Developer Installation

Though we plan to bundle all this together in the future, for now, to use PSST, you must install:

Python 2.7 or 3.5+ (preferably the distribution from Continuum Analytics)
Git
PSST source code
The additional Python packages Graphviz and Scipy
A mixed-integer solver like CBC

Installing the proper Python

Python is a free, open source language that can be downloaded from Python.org. Python comes with a package manager called Pip which allows you to download pacages to gain extended functionality.

However, we recomend that you do not use the default distribution of Python, but rather, install the Miniconda (or Anaconda) distribution from Continuum Analytics.

Why? Many of the Python packages used for scientific applications, like Numpy and Scipy, are written in a mixture of pure Python plus other, lower-level languages, like C or Fortran. Anaconda comes with many of these packages already installed. You don't need all of them, which is why we recomend Miniconda instead of Anaconda. Both Anaconda and Miniconda come with an alternative package manager, conda, which lets you install them with greater ease and reliability than when using pip. For packages written entirely in Python, you should still use pip to install.

Python 2 or Python 3?

PSST should be fully compatible with Python 2.7, but we recomend Python 3 as it is faster, cleaner, more robust, and, well, the future!

Download the Miniconda distribution of Python 3

Find the download here
During installation, mark that you do want to add Miniconda to your path
On Windows, we recomend installing Miniconda directly under your C:\ directory

Verify your installation.

Open a command line console ("command prompt" on Windows or "Terminal" on Mac/ Ubuntu) and type:

macOs and Linux -- which python
Windows -- where python

You should see output like:

macOs and Linux -- /home/username/miniconda3/bin/python
Windows -- C:\Miniconda3\python.exe

You should also be able trun python from the command line by simply typing python or python3.

You should see something like this:

Troubleshooting

If this isn't working, you might still need to add the Miniconda distribution to your system "Path" or place it before an existing installation of Python in your path.

Look at your path from the command line by typing:

macOs and Linux -- echo $PATH
Windows -- echo %PATH%

Here are some tips from docs.continuum.io/:

The process is similar if you're on Windows. Try uninstalling and reinstalling, or manually add the Miniconda executable to your path, by going into Computer > Properties > Advanced System Settings > Environment Variables and then adding the path to your miniconda installation (something like c:\.....\anaconda) near the front of your PATH variable.

Installing Git

If you don't already have Git installed on your computer, you should install it. This is version control software, which will make it easier to install and update your version of PSST.

On Windows

Install Git Bash, and then launch this application from the start menu.

You can use this application in place of the default command line console, 'Command Prompt.' It gives you access not only to git commands (like git clone and git bash but also to most folder navigation commands that macOS/linux users have access to. That is, pwd to see the current directory, cd to change directory, ls to see the files and folders you can access from your current directory, and more.

On macOS/ Linux

You should already have git installed. If you wish, you could follow these steps to upgrade it.

Installing PSST

Open your command prompt console (Terminal on linux/macOS or Git Bash on Windows)
Navigate to the directory in which you want to store the code. You may wish to create a GitRepos directory in an easy-to-reach directory. For instance:
```
cd ~
mkdir GitRepos
cd GitRepos
```

Enter following to clone the project, then install it.

git clone https://github.com/power-system-simulation-toolbox/psst.git
cd psst
pip install -e .

Installing Additional Libraries

A few packages/libraries don't get installed with the previous command, you will have to add them manaully. These code include:

Scipy

conda install scipy

Jupyter

conda install jupyter

GraphViz

On Windows

pip install graphviz

On Mac

Install homebrew

brew install graphviz --with-gts
pip install graphviz
pip install pygraphviz

On Ubuntu

To install Graphviz with gts:

Remove any existing installation with conda uninstall graphviz. (If you did not install with Conda, you might need to do sudo apt purge graphviz and/or pip uninstall graphviz).
run sudo apt install libgts-dev
run sudo pkg-config --libs gts
run sudo pkg-config --cflags gts
Download graphviz-2.40.1.tar.gz from here
Navigate to directory containing download, and extract with tar -xvf graphviz-2.40.1.tar.gz (or newer whatever the download is named.)
cd into extracted folder (ie cd graphviz-2.40.1) and run sudo ./configure --with-gts
Run sudo make in the folder
Run sudo make install in the folder
Reinstall library using pip install graphviz

Note: It is possible you will get an error pertaining to strings vs. bytes. If so, you can edit a file in the graphviz pacakge by hand.

See this github issue.

Installing a solver

Finally, you need a linear programming (LP) and mixed integer programming (MIP) solver to solve optimization problems. Any one will due. You can install multiple, and specify which you want to use in the psst code.

Installing COIN/CBC solver

There are some install instructions here. These procedures seem to work best:

On Windows

Install the most recent binary of COIN-OR from this list, ie
Add :/c/Program Files (x86)/COIN-OR/1.8.0/win32-msvc12/bin to the computer's path.
You may need to restart your computer.

On MacOS

brew tap coin-or-tools/coinor
brew install cbc

On Ubuntu

conda install -c conda-forge coincbc

Installing GLPK solver

On Ubuntu

sudo apt-get install glpk-doc glpk-utils libglpk-dev

On MacOS

brew install glpk



In [ ]: