MATHEMATICS 3: LESSON PLAN

(Approximate lecture time: 3 hours)

DIFFERENTIAL CALCULUS

Slope [FERM 7:1]

The slope of y(x) is y′ (x) for all x.

Examples: Slope

(1) Given: y(x) = 3x3 – 2x2 + 7. What is the slope of the function y(x) at x = 4?

y′(x) = 9x2 – 4x
y′(4) = (9)(4)2 – (4)(4)
= 128

(2) Given: y1′ = ½(1 + 4x – 7 + 2k). What is the value of k such that y1 is perpendicular to the
curve y2 = 2x at (1,2)?

Perpendicular implies that m1m2 = –1

Since y2′(1) = 2, then
y1′(1) = –½ = ½(1 + (4)(1) – 7 + 2k)
k=½

Maxima [FERM 7:1]

y = f(x) is maximum for x = a where f ′ (a) = 0 and f ″ (a) < 0

Example: Maxima

What is the maximum of the function y = –x3 + 3x for x ≥ –1?

(A) –2 (B) –1 (C) 0 (D) 2

y′ = –3x2 + 3
y″ = –6x
When y′ = 0 = –3x2 + 3
x2 = 1; x = ±1
y″(1) = –6 < 0; therefore this is a maximum.
y″(–1) = 6 > 0; therefore this is a minimum.
(Note: Since the minimum is less than the maximum, then all values of y will be less
than the value at the maximum. The point is that local maxima and minima are not
necessarily the maximum and minimum of the function. If the problem gives an

Copyright 2001 Professional Publications, Inc. Math 3 - 1

 interval, it’s usually necessary to check the limits of the interval if the problem asks for
the maximum or minimum value.)
y(1) = –(1)3 + 3 = 2
Therefore (D) is correct.

Minima [FERM 7:1]

y = f (x) is minimum for x = a where f ′ (a) = 0 and f ″ (a) > 0

Inflection points (EIT8 8:3)

y = f (x) is an inflection point for x = a where:

f ″ (a) = 0 and
f ″ (a) changes sign about x = a

Example: Inflection points

What is the point of inflection of the function y = –x3 + 3x – 2?

(A) (0,0) (B) (0,2) (C) (0,–2) (D) (2,0)

y′ = –3x2 + 3
y″ = –6x
y″ = 0 when x = 0 and y″ > 0 for x < 0; y″ < 0 for x > 0
Therefore this is an inflection point.
y(0) = –(0)3 + 3(0) – 2 = –2
Therefore (C) is correct.

Partial derivatives [FERM 7:2]

Treat everything that is not the variable being differentiated as a constant.

Example: Partial derivative

What is the partial derivative of P(R,S,T) taken with respect to T?

P = 2R3S2T½ + R¾Scos 2T

Everything that is not a function of T is treated as a constant.

P = 2R3S2(T½) + R¾S(cos 2T)
∂P/∂T = 2R3S2(½T-½) + R¾S(–2 sin 2T)
= R3S2T-½ – 2R¾S sin 2T

Copyright 2001 Professional Publications, Inc. Math 3 - 2

Curvature [FERM 7:2]

Curvature is defined as the limit of the angle over the length as the length goes to zero.
y ′′ − x ′′
K= 3 = 3

[1 + ( y ′) ]
2 2
[1 + ( x ′) ]
2 2

The radius of curvature is the radius of a circle that would be tangent to the function at any
point. The radius of curvature is the inverse of curvature magnitude, because the radius of
curvature must always be positive.
3 3

1 [1 + ( y ′) 2 ] 2 [1 + ( x ′) 2 ] 2
R= = =
K y ′′ x ′′

Example: Curvature

What is the curvature of y = –x3 + 3x for x = –1?

(A) –2 (B) –1 (C) 0 (D) 6

y′ = –3x2 + 3 y″ = –6x
y′(–1) = 0 y″(–1) = 6
y′′ 6
K= 3 = 3 =6
[1 + ( y′) ]2 2
[1 + (0) ]
2 2

Therefore (D) is correct.

Limits [FERM 7:2]

If the limit goes to plus or minus infinity, divide the numerator and the denominator by
largest power.

Factor out common terms and divide the numerator and the denominator

Look at what the function does as it approaches the limit.

If the limit goes to plus or minus infinity:
Look for constants that become irrelevant
Look for functions that blow up fast: a factorial, an exponential
If the limit goes to a finite number:
Look at what happens at both plus and minus a small number:

Copyright 2001 Professional Publications, Inc. Math 3 - 3

 lim  f ( x)   f (a )  0  f (a)  ∞
Looking for   but   = or   = : use L’Hôpital’s rule.
x → a  g ( x)   g ( a )  0  g (a)  ∞

lim  f ( x)  lim  f ′( x)  lim  f ′′( x)  lim  f ′′′( x) 

  =   =   =  ...
x → a  g ( x)  x → a  g ′( x)  x → a  g ′′( x)  x → a  g ′′′( x) 

provided the next derivative of f(x) and g(x) exist. If not, L’Hôpital’s rule
can’t be used.

Examples: Limits
lim  x + 4 
(1) What is the value of  ?
x → ∞ x − 4 
(A) 0 (B) 1 (C) ∞ (D) Undefined

Divide the numerator and denominator by x

 4
lim  x + 4  lim  1 + 
x  = 1+ 0 =1
 = 
x → ∞ x − 4  x → ∞ 4  1− 0
1 − 
 x
Therefore (B) is correct.

x2 − 4
(2) What is the limit as x → 2 of ?
x−2
(A) 0 (B) 2 (C) 4 (D) ∞

Factor out an (x – 2) term in the numerator

lim  x 2 − 4  lim  ( x − 2)( x + 2)  lim
 =   = ( x + 2) = 2 + 2 = 4
x → 2  x − 2  x → 2  x−2  x→2
Therefore (C) is correct.

Note: Although L’Hôpital’s rule does work on this problem, this solution is faster and
easier.

lim  1 − cos x 
(3) Compute the following limit:  
x → 0  x2 
(A) 0 (B) ¼ (C) ½ (D) ∞

Both the numerator and denominator approach 0, so use L’Hôpital’s rule:

lim  1 − cos x  lim  sin x 
 =  
x → 0 x 2
 x → 0  2x 

Copyright 2001 Professional Publications, Inc. Math 3 - 4

 Both the numerator and denominator are still approaching 0, so use L’Hôpital’s rule again:
lim  sin x  lim  cos x  cos(0) 1
 =  = =
x → 0  2x  x → 0  2  2 2
Therefore (C) is correct.

INTEGRAL CALCULUS

Constant of integration [FERM 7:3]

Derivatives of constants are zero, so when the derivative is undone, a constant must be
added. To restore the constant, we have to know some value of the original function for
some x = a.

Example: Constant of integration

What is the constant of integration for y(x) = ∫(e2x + 2x)dx if y = 1 when x = 1?

(A) C = 2 – e2 (B) C = – ½e2 (C) C = 4 – e2 (D) C = 1 + 2e2

y(x) = ½e2x + x2 + C
y(1) = ½e2 + 1 + C = 1
C = – ½e2
Therefore (B) is correct.

Indefinite integrals [FERM [7:3]

Look for ways to simplify the formula with algebra before integrating.
Plug in initial value(s).
Solve for constant(s).
Indefinite integrals can be solved by differentiating the answers, but this is usually the hard
way.

Integration by parts

Complicated functions that don’t have integrals listed in the NCEES Handbook can be
separated into a product minus an integral that can be integrated with the NCEES
Handbook, using the formula for integration by parts:

∫ u(x) dv(x) = u(x)v(x) – ∫ v(x) du(x)

The trick to integration by parts is recognizing the parts that should be u(x) and v(x).
v(x) should be a trigonometric function or exponential.
u(x) should be a function that becomes a lower order when differentiated. Usually
u(x) is a polynomial.

Copyright 2001 Professional Publications, Inc. Math 3 - 5

Example: Integration by parts

Find ∫x2ex dx

Let v(x) = ex and u(x) = x2

so dv(x) = ex dx and ∫ x2ex dx = x2ex – ∫ 2 xex dx
e ax
From the NCEES Handbook: ∫ xe dx = 2 (ax − 1)
ax

a
Therefore, ∫x2ex dx = x2ex – 2(xex – ex) + C

Notice that if we had chosen dv(x) = x2 dx and u(x) = ex, the integral gets no better.

Integration by substitution

Integrals that cannot be solved with the formulas in the NCEES Handbook directly can be
solved by substituting some variable for an expression that makes the formula match a
known integral.

Example: Integration by substitution

Find ∫ (ex + 2x)2 (ex + 2) dx

let u(x) = ex + 2x
so du = (ex + 2) dx
and ∫ (ex + 2x)2 (ex + 2) dx = ∫ u2 du = αu3 + C
= α (ex + 2x)3 + C

Partial fractions

If the integral is the quotient of two complicated polynomials, and the denominator has the
higher order, the integral can be simplified by using partial fraction expansion. (Some ratios
of polynomials can only be simplified with factors that have a linear function over a
quadratic, but this is not of interest here because we can’t integrate such functions with the
formulas in the NCEES Handbook, so they will not be on the exam.)

Examples: Partial fractions

6x 2 + 9x − 3
(1) Find ∫ x(x + 3)(x − 1)
dx

Copyright 2001 Professional Publications, Inc. Math 3 - 6

 Using the partial fraction expansion

6x 2 + 9x − 3 A B C A (x + 3)( x − 1) B (x)(x − 1) C (x)(x + 3)

= + + = + +
x(x + 3)( x − 1) x x + 3 x − 1 x(x + 3)(x − 1) x(x + 3)(x − 1) x(x + 3)(x − 1)

So 6x2 + 9x – 3 = A(x + 3)(x –1) + B(x)(x – 1) + C(x)(x + 3)

which gives us three simultaneous equations:
A+B+C=6
2A – B + 3C = 9
–3A = –3

Therefore A = 1, B = 2, and C = 3

6x 2 + 9 x − 3 1 2 3
∫ x ( x + 3)( x − 1)
dx = ∫ dx + ∫
x x+3
dx + ∫
x −1
dx = ln x + 2 ln x + 3 + 3 ln x − 1 + C

If the denominator has repeated roots, then the partial fraction expansion will have all the
powers of that root.

4x − 9
(2) Find the partial fraction expansion of
(x − 3)
2

4x − 9 A B A ( x − 3) B
= + = +
( x − 3) x − 3 ( x − 3) x − 3 ( x − 3) ( x − 3) 2
2 2

so 4x – 9 = Ax – 3A + B
which gives us two simultaneous equations:
A =4
–9 = –3A + B

Thus A = 4 and B = 3

4x − 9 4 3
Therefore = +
( x − 3) x − 3 ( x − 3) 2
2

Definite integrals (EIT8 9:7)

Solve the indefinite integral (without the constant of integration).

Evaluate at upper and lower bounds.
Subtract lower bound value from upper bound value.

Copyright 2001 Professional Publications, Inc. Math 3 - 7

Example: Definite integrals

Find the integral between π/3 and π/4 of f(x) = sin x

π π
π  π
∫π
3
sin xdx = [ − cos x] 3
π = − cos −  − cos 
3  4
4 4
= –0.5 + 0.707 = 0.207

Average value (EIT8 9:8)

The average value is the definite integral divided by the width of boundary.
∫ f ( x)dx
1 b
Average =
b−a a

Example: Average value

What is the average value of y(x) = 2x + 4 between x = 0 and x = 4?

[ ]
4
1 4 1  2x 2 
∫
1
Average = ( 2 x + 4) dx =  + 4 x  = 4 2 + 4( 4) = 8
4−0 0 4  2  0 4

Area problems (EIT8 9:9)

The first definite integral of the difference between two functions is the area between the
functions.

Area = ∫a( f 1 ( x ) − f 2 ( x ))dx

b

f1(x)

f2(x)

a b x

Example: Area problems

What is the area between y1 = ¼x + 3 and y2 = 6x – 1 between x = 0 and x = ½?

Copyright 2001 Professional Publications, Inc. Math 3 - 8

 1
 1   1
 23 
Area = ∫ 2
 x + 3 − ( 6 x − 1) dx = ∫0 − x + 4 dx
2
0  4    4 
1
2
 23 2 2 23  1  4 41
= − x + 4 x = −   + =
 8 0 8 2 2 32

CENTROIDS AND MOMENTS OF INERTIA (FERM 7:4]

Centroids and moments of inertia are mainly discussed in the Statics lecture. We are
discussing them here because the formulas in the Statics section require integration. It’s
worth while to take a moment to talk about the integration formulas. (It is unlikely that the
exam will have any problems that require integrating centroids and moments of inertia
because the tables in the Statics chapter of the NCEES Handbook show the results for most
common shapes.)

Centroid
∫ xf (x)dx ∫ yg ( y ) dy
xc = yc =
A A
First moment of area
My = ∫ xf(x)dx = xcA
Mx = ∫ yg(y)dy = ycA
Moment of inertia (EIT8 39:5)
Iy = ∫ x2f(x)dx
Ix = ∫ y2g(y)dy

DIFFERENTIAL EQUATIONS [FERM 8:1]

First-order homogeneous equations

General form: y′ + ay = 0
General solution: y = Ce-ax
Initial condition
Usually y(b) = constant or y′ (b) = constant
y ( b) y ′(b)
C = − ab or C = − ab
e e

Example: First-order homogeneous equations

Find the solution to the differential equation y = 4y′ if y(0) = 1.

(A) 4e-4t (B) ¼e-¼t (C) e-¼t (D) e¼t

Copyright 2001 Professional Publications, Inc. Math 3 - 9

 Rearranging in standard form
4y′ – y = 0
y′ – ¼y = 0
General solution: y = Ce-at

y(b) y(0)
C= −ab
= 1
=1
e (0)
e4
Since a = –¼ and C = 1 then
y = e¼t
Therefore (D) is correct.

Separable equations (EIT8 10:4)

Some differential equations can be separated with all x terms on one side and all y terms on
the other. Then the equation can be solved by integrating both sides.
m(x)dx = n(y)dy

Example: Separable equations

Reduce y′ + 3(2y – sin x) – (x (sin x) + 6y) = 0 to a separable equation.

y′ + 3(2)y – 6y – 3sin x – x(sin x) = 0

dy
= 3 sin x + x (sin x )
dx
dy = (3sin x + x(sin x))dx

Then both sides can be integrated:

y = –3cos x + (sin x – x cos x) + C

Second-order homogeneous equations [FERM 8:1-2]

y″ + 2ay′ + by = 0

Characteristic equation: r2 + 2ar + b = 0

Roots = −a ± a 2 − b

The characteristic equation and the equation for the roots are variations of the quadratic
formula. Be careful to put the polynomial in the right form. If you use the form above, use
the formula for the roots above. You can also find the roots using the usual form for
quadratics and the quadratic equation. Just don’t confuse the two.

Copyright 2001 Professional Publications, Inc. Math 3 - 10

 General solutions:
Overdamped: Real roots (a2 > b)
y = C1 e r1x + C2 e r2 x
Critically damped: equal roots (a2 = b)
y = ( C1 + C2 x) e r1x
Underdamped: Complex roots (a2 < b)
y = eax(C1 cos βx + C2 sin βx)
β = b − a2
Initial conditions
Usually y(constant) = constant and y′ (constant) = constant
Results in two simultaneous equations and two unknowns

Example: Second-order homogeneous equations

y″ + 6y′ + 5y = 0
y(0) = 1
y′(0) = 0

Write the equation in standard form:

y″ + (2)(3)y′ + 5y = 0

The characteristic equation is:

r2 + (2)(3)r + 5 = 0
Roots = −3 ± 32 − 5 = −3 ± 2 = −1,−5

This is the overdamped case because there are two real roots, so the general solution is:

y = C1e −1 x + C 2 e − 5 x
y(0) = 1 = C1 + C2
y′(0) = 0 = –C1 – 5C2
1 = –4C2
C2 = –¼
C1 = 1¼
y = 1¼e-x – ¼e-5x

Nonhomogeneous equations

The solution to a nonhomogeneous differential equation is the sum of the particular solution
and the homogeneous solution.

Copyright 2001 Professional Publications, Inc. Math 3 - 11

 General solution = particular solution + homogeneous solution
To solve the particular solution, you must know the form of the solution. The form of the
particular solution is given in the NCEES Handbook for certain possible forcing
functions, f(x).
To solve the particular solution:
differentiate and then plug into the original equation
collect like terms.
The coefficients of the like terms must sum to zero, giving simultaneous equations.
Solve the equations and determine what the constant(s) is (are).

f(x) yp(x)
A B
Aeαx Beαx α ≠ root of characteristic equation
A1 sin ωx + A2 cos ωx B1 sin ωx + B2 cos ωx

Example: Nonhomogeneous equations

Find the particular solution for the differential equation y″ – y′ – 2y = 10 cos x

From the table in the NCEES Handbook, the particular solution has the form:
yp = B1cos x + B2 sin x
yp′ = – B1 sin x + B2 cos x
yp″ = – B1 cos x – B2 sin x
Plugging in, we get:
– B1 cos x – B2 sin x – (–B1 sin x + B2 cos x) – 2(B1 cos x + B2 sin x) = 10 cos x
(–3 B1 – B2)cos x + (B1 – 3 B2)sin x = 10 cos x
Isolating the sin and cos coefficients, we get the following simultaneous equations:
–3 B1 – B2 = 10
B1– 3 B2 = 0
B1 = –3
B2 = –1
yp = –3 cos x – sin x

Fourier series [FERM 8:2]

The Fourier series is a method for representing any periodic function as an infinite
summation of sin and cos functions plus a constant. If the series is infinite, the Fourier
series is an exact representation of the original function. If the series is finite, the
representation gets closer the more terms there are. With enough terms, the series can be
close enough to the original function for practical purposes, allowing us to perform some
mathematics on the finite series that we could not do with the original function.

a0 ∞
F (t ) = + ∑ (a cos(nωt ) + bn sin( nωt ))
2 n =1 n

Copyright 2001 Professional Publications, Inc. Math 3 - 12

 2π 2π
τ = period = =
ω natural frequency

∫ F (t ) cos(nωt )dt
2 τ
an =
τ 0
bn = ∫0F ( t ) sin(nωt )dt
2 τ
τ

Example: Fourier series

Find the Fourier coefficients for a square wave function f(t) with a period of 2π.

f(t) = –2 when –π < x < 0

f(t) = 2 when 0 < x < π
2 π 1 0 π 
an =
∫
2π −π
F(t ) cos(nt)dt = 
π ∫−π ∫
− 2 cosnxdx+ 2cosnxdx = 0
0 
π 1 0 π 
∫ ∫ ∫
2
bn = F (t ) sin( nt ) dt =  − 2 sin nxdx + 2 sin nxdx 
2π −π π  −π 0 
2( 2)
bn = (1 − cos nπ )
nπ

Laplace transforms [FERM 8:3]

The Laplace transform is a method for solving differential equations. To solve differential
equations with Laplace transforms:

Put the equation in standard form: ay″ + by′ + cy = f(y)

Take the Laplace transform of both sides
Expand terms
 d n f (t )  n −1 m
n − m −1 d f ( 0)
/
 dt n 
 = s n
F ( s ) − ∑ s
dt m
m= 0

Note: It’s worthwhile to know the first two expansions of the Laplace transform because
those are the most complicated forms that might appear on the exam.
/(y″) = s2 / (y) – sy(0) – y′(0) = s2 y – sy(0) – y′(0)
/(y′) = s /(y) – y(0) = sy – y(0)
Plug in the initial conditions y(0) = c; y′(0) = k
Manipulate into a form that has an inverse transform
Take inverse transform

Example: Laplace transforms

Solve by Laplace transform: y″ + 4y′ + 3y = 0, y(0) = 3, y′(0) = 1

Copyright 2001 Professional Publications, Inc. Math 3 - 13

 The equation is already in standard form.
Take the Laplace transform of both sides:
s2y – sy(0) – y′(0) + 4(sy – y(0)) + 3(y) = 0
Plugging in initial conditions and rearranging:
s2y + 4sy + 3y = 3s + 1 + (3)(4)
(s + 3)(s + 1) y = 3s + 13
Solving for y and separating by partial fractions:
3s + 13
y=
( s + 3)( s + 1)
Partial fraction expansion
3s + 13 A B
= +
( s + 3)( s + 1) s + 3 s + 1
A( s + 1) + B ( s + 3)
=
( s + 3)( s + 1)
( A + B ) s + ( A + 3B )
=
( s + 3)( s + 1)
A+B=3
A + 3B = 13
A = –2; B = 5
−2 5
y= +
s + 3 s +1
Taking the inverse Laplace transform of y:
y(t) = –2e-3t + 5e-t
/-1(1/(s + α)) = e-α t

Difference equations [FERM 8:4]

Difference equations are used to model discrete systems where there is an output that
depends on the input over a finite number of steps. The most familiar difference
equations occur in engineering economics. In fact, all the discount factors for discrete
compounding and the factor tables are nothing more than solved difference equations.

The balance on a loan is an example of a first-order difference equation. The principle is

increased at discrete intervals by the interest and decreased by the payment, so the
difference equation for a loan balance is
Pk = Pk–1(1 + i) – A
where Pk is the balance after the kth period, Pk–1 is the balance after (k–1)th period, i is
the interest rate, and A is payment made at the end of the kth period.

The Fibonacci number sequence is an example of a second-order difference equation.

The kth number in the sequence is the sum of the (k–1)th and (k–2)th number in the
sequence. The first two numbers of the Fibonacci sequence have to be defined because

Copyright 2001 Professional Publications, Inc. Math 3 - 14

 they cannot be determined. Usually, the first two numbers are 1 and 1, so the classic
Fibonacci number sequence is:
1, 1, 2, 3, 5, 8, 13, 21, 34, 55,...

The formula for the Fibonacci sequence can be written

y(k) = y(k – 1) + y(k – 2)
where y(–1) = 1 and y(–2) = 1
or
f (k + 2) = f (k + 1) + f (k)
where f (0) = 1 and f (1) = 1

For simple difference equations, it’s probably best to work backward from the possible
answers.

Example: Difference equations

What is the solution to the linear difference equation y(k + 1) = 15y(k)?

15 15k
(A) y ( k ) = (B) y ( k ) = (C) y(k) = C + 15k, C is a constant (D) y(k) =15k
1 + 15k 16

Try (D) by plugging in a (k + 1) for every k

y(k + 1) = 15k+1
y(k + 1) = 15(15k)
y(k + 1) = 15y(k)
so y(k) =15k
Therefore (D) is correct.
Note: If y(k) = C + 15k, then y(k+1) = C + 15k+1 ≠ 15 y(k)

z-transforms [FERM 8:5]

The z-transform is a method for solving difference equations similar to the Laplace
transform used to solve differential equations. To solve difference equations using the z-
transform:
Convert to standard form: y(k + 1) = ay(k)
Take the z-transform of both sides of the equation
Expand terms
Plug in terms: y(0), y(1), y(–1), etc.
Manipulate into a form that has an inverse transform
Take the inverse transform.

Example: z-transforms

Solve the linear difference equation y(k + 1) = 15y(k) by z-transform, given that y(0) = 1

Copyright 2001 Professional Publications, Inc. Math 3 - 15

 Convert to standard form
y(k + 1) – 15y(k) = 0

Take the z-transform

zY(z) – zy(0) – 15Y(z) = 0
Y(z)(z – 15) = z
z 1
Y ( z) = =
z − 15 1 − 15z −1

Take the inverse transform

y(k) = 15k

END OF MATHEMATICS 3 LESSON

Copyright 2001 Professional Publications, Inc. Math 3 - 16