Due to the integration of wireless technology in power systems, the issue of cybersecurity becomes very critical. Cyberattacks mainly cause changes in power system measurement data like GPS spoofing attack (GSA), which modifies the phase angle of Phasor Measurement Units (PMUs) data. The GSA may lead to disruption in monitoring and protection systems. In this paper, a neural network GPS spoofing detection (NNGSD) with employing PMU data from the dynamic power system is proposed to detect GSAs. Due to the use of an artificial intelligence-based kernel, the proposed structure can be used in various power grid conditions, such as load changes in the grid, in the presence of noise and multi-GSA on PMUs. NNGSD is applied to the standard IEEE 14-buses power system. Numerical results show the real-time performance of the proposed detection method in different conditions.
23 July 2023 in IET Generation, Transmission & Distribution
The growing use of wireless technologies in power systems has raised concerns about cybersecurity, particularly regarding GPS spoofing attacks (GSAs). These attacks manipulate GPS data, leading to modifications in the phase angle of phasor measurement units (PMUs). In this paper, a Deep‐learning GPS‐Spoofing Counteraction (DLGSC) algorithm is proposed, utilizing PMU data for GSA detection and PMU data correction. The algorithm incorporates a recurrent neural network (RNN) and a set of long short‐term memory (LSTM) units separately, for signal correction after attack detection. Unlike existing methods that struggle with simultaneous attacks or they are static methods, DLGSC tackles these challenges by leveraging deep learning techniques. By selecting appropriate features for GSA detection, DLGSC achieves accurate results. The algorithm is evaluated on standard IEEE 14‐bus and IEEE 39‐bus power systems, and its performance is compared to statistical, dynamic, and Deep Learning (DL) methods in the literature. Additionally, an experimental setup is designed to validate the algorithm in a laboratory environment. Results demonstrate the easy‐implementable DLGSC algorithm's satisfactory real‐time performance in various scenarios, such as load variations and noise, achieving over 98% accuracy. Notably, DLGSC is cable of detecting multiple GSAs on different PMUs.
Hand Gesture Recognition (HGR) is one of the most interesting branches in computer vision with lots of applications in mobiles, tablets, personal computers, and interactive platforms. The aim of this technology is to efficiently relate facilitate the people's communication with gadgets. Tracking hand movement in order to understand the desired operation can play an important role in the recognition of the hand gesture. In this paper, we propose a new approach for handwritten recognition in the context of hand gesture recognition. Our approach relies on the hand movements by means of its edge information and the movements directions. The next contribution is also minimizing the uncertainty of the handwritten recognition using fuzzy logic and deterministic finite automata (DFA). The uncertainty of the handwritten is due to the similar movements of the hands during writing the characters that we decreased it using our technique. The experimental result on real-world activities shows the success and usefulness of our approach.
Wheeled Mobile Robots (WMR) are often used to move and work in outdoor environments, often unstructured. The robots' operation in these environments was disrupted due to the ground's unevenness, and slippage occurs. Slipping compared to moving wheel robots at normal levels causes large accumulated errors in the robot's position. Therefore, diagnosing the amount of wheel slip is necessary to improve the control of robot movement. This paper presents an LPV controller with an LPV observer that allows the robot to track the path if there is uncertainty due to the kinematic model's linearization and the occurrence of an unknown longitudinal slip. To evaluate the performance of the proposed method, the controller and observer are designed using the linear matrix inequality (LMI) approach and are simulated in MATLAB software, and superior tracking results are presented.
In nonlinear dynamical systems such as complex and high-order systems, it is difficult to analytically compute the Lyapunov function considering the nonlinear dynamics. Hence, in such cases, numerical methods offer viable alternatives. This paper presents a new approach to construct appropriate Lyapunov functions based on evolutionary theory and biological mathematics, using the multi-objective genetic algorithm (MO-GA), which is a meta-heuristic algorithm for nonlinear dynamical systems. The use of genetic algorithm allows for direct generation of initial data from the Lyapunov theory, leading to the selection and construction of suitable candidates in the shortest time possible. The proposed method has been implemented and validated through several examples.
Power transformers are used throughout the interconnected power grid. These transformers significantly affect the reliability of the power grid. One of the most critical components of a transformer is insulating oil. Insulating oil is responsible for heat transfer and electrical insulation. Scheduled monitoring of transformer oil ageing can improve the status of the transformers and make significant changes to increase the reliability of the power grid. The main reasons for oil ageing are thermal and electrical pressures. These tensions cause chemical degradation of the oil. Therefore, by several types of tests, the relative ageing of the oil can be detected. Current industrial methods are generally offline and costly. In this paper, a laboratory-based method for online transformer oil analysis based on spectroscopy technique has been proposed and implemented. Furthermore, machine learning algorithms implemented on system results to distinguish between oils with different ageing. According to the obtained results, the presented method is effective with high accuracy.