Majlesi Journal of Electrical Engineering
Volume:16 Issue: 4, Dec 2022

As the proportion of total generation by renewable sources compared to non-renewable sources increases, the relative inertial stability provided by large rotating generators in electricity grids is found to shrink and is not being replaced by sources such as photovoltaic and wind power, which are already known for their inherent variability. This leads to electricity generation systems being less stable, less flexible, and less adequate in applications with a high diversity factor, and literature shows that the penetration of renewable energy sources in distribution-generation/microgrid system frequently presents several technical and economic challenges in their usual applications. This work examines how increased renewable energy penetration impacts the distribution-generation system and suggests approaches and measures for tackling the challenges that are associated with it.

Electric machines, particularly induction motors, are widely used in industry as they are robust in construction and require less maintenance. These machines are continuously operated under high loading conditions, which may result in machine vibrations. The Experimental-Modal-Analysis (EMA) technique is employed in this research to determine electrical machine vibrations. The hammer test is a popular EMA approach for locating the exact location of winding looseness. EMA test is executed on the stator winding to extract the modal parameters to find exact deformation in machine windings. Mathematical calculations and a numerical model are developed to validate the experimental data. The EMA’s Operational Deformation Structure (ODS) validates the winding’s looseness precisely. Two test machines’ MA and MB windings are tested with EMA to evaluate the stator slot structure’s looseness. Finally, the proposed technique is compared with the finite element technique along with mathematical calculations for verification.

Recently, many studies have examined filters for reducing or removing speckle noise, which is inherent to different images types such as Porous Silicon (PS) images, in order to ameliorate the metrological evaluation of their applications. In the case of digital images, noise can produce difficulties in the diagnosis of images details, such as edges and limits, should be preserved. Most algorithms can reduce or remove speckle noise, but they do not consider the conservation of these details. This paper describes in detail, the different techniques that focus mainly on the smoothing or elimination of speckle noise in images, as the aim of this study is to achieve the improvement of this smoothing and elimination, which is directly related to different processes (such as the detection of interest regions). Furthermore, the description of these techniques facilitates the operations of evaluations and research with a more specific scope. This study initially covers the definition and modeling of speckle noise. Then we elaborated in detail the different types of filters used in this study, finally, five statistical parameters such as Root Mean Square Error (RMSE), Mean Square Error (MSE), Structural Similarity Index (SSIM), Peak Signal to Noise Ratio (PSNR), Signal to Noise Ratio (SNR)  are calculated, compared and the results are tabulated, common in filter evaluation processes. Trough the calculation of the statistical parameters, we can classify the filters in terms of perceptual quality by providing greater certainty.

Super resolution algorithms attempt to reconstruct high resolution images from low resolution images and it can be considered as a preprocessing step for object recognition and image classification. Various algorithms have been introduced for single-image super resolution, but these algorithms often face important challenges such as poorly matching the reconstructed image with the original image. This article introduces a preprocessing operation to improve the performance of the super resolution process. In the proposed method, the low-resolution images are enhanced before entering to the resolution change module. Calculating the brightness of the pixels in the image channels, creating the luminance map and removing atmospheric light, applying the transmittance map by using the luminance coefficients, and recovering the natural image in all three-color channels are the above preprocessing steps. The proposed method succeeded in increasing the PSNR parameter by 4.35%, 10.62%, and 8.31%, as well as 0.23%, 3.10%, and 7.91% of the SSIM parameter for Set5, Set14, and BSD100 benchmark datasets compared to its closest state-of-the-art methods.

A historical review of Radio Frequency Energy Harvesting (RFEH) Rectenna (Rectifier Antenna) Systems without a matching network is performed, with emphasis on the antenna part. As the antenna, matching network and rectifier are the main parts of the rectenna systems, the reasons behind the elimination of the matching network are presented and different special antennas suitable for direct matching to the rectifier, without using a matching network, are reviewed. Since the diode in the rectifier is a nonlinear element, its input impedance is changed with varying operating conditions such as input power, frequency and output load impedance of the rectifier. So, it is a challenge for researchers to match the antenna impedance directly to the rectifier in variable operational conditions.

Breast cancer is one the most ubiquitous types of cancer which affect a considerable number of women around the globe. It is a malignant tumor, whose origin is in the glandular epithelium of the breast and causes serious health-related problems for patients. Although there is no known way of curing this disease, early detection of it can be very fruitful in terms of reducing the negative ramifications. Thus, accurate diagnosis of breast cancer based on automatic approaches is demanded immediately. Computer vision-based techniques in the analysis of medical images, especially histopathological images, have proved to be extremely performant. In this paper, we propose a novel approach for classifying malignant or non-malignant images. Our approach is based on the latent space embeddings learned by convolutional autoencoders. This network takes a histopathological image and learns to reconstruct it and by compressing the input into the latent space, we can obtain a compressed representation of the input. These embeddings are fed to a reinforcement learning-based feature selection module which extracts the best features for distinguishing the normal from the malicious images. We have evaluated our approach on a well-known dataset, named BreakHis, and used the K-Fold Cross Validation technique to obtain more reliable results. The accuracy, achieved by the proposed model, is 96.8% which exhibits great performance.

The frighteningly high levels of power consumption at present are caused mainly by the expanding global population and the accessibility of energy-hungry smart technologies. So far, various simulation tools, engineering- and AI-based methodologies have been utilized to anticipate power consumption effectively. While engineering approaches forecast using dynamic equations, AI-based methods forecast using historical data. The modeling of nonlinear electrical demand patterns is still lacking for durable solutions, however, the available approaches are only effective for resolving transient dependencies. Furthermore, because they are only based on historical data, the current methodologies are static in nature. In this research, we present a system based on deep learning to anticipate power consumption while accounting for long-term historical relationships. In our approach, a transformer-based model is used for the prediction of electricity demand on data collected from the regional facilities in Iraq. According to the conducted experiments, our approach claims competitive performance, achieving an error rate of 2.0 in predicting 1-day-ahead of electricity demand in the test samples.

Today, electricity theft is one of the main challenges for energy distribution and transmission companies around the world. Early detection of abnormal consumers can prevent security and financial losses. Extensive research studies have been done to detect electricity theft by analyzing customer consumption patterns. Today, one of the most widely used methods is convolutional neural networks (CNNs). These networks contain a large number of hyper-parameters.  The accuracy of these networks is low in most studies due to the lack of attention to the adjustment of these hyper-parameters.  Network accuracy and achieving a robust learning model are influenced by the optimal adjusting of these hyper-parameters, which requires exploring a complex and large search space. Meta-heuristic-based search methods are suitable for solving these problems. Therefore, the main contribution of this paper is to use the high ability of the cheetah optimization algorithm (CHOA) to optimally extract CNN hyper-parameters. In this paper, in order to balance the dataset, abnormal samples are created using artificial attacks and added to the dataset. Also, in order to increase the accuracy of the network, abnormal data are clustered using the CHOA algorithm. ISSDA dataset is used to test and evaluate the results. Based on the results obtained and comparing them with the other works, it was proved that the proposed framework with high accuracy identifies abnormal consumers.

To stop vision loss from glaucoma, early identification and regular screening are crucial. Convolutional neural networks (CNN) have been effectively used in recent years to diagnose glaucoma automatically from color fundus pictures. CNNs can extract distinctive characteristics directly from the fundus pictures, as opposed to the current automatic screening techniques. In this study, a CNN-based deep learning architecture is created for the categorization of normal and glaucomatous fundus pictures. In this paper, we propose a deep learning-based framework for the detection of glaucoma based on retinal images. Our proposed approach utilizes the two CNN-based models, namely Inception and DenseNet, in order to classify the input images. We also show the impact of transfer learning on the training and the validation processes and put forward an effective pipeline with lower trainable parameters for the target task. Our experiments on a collected dataset demonstrate the efficacy of the proposed model by achieving an accuracy of 93.84%, a precision of 92.83%, and a recall of 95.00%.

Through the use of malware, particularly JavaScript, cybercriminals have turned online applications into one of their main targets for impersonation. Detection of such dangerous code in real-time, therefore, becomes crucial in order to prevent any harmful action. By categorizing the salient characteristics of the malicious code, this study suggests an effective technique for identifying malicious Java scripts that were previously unknown, employing an interceptor on the client side. By employing the wrapper approach for dimensionality reduction, a feature subset was generated. In this paper, we propose an approach for handling the malware detection task in imbalanced data situations. Our approach utilizes two main imbalanced solutions namely, Synthetic Minority Over Sampling Technique (SMOTE) and Tomek Links with the object of augmenting the data and then applying a Deep Neural Network (DNN) for classifying the scripts. The conducted experiments demonstrate the efficient performance of our approach and it achieves an accuracy of 94.00%.

Although theoretical models of manufacturing processes are useful for understanding physical events, it can be challenging to apply them in real-world industrial settings. When huge data are accessible, artificial intelligence approaches in the context of Industry 4.0 can offer effective answers to real production challenges. Deep learning is increasingly being used in the realm of artificial intelligence to address a variety of issues relating to information and communication technology, but it is still limited or perhaps nonexistent in the industrial sector. In this study, wire electrical discharge machining—a sophisticated machining technique primarily used for computer hardware components—is applied to effectively forecast unforeseen occurrences. By identifying hidden patterns in process signals, anomalies, such as changes in the thickness of a machined item, may be efficiently anticipated before they occur. In this study, a model for anomaly detection in the sequence of thickness change in the machined component based on transformers is suggested. Our method is able to achieve 94.32 % and 94.16 % accuracy in Z 135 and Z 15 datasets, respectively. Also, it forecasts the abnormalities inside the sequence 1.1 seconds in advance, according to our tests on a dataset that has been introduced.

Un-sharp masking method improves the images contrast without requiring any prior knowledge. In this method, a sharper image can be achieved by empowering the high frequency components of the input image. Un-sharp masking has a parameter named gain factor which has a high effect on the enhanced image quality. In this paper, an approach is proposed to adaptively estimate the appropriate value of this parameter in order to effectively enhance an image with local blur, or an image with non-uniform blur. In proposed method, first, the input image is segmented into blur and non-blur regions. Then the gain factor is estimated for each region adaptively. In this approach, the influence of the image blurriness on its gradient information is used to estimate the value for the gain factor. The image quality assessments are applied to evaluate the performance of proposed un-sharp masking method in image enhancement. Experimental results demonstrate that the performance of our proposed method is better than the performance of existing un-sharp masking methods in image enhancement.

This paper presents a broadband low-power CMOS low noise amplifier (LNA) in 130 nm technology for sub-GHz Internet of Things (IoT) applications. The proposed circuit consists of a current reuse common source amplifier (CSA) in the forward path, and a positive simple transconductance amplifier (PSTA) in the feedback path. Theoretical calculation of the input admittance shows a positive part that presents a parallel inductance. This equivalent parallel inductance in the input can cancel out the input capacitance of CSA and electrostatic discharge (ESD) pad, enhancing the frequency bandwidth in the sub-GHz frequency band. Post-layout was simulated including ESD pads and package model in 130 nm CMOS technology, LNA achieves a voltage gain of 16.5 dB in a frequency bandwidth of 50 MHz to 1.1 GHz, noise figure (NF) of less than 2.4 dB, input return loss (S11) of -11 dB, input third order intercept point (IIP3) of -11 dBm and 1 mW power consumption from a 1 V power supply, showing a good figure of merit compared to other works. The occupied core area is less than 0.002 mm2</sup>.

Wildfire detection is a time-critical application since it can be challenging to identify the source of ignition in a short amount of time, which frequently causes the intensity of fire incidents to increase. The development of precise early-warning applications has sparked significant interest in expert systems research due to this issue, and recent advances in deep learning for challenging visual interpretation tasks have created new study avenues. In recent years, the power of deep learning-based models sparked the researcher’s interests from a variety of fields. Specially, Convolutional Neural Networks (CNN) have become the most suited approach for computer vision tasks. As a result, in this paper we propose a CNN-based pipeline for classifying and verifying fire-related images. Our approach consists of two models, first of which classifies the input data and then the second model verifies the decision made by the first one by learning more robust representations obtained from a large masked auto encoder-based model. The verification step boosts the performance of the classifier with respect to false positives and false negatives. Based on extensive experiments, our approach proves to improve previous state-of-the-art algorithms by 3 to 4% in terms of accuracy.

Photovoltaic (PV) panel produces electricity depending on a variety of characteristics, including the PV module model, design specifications, and ambient circumstances such as temperature and sun irradiation. To analyze and model the effect of these factors on PV performance, a PV model is significant to be studied and modeled in advance. It is desirable to be compatible with the real-physical behavior of the PV panel. This paper presents mathematical modeling, design, and simulation of the three-diode model (3DM) MPPT controller instead of using conventional single/double diode PV models. The proposed PV model is analyzed, verified, and simulated at various temperature and irradiance levels. Furthermore, Particle Swarm Optimization (PSO) as a multi-objective algorithm is used for the Maximum Power Point Tracking MPPT controller to enhance the performance of the module and PV array system. A DC/DC boost converter is combined with the proposed 3DM model and connected through a resistive load. Results show that adopting PSO-based MPPT improves the performance of the PV panel compared to the traditional MPPT and verified the theoretical background.