Stochastic optimization for Wind Farm Cable Routing

About

In this project, we introduce uncertainty in the power production for the Wind Farm Cable Routing problem. Please find details about this project in the report.

Requirements

Please install AMPL, Python3 and the following modules :

• amplpy
• numpy
• matplotlib
• networkx

How to run

First find the path to you AMPL installation.

To run the classic deterministic optimization (no uncertainty):

python3 script_deterministic.py 'paht_to_your_ampl_installation'

To run the our pipeline (with uncertainty):

python3 script.py 'paht_to_your_ampl_installation'

Personalization

Number of scenarios

You can change the number of scenarios to consider for stochastic optimization in the file cable_optimization_stochastic.mod by changing the value of n. More scenarios is more accurate but also more computationally expensive.

Data change

The data used data is from the wind farm Kentish Flats, located close to Kent in South East England. If you were to change the data, put the name of you .dat file in both script_deterministic.py and script.py, run first script_deterministic.py and then run script.py.