World Health Organization (WHO) has identified depression as a significant contributor to global disability, creating a complex thread in both public and private health. Electroencephalogram (EEG) can accurately reveal the working condition of the human brain, and it is considered an effective tool for analyzing depression. However, manual depression detection using EEG signals is time-consuming and tedious. To address this, fully automatic depression identification models have been designed using EEG signals to assist clinicians. In this study, we propose a novel automated deep learning-based depression detection system using EEG signals. The required EEG signals are gathered from publicly available databases, and three sets of features are extracted from the original EEG signal. Firstly, spectrogram images are generated from the original EEG signal, and 3-dimensional Convolutional Neural Networks (3D-CNN) are employed to extract deep features. Secondly, 1D-CNN is utilized to extract deep features from the collected EEG signal. Thirdly, spectral features are extracted from the collected EEG signal. Following feature extraction, optimal weights are fused with the three sets of features. The selection of optimal features is carried out using the developed Chaotic Owl Invasive Weed Search Optimization (COIWSO) algorithm. Subsequently, the fused features undergo analysis using the Self-Attention-based Gated Densenet (SA-GDensenet) for depression detection. The parameters within the detection network are optimized with the assistance of the same COIWSO. Finally, implementation results are analyzed in comparison to existing detection models. The experimentation findings of the developed model show 96% of accuracy. Throughout the empirical result, the findings of the developed model show better performance than traditional approaches.
The growing demand for Magnesium in the automotive and aviation industries has enticed the need to improve its corrosive properties. In this study, the WE43 magnesium alloys were friction stir welded (FSW) by varying the traverse speed. FSW eliminates defects such as liquefication cracking, expulsion, and voids in joints encountered frequently in fusion welding of magnesium alloys. The microstructural properties were scrutinized using light microscopy (LM) and scanning electron microscopy (SEM). Additionally, the elemental makeup of precipitates was studied using electron dispersive X-ray spectroscopy (EDS). The electrochemical behavior of specimens was evaluated by employing potentiodynamic polarization tests and was correlated with the microstructural properties. A defect-free weldment was obtained at a traverse and rotational speed of 100 mm/min and 710 rpm, respectively. All weldments significantly improved corrosion resistance compared to the base alloy. Moreover, a highly refined microstructure with redistribution/dissolution of precipitates was obtained. The grain size was reduced from 256 µm to around 13 µm. The corrosion resistance of the welded sample was enhanced by 22 times as compared to the base alloy. Hence, the reduction in grain size and the dissolution/distribution of secondary-phase particles within the Mg matrix are the primary factors for the enhancement of anti-corrosion properties.
In 2020, coronavirus (COVID-19) was declared a global pandemic and it remains prevalent today. A necessity to model the transmission of the virus has emerged as a result of COVID-19's exceedingly contagious characteristics and its rapid propagation throughout the world. Assessing the incidence of infection could enable policymakers to identify measures to halt the pandemic and gauge the required capacity of healthcare centers. Therefore, modeling the susceptibility, exposure, infection, and recovery in relation to the COVID-19 pandemic is crucial for the adoption of interventions by regulatory authorities. Fundamental factors, such as the infection rate, mortality rate, and recovery rate, must be considered in order to accurately represent the behavior of the pandemic using mathematical models. The difficulty in creating a mathematical model is in identifying the real model variables. Parameters might vary significantly across models, which can result in variations in the simulation results because projections primarily rely on a particular dataset. The purpose of this work was to establish a susceptible-exposed-infected-recovered (SEIR) model describing the propagation of the COVID-19 outbreak throughout the Kingdom of Saudi Arabia (KSA). The goal of this study was to derive the essential COVID-19 epidemiological factors from actual data. System dynamics modeling and design of experiment approaches were used to determine the most appropriate combination of epidemiological parameters and the influence of COVID-19. This study investigates how epidemiological variables such as seasonal amplitude, social awareness impact, and waning time can be adapted to correctly estimate COVID-19 scenarios such as the number of infected persons on a daily basis in KSA. This model can also be utilized to ascertain how stress (or hospital capacity) affects the percentage of hospitalizations and the number of deaths. Additionally, the results of this study can be used to establish policies or strategies for monitoring or restricting COVID-19 in Saudi Arabia.
This study investigates self-propelled rotary tool (SPRT) performance in hard turning using 3D finite element (FE) models. The FE models developed in this study are based on coupled temperature-displacement analysis using an explicit time-integration scheme. The developed FE models can predict chip morphology, cutting forces, tool and workpiece stresses and temperatures. For model verification, hard turning experiments were conducted using an SPRT on AISI 4340 bars. Cutting forces and maximum tool-chip interface temperatures were recorded and compared with the model findings. The effects of different process parameters were analyzed and discussed using the developed FE models. The FE models were run with a central composite design (CCD-25) matrix with four input variables, i.e., the cutting speed, the feed rate, the depth of the cut and the inclination angle. Response surfaces based on the Gaussian process were generated for each performance variable in order to predict design points not available in the original design of the experiment matrix. An optimization study was carried out to minimize tool stress and temperature while setting limits for the material removal rate (MRR) and specific cutting energy for the process. Optimized processes were found with moderate cutting speeds and feed rates and high depths of cut and inclination angles.
The performance of a self-propelled rotary carbide tool when cutting hardened steel is evaluated in this study. Although various models for evaluating tool wear in traditional (fixed) tools have been introduced and deployed, there have been no efforts in the existing literature to predict the progression of tool wear while employing self-propelled rotary tools. The work-tool geometric relationship and the empirical function are used to build a flank wear model for self-propelled rotary cutting tools. Cutting experiments are conducted on AISI 4340 steel, which has a hardness of 54-56 HRC, at various cutting speeds and feeds. The rate of tool wear is measured at various intervals of time. The constant in the proposed model is obtained using genetic programming. When experimental and predicted flank wear are examined, the established model is found to be competent in estimating the rate of rotary tool flank wear progression.
Applications of non-ferrous light metal alloys are especially popular in the field of aerospace. Hence it is important to investigate their properties in joining processes such as welding. Solid state joining process such as friction stir welding (FSW) is quite efficient for joining non-ferrous alloys, but with thick plates, challenges increase. In this study, Mg alloy plates of thickness 11.5 mm were successfully welded via single-pass FSW. Due to the dynamic recrystallization, grain size in the stir zone was reduced to 16 µm which is ≈15 times smaller than the parent material. The optimized rotational speed and traverse speed for optimum weld integrity were found to be 710 rpm and 100 mm/min, respectively. A sound weld with 98.96% joint efficiency, having an Ultimate Tensile Strength (UTS) of 161.8 MPa and elongation of 27.83%, was accomplished. Microhardness of the nugget was increased by 14.3%.
Nickel-Titanium (NiTi)-based shape-memory alloys (SMA) are utilized in automotive, biomedical, microsystem applications because of their excellent shape memory effect, biocompatibility and super elastic properties. These alloys are considered difficult to cut-especially with conventional technologies because of the work hardening and residual stresses. Laser-machining is one of the most effective tools for processing of these alloys especially for microsystem applications. In this work, a thorough investigation of effect of process parameters on machining of microchannels in NiTi SMA is presented. In addition, a multi-objective optimization is carried out in order to find the optimal input parameter settings for the desired output performances. The results show that the quality of microchannels is significantly affected by input parameters. Layer thickness was found to have a significant effect on taper angle of the microchannel. Scan speed, layer thickness and scan strategy were found to have significant effects on both spatter thickness and top-width error, but in opposite directions. The multi-objective optimization-minimizing taper angle and spatter thickness revealed an optimal solution that was characterized by high frequency, moderate speed and low layer-thickness and track displacement.
Ceramic microchannels have important applications in different microscale systems like microreactors, microfluidic devices and microchemical systems. However, ceramics are considered difficult to manufacture owing to their wear and heat resistance capabilities. In this study, microchannels are developed in alumina ceramic using direct Nd:YAG laser writing. The laser beam with a characteristic pulse width of 10 µs and a beam spot diameter of 30 µm is used to make 200 µm width microchannels with different depths. The effects of laser beam intensity and pulse overlaps on dimensional accuracy and material removal rate have been investigated using different scanning patterns. It is found that beam intensity has a major influence on dimensional accuracy and material removal rate. Optimum parameter settings are found using grey relational grade analysis. It is concluded that low intensity and low to medium pulse overlap should be used for better dimensional accuracy. This study facilitates further understanding of laser material interaction for different process parameters and presents optimum laser process parameters for the fabrication of microchannel in alumina ceramic.
Titanium aluminides qualify adequately for advanced aero-engine applications in place of conventional nickel based superalloys. The combination of high temperature properties and lower density gives an edge to the titanium aluminide alloys. Nevertheless, challenges remain on how to process these essentially intermetallic alloys in to an actual product. Electron Beam Melting (EBM), an Additive Manufacturing Method, can build complex shaped solid parts from a given feedstock powder, thus overcoming the shortcomings of the conventional processing techniques such as machining and forging. The amount of energy supplied by the electron beam has considerable influence on the final build quality in the EBM process. Energy input is decided by the beam voltage, beam scan speed, beam current, and track offset distance. In the current work, beam current and track offset were varied to reflect three levels of energy input. Microstructural and mechanical properties were evaluated for these samples. The microstructure gradually coarsened from top to bottom along the build direction. Whereas higher energy favored lath microstructure, lower energy tended toward equiaxed grains. Computed tomography analysis revealed a greater amount of porosity in low energy samples. In addition, the lack of bonding defects led to premature failure in the tension test of low energy samples. Increase in energy to a medium level largely cancelled out the porosity, thereby increasing the strength. However, this trend did not continue with the high energy samples. Electron microscopy and X-ray diffraction investigations were carried out to understand this non-linear behavior of the strength in the three samples. Overall, the results of this work suggest that the input energy should be considered primarily whenever any new alloy system has to be processed through the EBM route.
