|
Wed Sept 1: E1.1
|
|
Fri Sept 3: E1.3
|
|
Comments for Week 1:
|
| Some terminology learned: |
| order of a differential equation |
| n-parameter family of solutions |
| linear and nonlinear equations |
| slope fields |
| Slope field example handed out in class on Friday. |
|
|
|
|
|
|
|
|
Wed Sept 8: E1.4
|
|
Fri Sept 10: E1.5
|
|
Comments for Week 2:
|
| Learned how to solve separable equations |
| (which are a particular type of first-order equation). |
| Learned how to find an integrating factor |
| (for a 1st order linear eqn with non-constant coeffs). |
|
|
|
|
|
|
|
|
Mon Sept 13: E2.1
|
|
Wed Sept 15: E3.1
|
|
Fri Sept 17: E3.2 / E3.3
|
|
Comments for Week 3:
|
| Some population models, including the logistic equation |
| The logistic equation can be solved explicitly by hand, though it's cumbersome. |
| Brief mention of stable and unstable equilibrium points |
| Handout illustrating stable and unstable equilibria |
| The characteristic equation of a lin. homog. diff eqn with constant coeffs |
| Using the char. eqn. to find general solution of the diff. eqn., including: |
| what to do if there are repeated roots or complex roots |
| A brief introduction to differential operators |
|
|
|
|
|
|
|
|
Mon Sept 20: E3.2 / E3.3
|
|
Wed Sept 22: E3.4
|
|
Fri Sept 24: E3.5
|
|
Comments for Week 4:
|
| There's an existence-uniqueness theorem for higher order linear diff. eqns. |
| The Wronskian is a determinant that tests if functions are linearly dependent. |
| Defintion of the complementary function of a differential equation |
| Motion of a mass attached to a spring and dashpot: |
| undamped versus damped motion, and free versus forced |
| Using undetermined coefficients to find particular solns to nonhomog. diff. eqns. |
|
|
|
|
|
|
|
|
Mon Sept 27: E3.7
|
|
Wed Sept 29: L1.1
|
|
Fri Oct 1: TEST 1
|
|
Comments for Week 5:
|
| Simple RLC series circuit and corresponding differential equation: |
| RLC equation is mathematically the same as for mass-spring-dashpot. |
| Solution is sum of transient current and steady periodic current. |
| Systems of linear equations, coefficient matrix, augmented matrix |
| Elementary row operations, pivot, strict triangular form |
|
|
|
|
|
|
|
|
Mon Oct 4: L1.2
|
|
Wed Oct 6: L1.3
|
|
Fri Oct 8: L1.4
|
|
Comments for Week 6:
|
| Row echelon form, homogeneous and nonhomogeneous linear systems |
| Leading 1's allow us to define leading variables and free variables. |
| REDUCED row echelon form (where you also have zeros ABOVE leading ones) |
| Overdetermined and underdetermined systems (how many solutions can they have?) |
| Example of application of linear algebra to electric circuits |
| Definition of Rn, matrix addition, scalar multiplication, matrix multiplication |
| Powers of a square matrix, n by n identity matrix, transpose of a matrix |
| Inverse of a square matrix -- A given matrix may or may not have an inverse. |
| A square matrix can be nonsingular (invertible) or singular (noninvertible) |
|
|
|
|
|
|
|
|
Mon Oct 11: L2.1
|
|
Wed Oct 13: L2.2
|
|
Fri Oct 15: L3.1
|
|
Comments for Week 7:
|
| Determinant of a square matrix. 2x2 and 3x3 have nice formulas, larger ones don't. |
| Minors and cofactors. Minor or cofactor expansion along a row or column. |
| Rules for how the three types of row operations affect the determinant of a matrix. |
| Rules for the determinant of the transpose or the inverse of a matrix. |
| Rule for the determinant of a triangular matrix. |
| A vector space must be closed under vector addition and scalar multiplication. |
| Vectors can be n-tuples in Rn, functions, or other things. |
| Some subsets of a vector space can also be a vector space. |
|
|
|
|
|
|
|
|
Mon Oct 18: L3.2
|
|
Wed Oct 20: L3.3
|
|
Fri Oct 22: L3.4
|
|
Comments for Week 8:
|
| Some subsets of a vector space can also be a vector space -- i.e., a subspace. |
| A subset is a subspace if it is closed under addition and scalar multiplication. |
| The solutions of a linear homogeneous differential equation are a subspace |
| of the space of all continuous functions -- even if we can't solve the equation! |
| Definition of the null space of a matrix. |
| An m by n matrix can define a function from Rn to Rm. |
| The null space of an m by n matrix is a subspace of Rn. |
| Definition of the span of a set of vectors. |
| Definition of a linearly independent set of vectors. |
| Definition of a basis of a vector space. |
| In Rn, what can and can't happen if you have a set containing: |
| fewer than n vectors, more than n vectors, or exactly n vectors? |
|
|
|
|
|
|
|
|
Mon Oct 25: TEST 2
|
|
Wed Oct 27: L3.5
|
|
Fri Oct 29: L3.6
|
|
Comments for Week 9:
|
| There are many possible bases for Rn (or for any n-dimensional space). |
| Given a basis for Rn, any vector in Rn has coordinates
with respect to that basis. |
| A transition matrix can be used to convert between two different bases. |
| For any matrix, we can define its row space and its column space. |
| The row space of an m by n matrix is a subspace of Rn. |
| The column space of an m by n matrix is a subspace of Rm. |
| To find a basis for the row space or the column space, do row operations. |
| (For a basis for the row space, use appropriate rows of the reduced matrix.) |
| (For a basis for the column space, use appropriate rows of the original matrix.) |
|
|
|
|
|
|
|
|
Mon Nov 1: L6.1
|
|
Wed Nov 3: L6.3
|
|
Fri Nov 5: L6.3
|
|
Comments for Week 10:
|
| Definition of eigenvalue and eigenvector. |
| Eigenvalues and eigenvectors are defined only for square matrices. |
| Eigenvectors, by definition, must be nonzero vectors. |
| The eigenvalues of a matrix A are the numbers λ |
| that make the matrix (A−λI) have determinant zero. |
| We refer to det(A−λI) as the characteristic polynomial. |
| The algebraic multiplicity of an eigenvalue is the number |
| of times it appears as a root of the characteristic polynomial. |
| The geometric multiplicity of an eigenvalue is the number |
| of independent eigenvectors belonging to that eigenvalue, |
| or the dimension of the corresponding eigenspace. |
| We can diagonalize a square matrix if every eigenvalue |
| has geometric multiplicity equal to its algebraic multiplicity. |
| (Lines 4, 5, and 6 contain the letter lambda -- hope it works in all browsers) |
|
|
|
|
|
|
|
|
Mon Nov 8: L6.3
|
|
Wed Nov 10: E4.1
|
|
Fri Nov 12: E5.1
|
|
Comments for Week 11:
|
| Diagonalization, when possible, helps us compute powers of a matrix |
| as well as the exponential of a matrix. |
| We can find both specific powers and a general formula for the kth power. |
| We skipped most of 6.4. We just singled out a few facts about eigenvalues. |
| SYSTEMS of first-order differential equations: |
| There are standard tricks for rewriting a high-order differential equation |
| as a system of first-order differential equations. |
| The system can be written in matrix form, and may be either |
| homogeneous or nonhomogeneous. |
| For a homogeneous system, there's a technique for finding the general solution |
| using eigenvalues and eigenvectors. |
|
|
|
|
|
|
|
|
Mon Nov 15: E5.2
|
|
Wed Nov 17: TEST 3
|
|
Fri Nov 19: E5.4
|
|
Comments for Week 12:
|
| Initial conditions for a system of diff eqns are most often at t=0, |
| which makes the math fairly easy, but you still do a bit of linear algebra. |
| Complex eigenvalues are a little tricky: |
| -- eigenvector will contain complex numbers |
| -- one complex eigenvalue will lead to two real solutions |
| -- use Euler's formula to split complex solution into real and imaginary parts. |
| Repeated eigenvalues are a little tricky: |
| -- form of solution is a bit more complicated |
| -- solution contains t times eigenvector plus generalized eigenvector |
|
|
|
|
|
|
|
|
Mon Nov 22: E5.4
|
|
Comments for Week 13:
|
see above |
|
|
|
|
|
|
|
Mon Nov 29: E5.5
|
|
Wed Dec 1: E6.1
|
|
Fri Dec 3: E6.1
|
|
Comments for Week 14:
|
| Definition of a fundamental matrix Phi(t) for a system of diff eqns. |
| The product Phi(t) times the inverse of Phi(0) is useful for initial value problems. |
| There is a relationship between that product and matrix exponentials. |
| Definition of an autonomous system of diff eqns. |
| Given a second-order autonomous system with x = x(t), y = y(t), |
| we can form a slope field, which might tell us something |
| about the solutions even if we can't solve the system explicitly. |
| Definition of a critical point or equilibrium point of a |
| (not necessarily linear) system of diff eqs. Finding critical points is |
| "just algebra", and typically there are a small finite number of them. |
| One can try to determine the behavior of a system near a critical point |
| by looking at the slope field, or by looking at the linear approximation |
| of the system near the critical point. |
| In fact, the eigenvalues of the linear approximation can help you classify a |
| critical point as a sink, source, saddle point, etc., but we didn't get into the details. |
|
|
|
|
|
|
|
|
Mon Dec 6: REVIEW
|
|
Wed Dec 8: REVIEW
|
|
Comments for Week 15:
|
see above |