g-bulgarit/pyPowerFlow

Use numeric methods to solve the power-flow equations in a power distribution network. Code is partially a port of the MATLAB toolbox provided by Hadi Sa'adat along with his book 'Power System Analysis', published by McGraw-Hill, along with other implementations of numeric methods and other visualization tools.

pyPowerFlow

Python tool to solve power flow problems in power networks using numeric methods, given line and bus parameters.

Suppored Numeric Methods

• Newton-Raphson (Port of Hadi Sa'adat's implementation)
• Gauss-Seidel

Supported Generator Policies

• "Tap-Changer": Play with voltages of busses that have generators on them, such that the reactive power Q_gen stays within the realm of Q_min and Q_max.
• "Clamp": Keep the voltages of busses with generators on them constant, and clamp Q_gen to the closest of {Q_min, Q_max}.

Note

Code adapted from Hadi Sa'adat's MATLAB toolbox, published along with his book Power System Analysis, third edition, McGraw-Hill, 1999.

This code was ported and edited to solve a graded assignment in the course Techno-Economical Problems in Power Systems, in Tel-Aviv University, 2022. The changes from the original MATLAB implementation are detailed below.

Changes (from the original implementation)

• No tap-changer support (apart from the policy described earlier), i.e can't input tap changer stops for each generator.
• Added all line currents and losses to the outputs
• Added a network plot, a voltage plot and convergence statistics for each mode.

---

Installation

1. Create a virtual environment python -m venv venv
2. Activate it (On unix: source venv/Scripts/activate, On windows: .\venv\Scripts\activate)
3. Install the requirements: pip install -r requirements.txt
4. (Modify the line and bus parameters)
5. Run the code python main.py

Usage (main.py)

• Input the bus and line data according to the example and headers. The format is consistent with Hadi Sa'adat's format.

• Set the accuracy constraint and the maximum number of iterations

• Set the mode: newton for Newton-Raphson, gauss for Gauss-Seidel.

Example initialization

my_pf =     PowerFlowNetwork(bus_params,
                             line_params,
                             base_power_reference,
                             accuracy, 
                             max_iterations,
                             mode="newton",
                             policy="tap_changer",
                             bus_csv_name="newton_bus_data.csv",
                             line_csv_name="newton_line_data.csv")