Introduction to Differential Equations MAT26600


Important information: Main course page , where you can find info about the assignments and projects.
Office hours Wednesday 2:00pm, Thursday 3:00pm.

Some useful links:
Khan Academy on Differential Equations Online lectures (first one here ) it basically goes through our course in several 10 minute videos - also available on youtube.
Paul's online notes on differential equations.
MIT's open courseware: Differential equations and its honors version
Important Dates, and rough grading scheme: I will assume you will be free on each of these dates to attend class; any class work missed on those days will be forfeited.
Meeting times and textbook
Overview A typical class will involve the presentation of the material and attempting some problems on the board. Student participation in class is strongly encouraged. If at anytime you have a question regarding the material, please do not hesitate to interrupt. As well, if you cannot understand me when I talk too fast, stumble over my words, or speak into the chalkboard, be sure to complain loudly.
There will probably be two midterms, one final, weekly homework, and several in-class quizzes. Midterms will be based on work presented in class and done on homework, but you should not expect it to be identical. It is expected that you can extrapolate from what is taught in lecture and what you do on your homework to parallel problems. This may be true to an even greater extent for the final.
If you wish to contact me out of class it is best to do so by e-mail. For help outside of office hours please schedule an appointment with me this way. Even if you're coming to office hours please let me know beforehand, as I tend to wander away from my office even when I'm supposed to be there.
Topics covered
Chapter 1 - Introduction to Differential Equations.

In this chapter, we will explain what a differential equation is, introduce some terminology to categorize differential equations (terms like ordinary differential equation, partial differential equation, order, linear, non-linear), talk briefly about mathematical modeling and solve a particular equation explicitly. We will also describe direction fields and use them a little to see qualitative behaviour - this can be useful if an analytic expression cannot be found (and even when they can be found).

Chapter 2 - First order differential equations.

We will discuss first order (ordinary!) differential equations (we will deal exclusively with ordinary differential equations; partial differential equations are beyond the scope of this course). Even within this relatively simple class we will have to restrict our attention to classes of equations having particular forms - linear equations, autonomous equations, exact equations (and discuss integrating factors which can sometimes be used to transform an equation into an exact equation).

Chapter 3 - Second order linear differential equations

We will carefully study the solution of this class of equations, mostly the case of constant coefficients. If the coefficients are constant, a very simple and complete analytic description is possible and we show how to do this (in the 'homogeneous' case). In the non-homogeneous case, we discuss some ways to get solutions from solutions to the corresponding homogeneous equation. The case of non-constant coefficients is very briefly discussed (it is discussed more carefully in Chapter 5, which we will omit in this course). Finally, two applications are discussed: simple electrical circuits, and harmonic motion (e.g. a spring satisfying Hooke's law) with damping and forcing.

Chapter 4 - Higher order linear differential equations

We extend the results of Chapter 3 to higher order; the main point is that many of the same techniques still work with small modification.

Chapter 6 - Laplace Transform

The Laplace transform can be useful for dealing with certain forcing functions (we will concentrate on the setting of second order linear constant coefficient equations, though it can be used for higher orders and even for systems of equations) which have some discontinuity, such as a Heaviside function or even a Dirac delta function. This may arise, for example, when studying circuits where one drives the circuit with an applied voltage that is constant on intervals but jumps every once in a while (for example, a square wave). It is also an opportune time to introduce this integral transform, of which there are others (e.g. the Fourier transform) which are indispensable tools in their own right in mathematical applications beyond ordinary differential equations (PDE, number theory, signal processing, etc.).

Chapter 7 - Systems of first order linear equations

We study coupled systems of first order linear ODE's with constant coefficients - the problem is very well understood (at least theoretically, though computationally it can be difficult) and we discuss how to solve these systems in some detail. If you took MA26500 then you will be familiar with much of what we will discuss in this chapter.
It so happens that the results of Chapters 3, 4 can be thought of as particular cases of what we will do here, because an nth order equation can be solved by writing an equivalent system of n first order equations. However, it is, perhaps, worthwhile to have done things as we did since the solutions we found earlier can be more simply described and found than the solutions of general first order systems.


Goals: It would seem the main goal of the class is to understand how to solve linear equations with constant coefficients, and then more generally first order systems of equations. We also discuss a few techniques to solve non-homogeneous problems (undetermined coefficients, variation of parameters), and a few methods for solving some simple non-linear equations (separation of variables, integrating factors).
Forward Euler for section 2.7
Plotting Example 1 from 6.4 for section 6.4
Plotting Example 2 from 6.4 for section 6.4
Example of how to use eul.m with the function defined my myfun.m .
F2.m Example of a function definition for modeling the equation y'' + 0.02y' + 4y = 2*cos(2*t).