This folder contains the respective methods and supporting files needed for identifying static defense strategies with substation configuration.

Static attacks are those attacks that takes place simultaneously. Here, we consider a power system model representing a cyber-physical energy system. This system consists of both physical and cyber components. Physical components represent transmission lines, loads, generators, transformers, buses, etc. However, cyber components represents the protection assemblies such as distance relay, over-current relay and circuit breakers. A power system consists of a large number of substations. Each substation consists of several protection assemblies. These protection assemblies can be remotely monitored and controlled using the remote terminal units located in every substations. If an attacker gain access to a substation then he can manipulate these protection assemblies to disconnect transmission lines from the power system network and cause system damage. However, considering the scale of the power system network it is impossible to exhaustively identify the substations and protection assemblies to attack in order to maximize system damage. Similarly, with a limited defense budget, it is impossible to protect all the necessary substations of the power system network. Hence, we use our game-theoretic approach to model this as an optimization problem and use our "static_defense_substations_v1_test.py" method to obtain an efficient and resonable solution that provides the details on which substations to protect that will minimize the system damage when the system is under attack.

Note: For more details about the problem, system model, etc., please refer to the paper titled "Vulnerability Analysis of Power Systems Based on Cyber-Attack and Defense Models". The document is available on Vanderbilt Universities, Institute for software integrated systems under "Publications".

To run the defense method, the following steps are needed:

1. Install the OpenDSS from Download Link on the Windows machine as OpenDSS is not mac compatible.
2. Download and save the system models from the "System Models" folder to the local drive.
3. Install Python 2.7.
4. Install Spyder, PyCharm or any other Python IDE.
5. Install all the necessary inbuilt python packages using pip install package name. Example pip install numpy.
6. Open the static_defense_substations_v1_test.py method and set the paths for the filepath, comp_filename, load_file_name as per the directory where the downloaded files are stored from the "System Models" folder.

filepath is the path where .dss file is stored on the local drive.

comp_filename is the local drive path where component_data_subs.txt file is stored.

load_file_name is the path for the load_data.txt file stored on the local drive. For certain models the file name can be load_data1.txt or load_data_testing1.txt, etc.

7. Run static_defense_substations_v1_test.py.

1. static_attack_subs_v2.py
2. static_defense_substations_v1_support.py
3. maptest_testing_defense_substations1_v1_support.py
4. trimmed_list_defense_substations1_v1_support.py
5. maptest_testing.py
6. trimmed_list.py
7. static_defense_substations1_v1_support.py
8. post_defense_static_attack_subs.py

Each of these method have their own supporting methods that are not listed here.

more-itertools

pywin32

pypiwin32

numpy

To install dependencies: pip install <dependencyname>; ex: pip install numpy.