Scribd
Upload
45 views4 pages

MTH5109: Revision Lectures

- The document outlines the key topics and formulas that will be covered in the revision lectures before the midterm exam and in April. It provides summaries of important theorems and results in knots, curves, surfaces, geodesics, and the Gauss-Bonnet theorem. - Students are advised to practice all examples from lectures and homework, check past midterm solutions, and attempt prior exams under timed conditions to prepare for the format of the upcoming exam. - While the document does not constitute a complete list of exam topics, it highlights many of the main concepts and formulas students should ensure they understand to succeed on the exam.

Uploaded by

Jimmy Ly
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
45 views4 pages

MTH5109: Revision Lectures

- The document outlines the key topics and formulas that will be covered in the revision lectures before the midterm exam and in April. It provides summaries of important theorems and results in knots, curves, surfaces, geodesics, and the Gauss-Bonnet theorem. - Students are advised to practice all examples from lectures and homework, check past midterm solutions, and attempt prior exams under timed conditions to prepare for the format of the upcoming exam. - While the document does not constitute a complete list of exam topics, it highlights many of the main concepts and formulas students should ensure they understand to succeed on the exam.

Uploaded by

Jimmy Ly
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

11

MTH5109: Revision Lectures


This is a sketch of my two revision lectures the one before the midterm and the one
in April. I outlined some of the main theorems and formulae in each segment but you
should not treat this either as a complete list or as an indication that these particular
items will come up. This is not about the exam.
Exam: what I did say was that this would have similar format and style to previous years
eg all questions compulsory. I said that this year the questions would be more dierent
from previous years than in the past. I will not expect you to memorise long proofs or
detailed examples but I may ask you to repeat smaller proofs or parts of proofs.
Revision: as the exam has no options my advice is to cover all the material and not skip
any of the segments. Do the cwks that you missed and check over the solutions that you
did, and not only the * questions. The way to be able to sketch curves and surfaces is to
practice, so try to understand all the examples given in lectures and cwks. You should
also check over your midterm solutions to see where you went wrong. Another part of
revision is exam technique and I suggested to try at least a couple of past exam papers
preferably under exam conditions for practice (solutions in a form sucient to see if you
got it right will be on the web for some recent exams). In general my advice is to focus
on understanding the material, but also be able to repeat denitions, statements of main
results and smaller elements of bookwork.
No oce hours as I am away but there should be some resources organised centrally by
the department and in addition I will try to respond to specic questions by email eg
about solutions to cwks or points in lectures (but be very specic eg the page number of
the sentence in question).
Here is my sketch of some main theorems and formulae in the module.
knots: the concept of a knot, Reidemeister moves I,II,III aect a region with resp.
1,2,3 crossings leaving rest of the knot unchanged. They also aect respectively
1,2,3 strands leaving the rest of the knot unchanged:
A knot is chiral if it is not equal as a knot to its mirror image.
A knot is tricolourable if the arcs of its knot diagram can be coloured using
three colours (all of them used somewhere in the diagram) such that at each
crossing either all three arcs there have the same colour or all three have dierent
colours. If you think a knot is tricolourable then tricolour it. If you think its not
tricolourable then construct a logical proof that it never can be tricoloured.
Writhe W is the sum over all crossings in the knot diagram of 1 according
to a sign convention. There are many ways to remember that, my suggested is
12
that +1 is when the upper arrow has to rotate anticlockwise to line up with the
lower arrow. W is invariant under moves II, III (and behaves simply under I).
Kauman bracket B(x) has inductive denition
(make sure you know what the notation means). This is invariant under moves
II, III and behaves under move I as
There are many ways to remember the powers of x, my suggested is that when
the upper arc goes down and to the right over the lower strand of the knot diagram
(a forward crossing) this tends to be associated with positive powers of x.
The Jones polynomial J(t) = (x)
3W
B(x) with t = x
4
is unchanged under all
three moves (a knot invariant). If a J(t) computation comes out with fractional
powers of t, i.e. powers of x not divisible by 4, then you made a mistake. One
result was that J(t) J(t
1
) under mirror-reversal of the knot, which can be a
way to prove a knot is chiral.
curves: denitions of unit speed, regularity; curvature for a regular curve and
torsion for a regular curve with nonvanishing curvature:
K =
|| ||
|| ||
3
, T =
( )
...

|| ||
2
K measures deviation from a straight line and T measures deviation from being
planar.
reparametrization just means a change of variables from t to

t where t = (

t).
The same curve (t) is then (

t) = (

t). Key to proofs was the chain


rule d /d

t = (d/dt)(d/d

t). Curvature, torsion etc are all invariant under


reparametrization. Formulae simplify for unit speed curves and every regular
curve has a unit speed reparametrization.
An orthonormal triple t, n, b at each point of the curve and the Serret-Frenet
equations:
t =

|| ||
, n = b t, b =

|| ||
d
dt

t
n
b

= || ||

0 K 0
K 0 T
0 T 0

t
n
b

Using the Serret-Frenet we showed that (1) T = 0 the curve is planar (2) a
unit speed curve is uniquely determined up to translations and rotations by its
curvature and torsion.
Make sure your equations make sense, you cant equate a vector with a scalar.
We often used dot and cross products identities such as
(a b) c = (a c)b (b c)a, (a b) c = a (b c)
13
a b = 0 i a b; a, b a b.
For a planar curve in the x y plane and k = (0, 0, 1) the signed curvature
and a result about it:
K
S
=
( ) k
|| ||
3
,
anticlockwise angular change
(t
0
) to (t
1
)
=

t
1
t
0
K
S
|| ||dt.
This also applies to a closed curves of period a: the anticlockwise angular change
in direction going round the curve (which will be some integer multiple of 2) is

a
0
K
S
|| ||dt.
surfaces: regular surface, the normal vector N

of a surface patch (normal to the


tangent plane), orientable and non-orientable surfaces; fundamental forms F
I
, F
II
:
N

=

u

v
||
u

v
||
, F
I
=

u

u

u

v

u

v

v

v

, F
II
=


uu
N


uv
N


uv
N


vv

|| ||
2
= ( u, v)F
I

u
v

R
dA =

det(F
I
)dudv
shows the role of F
I
to measure speed of a curve (t) = (u(t), v(t)) in a surface
and area of a region R = (

R) of a surface in terms of u, v. Similarly,
K
n
=
( u, v)F
II

u
v

|| ||
2
=
N
|| ||
2
, K
G
= K
1
K
2
=
det(F
II
)
det(F
I
)

> 0 elliptic
< 0 hyperbolic
= 0 parabolic
shows the role of F
II
to dene the normal curvature of a curve in a surface and
the Gauss curvature (and hence type) of a point of a surface.
We had some more advanced theory; principal curvatures K
1
, K
2
, principal
directions t
1
, t
2
and Eulers theorem that K
n
lies between K
1
, K
2
.
Being able to sketch and compute unseen examples is an important part of
the course. We had examples like the sphere, torus, paraboloid, cone, or more
generally a surface of revolution
(u, v) = (f(u) cos(v), f(u) sin(v), g(u))
generated by a curve = (f, 0, g) in the x z plane rotated about the z-axis (we
particularly looked at the case where was constant speed i.e. ||

||
2
= f
2
+ g
2
constant, eg unit speed.)
geodesics: means the geodesic curvature K
g
of the curve in a surface vanishes or
that n is parallel or antiparallel to N:
Kn = K
n
N+ K
g
Nt, K
g
=
(t ) N
|| ||
2
geodesic n||N K
g
= 0
d
dt
(E u + F v) =
1
2
(E
u
u
2
+ 2F
u
u v + G
u
v
2
)
d
dt
(F u + G v) =
1
2
(E
v
u
2
+ 2F
v
u v + G
v
v
2
)
where the explicit formula assumes F
I
=

E F
F G

and that constant speed.


14
As a main result we showed that for unit speed geodesics on a surface of
revolution where has unit speed,
u = ff

v
2
,
d
dt
(f
2
v) = 0, ( f
2
v is constant), u =

1

2
f
2
.
Meridians are given by constant v (constant angle of rotation) and are al-
ways geodesics. Parallels are given by constant u (hence constant height) and
are geodesics i f

= 0 (locally maximum or minimum distance from the axis).


In general = f sin gives a geometric interpretation of in terms of distance
f from the axis and angle to the relevant meridian. (We also explained that
geodesics have shortest/longest (i.e. extremal) distance between two points.)
Gauss-Bonnet theorem: statement of the three (increasingly more general) versions
of the theorem. Make sure you understand the terms in the theorems.
(1) for a compact (oriented) surface S with g holes

S
K
G
dA = 4(1 g), dA =

det(F
I
)du dv
(2) for a positively oriented s.c.c. of period a and int() = (int()) where
(t) = ((t))

int()
K
G
dA = 2

K
g
ds,

K
g
ds =

a
0
K
g
|| ||dt
(3) for a positively oriented curvilinear polygon with n segments and interior
angles
i
.

int()
K
G
dA =
n

i=1

i
(n 2)

K
g
ds
I said that I do not expect you to memorise the (long) proof of the full GB
theorem and that emphasis assuming the topic comes up would be on being able
to state and apply the theorems. I reminded you to collect cwk 10 scripts from
maths 101 and that its solutions are on the web.
I said that following these later segments of the module (geodesics and GB)
was also revision of the earlier more elementary sections, so worth the eort even
if you are struggling.
Good luck!
S.M. 22/4/14

You might also like