Mohsen Azizi

Dr. Mohsen Azizi

orcid

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 November 2022 in Artificial Intelligence Review
Verified
Journal Article
Closed access
PDF file
FSFatemeh Mohammadi Shakiba
Fatemeh Mohammadi Shakiba
MAMohsen Azizi
Mohsen Azizi
Mengchu ZhouMengchu Zhou
Mengchu Zhou
AAAbdullah Abusorrah
Abdullah Abusorrah

1 August 2021 in Sustainable Cities and Society
Verified
Journal Article
Closed access
SZSamaneh Zamanloo
Samaneh Zamanloo
HAHossein Askarian Abyaneh
Hossein Askarian Abyaneh
HNHamed Nafisi
Hamed Nafisi
MAMohsen Azizi
Mohsen Azizi

17 October 2022 in IEEE Systems, Man, and Cybernetics Magazine
Verified
Journal Article
Closed access
FSFatemeh Mohammadi Shakiba
Fatemeh Mohammadi Shakiba
MAMohsen Azizi
Mohsen Azizi
Mengchu ZhouMengchu Zhou
Mengchu Zhou
Deep learning methods have shown great promise in high-voltage transmission lines’ (TLs’) intelligent inspections. The expansion of power systems, including TLs, has brought the problem of insulator fault detection into account more than before. In this article, a novel transfer learning framework based on a pretrained VGG-19 deep convolutional neural network (CNN) is proposed to detect “missing faults” (broken insulators) in aerial images. In this procedure, a well-known large imagery dataset called ImageNet is used to train VGG-19, and then the knowledge of this deep CNN is transferred. By using a few layers for a fine-tuning purpose, the newly built deep CNN is capable of distinguishing the corrupted and intact insulators. This method is able to diagnose these faults using the aerial images taken from TLs in different environments. The original dataset used in this article is the Chinese Power Line Insulator Dataset (CPLID), which is an imbalanced dataset and includes only 3,808 insulator images. Therefore, a random image-augmentation procedure is proposed and applied to generate a more suitable dataset with 16,720 images. This new dataset allows us to offer higher detection accuracy than the original one because it is a balanced dataset. Training a deep CNN by using it gives more power to the system for detecting the corrupted insulators in different situations such as rotated, dark, and blurry images with complex backgrounds. The comparison results of this study show the advantages of the proposed method over various existing ones.

24 September 2018 in IEEE Transactions on Control Systems Technology
Verified
Journal Article
Closed access
NANima Amouzegar Ashtiani
Nima Amouzegar Ashtiani
MAMohsen Azizi
Mohsen Azizi
SKS. Ali Khajehoddin
S. Ali Khajehoddin
This paper proposes a robust current controller (CC) and dc-link voltage controller based on μ-synthesis and H∞ method, respectively, for a single-phase photovoltaic (PV) converter with an LCL-filter under weak grid operation. The μ-synthesis CC guarantees system robustness against weak grid uncertainties such as grid impedance variations and voltage harmonics. Moreover, the controller compensates the system delays arising from the calculation, pulsewidth modulation, and zero order hold, as well as delay uncertainties. The CC only uses the grid current feedback which eliminates the requirement for additional loops or state measurements utilized in partial or full state-feedback systems. In addition, the H∞ dc-link voltage controller minimizes the bus voltage fluctuations caused by variations of PV system power generation. Also, the controller ensures system robustness against such power variations, which conventional controllers fail to perform. Moreover, power decoupling is achieved by the control system without any auxiliary circuit. In addition, the stability and robust performance of the phase-locked loop in a weak grid with major impedance changes is verified. Simulation and experimental results prove the superior performance of the proposed controllers in both cases of a nominal system and a system with deviated parameters.

29 December 2023 in IEEE Transactions on Circuits and Systems I: Regular Papers
Verified
Journal Article
Closed access
MSMilad Shojaee
Milad Shojaee
MAMohsen Azizi
Mohsen Azizi
In recent years, hybrid AC/DC microgrids have become popular due to their combined benefits from both AC and DC microgrids. One of the challenges in such a system stems from high AC load demands, which can jeopardize the stability of the system. The existing droop controllers, which are based on the per-unit values of DC voltage and AC frequency, have poor performances in sharing the active power between the DC and AC microgrids in the case of high AC load demands. These controllers are required to provide very high gains to mitigate the power sharing problem in overloaded cases, which make the system more sensitive to overload disturbances. Therefore, this paper proposes a novel control scheme based on model predictive controllers to compensate for the poor performance of the existing power sharing droop controllers. The proposed controller minimizes the active power difference between the DC and AC microgrids by transferring more power from the DC to the AC microgrid. The simulation results illustrate the outperformance of the proposed model predictive control method as compared to the benchmark techniques including Fuzzy-proportional-integral-derivative, adaptive-fractional-two-degree-of-freedom proportional-integral-derivative, and virtual impedance in terms of steady-state overload power sharing, transient overload damping, nonlinear load harmonic attenuation, and integral time absolute error analyses.

14 April 2022 in IEEE Transactions on Power Systems
Verified
Journal Article
Closed access
MSMilad Shojaee
Milad Shojaee
MAMohsen Azizi
Mohsen Azizi
This paper presents the design of a robust decentralized controller for the frequency and active power control of a single stand-alone (off-grid) mechanically-coupled generator system including wind turbine and diesel engine generator subsystems. Two robust decentralized controllers are designed and optimized for the two subsystems by using the $\mu$ -synthesis technique. The decentralized controllers address the mechanical coupling between the two subsystems without communicating with each other. In addition, both controllers are designed in such a way that they are robust against the concurrent variations of all the system parameters, which in turn improves the frequency and power control performances significantly. Simulation results indicate that the proposed decentralized controllers have a better performance as compared to the benchmark decentralized model predictive controllers and decentralized linear-quadratic Gaussian controllers. It is also demonstrated that the proposed control scheme is able to damp out mild and large disturbances originating from the load and wind power changes, and has an acceptable robust performance against the concurrent variations in all parameters of the system.