.png){kind=icon}
.png){kind=icon}
Return
Robust Load Forecasting Towards Adversarial Attacks via Bayesian Learning
.png){kind=icon}
PRE
AI
DOI：10.1109/TPWRS.2022.3175252.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Electric load forecasting is an essential problem for the power industry, which has a significant impact on power system operation. Currently, deep learning is proved to be an effective tool for load forecasting. However, those learning-based models are vulnerable towards adversarial attacks, which raises concerns about the robustness of load forecasting models. In this study, we propose a Bayesian training method to enhance the robustness of deep learning-based load forecasting models towards adversarial attacks. We theoretically prove that the proposed method can improve the load forecasting robustness against various attacking objectives without compromising the prediction performance. An approximation-based training scheme is applied to reduce the computing burden so as to make the method better applied in practice. The experimental results show that such an approximation still yields higher robustness compared to four recently proposed benchmark robust forecasting methods while maintaining the prediction performance under no attack.
Keywords:
Load forecasting
Load modeling
Predictive models
Robustness
Forecasting
Bayes methods
Power systems
Adversarial attacks
bayesian method
deep learning
load forecasting
robustness
Journal
IF:
7.2
Papers:
1.1W
Citations:
5.0W