Skip to content

Latest commit

 

History

History
executable file
·
177 lines (119 loc) · 5.51 KB

README.md

File metadata and controls

executable file
·
177 lines (119 loc) · 5.51 KB

Matlab Tools

Installation

Using SSH or Linux

Use an SSH client to log in to a Linux server if you are not on one, or just open a terminal. To clone this repository into your Matlab folder (assuming it is at ~/Personal/MATLAB), enter the following commands:

cd ~/Personal/MATLAB
git clone git://github.com/tgvoskuilen/MatlabTools.git

To update your version of the code if you have already cloned the repository:

cd ~/Personal/MATLAB/MatlabTools
git pull

To have the MatlabTools folder added to your Matlab path by default, add the following line to your startup.m file in your default Matlab path (create startup.m if it does not exist):

addpath(fullfile(fileparts(userpath),'MATLAB','MatlabTools'));

Using the GitHub Windows Client

Download GitHub for Windows and install it. Click the Clone in Desktop button on the right to clone this repository onto your local machine. The repository will be saved in your GitHub folder by default.

To update your version of the code, use the 'Sync' button in the GitHub program.

To have the MatlabTools folder added to your Matlab path by default, add the following line to your startup.m file in your default Matlab path (create startup.m if it does not exist):

addpath(fullfile(fileparts(userpath),'GitHub','MatlabTools'));

Using the Source Download

Download the source zip file. Extract its contents to your default Matlab folder and rename the newly created MatlabTools-master folder to MatlabTools.

To update your version of the code, delete the existing MatlabTools folder and repeat the above procedure.

To have the MatlabTools folder added to your Matlab path by default, add the following line to your startup.m file in your default Matlab path (create startup.m if it does not exist):

addpath(fullfile(fileparts(userpath),'MATLAB','MatlabTools'));

Classes

@DimVar

The DimVar class (shorthand for Dimensioned Variable) lets you use quantities with units in Matlab with minimal tedious unit conversion. An example use to calculate heat conduction is:

k = DimVar(2,'BTU-in/hr-ft^2-F');
A = DimVar(1,'cm2');
L = DimVar(1,'in');
DT = DimVar(300,'K') - DimVar(0,'F');
Q = k*A*DT/L

will show

0.05068 [kg-m^2/s^3]

In this example, the units from the individual components are combined and the resulting units on Q are in power (Watts, all derived quantities will be in SI units). The class also checks unit compatibility during addition, subtraction, comparison, and other math operations. For example, the following command would generate an error since k and L have different units:

x = k + L;

However, these commands are valid:

x = L + sqrt(A);
V = L*A;
y = exp(-x/L);
z = 10*A;

If you want to do a unit conversion, for example to convert Q from the example above into BTU/hr, you can do

Qbtu = Q.Convert('BTU/hr');

and Qbtu will be a dimensionless number whose magnitude is now in BTU/hr. If you attempt to convert Q to meters, it will generate an error.

You can combine this class with the features of the @UC class below as well. For example:

k = DimVar(UC(16,2),'W/m-K');
L = DimVar(UC(5,1),'mm');
A = DimVar(UC(10,1),'cm^2');
DT = DimVar(500,'R') - DimVar(200,'K');

Q = k*A/L*DT;
fprintf('Q = %s\n',num2str(Q,4));

will print out

Q = 248.9 ± 63.76 [kg-m^2/s^3]

which has not only propogated the units, but also calculated the resulting uncertainty in Q. For more examples, look in DimVarDemo.m.

@UC

The UC class is for manipulating and tracking uncertainty through operator overloading of basic Matlab scalars and arrays. For usage examples, see UCDemo.m.

You can assign names to UC objects z = UC(15, 10, 'z') and then their contribution to derived uncertainties will be tracked automatically.

If you plot a UC object, it plots the values with both X and Y error bars if the error is nonzero. To plot only the values of an array of UC objects, just use brackets to extract the values as arrays (plot([x.Value],[y.Value]))

This is a continual work in progress, so if you encounter bugs or things are not working the way you expect, let me know.

@Fluid

The Fluid class downloads gas properties from the online NIST database to allow lookup with temperature and pressure inputs. See FluidDemo.m for usage examples.

Functions

GetEvenSpacing.m

If you have densely spaced data points, sometimes it is useful to down-sample the data for plotting markers that are visible. If your data is not evenly spaced, or is plotted on a log scale, this can become more difficult.

This function gets evenly spaced down-sampled points with the option of specifying log or linear plotting scales. For example:

r = GetEvenSpacing(x, y, 50);
plot(x, y, '-k');
hold on;
plot(x(r), y(r), 'sk','MarkerFaceColor','w');

or

r = GetEvenSpacing(x, y, 50, 'linear', 'log');
semilogy(x, y, '-k');
hold on;
semilogy(x(r), y(r), 'sk','MarkerFaceColor','w');

errorbars_xy.m

This can be used to plot data with both X and Y error bars (which Matlab does not currently support).

Demo Scripts

FluidDemo.m

Usage examples for the Fluid class.

GraphDemo.m

General Matlab tutorial on making graphs.

UCDemo.m

Usage examples for the UC class.

DimVarDemo.m

Usage examples for the DimVar class.