nklomp / wind_estimation

Estimation of Multi-Variate Ornstein-Uhlenbeck Wind Process

This repository contains a dataset and a script for modeling wind as a stochastic process. It consists of three folders, described below.

Dataset

Wind data from Hollandse Kust Noord (site B).

Description

Wind data from Hollandse Kust Noord (site B), originally published by Netherlands Enterprise Agency (RVO). The dataset includes measurements from 2017-04-10 to 2019-04-10, in 10-minute intervals. The data includes:

• time of the measurement (in "%Y-%m-%dT%H:%M:%OSZ" format, e.g., "2017-04-10T00:00:00Z"),
• direction of the wind (in degrees, from north)
• speed (in m/s)
• ti (turbulence intensity, unitless)
• veer (in degrees/m)
• shear (in 1/s)

Tags

wind, offshore wind, Hollandse Kust, the Netherlands

Script

Multi-variate Ornstein-Uhlenbeck wind estimator

Description

This script is used for modeling wind as a stochastic process. The estimation is based on historical data and uses a model called multi-variate Ornstein-Uhlenbeck (MVOU) process. This wind process is used in experiments presented in Neustroev, G., Andringa, S. P., Verzijlbergh, R. A., & De Weerdt, M. M. (2022, May). Deep Reinforcement Learning for Active Wake Control. In Proceedings of the 21st International Conference on Autonomous Agents and Multiagent Systems (pp. 944-953).

Tags

wind, stochastic process, wind simulation, multi-variate Ornstein-Uhlenbeck process, wind modeling

Outputs

The output folder contains a sample output file.

How to Use This Repository

You need to install R and two libraries, yaml (for writing the output), and pracma (for matrix algebra). The main script is script/estimate_wind.R. It takes a single command line argument, which is the path to the input file. To run the script, use the following command from the script directory:

Rscript estimate_wind.R PATH_TO_THE_DATASET

Here PATH_TO_THE_DATASET is either "../dataset/HKNB.csv" or another file in the same format.

If you want to run and manage this from the EnergySHR platform, you would have to provide the following arguments for the algorithm during publishing, where $ALGO will be replaced with the link from the algorithm file (estimate_wind.R) and $INPUT with the dataset file

Rscript $ALGO /data/inputs/<id_of_dataset>/0

Citation

Please, cite the paper if you use it:

@inproceedings{Neustroev2022,
  title     = {Deep Reinforcement Learning for Active Wake Control},
  author    = {Neustroev, Grigory and Andringa, Sytze P.E. and Verzijlbergh, Remco A. and de~Weerdt, Mathijs M.},
  booktitle = {International Conference on Autonomous Agents and Multi-Agent Systems},
  year      = {2022},
  address   = {Online},
  publisher = {IFAAMAS},
  month     = {May},
  numpages  = {10}
}