Arabian Journal for Science and Engineering最新文献

This paper presents an ultra-low-voltage 10-bit successive approximation-register analog-to-digital converter (SAR ADC) based on the binary search algorithm for biomedical applications. An energy-efficient DAC switching scheme for a fully differential SAR ADC is proposed, which achieves a 99.8% reduction in DAC switching energy compared to conventional SAR ADC. In this design, by using a dummy capacitor split technique, an attempt has been made to reduce the capacitor of the most significant bit, resulting in a 92.87% reduction in the total number of capacitors compared to conventional design. In the proposed structure, the common-mode voltage of the comparator is approximately constant. The maximum voltage variation in the proposed switching scheme is Vref/2. Additionally, power consumption has been reduced by implementing the power gating technique in the control logic part. The proposed converter with a sampling frequency of 1 kS/s and a supply voltage of 0.2 V has been designed and simulated in TSMC 65nm CMOS technology. Both analytical calculations and simulation results confirm the effectiveness of the proposed switching scheme. Ultimately, the proposed scheme achieves a power consumption of 2.08 nW and a Figure of Merit (FoM) of 5.39 fJ/conversion-step. In comparison with the state-of-the-art, the proposed design has demonstrated excellent performance in achieving optimal power.

This paper deals with the discrete-time antiwindup compensator (AWC) synthesis for nonlinear discrete-time systems under input saturation. The proposed method considers the objective of an optimal AWC design for fast convergence and for improved performance against the saturation nonlinearity. A discrete-time full-order AWC architecture is presented for nonlinear discrete-time systems to achieve an improved performance against the saturation nonlinearity. Additionally, an equivalent decoupled AWC architecture for nonlinear discrete-time system is derived through algebraic analysis and transformation of saturation to dead-zone function. To achieve fast convergence, a more generic Lyapunov function has been applied for the AWC design by incorporating an exponential term in the Lyapunov function. Then, new conditions for the AWC synthesis are revealed by application of the resultant decoupled discrete-time architecture, nonlinearity condition, a modified quadratic-exponential Lyapunov function, optimally exponential (L_{2}) approach, and input saturation properties. The design conditions are provided for both global and local design scenarios, which can be applied to both stable and unstable plants. Compared with the conventional methods, the proposed approach deals with nonlinear systems, can be more practical due to discrete-time scenario, provides an optimal design for both fast convergence and performance, and applicable to both stable and unstable plants. A simulation example has been provided to demonstrate the efficacy of the proposed nonlinear AWC design.

The health status of bearings seriously affects the operational efficiency of equipment, and it is important to carry out bearing health status detection. A bearing fault diagnosis method based on stepwise variational modal decomposition (SVMD) with adaptive initialization center frequency and Gramian angular difference field is proposed. Firstly, a method of center frequency initialization base on frequency energy distribution characteristics is proposed to improve the decomposition speed and stability. Secondly, SVMD with single component decomposition and local decomposition is proposed to improve decomposition efficiency. It can effectively avoid inconsistency in different signal parameter settings and ensures consistency in the number of signal components, which is very suitable for batch processing of signals. Finally, Gramian angular field (GAF) and convolutional neural networks (CNNs) are combined to extract features of the reconstructed signal spectrum and enhance the differential characteristics between different signal spectrum. The experiment shows that the center frequency initialization method can shorten the single decomposition time from 11.13 to 6.71 s. The overall recognition rate can reach 95.2%, which is at least 1.9% higher than other decomposition methods.

Wire arc additive manufacturing (WAAM) causes significant heat accumulation within the deposited layers making it difficult to achieve necessary geometrical accuracy and built properties. A cost-effective technique to mitigate build-up heat and improve the thermal cycle is accomplished with the introduction of an inter-layer dwell time (ILDT). The study investigates the impact of ILDTs of 60, 90, 120, 150, and 180 s on the properties of the printed parts. The grain refinements were achieved at higher ILDT with no appreciable changes in the compositions. However, a slight increase in oxides and (delta )-ferrite was detected. The study found improved part density, mechanical and tribological performance while reduced corrosion-resistant properties with increased ILDT. Furthermore, to understand the effect of ILDTs on location-based thermal cycling and distribution, numerical models of the process were established. The peak temperature gradually reduces with an increase in ILDT. Heat accumulations within the three-layered WAAM-built structure lessen by 70% approximately with increases in ILDT from the 60 s to 180 s. The results observed repeated thermal cycles of the previously deposited layers and reduction of peak temperatures with subsequent layer deposition irrespective of the ILDTs. The trough temperature slightly reduced with ILDT; however, at higher delay periods the variations were found to be more stable. The study discovered too low and too high ILDT is detrimental and therefore an appropriate mid-range ILDT of 120 s is suggested to fabricate medium-size WAAM-built components based on experimental and numerical investigation considering overall production time and functional properties.

Adversarial attacks involve introducing minimal perturbations into the original input to manipulate deep learning models into making incorrect network predictions. Despite substantial interest, there remains insufficient research investigating the impact of adversarial attacks in real-world scenarios. Moreover, adversarial attacks have been extensively examined within the digital domain, but adapting them to realistic scenarios brings new challenges and opportunities. Existing physical world adversarial attacks often look perceptible and attention-grabbing, failing to imitate real-world scenarios credibly when tested on object detectors. This research attempts to craft a physical world adversarial attack that deceives object recognition systems and human observers to address the mentioned issues. The devised attacking approach tried to simulate the realistic appearance of stains left by rain particles on traffic signs, making the adversarial examples blend seamlessly into their environment. This work proposed a region reflection algorithm to localize the optimal perturbation points that reflected the trusted regions by employing the trust region optimization with a multi-quadratic function. The experimental evaluation reveals that the proposed work achieved an average attack success rate (ASR) of 94.18%. Experimentation underscores its applicability in a dynamic range of real-world settings through experiments involving distance and angle variations in physical world settings. However, the performance evaluation across various detection models reveals its generalizable and transferable nature. The outcomes of this study help to understand the vulnerabilities of object detectors and inspire AI (artificial intelligence) researchers to develop more robust and resilient defensive mechanisms.

Rapid advancements in battery technologies led to dramatic growth in adoption of electric vehicles (EVs) all over the world. On the other hand, ever-increasing renewable energy sources (RES) in microgrids (MGs) posing numerous challenges ahead. In this context, EVs can be used as virtual storage units to confront the intermittency aspect of RES in MG scenarios. This work proposes an EV fleet control strategy to implement a three-layer energy management system: Optimal storage distribution (OSD), optimal power exchange (OPE) and smart EV ranking (SER). The key objectives are minimizing grid dependency, energy cost, EV battery degradation and to maximize EV storage usage. Water filling algorithm is used to obtain OSD and multi-objective optimization problem is formulated and solved by e-constraint method to obtain OPE. SER is implemented using a fuzzy logic controller where a number of decision variables are involved. EV battery degradation has been considered through SER by including a key decision variable, EV usage probability (EUP). EUP has been obtained using a probabilistic approach that accounts all possible state transitions of each EV in the given time interval. An on-grid MG scenario with EV fleets and RES is considered to implement the proposed EMS.

Due to the high deployment flexibility and strong maneuverability, unmanned aerial vehicles (UAVs) have gained a significant attention in civilian and military applications. One of the main essential aspects of UAV is coverage path planning (CPP), which autonomously obtains sufficient paths to cover the entire region of interest (RoI). Several advantages have been offered by UAVs’ CPP such as cost and time efficiency, reduced human intervention, resource optimization, data collection, scalability, and adaptability. However, the flight time of UAVs is constrained by battery capacity, necessitating energy-efficient solutions to prolong flight duration. This paper introduces a novel approach for energy-efficient multi-UAV multi-region CPP to generate appropriate paths that cover multiple disjoint regions, aiming to minimize overall energy consumption. First, we employ a back-and-forth strategy to generate intra-region path patterns with minimum turns and propose a smoothing turns approach (STA) based on Bezier curves to effectively reduce an energy consumption due to taking turns. Then, the inter-region path planning is formulated as a multi-constraint optimization problem and solved utilizing the CPLEX solver for small-scale problems and heuristic approaches for large-scale ones. A region allocation approach is proposed to assign RoIs to appropriate UAVs. Simulations are conducted to evaluate the performance of the proposed approach in terms of energy consumption. Comparative results against EECPPA and Nearest Neighbor (NN) approaches demonstrate advantages in energy consumption reduction. Besides, heuristic methods yield superior solutions for large-scale problems within shorter execution times compared to the CPLEX solver. These comparisons highlight the superiority of the proposed approach over existing methods in generating higher-quality solutions.

In this paper, we propose a new strategy to exploit the advantages of Deep Neural Network-based architectures for brain tumor classification using MRI images for a better diagnosis. This was achieved by analyzing and evaluating pre-trained models on three different datasets. To better design the optimal architecture for solving the classification of brain tumor using MRIs, we have conducted extensive experiment-based analysis on how different layers of Convolutional Neural Network (CNN) process the inputs. Four distinct architectures are then built, each with its specific hyperparameters and layers. The images are fed into the convolutional layers for feature extraction followed by a softmax function before applying the classification process. An extensive experimental study carried out clearly demonstrates that our novel CNN-based classification approach achieves state-of-the-art accuracy, precision, recall and an F1-score of 99.76% 99.64% 99.62% and 99.64%, respectively. Also, a higher performance in terms of Micro-Avg Matthew correlation coefficient (MCC) of 0.929 is achieved. This exceptional performance is achieved thanks to the new proposed model's architecture. Indeed, unlike conventional methods, that often rely on complex transfer learning models or hybrid architectures, our approach utilizes a custom and non-hybrid scheme. Consequently, this streamlined architecture offers a significant advantage of being remarkably lightweight, enabling efficient operation on resource-constrained computing systems.

The operational behavior of modern active distribution networks differs greatly from their passive predecessors. Active network management schemes, such as Conservation Voltage Reduction (CVR), are being widely explored to enhance energy conservation and network resiliency. CVR can potentially decrease energy consumption in distribution networks by reducing the voltage at load terminals. However, accurately assessing CVR suitability for distribution feeders is paramount, as it exploits the voltage dependency of load. Most CVR assessment schemes require extensive data from smart meters, micro-pmus, and detailed network topology information. This paper presents a novel synthetic data-based approach for probabilistic CVR assessment, which uses customer categorization to generate realistic synthetic data for probabilistic assessment, taking into account the voltage dependency of common household appliances. Additionally, an extended version of the probabilistic synthetic-data-based approach is introduced to enhance the accuracy of CVR assessment using limited distribution network information. MATLAB was used for data handling and customer categorization, while OpenDSS was used for power flow analysis to evaluate CVR effectiveness. The proposed CVR assessment methodologies are tested and validated in detail for the TOPI distribution network in KPK, Pakistan. The study’s findings shows that CVR can significantly reduce energy and cost, with a 10.13% reduction in energy use and approximately 4 million PKR in cost

The broad term ‘leukemia’ refers to different types of cancer related to blood cells. Detecting and identifying the specific type of leukemia continues to be a major challenge in the medical field. Diverse machine learning techniques can be vital in analyzing gene expression data from microarray experiments in cancer research related to leukemia. In particular, the Leukemia Gene Expression data from the Curated Microarray Database (CuMiDa) is used here. Microarrays can be challenging in determining expression patterns. In this work, we use Fisher’s linear discriminant analysis, a popular technique for dimensionality reduction, together with a new feature selection approach to predict leukemia using microarray data. Our machine learning model is used to predict five types of leukemia including AML, PBSC CD34, Bone Marrow, and CD34 from the bone marrow. This is achieved by first rescaling the data features. We then use a feature selection technique to obtain the 25 most significant features from the dataset’s 22,283 features, then further reduce the dimension to 5 features only, to reduce computational complexity. These features are then fed into a Fisher’s linear discriminant module and a likelihood-based index for classification. The overall performance of our model was excellent. We examine the results using 2, 4, 5, 6, and 7 selected features. The best classification accuracies are 89.6%, 96.92%, and 96.15%, for 2, 5, and 7 selected features, respectively. Our results outperform the state-of-the-art by about 4%, with an excellent task completion time of less than 100 ms.