Ali Khaleghi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Ali Khaleghi has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Hadis Karimipour
Department of Electrical and Software Engineering, Smart Cyber-Physical (SCPS) Laboratory, University of Calgary, Calgary, AB, Canada
2 shared publications - Mohammad Reza Aghamohammadi
Shahid Beheshti University
2 shared publications - Mohammad Sadegh Ghazizadeh
Faculty of Electrical Engineering, Shahid Beheshti University, Tehran, Iran
2 shared publications - Soroush Oshnoei
Department of Electrical and Computer Engineering, Aarhus University, 8000 Aarhus, Denmark
2 shared publications Alireza Mehdipour Rabori
1 shared publication- Josep Maria Guerrero
Zhejiang University
1 shared publication
Latest Publications
- Open Access
- 11 October 2026
- Published by MDPI AG in Fractal and Fractional
- Vol. 10 (10) , 715
Physics-Informed Neural Network-Based Detection of False Data Injection Attacks on FACTS-Assisted Wide-Area Fractional-Order Damping Controllers
False data injection attacks (FDIAs) pose a serious threat to FACTS-assisted wide-area damping controllers because they corrupt remote measurements, degrade inter-area oscillation damping, and may evade conventional detection. Data-driven detectors that ignore closed-loop physics are particularly vulnerable under multiplicative gain-ramp attacks; the problem is further exacerbated by the memory-dependent nature of fractional-order controllers, whose long-term history effects can mask attack-induced inconsistencies in the residuals unless system dynamics, actuator behavior, attack-channel constraints, and fractional-order memory are jointly embedded in the learning objective. This paper addresses that gap through a physics-informed neural network (PINN) framework that jointly estimates latent system states and reconstructs the multiplicative attack gain by incorporating plant dynamics, SVC actuator dynamics, measurement consistency, attack-channel reconstruction, and the reconstructed fractional-order command inside the SVC actuator residual. Detection is declared when a smoothed composite residual exceeds a calibrated threshold. In addition, a supervisory switching mechanism is introduced to replace corrupted wide-area control signals with reconstructed secure equivalents after confirmed detection, thereby restoring closed-loop damping performance without interrupting controller operation. Simulation results on a FACTS-based damping control system under gain-based FDIA scenarios show that the proposed detector achieves up to 99.42% accuracy, 98.86% precision, 99.18% recall, and a 99.02% F1-score, with a false alarm rate as low as 0.53%. The recovery mechanism also substantially reduces post-attack speed-deviation overshoot relative to the unmitigated case.
- 11 October 2026
- Published by MDPI AG in Fractal and Fractional
- Vol. 10 (10) , 715
Open AccessPhysics-Informed Neural Network-Based Detection of False Data Injection Attacks on FACTS-Assisted Wide-Area Fractional-Order Damping Controllers
False data injection attacks (FDIAs) pose a serious threat to FACTS-assisted wide-area damping controllers because they corrupt remote measurements, degrade inter-area oscillation damping, and may evade conventional detection. Data-driven detectors that ignore closed-loop physics are particularly vulnerable under multiplicative gain-ramp attacks; the problem is further exacerbated by the memory-dependent nature of fractional-order controllers, whose long-term history effects can mask attack-induced inconsistencies in the residuals unless system dynamics, actuator behavior, attack-channel constraints, and fractional-order memory are jointly embedded in the learning objective. This paper addresses that gap through a physics-informed neural network (PINN) framework that jointly estimates latent system states and reconstructs the multiplicative attack gain by incorporating plant dynamics, SVC actuator dynamics, measurement consistency, attack-channel reconstruction, and the reconstructed fractional-order command inside the SVC actuator residual. Detection is declared when a smoothed composite residual exceeds a calibrated threshold. In addition, a supervisory switching mechanism is introduced to replace corrupted wide-area control signals with reconstructed secure equivalents after confirmed detection, thereby restoring closed-loop damping performance without interrupting controller operation. Simulation results on a FACTS-based damping control system under gain-based FDIA scenarios show that the proposed detector achieves up to 99.42% accuracy, 98.86% precision, 99.18% recall, and a 99.02% F1-score, with a false alarm rate as low as 0.53%. The recovery mechanism also substantially reduces post-attack speed-deviation overshoot relative to the unmitigated case.
- Open Access
- 29 August 2025
- Published by MDPI AG in Fractal and Fractional
- Vol. 9 (9) , 569
Federated Learning Detection of Cyberattacks on Virtual Synchronous Machines Under Grid-Forming Control Using Physics-Informed LSTM
The global shift toward clean production, like using renewable energy, has significantly decreased the use of synchronous machines (SMs), which help maintain stability and control, causing serious frequency stability issues in power systems with low inertia. Fractional order controller-based virtual synchronous machines (FOC-VSMs) have become a promising option, but they rely on communication networks to work together in real time, causing them to be at risk of cyberattacks, especially from false data injection attacks (FDIAs). This paper suggests a new way to detect FDI attacks using a federated physics-informed long short-term memory (PI-LSTM) network. Each FOC-VSM uses its data to train a PI-LSTM, which keeps the information private but still helps it learn from a common model that understands various operating conditions. The PI-LSTM incorporates physical constraints derived from the FOC-VSM swing equation, facilitating residual-based anomaly detection that is sensitive to minor deviations in control dynamics, such as altered inertia or falsified frequency signals. Unlike traditional LSTMs, the physics-informed architecture minimizes false positives arising from benign disturbances. We assessed the proposed method on an IEEE 9-bus test system featuring two FOC-VSMs. The results show that our method can successfully detect FDI attacks while handling regular changes, proving it could be a strong solution.
- 29 August 2025
- Published by MDPI AG in Fractal and Fractional
- Vol. 9 (9) , 569
Open AccessFederated Learning Detection of Cyberattacks on Virtual Synchronous Machines Under Grid-Forming Control Using Physics-Informed LSTM
The global shift toward clean production, like using renewable energy, has significantly decreased the use of synchronous machines (SMs), which help maintain stability and control, causing serious frequency stability issues in power systems with low inertia. Fractional order controller-based virtual synchronous machines (FOC-VSMs) have become a promising option, but they rely on communication networks to work together in real time, causing them to be at risk of cyberattacks, especially from false data injection attacks (FDIAs). This paper suggests a new way to detect FDI attacks using a federated physics-informed long short-term memory (PI-LSTM) network. Each FOC-VSM uses its data to train a PI-LSTM, which keeps the information private but still helps it learn from a common model that understands various operating conditions. The PI-LSTM incorporates physical constraints derived from the FOC-VSM swing equation, facilitating residual-based anomaly detection that is sensitive to minor deviations in control dynamics, such as altered inertia or falsified frequency signals. Unlike traditional LSTMs, the physics-informed architecture minimizes false positives arising from benign disturbances. We assessed the proposed method on an IEEE 9-bus test system featuring two FOC-VSMs. The results show that our method can successfully detect FDI attacks while handling regular changes, proving it could be a strong solution.
- 31 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Smart Grid
- Vol. 16 (2) , 1851-1861
A Probabilistic-Based Approach for Detecting Simultaneous Load Redistribution Attacks Through Entropy Analysis and Deep Learning
Load redistribution attacks (LRAs) are one of the most sneaky and realistic types of false data injection attacks (FDIAs), in which the attacker manipulates the measurements in a way that depicts a false image of the system situation for the operator. Due to the uncertainty in the system’s parameters, system modeling (AC or DC), and so on, detection LRAs have a lot of challenges. To overcome the difficulty of devising a general mechanism for LRA detection based on deterministic methods, we propose a probabilistic approach based on entropy analysis and deep learning. The ratio of cyber loads to real loads (RCLRLs) is the major input of the proposed detection algorithm to make the presented method applicable for different load levels in the system. By extracting the entropy of RCLRLs under LRAs, our method reduces dependency on system modeling and the system’s parameters. We employ the bias correction method on forecasted loads to approximate the real load in the system, enhancing our approach’s accuracy. The framework is a decentralized algorithm that detects simultaneous LRAs in different areas and ensures scalability for large systems. Simulations on the IEEE 118-bus systems demonstrate the proposed method’s high accuracy and rapid response.
- 31 December 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Smart Grid
- Vol. 16 (2) , 1851-1861
A Probabilistic-Based Approach for Detecting Simultaneous Load Redistribution Attacks Through Entropy Analysis and Deep Learning
Load redistribution attacks (LRAs) are one of the most sneaky and realistic types of false data injection attacks (FDIAs), in which the attacker manipulates the measurements in a way that depicts a false image of the system situation for the operator. Due to the uncertainty in the system’s parameters, system modeling (AC or DC), and so on, detection LRAs have a lot of challenges. To overcome the difficulty of devising a general mechanism for LRA detection based on deterministic methods, we propose a probabilistic approach based on entropy analysis and deep learning. The ratio of cyber loads to real loads (RCLRLs) is the major input of the proposed detection algorithm to make the presented method applicable for different load levels in the system. By extracting the entropy of RCLRLs under LRAs, our method reduces dependency on system modeling and the system’s parameters. We employ the bias correction method on forecasted loads to approximate the real load in the system, enhancing our approach’s accuracy. The framework is a decentralized algorithm that detects simultaneous LRAs in different areas and ensures scalability for large systems. Simulations on the IEEE 118-bus systems demonstrate the proposed method’s high accuracy and rapid response.
- 16 November 2024
- Published by Springer Nature
- p. 209-231
Investigation of Detection Mechanisms Against False Data Injection Attacks Based on Machine Learning Approaches
- 16 November 2024
- Published by Springer Nature
- p. 209-231
Investigation of Detection Mechanisms Against False Data Injection Attacks Based on Machine Learning Approaches
- 25 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Systems
- Vol. 39 (4) , 5806-5817
A Probabilistic Data Recovery Framework Against Load Redistribution Attacks Based on Bayesian Network and Bias Correction Method
In the realm of power systems, load redistribution attacks (LRAs) involve the manipulation of measurements by attackers, leading operators to make decisions based on falsified estimated loads. Detecting LRAs is a well-explored area, but the critical post-detection phase, which requires operators to understand the attack's nature, remains under-addressed. Our paper introduces a novel probabilistic mechanism based on the Bayesian network model. It excels at not only detecting affected areas but also pinpointing the buses with manipulated loads under LRAs. Our work prioritizes simplicity by offering a straightforward calculation method that doesn't rely on complex tools or extensive system parameters. It also exhibits robustness against measurement noise. Acknowledging the paramount importance of data recovery, we highlight that post-detection measures are essential for minimizing errors in optimal power flow calculations. However, this data recovery is challenging due to the complexity of understanding the attacker's behavior. Our methodology employs kernel density estimation for precise probability density calculations and integrates the Bayesian Markov model to reduce computational complexity. Rigorous testing on the IEEE 118 bus system demonstrates its strong performance in producing consistently satisfactory results.
- 25 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Power Systems
- Vol. 39 (4) , 5806-5817
A Probabilistic Data Recovery Framework Against Load Redistribution Attacks Based on Bayesian Network and Bias Correction Method
In the realm of power systems, load redistribution attacks (LRAs) involve the manipulation of measurements by attackers, leading operators to make decisions based on falsified estimated loads. Detecting LRAs is a well-explored area, but the critical post-detection phase, which requires operators to understand the attack's nature, remains under-addressed. Our paper introduces a novel probabilistic mechanism based on the Bayesian network model. It excels at not only detecting affected areas but also pinpointing the buses with manipulated loads under LRAs. Our work prioritizes simplicity by offering a straightforward calculation method that doesn't rely on complex tools or extensive system parameters. It also exhibits robustness against measurement noise. Acknowledging the paramount importance of data recovery, we highlight that post-detection measures are essential for minimizing errors in optimal power flow calculations. However, this data recovery is challenging due to the complexity of understanding the attacker's behavior. Our methodology employs kernel density estimation for precise probability density calculations and integrates the Bayesian Markov model to reduce computational complexity. Rigorous testing on the IEEE 118 bus system demonstrates its strong performance in producing consistently satisfactory results.
- 14 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Smart Grid
- Vol. 14 (6) , 4772-4783
A Deep Learning-Based Attack Detection Mechanism Against Potential Cascading Failure Induced by Load Redistribution Attacks
The occurrence of load redistribution (LR) attacks has disastrous consequences for the power system, but these attacks have a significant impact when they cause cascading failures in the system. The mechanisms and strategies for detecting and designing LR attacks resulting in overflow on lines have been the focus of different studies. But fewer studies have been done to detect LR attacks that cause cascading failures. This paper presents a mechanism for identifying LR attacks that, besides causing overflow on lines, have the potential to generate cascading failure. A novel LR attack scheme with cascading failure potential is first proposed for this purpose. The detection mechanism has a basic exploitable structure based on analyzing the estimated cyber load data through the energy management system and a deep learning network. The performance evaluation of the detection mechanism is implemented with regard to the IEEE standard 118-bus system. Various attack scenarios under different conditions (topologies, target lines, and attack load level variations $(\alpha)$ ) are used to verify the effectiveness of the proposed framework. The results clearly show an acceptable level of accuracy for the proposed mechanism, which can distinguish between LR attacks via the overload purpose while also showing the system’s secure state.
- 14 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Smart Grid
- Vol. 14 (6) , 4772-4783
A Deep Learning-Based Attack Detection Mechanism Against Potential Cascading Failure Induced by Load Redistribution Attacks
The occurrence of load redistribution (LR) attacks has disastrous consequences for the power system, but these attacks have a significant impact when they cause cascading failures in the system. The mechanisms and strategies for detecting and designing LR attacks resulting in overflow on lines have been the focus of different studies. But fewer studies have been done to detect LR attacks that cause cascading failures. This paper presents a mechanism for identifying LR attacks that, besides causing overflow on lines, have the potential to generate cascading failure. A novel LR attack scheme with cascading failure potential is first proposed for this purpose. The detection mechanism has a basic exploitable structure based on analyzing the estimated cyber load data through the energy management system and a deep learning network. The performance evaluation of the detection mechanism is implemented with regard to the IEEE standard 118-bus system. Various attack scenarios under different conditions (topologies, target lines, and attack load level variations $(\alpha)$ ) are used to verify the effectiveness of the proposed framework. The results clearly show an acceptable level of accuracy for the proposed mechanism, which can distinguish between LR attacks via the overload purpose while also showing the system’s secure state.
- 22 May 2019
- Published by Springer Nature in Journal of Electrical Engineering & Technology
- Vol. 14 (4) , 1577-1584
Single-Phase Fault Location in Four-Circuit Transmission Lines Based on Wavelet Analysis Using ANFIS
- 22 May 2019
- Published by Springer Nature in Journal of Electrical Engineering & Technology
- Vol. 14 (4) , 1577-1584
Single-Phase Fault Location in Four-Circuit Transmission Lines Based on Wavelet Analysis Using ANFIS
- Open Access
- 1 October 2018
- Published by Institute of Advanced Engineering and Science in International Journal of Electrical and Computer Engineering (IJECE)
- Vol. 8 (5) , 2709-2720
Permanent Fault Location in Distribution System Using Phasor Measurement Units (PMU) in Phase Domain
This paper proposes a new method for locating high impedance fault in distribution systems using phasor measurement units (PMUs) installed at certain locations of the system. To implement this algorithm, at first a new method is suggested for the placement of PMUs. Taking information from the units, voltage and current of the entire distribution system are calculated. Then, the two buses in which the fault has been occurred is determined, and location and type of the fault are identified. The main characteristics of the proposed method are: the use of distributed parameter line model in phase domain, considering the presence of literals, and high precision in calculating the high impedance fault location. The results obtained from simulations in EMTP-RV and MATLAB software indicate high accuracy and independence of the proposed method from the fault type, fault location and fault resistance compared to previous methods, so that the maximum observed error was less than 0.15%
- 1 October 2018
- Published by Institute of Advanced Engineering and Science in International Journal of Electrical and Computer Engineering (IJECE)
- Vol. 8 (5) , 2709-2720
Open AccessPermanent Fault Location in Distribution System Using Phasor Measurement Units (PMU) in Phase Domain
This paper proposes a new method for locating high impedance fault in distribution systems using phasor measurement units (PMUs) installed at certain locations of the system. To implement this algorithm, at first a new method is suggested for the placement of PMUs. Taking information from the units, voltage and current of the entire distribution system are calculated. Then, the two buses in which the fault has been occurred is determined, and location and type of the fault are identified. The main characteristics of the proposed method are: the use of distributed parameter line model in phase domain, considering the presence of literals, and high precision in calculating the high impedance fault location. The results obtained from simulations in EMTP-RV and MATLAB software indicate high accuracy and independence of the proposed method from the fault type, fault location and fault resistance compared to previous methods, so that the maximum observed error was less than 0.15%
- Open Access
- 21 June 2018
- Published by VSB - Technical University of Ostrava in Advances in Electrical and Electronic Engineering
- Vol. 16 (2) , 155 - 166-166
Transient Fault Area Location and Fault Classification for Distribution Systems Based on Wavelet Transform and Adaptive Neuro-Fuzzy Inference System (ANFIS)
Transient Fault Area Location and Fault Classification for Distribution Systems Based on Wavelet Transform and Adaptive Neuro-Fuzzy Inference System (ANFIS)
- 21 June 2018
- Published by VSB - Technical University of Ostrava in Advances in Electrical and Electronic Engineering
- Vol. 16 (2) , 155 - 166-166
Open AccessTransient Fault Area Location and Fault Classification for Distribution Systems Based on Wavelet Transform and Adaptive Neuro-Fuzzy Inference System (ANFIS)
Transient Fault Area Location and Fault Classification for Distribution Systems Based on Wavelet Transform and Adaptive Neuro-Fuzzy Inference System (ANFIS)