Journal of Modeling and Simulation in Electrical and Electronics Engineering
Volume:2 Issue: 3, Summer 2022

Abstract Meaning Representation (AMR) is a representation model in which AMRs are rooted and labeled graphs that capture semantics on the sentence level while abstracting away from Morpho-Syntactic properties. The nodes of the graph represent meaning concepts and the edge labels show relationships between them. The application of AMR, as a principal form of structured sentence semantics, in Natural Language Processing (NLP) tasks is widely increasing, and it is considered a turning point for NLP research. The present study gives a brief review of the existing AMR applications in various NLP tasks. Moreover, they are compared and some of their basic features are discussed.

One of the most common causes of vibration in rotating machines is the misalignment fault. The Motor Current Signature Analysis (MCSA) is an excellent method for the detection of the misalignment fault on those electric machines whose current signals are practically available. This paper aims to extend the application of the MCSA method to non-electric rotating systems for the detection of the misalignment fault between the driver machine and the driven machine. For this, a small brushless direct current (BLDC) motor was connected to the driver machine. Then, by using the Fast Fourier Transform and Wavelet Packet Transform the current signal of the BLDC motor was analyzed to detect the misalignment fault. In addition, a fault detection indicator was provided using the energy of the current signal. For the evaluation of the proposed method, an experimental setup was provided. The driver machine of the setup was an induction machine. So, it was possible to investigate the misalignment fault through both the BLDC motor and the induction motor. The results showed that the misalignment fault can be detected by the current signal of the BLDC motor as well as the current signal of the driver machine.

This paper proposes distance matrices, Euclidean, and offset translation methods in machine learning prediction of wind speed. The primary aim for this research is to design forecasting models for very short-term and short-term wind speed prediction based on these two methods by using historical data on wind speed. The test data is collected at a wind power station at 10 minutes intervals. Furthermore, we evaluate the output in different time horizons in comparison to the benchmark method (persistence). To ensure the output results, comparing this method with the persistence method is essential. The proposed method performance was evaluated and compared with the conventional persistence method performance in terms of mean absolute error.

Due to the increasing amount of video data, a lot of research has been done in the field of retrieving and categorizing this type of data. On the other hand, with the growing popularity of football and the increasing number of its audiences, the importance of automatic and real-time extraction of statistics and information about soccer matches has increased. One of the critical and challenging tasks in soccer video analysis is the detection of players’ blobs and regions, along with identifying the teams related to the players. This task encounters many challenges, including grass loss in the playfield, the presence of playfield lines and players' shadows, the overlapping of players with objects and other players, and different shapes of players in different situations. This paper proposes a framework for detecting players and their related teams. For this purpose, an object-sieve-based method is used to detect players’ blobs, and a genetic algorithm is used to identify their related teams. Each chromosome of the genetic algorithm is a window that lies on one blob whose fitness function shows how much its color and shape characteristics fit with the uniforms of each of the two teams. The proposed method was evaluated by 50 different frames of broadcast soccer videos, including 563 players, and 40 different sub-images, including 84 players. The results show 98% and 91.6% precision for player detection and labeling, respectively.

Embedding learning is an essential issue in Natural Language Processing (NLP) applications. Most existing methods measure the similarity between text chunks in a context using pre-trained word embedding. However, providing labeled data for model training is costly and time-consuming. So, these methods face downward performance when limited amounts of training data are available. This paper presents an unsupervised sentence embedding method that effectively integrates semantic hashing into the Kernel Principal Component Analysis (KPCA) to construct embeddings of lower dimensions that can be applied to any domain. The experiments conducted on benchmark datasets highlighted that the generated embeddings are general-purpose and can capture semantic meanings from both small and large corpora.

In this paper, a novel structure for silicon on insulator metal semiconductor field effect transistors (SOI MESFETs) is introduced using the heterogeneous Si/SiGe region. SiGe semiconductor is used to expand the effective width of the drift region inside the buried oxide (BOX) layer. Due to its properties such as high electron mobility, high electron drift velocity, and excellent radio frequency (RF) performance, it significantly increases the current density of drain and other DC and RF parameters. Also, to control the critical electric field, which determines the breakdown voltage of the device, as well as to reduce the parasitic capacitance to improve its frequency characteristics, an additional oxide region between the gate and drain and below a part of the gate region is used. Numerical simulation shows that the drain current density and breakdown voltage of the proposed device compared to the conventional structure has been improved by 120% and 37%, respectively, resulting in a 2 times increase in maximum output power density (Pmax). Also, the RF specifications of the new structure, including current gain (h21), unilateral power gain (U), and maximum available power gain (MAG), have been improved by 130%, 85%, and 65%, respectively. These specifications are proper for a device in high power and RF circuits like D-band applications.

Voltage drop and voltage fluctuations are one of the most important power quality problems of power networks. There are so many sensitive loads like electronic controllers, microprocessors and instrumentations like PLCs in power system that are very sensitive to disturbances and their operation could be affected easily by Power Quality Problems. Uninterruptable sources and static switch are used to prevent disturbance in sensitive loads. When a fault occurred in main source side, a static switch can disconnect the main source and connect the alternative source in few milliseconds. In this paper, the impact of static transfer switch performance on power quality in presence of any faults such as one phase, two phase and tree phase faults and motor starting in power system are studied.