Cascading Failure Learning

The relevant data and codes of our ISCAS2022 work "Predicting Onset Time of Cascading Failure in Power Systems Using A Neural Network-Based Classifier". [Paper]

Fig. 1. A systematic framework of the application of neural network-based classifier for predicting the degree of urgency of the onset time that is identified from failure propagation in a power system.

Requirements

• torch
• networkx
• numpy
• sklearn
• scipy # for loading matlab matrix

Data description

In the /data folder, we provide the following samples for training or testing.

• Nm2: All samples (i.e., power parameter matrices and corresponding labels) on N-2 security criteria.
• Nm2_changed: 5000 samples generated from changing original state on N-2 security criteria.
• Nm3 to Nm5: Corresponding samples generated on N-3/4/5 security criteria, 5000 samples for each subset.

Code description

• main_mlp_topo.py: implements the mlp-based prediction task.
• utils.py: implements some basic functions such as data loading, label transform, etc.

Run the demo

python main_mlp_topo.py

Cite

If you find this work is helpful, please cite our paper. Thank you.

@inproceedings{fang2022predicting,
  title={Predicting onset time of cascading failure in power systems using a neural network-based classifier},
  author={Fang, Junyuan and Liu, Dong and Tse, Chi, K},
  booktitle={2022 IEEE International Symposium on Circuits and Systems (ISCAS)},
  pages={3522--3526},
  year={2022},
  organization={IEEE}
}