Showing 1–5 of 5 results for author: El-Saadany, E

Enhancing Power Quality Event Classification with AI Transformer Models

Authors: Ahmad Mohammad Saber, Amr Youssef, Davor Svetinovic, Hatem Zeineldin, Deepa Kundur, Ehab El-Saadany

Abstract: Recently, there has been a growing interest in utilizing machine learning for accurate classification of power quality events (PQEs). However, most of these studies are performed assuming an ideal situation, while in reality, we can have measurement noise, DC offset, and variations in the voltage signal's amplitude and frequency. Building on the prior PQE classification works using deep learning,… ▽ More Recently, there has been a growing interest in utilizing machine learning for accurate classification of power quality events (PQEs). However, most of these studies are performed assuming an ideal situation, while in reality, we can have measurement noise, DC offset, and variations in the voltage signal's amplitude and frequency. Building on the prior PQE classification works using deep learning, this paper proposes a deep-learning framework that leverages attention-enabled Transformers as a tool to accurately classify PQEs under the aforementioned considerations. The proposed framework can operate directly on the voltage signals with no need for a separate feature extraction or calculation phase. Our results show that the proposed framework outperforms recently proposed learning-based techniques. It can accurately classify PQEs under the aforementioned conditions with an accuracy varying between 99.81%$-$91.43% depending on the signal-to-noise ratio, DC offsets, and variations in the signal amplitude and frequency. △ Less

Submitted 22 February, 2024; originally announced February 2024.

Comments: Accepted in the IEEE Power and Energy Society General Meeting, 2024

A Cyber-Physical Architecture for Microgrids based on Deep learning and LORA Technology

Authors: Mojtaba Mohammadi, Abdollah KavousiFard, Mortza Dabbaghjamanesh, Mostafa Shaaban, Hatem. H. Zeineldin, Ehab Fahmy El-Saadany

Abstract: This paper proposes a cyber-physical architecture for the secured social operation of isolated hybrid microgrids (HMGs). On the physical side of the proposed architecture, an optimal scheduling scheme considering various renewable energy sources (RESs) and fossil fuel-based distributed generation units (DGs) is proposed. Regarding the cyber layer of MGs, a wireless architecture based on low range… ▽ More This paper proposes a cyber-physical architecture for the secured social operation of isolated hybrid microgrids (HMGs). On the physical side of the proposed architecture, an optimal scheduling scheme considering various renewable energy sources (RESs) and fossil fuel-based distributed generation units (DGs) is proposed. Regarding the cyber layer of MGs, a wireless architecture based on low range wide area (LORA) technology is introduced for advanced metering infrastructure (AMI) in smart electricity grids. In the proposed architecture, the LORA data frame is described in detail and designed for the application of smart meters considering DGs and ac-dc converters. Additionally, since the cyber layer of smart grids is highly vulnerable to cyber-attacks, t1his paper proposes a deep-learning-based cyber-attack detection model (CADM) based on bidirectional long short-term memory (BLSTM) and sequential hypothesis testing (SHT) to detect false data injection attacks (FDIA) on the smart meters within AMI. The performance of the proposed energy management architecture is evaluated using the IEEE 33-bus test system. In order to investigate the effect of FDIA on the isolated HMGs and highlight the interactions between the cyber layer and physical layer, an FDIA is launched against the test system. The results showed that a successful attack can highly damage the system and cause widespread load shedding. Also, the performance of the proposed CADM is examined using a real-world dataset. Results prove the effectiveness of the proposed CADM in detecting the attacks using only two samples. △ Less

Submitted 15 December, 2023; v1 submitted 14 December, 2023; originally announced December 2023.

Learning-Based Detection of Malicious Volt-VAr Control Parameters in Smart Inverters

Authors: Ahmad Mohammad Saber, Amr Youssef, Davor Svetinovic, Hatem Zeineldin, Ehab El-Saadany

Abstract: Distributed Volt-Var Control (VVC) is a widely used control mode of smart inverters. However, necessary VVC curve parameters are remotely communicated to the smart inverter, which opens doors for cyberattacks. If the VVC curves of an inverter are maliciously manipulated, the attacked inverter's reactive power injection will oscillate, causing undesirable voltage oscillations to manifest in the dis… ▽ More Distributed Volt-Var Control (VVC) is a widely used control mode of smart inverters. However, necessary VVC curve parameters are remotely communicated to the smart inverter, which opens doors for cyberattacks. If the VVC curves of an inverter are maliciously manipulated, the attacked inverter's reactive power injection will oscillate, causing undesirable voltage oscillations to manifest in the distribution system, which, in turn, threatens the system's stability. In contrast with previous works that proposed methods to mitigate the oscillations after they are already present in the system, this paper presents an intrusion detection method to detect malicious VVC curves once they are communicated to the inverter. The proposed method utilizes a Multi-Layer Perceptron (MLP) that is trained on features extracted from only the local measurements of the inverter. After a smart inverter is equipped with the proposed method, any communicated VVC curve will be verified by the MLP once received. If the curve is found to be malicious, it will be rejected, thus preventing unwanted oscillations beforehand. Otherwise, legitimate curves will be permitted. The performance of the proposed scheme is verified using the 9-bus Canadian urban benchmark distribution system simulated in PSCAD/EMTDC environment. Our results show that the proposed solution can accurately detect malicious VVC curves. △ Less

Submitted 19 September, 2023; originally announced September 2023.

Comments: Accepted for publication in IECON 2023; the 49th Annual Conference of the IEEE Industrial Electronics Society (IES)

LSTM-based approach to detect cyber attacks on market-based congestion management methods

Authors: Omniyah Gul M Khan, Amr Youssef, Ehab El-Saadany, Magdy Salama

Abstract: Market-based congestion management methods adopt Demand Side Management (DSM) techniques to alleviate congestion in the day-ahead market. Reliance of these methods on the communication layer makes it prone to cyber attacks affecting the security, reliability, and economic operation of the distribution network. In this paper, we focus on Load Altering Attacks that would compromise the operation of… ▽ More Market-based congestion management methods adopt Demand Side Management (DSM) techniques to alleviate congestion in the day-ahead market. Reliance of these methods on the communication layer makes it prone to cyber attacks affecting the security, reliability, and economic operation of the distribution network. In this paper, we focus on Load Altering Attacks that would compromise the operation of market-based congestion management methods. A detection technique is proposed using Long Short-term Memory (LSTM) Recurrent Neural Networks (RNN). IEEE 33 bus system is used as a case study to demonstrate the effectiveness of the proposed technique. An accuracy of 97% was obtained proving the capability of using LSTM-RNN to detect a load altering cyber attack compromising aggregators in the network. △ Less

Submitted 27 September, 2021; originally announced September 2021.

Impact of Electric Vehicles Botnets on the Power Grid

Authors: Omniyah Gul M Khan, Ehab El-Saadany, Amr Youssef, Mostafa Shaaban

Abstract: The increased penetration of Electric Vehicles (EVs) in the transportation sector has increased the requirement of Fast Charging Direct Current (FCDC) stations to meet customer's speedy charging requirements. However, both charging stations and EVs connection to the communication infrastructure as well as the power grid makes it vulnerable to cyber attacks. In this paper, the vulnerability of the… ▽ More The increased penetration of Electric Vehicles (EVs) in the transportation sector has increased the requirement of Fast Charging Direct Current (FCDC) stations to meet customer's speedy charging requirements. However, both charging stations and EVs connection to the communication infrastructure as well as the power grid makes it vulnerable to cyber attacks. In this paper, the vulnerability of the EV charging process is initially studied. We then show how a botnet of compromised EVs and FCDC stations can be utilized to launch cyber attacks on the power grid resulting in an increase in the load at a specific time. The effect of such attacks on the distribution network in terms of line congestion and voltage limit violations is investigated. Moreover, the effect of the botnet of the transmission network is also studied. △ Less

Submitted 16 March, 2021; originally announced March 2021.

Comments: 5 pages, conference