https://github.com/openclimatefix/predict_pv_yield_nwp
https://github.com/openclimatefix/predict_pv_yield_nwp
nowcasting
Last synced: 5 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/openclimatefix/predict_pv_yield_nwp
- Owner: openclimatefix
- License: mit
- Archived: true
- Created: 2020-05-29T16:01:18.000Z (over 5 years ago)
- Default Branch: main
- Last Pushed: 2021-10-22T14:22:19.000Z (almost 4 years ago)
- Last Synced: 2025-03-03T08:45:44.047Z (7 months ago)
- Topics: nowcasting
- Language: Python
- Homepage:
- Size: 23.4 KB
- Stars: 2
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
Build a baseline model for predicting PV yield using NWP (numerical weather predictions), as opposed to satellite imagery. This model is intentionally very simple, so we can get an end-to-end system up and running quickly to interate on.
# Model
The model can be found in the _model_ directory. It is a CSV file with columns for the PV system ID, and the slope and intercept for the regression. Given a value for irradiance ("Downward short-wave radiation flux") from the NWP data at a PV system location, the PV output for the system can be predicted.
## Known limitations
* A linear model is a very big simplification.
* Trained on only a single day's worth of data, at hourly time points.
* Outlier data from the PV systems is included.
# Training
The training data is in a GCP bucket, and is not yet publicly available.
## Install and test
```bash
conda env create -f environment.yml
conda activate predict_pv_yield_nwp
pip install -e .
conda install black flake8 isort mypy pytest
pytest -s
```
Enforce/check code standards
```bash
black predict_pv_yield_nwp
flake8
isort -rc .
mypy predict_pv_yield_nwp
```
## Train a model locally
```bash
python predict_pv_yield_nwp/train.py
```