Start Date

23-10-2020 1:00 PM

End Date

23-10-2020 1:30 PM

Location

Zoom Session 1 (Main Papers Track)

Abstract

Cyber attacks threaten the security of distribution power grids, such as smart grids. The emerging renewable energy sources such as photovoltaics (PVs) with power electronics controllers introduce new potential vulnerabilities. Based on the electric waveform data measured by waveform sensors in the smart grids, we propose a novel cyber attack detection and identification approach. Firstly, we analyze the cyber attack impacts (including cyber attacks on the solar inverter causing unusual harmonics) on electric waveforms in distribution power grids. Then, we propose a novel deep learning based mechanism including attack detection and attack diagnosis. By leveraging the electric waveform sensor data structure, our approach does not need the training stage for both detection and the root cause diagnosis, which is needed for machine learning/deep learning-based methods. For comparison, we have evaluated classic data-driven methods, including -nearest neighbor (KNN), decision tree (DT), support vector machine (SVM), artificial neural network (ANN), and convolutional neural network (CNN). Comparison results verify the performance of the proposed method for detection and diagnosis of various cyber attacks on PV systems.

Comments

key words:

Attack diagnosis, smart grids, data integrity attack, machine learning, deep learning

Share

COinS
 
Oct 23rd, 1:00 PM Oct 23rd, 1:30 PM

Cybersecurity Strategy against Cyber Attacks towards Smart Grids with PVs

Zoom Session 1 (Main Papers Track)

Cyber attacks threaten the security of distribution power grids, such as smart grids. The emerging renewable energy sources such as photovoltaics (PVs) with power electronics controllers introduce new potential vulnerabilities. Based on the electric waveform data measured by waveform sensors in the smart grids, we propose a novel cyber attack detection and identification approach. Firstly, we analyze the cyber attack impacts (including cyber attacks on the solar inverter causing unusual harmonics) on electric waveforms in distribution power grids. Then, we propose a novel deep learning based mechanism including attack detection and attack diagnosis. By leveraging the electric waveform sensor data structure, our approach does not need the training stage for both detection and the root cause diagnosis, which is needed for machine learning/deep learning-based methods. For comparison, we have evaluated classic data-driven methods, including -nearest neighbor (KNN), decision tree (DT), support vector machine (SVM), artificial neural network (ANN), and convolutional neural network (CNN). Comparison results verify the performance of the proposed method for detection and diagnosis of various cyber attacks on PV systems.

 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.