Calculus#
Functions and models#
even functions: \(f(x) = f(-x)\)
odd functions: \(f(-x) = -f(x)\)
polynomials: \(f(x) = x^2 + 2x + 1\)
power functions: \(f(x) = x^a\)
rational functions: \(f(x) = \frac{P(x)}{Q(x)}\) where \(P(x)\) and \(Q(x)\) are polynomials
algebraic functions
transcendental functions: \(\sin x \cos x \, e^x \, \ln x\)
exponential functions: \(f(x) = b^x\)
logarithmic functions: \(f(x) = \log_b x\)
Translation:
up/down: \(y = f(x) \pm c\)
right/left: \(y = f(x \mp c)\)
Stretching:
vertical: \(y = cf(x)\)
horizontal \(y = f(c^{-1}x)\)
Reflection:
about the x-axis \(y = -f(x)\)
about the y-axis \(y = f(-x)\)
The composition of two functions:
Note that
Inverse functions: reflect about the line \(y = x\)
Laws of logarithms;
Limits and derivatives#
Limits#
One-sided limits:
The Heaviside function:
Strategy: Calculating limits
try to reduce the number of occurrences of \(x\) in \(f(x)\)
try to make the expression inside the limit as simple as possible by applying limit laws
consider using two one-sided limits to determine a limit
The precise definition of a limit:
if for every number \(\varepsilon > 0\) there is a number \(\delta > 0\) such that
Continuity:
\(f(x)\) is continuous at a number \(a\) if
Tip
Warning
Limit laws don’t apply to infinite limits!
The following is wrong!!!
Error
Derivatives#
For function
Its derivative is
Differentiation rules:
Important
The Chain Rule:
Or alternatively:
Tip
The Power Rule combined with the Chain Rule:
The derivative of any exponential function:
Tip
Using the Chain Rule repeatedly:
Recursively “peel” the function by differentiating the outmost function evaluated at the inner function and then multiplying the result by the derivative of the inner function.
Implicit differentiation#
Important
Remember \(y\) is a function of \(x\)
Warning
Failing to differentiate a constant \(c\) and leaving it non-zero is a common error!
Error
Logarithmic differentiation#
Example
Differentiate
analyze: \(e^{-x} > 0\); \(\cos^2 x \ge 0\); \(x^2 + x + 1 > 0\), therefore \(y \ge 0\) and we can take the logarithm without using absolute values
take the natural logarithm of both sides:
\[ \ln y = -x + \ln \cos^2 x - \ln(x^2 + x + 1) \]differentiate implicitly with respect to x:
\[ \frac{1}{y}\frac{dy}{dx} = -1 + \frac{1}{\cos^2 x}[2\cos x\cdot(-\sin x)] - \frac{1}{x^2 + x + 1}(2x + 1) \]substitute \(y\) with the original function and simplify
\[ \frac{dy}{dx} = -\frac{e^{-x}\cos^2x}{x^2 + x + 1}(1 + \frac{\sin 2x}{\cos^2 x} + \frac{2x + 1}{x^2 + x + 1}) \]
Rates of change in physics#
Position, velocity and acceleration
Instantaneous velocity is the derivative of the position function with respect to time:
Acceleration is the derivative of the velocity function with respect to time:
Charge and current
Exponential growth and decay#
Consider the differential equation
It has only one solution:
This function is extremely common in various sciences.
Applications of differentiation#
Maximum and minimum values
- Fermat’s theorem
If \(f\) has a local extremum at \(c\), and if \(f'(c)\) exists, then \(f'(c) = 0\).
- Critical number
a number \(c\) such that \(f'(c) = 0\) or \(f'(c)\) doesn’t exist
Strategy: The closed interval method
To find the absolute maximum/minimum of a continuous function \(f\) on a closed interval \([a, b]\),
calculate \(f(c)\) for every critical number \(c\) on the open interval \((a, b)\)
calculate the endpoints
The mean value theorem
Increasing/decreasing test (first derivative)
The first derivative test
Concavity test
Inflection point
The second derivative test
Important
L’Hospital’s Rule:
If limit
is of the indeterminate form (\(\frac{0}{0}\) or \(\frac{\infty}{\infty}\)), then
provided that \(f'(x)\) and \(g'(x)\) exist.
Curve sketching:
Domain
Intercepts (if the equation is easy to solve)
Symmetry (even/odd) / periodic
Asymptotes
horizontal: \(\lim_{x \to \pm\infty}\)
vertical: \(\lim = \pm\infty\)
slant: \(\lim_{x \to \pm\infty}[f(x)-(mx+b)] = 0\)
Intervals of increase/decrease
Local max/min values
Concavity and points of inflection
Newton’s method:
Integrals#
The definite and indefinite integrals#
Table of indefinite integrals:
Some more exotic ones:
The FUNDAMENTAL theorem of calculus#
Part 1
If
then
Part 2
where \(F\) is any antiderivative of \(f\) (\(F' = f\)).
Important
A few VERY useful trigonometric identities:
Let \(u = \tan\frac{x}{2}\), then
Integration techniques#
This section is particularly important because integration is a tricky process and requires a significant amount of skill as well as ingenuity.
try the formulas given above
simplify the integrand if possible, e.g.
\[ \int \frac{\tan x}{\sec^2 x}\,dx = \int \sin x\cos x\,dx = \frac{1}{2}\int\sin 2x\,dx \]try the substitution rule:
look for a part in the expression whose derivative also exists in the integrand
let the part be \(u\) where \(u = g(x)\)
evaluate \(du = g'(x)\,dx\)
substitue every \(x\) in the integrand with appropriate forms of \(u\)
evaluate the integral \(\int f(u)\,du\)
If it is an indefinite integral, substitute \(x\) back in. Otherwise, replace the limits of integration for \(x\) with the corresponding values for \(u\).
classify the integrand:
trigonometric functions:
powers of \(\sin x\) and \(\cos x\), or powers of \(\tan x\) and \(\sec x\), or powers of \(\cot x\) and \(\csc x\):
use trigonometric identities to transform the expression and then use the substitution rule
others:
apply the VERY useful identity above to eliminate everything but \(\tan\frac{x}{2}\)
rational functions:
use partial fractions:
long division
\[ f(x) = \frac{P(x)}{Q(x)} = S(x) + \frac{R(x)}{Q(x)} \]write \(f(x)\) in this way
\[ f(x) = \sum_{i=1}^m\sum_{u=1}^{u_i}\frac{A_{iu}}{(a_ix + b_i)^u} + \sum_{j=1}^n\sum_{v=1}^{v_j}\frac{A_{jv}x + B_{jv}}{(a_jx^2 + b_jx + c_j)^v} \]evaluate the transformed integral section by section, completing squares and substituting with \(u\) when necessary
radicals:
\(\sqrt{\pm x^2 \pm a^2}\):
Expression
Substitution
Identity
\(\sqrt{a^2-x^2}\)
\(x = a\sin\theta, \;-\pi/2 \le \theta \le \pi/2\)
\(1 - \sin^2\theta = \cos^2\theta\)
\(\sqrt{a^2+x^2}\)
\(x = a\tan\theta, \;-\pi/2 < \theta < \pi/2\)
\(1 + \tan^2\theta = \sec^2\theta\)
\(\sqrt{x^2-a^2}\)
\(x = a\sec\theta, \;0 \le \theta < \pi/2\;\text{or}\;\pi \le \theta < 3\pi/2\)
\(\sec^2\theta - 1 = \tan^2\theta\)
\(\sqrt[n]{ax + b}\):
use substitution \(u = \sqrt[n]{ax + b}\)
other:
try integration by parts:
separate the integrand into two parts \(u\) and \(dv\) keeping in mind that \(u\) should be as simple as possible while \(dv\) should be easy to integrate
plug everything in:
\[ \int u\,dv = uv - \int v\,du \]
Differential equations#
First-order ODEs#
- Separable equations
Differential equations that can be written in the form
To solve separable equations,
put all of \(y\) on one side and all of \(x\) on the other side
\[ h(y)\,dy = g(x)\,dx \]integrate both sides
\[ \int h(y)\,dy = \int g(x)\,dx \]
- Linear equations
Differential equations that can be put into the form
To solve linear equations,
let
\[ I(x) = e^{\int P(x)\,dx} \]multiply both sides by \(I(x)\)
transform the left side into a derivative with the product rule
integrate both sides
Second-order ODEs#
- Second-order linear homogeneous equations
equations in the form
If \(P(x)\), \(Q(x)\) and \(R(x)\) are constant-valued functions (\(a\), \(b\) and \(c\)), the second-order differential equation can be easily solved by using auxiliary equations.
- Auxiliary equations
the quadratic equations
After solving the auxiliary equation, we can get the solution for the original second-order homogeneous linear differential equation (OMG, that’s really a mouthful):
Roots of the auxiliary equation |
General solution |
---|---|
\(r_1\), \(r_2\), real and distinct |
\(y = c_1e^{r_1x} + c_2e^{r_2x}\) |
\(r_1 = r_2 = r\) |
\(y = c_1e^{rx} + c_2xe^{rx}\) |
\(r_1\), \(r_2\), complex: \(\alpha + i\beta\) |
\(y = e^{\alpha x}(c_1\cos\beta x + c_2\sin\beta x)\) |
For non-homogeneous linear equations such as
it can be proven that the general solution is
where \(y_c\) is the general solution of the complementary homogeneous equation
and \(y_p\) is a particular solution of the original equation.
To find a particular solution, use either the method of undetermined coefficients or the method of variation of parameters.
Method: Undetermined coefficients
guess the form of the solution by classifying the function \(G(x)\)
\(Ce^{kx}\;\to\;Ae^{kx}\)
a polynomial \(P(x)\;\to\;Q(x)\), where \(Q(x)\) is a polynomial of the same degree as \(P(x)\)
\(C\sin kx \;\text{or}\; C\cos kx\;\to\;A\cos kx + B\sin kx\)
Everything above can be freely combined or superimposed together.
If the guessed form is already a solution of the complementary equation, multiply it by \(x\) or \(x^2\)
differentiate the function and substitute it into the differential equation
solve for the coefficients
Method: Variation of parameters
replace the constants \(c_1\) and \(c_2\) in \(y_c\) with functions \(u_1(x)\) and \(u_2(x)\)
differentiate the altered function \(y_p\) and simplify it by imposing conditions
Tip
The condition
\[ u_1'y_1 + u_2'y_2 = 0 \]often works well.
plug everything into the original equation and let
\[ ay'' + by' + cy \]be zero, because \(y_1\) and \(y_2\) are particular solutions of the complementary equation.
solve for \(u_1\) and \(u_2\)
If, however, the coefficients of \(y\) are not constant-valued functions, some of the second-order ODEs can still be solved by downgrading them to first-order ODEs.
If the equation is in the form
then let \(p = y'\). Therefore
and the original equation turns into a first-order ODE:
If the equation is in the form
then simply substitute \(y'\) with \(p\) and integrate \(p\) after solving the transformed equation.
Conic sections#
Parabolas#
- Parabola, focus, directrix
The set of points in a plane that are equidistant from a fixed point \(F\) (called the focus) and a fixed line (called the directrix)
- Vertex
The point halfway between the focus and the directrix
- Axis
The line through the focus perpendicular to the directrix
An equation of the parabola with focus \((0, p)\) and directrix \(y = -p\) is
Or equivalently:
Ellipses#
- Ellpise, foci
An ellipse is the set of points in a plane the sum of whose distances from two fixed points \(F_1\) and \(F_2\) is a constant. The two fixed points are the foci.
- Vertices
The points \((a, 0)\) and \((-a, 0)\) in the equation above
Foci:
Hyperbolas#
- Hyperbola
The set of points in a plane the difference of whose distances from two fixed points \(F_1\) and \(F_2\) is a constant.
The hyperbola
has foci \((\pm c, 0)\) where \(c^2 = a^2 + b^ 2\), vertices \((\pm a, 0)\) and asymptotes \(y = \pm(b/a)x\).
Conic sections in polar forms#
Eccentricity \(e\), directrix \(x = \pm d\)
Eccentricity \(e\), directrix \(y = \pm d\)
Specifically, the polar equation of an ellipse with focus at the origin, semimajor axis \(a\), eccentricity \(e\), and directrix \(x = d\) can be written as:
In the equation above, the perihelion distance is \(a(1 - e)\) and the aphelion distance is \(a(1 + e)\).
Multivariable calculus#
TODO
Vector calculus#
TODO
Complex numbers#
A complex number
has a real part and an imaginary part:
Its complex conjugate is
The modulus of \(z\) is
The polar form of a complex number \(z\):
To multiply (or divide) complex numbers in polar form, simply multiply (or divide) the moduli and add (or subtract) the arguments.
Important
Euler’s formula:
Therefore