑡)𝑦𝑦′ + 𝑎𝑎0 (𝑡𝑡)𝑦𝑦 = 𝑔𝑔(𝑡𝑡),
or simply in the summation form
𝑛𝑛
� 𝑎𝑎𝑘𝑘 𝑡𝑡 𝑦𝑦 (𝑘𝑘) = 𝑔𝑔 𝑡𝑡 ,
𝑘𝑘=0
𝑎𝑎0 , … , 𝑎𝑎𝑛𝑛 : 𝑎𝑎, 𝑏𝑏 → ℝ are real valued functions, which are called coefficients and
𝑔𝑔: 𝑎𝑎, 𝑏𝑏 → ℝ is a real valued function, which is sometimes called the forcing.
If the forcing
𝑔𝑔 𝑡𝑡 = 0,
then we say the ODE is homogeneous. Otherwise, if
𝑔𝑔 𝑡𝑡 ≠ 0,
the ODE is called nonhomogeneous .
Classification of ODEs:
Linearity and Homogeneity
■ Remark
Nonlinear equations are harder to solve comparing to linear equations and
nonhomogeneous equations are harder to solve comparing to homogeneous
equations.
■ Example
Check if each of the following ODEs are linear or nonlinear and
homogeneous or nonhomogeneous:
𝑦𝑦 (4) + 𝑦𝑦 ′ = sin 𝑡𝑡 ; linear, nonhomogeneous 𝑦𝑦 (3) + sin 𝑡𝑡 = 0; linear, nonhomogeneous
𝑦𝑦 4 + 𝑦𝑦 ′′ = sin 𝑡𝑡 ; nonlinear 𝑡𝑡 4 𝑦𝑦 ′′′ + 𝑒𝑒 𝑡𝑡 𝑦𝑦 = cos 𝑡𝑡 ; linear, nonhomogeneous
𝑦𝑦 (3) + sin 𝑡𝑡 𝑦𝑦 = 0; linear, homogeneous 𝑡𝑡 4 𝑦𝑦 ′′′ + 𝑒𝑒 𝑡𝑡 𝑦𝑦 = cos 𝑦𝑦 . nonlinear
𝑦𝑦 (3) + sin 𝑦𝑦 𝑦𝑦 = 0; nonlinear
Linearity and Homogeneity
The words linearity and homogeneity come from Linear Algebra.
Recall:
An operator/transformation/function
𝑇𝑇: 𝑋𝑋 → 𝑌𝑌
between vector spaces, e.g., ℝ𝑑𝑑 , is linear if it satisfies:
𝑇𝑇 𝑐𝑐1 𝑥𝑥 + 𝑐𝑐2 𝑦𝑦 = 𝑐𝑐1 𝑇𝑇 𝑥𝑥 + 𝑐𝑐2 𝑇𝑇 𝑦𝑦
for all c1 , c2 ∈ ℝ and 𝑥𝑥, 𝑦𝑦 ∈ 𝑋𝑋.
An equation 𝑇𝑇𝑇𝑇 = 𝑏𝑏 is homogeneous if 𝑏𝑏 = 0 (a zero vector).
A key property of homogeneous, linear equations is that if 𝑦𝑦1 , 𝑦𝑦2 are two solutions to the
equation, then so is 𝑐𝑐1 𝑦𝑦1 + 𝑐𝑐2 𝑦𝑦2 for any c1 , c2 ∈ ℝ, (superposition).
Classification of ODEs: Linearity and
Homogeneity
■ Exercise
Consider the differential equation
𝑦𝑦′′ + 4𝑦𝑦 = 0.
a. Verify that the ODE is linear and homogeneous. It is straightforward to check that.
b. Show that sin(2𝑥𝑥), cos(2𝑥𝑥) are solutions to the equation.
c. Show that 3sin(2𝑥𝑥) + 4 cos 2𝑥𝑥 is also a solution to the ODE.
b. Let 𝑦𝑦1 = sin 2𝑥𝑥 , then 𝑦𝑦1′ = 2 cos 2𝑥𝑥 and 𝑦𝑦1′′ = −4sin 2𝑥𝑥 . Substituting, we have
𝑦𝑦 ′′ + 4𝑦𝑦 = −4sin 2𝑥𝑥 + 4sin 2𝑥𝑥 = 0. Therefore, 𝑦𝑦1 is a solution.
Let 𝑦𝑦2 = cos 2𝑥𝑥 , then 𝑦𝑦2′ = −2 sin 2𝑥𝑥 and 𝑦𝑦2′′ = −4 cos 2𝑥𝑥. Substituting, we have
𝑦𝑦 ′′ + 4𝑦𝑦 = −4 cos 2𝑥𝑥+4 cos 2𝑥𝑥 = 0. Therefore, 𝑦𝑦2 is a solution.
Let 𝑦𝑦3 = 3sin(2𝑥𝑥) + 4 cos 2𝑥𝑥 , then 𝑦𝑦3′ = 6 cos 2𝑥𝑥 − 8 sin 2𝑥𝑥 and
𝑦𝑦3′′ = −12 sin 2𝑥𝑥 − 16 cos 2𝑥𝑥.
Substituting, we have
𝑦𝑦 ′′ + 4𝑦𝑦 = −12 sin 2𝑥𝑥 − 16 cos 2𝑥𝑥 + 4 3 sin 2𝑥𝑥 + 4 cos 2𝑥𝑥 =
−12 sin 2𝑥𝑥 − 16 cos 2𝑥𝑥 + 12 sin 2𝑥𝑥 + 16 cos 2𝑥𝑥 = 0.
Therefore, 𝑦𝑦3 is also a solution.
Remark
Any solution of the form
𝑦𝑦 = 𝑐𝑐1 sin 2𝑥𝑥 + 𝑐𝑐2 cos 2𝑥𝑥 ,
is a solution to the ODE
𝑦𝑦 ′′ + 4𝑦𝑦 = 0.
We will discuss this in more detail later!
