Thermographic Fault Diagnosis of Shaft of BLDC Motor.

A technique of thermographic fault diagnosis of the shaft of a BLDC (Brushless Direct Current Electric) motor is presented in this article. The technique works for the shivering of the thermal imaging camera in the range of 0-1.5 [m/s2]. An electric shaver was used as the source of the BLDC motor. The following states of the BLDC motor were analyzed: Healthy BLDC motor (HB), BLDC motor with one faulty shaft (1FSB), BLDC motor with two faulty shafts (2FSB), and BLDC motor with three faulty shafts (3FSB). A new method of feature extraction named PNID (power of normalized image difference) was presented. Deep neural networks were used for the analysis of thermal images of the faulty shaft of the BLDC motor: GoogLeNet, ResNet50, and EfficientNet-b0. The results of the proposed technique were very good. PNID, GoogLeNet, ResNet50, and EfficientNet-b0 have an efficiency of recognition equal to 100% for four classes.

A Novel Routing Protocol Based on Elliptical Shaped Movement of Autonomous Underwater Vehicles in Data Gathering Process for Underwater Wireless Sensor Network.

High end-to-end delay is a significant challenge in the data collection process in the underwater environment. Autonomous Underwater Vehicles (AUVs) are a considerably reliable source of data collection if they have significant trajectory movement. Therefore, in this paper, a new routing algorithm known as Elliptical Shaped Efficient Data Gathering (ESEDG) is introduced for the AUV movement. ESEDG is divided into two phases: first, an elliptical trajectory has been designed for the horizontal movement of the AUV. In the second phase, the AUV gathers data from Gateway Nodes (GNs) which are associated with Member Nodes (MNs). For their association, an end-to-end delay model is also presented in ESEDG. The hierarchy of data collection is as follows: MNs send data to GNs, the AUV receives data from GNs, and forwards it to the sink node. Furthermore, the ESEDG was evaluated on the network simulator NS-3 version 3.35, and the results were compared to existing data collection routing protocols DSG-DGA, AEEDCO, AEEDCO-A, ALP, SEDG, and AEDG. In terms of network throughput, end-to-end delay, lifetime, path loss, and energy consumption, the results showed that ESEDG outperformed the baseline routing protocols.

A Novel Design and Development of a Strip-Fed Circularly Polarized Rectangular Dielectric Resonator Antenna for 5G NR Sub-6 GHz Band Applications.

In this article, a rectangular dielectric resonator antenna (RDRA) with circularly polarized (CP) response is presented for 5G NR (New Radio) Sub-6 GHz band applications. A uniquely shaped conformal metal feeding strip is proposed to excite the RDRA in higher-order mode for high gain utilization. By using the proposed feeding mechanism, the degenerate mode pair of the first higher-order, i.e., TEδ13x at 4.13 GHz and TE1δ3y, at 4.52 GHz is excited to achieve a circularly polarized response. A circular polarization over a bandwidth of ~10%, in conjunction with a wide impedance matching over a bandwidth of ~17%, were attained by the antenna. The CP antenna proposed offers a useful gain of ~6.2 dBic. The achieved CP bandwidth of the RDRA is good enough to cover the targeted 5G NR bands around 4.4−4.8 GHz, such as n79. The proposed antenna configuration is modelled and optimized using computer simulation technology (CST). A prototype was built to confirm (validate) the performance estimated through simulation. A good agreement was observed between simulated and measured results.

Non-Thermal Plasma Reduction of Ag+ Ions into Silver Nanoparticles in Open Atmosphere under Statistically Optimized Conditions for Biological and Photocatalytic Applications.

An environmentally friendly non-thermal DC plasma reduction route was adopted to reduce Ag+ ions at the plasma−liquid interface into silver nanoparticles (AgNPs) under statistically optimized conditions for biological and photocatalytic applications. The efficiency and reactivity of AgNPs were improved by statistically optimizing the reaction parameters with a Box−Behnken Design (BBD). The size of the AgNPs was chosen as a statistical response parameter, while the concentration of the stabilizer, the concentration of the silver salt, and the plasma reaction time were chosen as independent factors. The optimized parameters for the plasma production of AgNPs were estimated using a response surface methodology and a significant model p < 0.05. The AgNPs, prepared under optimized conditions, were characterized and then tested for their antibacterial, antioxidant, and photocatalytic potentials. The optimal conditions for these three activities were 3 mM of stabilizing agent, 5 mM of AgNO3, and 30 min of reaction time. Having particles size of 19 to 37 nm under optimized conditions, the AgNPs revealed a 82.3% degradation of methyl orange dye under UV light irradiation. The antibacterial response of the optimized AgNPs against S. aureus and E. coli strains revealed inhabitation zones of 15 mm and 12 mm, respectively, which demonstrate an antioxidant activity of 81.2%.

Adaptive Contrastive Learning with Label Consistency for Source Data Free Unsupervised Domain Adaptation.

Unsupervised domain adaptation, which aims to alleviate the domain shift between source domain and target domain, has attracted extensive research interest; however, this is unlikely in practical application scenarios, which may be due to privacy issues and intellectual rights. In this paper, we discuss a more challenging and practical source-free unsupervised domain adaptation, which needs to adapt the source domain model to the target domain without the aid of source domain data. We propose label consistent contrastive learning (LCCL), an adaptive contrastive learning framework for source-free unsupervised domain adaptation, which encourages target domain samples to learn class-level discriminative features. Considering that the data in the source domain are unavailable, we introduce the memory bank to store the samples with the same pseudo label output and the samples obtained by clustering, and the trusted historical samples are involved in contrastive learning. In addition, we demonstrate that LCCL is a general framework that can be applied to unsupervised domain adaptation. Extensive experiments on digit recognition and image classification benchmark datasets demonstrate the effectiveness of the proposed method.

A New Kinect V2-Based Method for Visual Recognition and Grasping of a Yarn-Bobbin-Handling Robot.

This work proposes a Kinect V2-based visual method to solve the human dependence on the yarn bobbin robot in the grabbing operation. In this new method, a Kinect V2 camera is used to produce three-dimensional (3D) yarn-bobbin point cloud data for the robot in a work scenario. After removing the noise point cloud through a proper filtering process, the M-estimator sample consensus (MSAC) algorithm is employed to find the fitting plane of the 3D cloud data; then, the principal component analysis (PCA) is adopted to roughly register the template point cloud and the yarn-bobbin point cloud to define the initial position of the yarn bobbin. Lastly, the iterative closest point (ICP) algorithm is used to achieve precise registration of the 3D cloud data to determine the precise pose of the yarn bobbin. To evaluate the performance of the proposed method, an experimental platform is developed to validate the grabbing operation of the yarn bobbin robot in different scenarios. The analysis results show that the average working time of the robot system is within 10 s, and the grasping success rate is above 80%, which meets the industrial production requirements.

Influence of Elbow Angle on Erosion-Corrosion of 1018 Steel for Gas-Liquid-Solid Three Phase Flow.

Erosive wear due to the fact of sand severely affects hydrocarbon production industries and, consequently, various sectors of the mineral processing industry. In this study, the effect of the elbow geometrical configuration on the erosive wear of carbon steel for silt-water-air flow conditions were investigated using material loss analysis, surface roughness analysis, and microscopic imaging technique. Experiments were performed under the plug flow conditions in a closed flow loop at standard atmospheric pressure. Water and air plug flow and the disperse phase was silt (silica sand) with a 2.5 wt % concentration, and a silt grain size of 70 µm was used for performing the tests. The experimental analysis showed that silt impact increases material disintegration up to 1.8 times with a change in the elbow configuration from 60° to 90° in plug flow conditions. The primary erosive wear mechanisms of the internal elbow surface were sliding, cutting, and pit propagation. The maximum silt particle impaction was located at the outer curvature in the 50° position in 60° elbows and the 80° position in 90° elbows in plug flow. The erosion rate decreased from 10.23 to 5.67 mm/year with a change in the elbow angle from 90° to 60°. Moreover, the microhardness on the Vickers scale increased from 168 to 199 in the 90° elbow and from 168 to 184 in the 60° elbow.

Damage Detection for Conveyor Belt Surface Based on Conditional Cycle Generative Adversarial Network.

The belt conveyor is an essential piece of equipment in coal mining for coal transportation, and its stable operation is key to efficient production. Belt surface of the conveyor is vulnerable to foreign bodies which can be extremely destructive. In the past decades, much research and numerous approaches to inspect belt status have been proposed, and machine learning-based non-destructive testing (NDT) methods are becoming more and more popular. Deep learning (DL), as a branch of machine learning (ML), has been widely applied in data mining, natural language processing, pattern recognition, image processing, etc. Generative adversarial networks (GAN) are one of the deep learning methods based on generative models and have been proved to be of great potential. In this paper, a novel multi-classification conditional CycleGAN (MCC-CycleGAN) method is proposed to generate and discriminate surface images of damages of conveyor belt. A novel architecture of improved CycleGAN is designed to enhance the classification performance using a limited capacity images dataset. Experimental results show that the proposed deep learning network can generate realistic belt surface images with defects and efficiently classify different damaged images of the conveyor belt surface.

An Embedded System Using Convolutional Neural Network Model for Online and Real-Time ECG Signal Classification and Prediction.

This paper presents an automatic ECG signal classification system that applied the Deep Learning (DL) model to classify four types of ECG signals. In the first part of our work, we present the model development. Four different classes of ECG signals from the PhysioNet open-source database were selected and used. This preliminary study used a Deep Learning (DL) technique namely Convolutional Neural Network (CNN) to classify and predict the ECG signals from four different classes: normal, sudden death, arrhythmia, and supraventricular arrhythmia. The classification and prediction process includes pulse extraction, image reshaping, training dataset, and testing process. In general, the training accuracy achieved up to 95% after 100 epochs. However, the prediction of each ECG single type shows a differentiation. Among the four classes, the results show that the predictions for sudden death ECG waveforms are the highest, i.e., 80 out of 80 samples are correct (100% accuracy). In contrast, the lowest is the prediction for normal sinus ECG waveforms, i.e., 74 out of 80 samples are correct (92.5% accuracy). This is due to the image features of normal sinus ECG waveforms being almost similar to the image features of supraventricular arrhythmia ECG waveforms. However, the model has been tuned to achieve an optimal prediction. In the second part, we presented the hardware implementation with the predictive model embedded in an NVIDIA Jetson Nanoprocessor for the online and real-time classification of ECG waveforms.

Navigating an Automated Driving Vehicle via the Early Fusion of Multi-Modality.

The ability of artificial intelligence to drive toward an intended destination is a key component of an autonomous vehicle. Different paradigms are now being employed to address artificial intelligence advancement. On the one hand, modular pipelines break down the driving model into submodels, such as perception, maneuver planning and control. On the other hand, we used the end-to-end driving method to assign raw sensor data directly to vehicle control signals. The latter is less well-studied but is becoming more popular since it is easier to use. This article focuses on end-to-end autonomous driving, using RGB pictures as the primary sensor input data. The autonomous vehicle is equipped with a camera and active sensors, such as LiDAR and Radar, for safe navigation. Active sensors (e.g., LiDAR) provide more accurate depth information than passive sensors. As a result, this paper examines whether combining the RGB from the camera and active depth information from LiDAR has better results in end-to-end artificial driving than using only a single modality. This paper focuses on the early fusion of multi-modality and demonstrates how it outperforms a single modality using the CARLA simulator.

Analyzing Distributed Vibrating Sensing Technologies in Optical Meshes.

Hundreds of kilometers of optical fibers are installed for optical meshes (OMs) to transmit data over long distances. The visualization of these deployed optical fibers is a highlighted issue because the conventional procedure can only measure the optical losses. Thus, this paper presents distributed vibration sensing (DVS) estimation mechanisms to visualize the optical fiber behavior installed for OMs which is not possible by conventional measurements. The proposed technique will detect the power of light inside the optical fiber, as well as different physical parameters such as the phase of transmitted light inside the thread, the frequency of vibration, and optical losses. The applicability of optical frequency domain reflectometry (OFDR) and optical time-domain reflectometry (OTDR) DVS techniques are validated theoretically for various state detection procedures in optical fibers. The simulation model is investigated in terms of elapsed time, the spectrum of a light signal, frequency, and the impact of many external physical accidents with optical fibers.

Thermographic Fault Diagnosis of Ventilation in BLDC Motors.

Thermographic fault diagnosis of ventilation in BLDC (brushless DC) motors is described. The following states of BLDC motors were analyzed: a healthy BLDC motor running at 1450 rpm, a healthy BLDC motor at 2100 rpm, blocked ventilation of the BLDC motor at 1450 rpm, blocked ventilation of the BLDC motor at 2100 rpm, healthy clipper, and blocked ventilation of the clipper. A feature extraction method called the Common Part of Arithmetic Mean of Thermographic Images (CPoAMoTI) was proposed. Test thermal images were analyzed successfully. The developed method, CPoAMoTI is useful for industry and society. Electric cars, trains, fans, clippers, computers, cordless power tools can be diagnosed using the developed method.

Statistical Study of Nonthermal Plasma-Assisted ZnO Coating of Cotton Fabric through Ultrasonic-Assisted Green Synthesis for Improved Self-Cleaning and Antimicrobial Properties.

Nonthermal plasma processing is a dry, environment-friendly and chemical-free method of improving the wettability, adhesion, self-cleaning and dying quality of fabrics without affecting their bulk properties. This study presents a green synthesis and coating method for the immobilization of nanoparticles of ZnO on the nonthermal plasma functionalized cotton fabric. The self-cleaning activity of ZnO-coated cotton was then optimized statistically. The ultraviolet protection and antimicrobial activity of the optimized and a control sample were also elaborated in this study. Psidium guajava Linn (guava) plant extract and zinc chloride were used in the ultrasonic biosynthesis of ZnO nanoparticles and concurrent immobilization over plasma functionalized cotton. Sodium hydroxide was used as a reaction accelerator. Statistical complete composite design (CCD) based on the amount of ZnCl2, NaOH and plasma exposure time was used to optimize the role of input parameters on the self-cleaning ability of the coated cotton. Methylene blue in water was used as a sample pollutant in the self-cleaning study. The ZnO-coated cotton showed notably high self-cleaning activity of 94% and a UV protection factor of 69.87. The antimicrobial activity against E. Coli and S. Aureus bacteria was also appreciably high compared to the control.

Statistically Optimized Production of Saccharides Stabilized Silver Nanoparticles Using Liquid-Plasma Reduction Approach for Antibacterial Treatment of Water.

Various conventional approaches have been reported for the synthesis of nanomaterials without optimizing the role of synthesis parameters. The unoptimized studies not only raise the process cost but also complicate the physicochemical characteristics of the nanostructures. The liquid-plasma reduction with optimized synthesis parameters is an environmentally friendly and low-cost technique for the synthesis of a range of nanomaterials. This work is focused on the statistically optimized production of silver nanoparticles (AgNPs) by using a liquid-plasma reduction process sustained with an argon plasma jet. A simplex centroid design (SCD) was made in Minitab statistical package to optimize the combined effect of stabilizers on the structural growth and UV absorbance of AgNPs. Different combinations of glucose, fructose, sucrose and lactose stabilizers were tested at five different levels (-2, -1, 0, 1, 2) in SCD. The effect of individual and mixed stabilizers on AgNPs growth parameters was assumed significant when p-value in SCD is less than 0.05. A surface plasmon resonance band was fixed at 302 nm after SCD optimization of UV results. A bond stretching at 1633 cm-1 in FTIR spectra was assigned to C=O, which slightly shifts towards a larger wavelength in the presence of saccharides in the solution. The presence of FCC structured AgNPs with an average size of 15 nm was confirmed from XRD and EDX spectra under optimized conditions. The antibacterial activity of these nanoparticles was checked against Staphylococcus aureus and Escherichia coli strains by adopting the shake flask method. The antibacterial study revealed the slightly better performance of AgNPs against Staph. aureus strain than Escherichia coli.

Concurrent Synthesis and Immobilization of Ag Nanoparticles over TiO2 via Plasma Reduction for Photocatalytic Treatment of Methyl Blue in Water.

Pure TiO2 nanoparticles (TiO2NPs) were produced via the sol-gel method and then coated with silver nanoparticles (AgNPs) to reduce their optical band gap. The concurrent synthesis and immobilization of AgNPs over TiO2NPs was achieved through the interaction of an open-air argon plasma jet with a solution of silver nitrate/stabilizer/TiO2NPs. The one-pot plasma synthesis and coating of AgNPs over TiO2NPs is a more straightforward and environmentally friendly method than others. The plasma-produced Ag/TiO2 nanocomposites were characterized and tested for their photocatalytic potential by degrading different concentrations of methyl blue (MB) in water. The dye concentration, oxidant dose, catalyst dose, and reaction time were also optimized for MB degradation. XRD results revealed the formation of pure AgNPs, pure TiO2NPs, and Ag/TiO2 nanocomposites with an average grain size of 12.36 nm, 18.09 nm, and 15.66 nm, respectively. The immobilization of AgNPs over TiO2NPs was also checked by producing SEM and TEM images. The band gap of AgNPs, TiO2NPs, and Ag/TiO2 nanoparticles was measured about 2.58 eV, 3.36 eV, and 2.86 eV, respectively. The ultraviolet (UV) results of the nanocomposites were supportive of the degradation of synthetic dyes in the visible light spectrum. The AgNPs in the composite not only lowered the band gap but also obstructed the electron-hole recombinations. The Ag/TiO2 composite catalyst showed 90.9% degradation efficiency with a 5 ppm dye concentration after 120 min of light exposure.

Synthesis and Characterization of Carbon and Carbon-Nitrogen Doped Black TiO2 Nanomaterials and Their Application in Sonophotocatalytic Remediation of Treated Agro-Industrial Wastewater.

The conventional open ponding system employed for palm oil mill agro-effluent (POME) treatment fails to lower the levels of organic pollutants to the mandatory standard discharge limits. In this work, carbon doped black TiO2 (CB-TiO2) and carbon-nitrogen co-doped black TiO2 (CNB-TiO2) were synthesized via glycerol assisted sol-gel techniques and employed for the remediation of treated palm oil mill effluent (TPOME). Both the samples were anatase phase, with a crystallite size of 11.09-22.18 nm, lower bandgap of 2.06-2.63 eV, superior visible light absorption ability, and a high surface area of 239.99-347.26 m2/g. The performance of CNB-TiO2 was higher (51.48%) compared to only (45.72%) CB-TiO2. Thus, the CNB-TiO2 is employed in sonophotocatalytic reactions. Sonophotocatalytic process based on CNB-TiO2, assisted by hydrogen peroxide (H2O2), and operated at an ultrasonication (US) frequency of 30 kHz and 40 W power under visible light irradiation proved to be the most efficient for chemical oxygen demand (COD) removal. More than 90% of COD was removed within 60 min of sonophotocatalytic reaction, producing the effluent with the COD concentration well below the stipulated permissible limit of 50 mg/L. The electrical energy required per order of magnitude was estimated to be only 177.59 kWh/m3, indicating extreme viability of the proposed process for the remediation of TPOME.

Severity Grading and Early Retinopathy Lesion Detection through Hybrid Inception-ResNet Architecture.

Diabetic retinopathy (DR) is a diabetes disorder that disturbs human vision. It starts due to the damage in the light-sensitive tissues of blood vessels at the retina. In the beginning, DR may show no symptoms or only slight vision issues, but in the long run, it could be a permanent source of impaired vision, simply known as blindness in the advanced as well as in developing nations. This could be prevented if DR is identified early enough, but it can be challenging as we know the disease frequently shows rare signs until it is too late to deliver an effective cure. In our work, we recommend a framework for severity grading and early DR detection through hybrid deep learning Inception-ResNet architecture with smart data preprocessing. Our proposed method is composed of three steps. Firstly, the retinal images are preprocessed with the help of augmentation and intensity normalization. Secondly, the preprocessed images are given to the hybrid Inception-ResNet architecture to extract the vector image features for the categorization of different stages. Lastly, to identify DR and decide its stage (e.g., mild DR, moderate DR, severe DR, or proliferative DR), a classification step is used. The studies and trials have to reveal suitable outcomes when equated with some other previously deployed approaches. However, there are specific constraints in our study that are also discussed and we suggest methods to enhance further research in this field.

Global Sensitivity Analysis of Chosen Harmonic Drive Parameters Affecting Its Lost Motion.

The aim of the presented study was to perform a global sensitivity analysis of various design parameters affecting the lost motion of the harmonic drive. A detailed virtual model of a harmonic drive was developed, including the wave generator, the flexible ball bearing, the flexible spline and the circular spline. Finite element analyses were performed to observe which parameter from the harmonic drive geometry parameter group affects the lost motion value most. The analyses were carried out using 4% of the rated harmonic drive output torque by the locked wave generator and fixed circular spline according the requirements for the high accuracy harmonic drive units. The described approach was applied to two harmonic drive units with the same ratio, but various dimensions and rated power were used to generalize and interpret the global sensitivity analysis results properly. The most important variable was for both harmonic drives the offset from the nominal tooth shape.

Real Time Multipurpose Smart Waste Classification Model for Efficient Recycling in Smart Cities Using Multilayer Convolutional Neural Network and Perceptron.

Urbanization is a big concern for both developed and developing countries in recent years. People shift themselves and their families to urban areas for the sake of better education and a modern lifestyle. Due to rapid urbanization, cities are facing huge challenges, one of which is waste management, as the volume of waste is directly proportional to the people living in the city. The municipalities and the city administrations use the traditional wastage classification techniques which are manual, very slow, inefficient and costly. Therefore, automatic waste classification and management is essential for the cities that are being urbanized for the better recycling of waste. Better recycling of waste gives the opportunity to reduce the amount of waste sent to landfills by reducing the need to collect new raw material. In this paper, the idea of a real-time smart waste classification model is presented that uses a hybrid approach to classify waste into various classes. Two machine learning models, a multilayer perceptron and multilayer convolutional neural network (ML-CNN), are implemented. The multilayer perceptron is used to provide binary classification, i.e., metal or non-metal waste, and the CNN identifies the class of non-metal waste. A camera is placed in front of the waste conveyor belt, which takes a picture of the waste and classifies it. Upon successful classification, an automatic hand hammer is used to push the waste into the assigned labeled bucket. Experiments were carried out in a real-time environment with image segmentation. The training, testing, and validation accuracy of the purposed model was 0.99% under different training batches with different input features.

A Novel Feature Extraction and Fault Detection Technique for the Intelligent Fault Identification of Water Pump Bearings.

The reliable and cost-effective condition monitoring of the bearings installed in water pumps is a real challenge in the industry. This paper presents a novel strong feature selection and extraction algorithm (SFSEA) to extract fault-related features from the instantaneous power spectrum (IPS). The three features extracted from the IPS using the SFSEA are fed to an extreme gradient boosting (XBG) classifier to reliably detect and classify the minor bearing faults. The experiments performed on a lab-scale test setup demonstrated classification accuracy up to 100%, which is better than the previously reported fault classification accuracies and indicates the effectiveness of the proposed method.