https://github.com/longxingtan/kddcup2022-wpf
3rd Place Solution of KDD Cup 2022-Spatial Dynamic Wind Power Forecasting
https://github.com/longxingtan/kddcup2022-wpf
bert deep-learning kdd2022 kddcup spatio-temporal time-series wind-power-forecasting
Last synced: 6 months ago
JSON representation
3rd Place Solution of KDD Cup 2022-Spatial Dynamic Wind Power Forecasting
- Host: GitHub
- URL: https://github.com/longxingtan/kddcup2022-wpf
- Owner: LongxingTan
- Created: 2022-07-21T01:58:48.000Z (about 3 years ago)
- Default Branch: master
- Last Pushed: 2023-09-21T05:23:16.000Z (about 2 years ago)
- Last Synced: 2025-03-28T13:21:16.081Z (7 months ago)
- Topics: bert, deep-learning, kdd2022, kddcup, spatio-temporal, time-series, wind-power-forecasting
- Language: Python
- Homepage:
- Size: 81.1 KB
- Stars: 116
- Watchers: 1
- Forks: 19
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# KDD Cup 2022 - Baidu Spatial Dynamic Wind Power Forecasting
[](https://www.kaggle.com/code/tanlongxing/kdd-cup-2022-wind-power-forecast)
[](https://arxiv.org/abs/2307.09248)
This is the 3rd place solution in Baidu KDD Cup 2022. The task is to predict the wind farm's future 48 hours active power for every 10 minutes.
## Solution summary
- A single Transformer/ BERT model is made from [the tfts library](https://github.com/LongxingTan/Time-series-prediction). Follow its latest development [here](https://github.com/LongxingTan/Time-series-prediction)
- Using sliding window to generate more samples
- Only 2 raw features are used, wind speed and direction
- The daily fluctuation is added by post-processing to make the predicted result in line with daily periodicity
## How to use it
0. Prepare the tensorflow environment
```shell
pip install -r requirements.txt
```
1. Download the data from [Baidu AI studio](https://aistudio.baidu.com/aistudio/competition/detail/152/0/introduction), and put it in `./data/raw`
2. Train the model, the file `result.zip` in `./weights/` can be used for submit.
```shell
cd src/train
python nn_train.py
```
## Citation
If you find it useful in your research, please consider cite:
```
@article{tan2023application,
title={Application of BERT in Wind Power Forecasting-Teletraan's Solution in Baidu KDD Cup 2022},
author={Tan, Longxing and Yue, Hongying},
journal={arXiv preprint arXiv:2307.09248},
year={2023}
}
```
## Reference
```
[1] Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2018. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805 (2018).
[2] Haixu Wu, Jiehui Xu, Jianmin Wang, and Mingsheng Long. 2021. Autoformer: De-composition transformers with auto-correlation for long-term series forecasting. Advances in Neural Information Processing Systems 34 (2021), 22419–22430.
[3] JingboZhou,ShuangliLi,LiangHuang,HaoyiXiong,FanWang,TongXu,Hui Xiong, and Dejing Dou. 2020. Distance-aware molecule graph attention network for drug-target binding affinity prediction. arXiv preprint arXiv:2012.09624 (2020).
[4] HaoyiZhou,ShanghangZhang,JieqiPeng,ShuaiZhang,JianxinLi,HuiXiong, and Wancai Zhang. 2021. Informer: Beyond efficient transformer for long sequence time-series forecasting. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 35. 11106–11115.
```