AUS Repository (American University of Sharjah)
Not a member yet
2669 research outputs found
Sort by
Machine Learning Based Palm Farming: Harvesting and Disease Identification
No full textIn the culturally and economically vital date palm sector of the Arab world, precise assessment of fruit maturity, type, and disease is crucial for optimizing yield, quality, and palm health. This work pioneers a novel paradigm: machine learning (ML) frameworks for analysis of all three aspects using individual and merged datasets. Moreover, explainable AI (XAI) techniques are exploited to enhance result interpretability which has not been previously explored in this field. The purpose of this work is two-fold: 1) date fruit bunch type and ripeness classification, 2) classification of healthy and three stages of white-scale disease (WSD) infested date palm leaflets. For this purpose, we utilize deep learning (DL) models by adding additional layers and optimizing various parameters to enhance their performance for these specific tasks. Two publicly available datasets are used for both type and ripeness classification: Dataset 1 contains 8079 images, and Dataset 2 contains 9092 images of date fruit bunches. Furthermore, dataset 3 with 2161 images is used for healthy and WSD infestation stage identification. For individual datasets, the best performing model, VGG16, achieved the highest accuracy for date type classification (98%) and ripeness classification (93%), using dataset 1. The best performing classifier architecture on merged dataset, VGG16, achieved an accuracy of 97% and 94% for date fruit type and ripeness classification, respectively. The highest accuracy achieved for healthy and WSD classification was 99.7% using VGG16. These results were explained using several XAI techniques which were found to be useful in enhancing the models’ interpretability. Through this work, precision agriculture in the date palm sector stands to benefit from informed decision-making, optimized resource allocation, and the adoption of sustainable practices. This work contributes significantly to the sector's advancement, ensuring a thriving and resilient date palm industry in the region.American University of Sharja
INScription: Department of International Studies (INS) Issue #19 (January 25, 2024, Issue 5)
No full textCollege of Arts and SciencesDepartment of International Studie
Reliability-Centered Maintenance Implementation for Water Stations
No full textA Master of Science thesis in Engineering Systems Management by Dalal Abdulla Ali Mohammed Abdulla entitled, “Reliability-Centered Maintenance Implementation for Water Stations”, submitted in November 2024. Thesis advisor is Dr. Mahmoud Awad and thesis co-advisor is Dr. Hussam Alshraideh. Soft copy is available (Thesis, Completion Certificate, Approval Signatures, and AUS Archives Consent Form).Ensuring equipment availability and reliability is critical in utility systems, particularly in water distribution networks. Failures in these systems can have severe consequences, including substantial operational costs and widespread service disruptions affecting large customer populations. Reliability Centred Maintenance (RCM) is a proven and systematic methodology used to determine the requirement and effective maintenance strategy for an asset. Despite the wide implementation of RCM in many industries, RCM implementation in water pump stations (WPSs) is still limited. This is due to several challenges, such as the redundancy nature of WPS’s systems, data accessibility, and difficulty of estimating the benefits of RCM implementation beforehand. The objective of this study is to customise the existing RCM framework for WPS’s and develop a method for estimating the expected cost and benefits of RCM implementation. Water stations can greatly benefit from the implementation of RCM by saving time and money in predicting and mitigating failure risks. The customised approach is demonstrated using a major pumping station located in Dubai. Based on data collected from six failure case studies, changing the maintenance technique to one advocated by customized RCM will reduce cost and improve service availability. Results suggest that benefit to cost ratio gained from maintenance technique change is between 1.15 to 9.06.College of EngineeringDepartment of Industrial EngineeringMaster of Science in Engineering Systems Management (MSESM
Aeroelastic Behavior of a Piezoelectric Energy Harvesting Flag Under Wind Excitation
No full textThis study aims to advance the understanding of alternative energy sources by exploring flag-based flutter energy harvesters. A smart piezoelectric flag equipped with an MFC patch on a steel substrate is designed and subjected to experimental aeroelastic and energy harvesting analyses under various inflow conditions. The research reveals that under streamlined flow, flutter occurs at higher flow speeds, but can be induced at lower speeds if subcriticality is effectively triggered, thus potentially enhancing energy harvesting at reduced flow speeds. Conversely, under vortex-induced excitation, the flag demonstrates limited energy harvesting potential at smaller gap distances from the upstream bluff body. However, larger gap distances facilitate energy capture at lower flow speeds. The study examines the impact of bluff body diameter on the performance of the piezoelectric flags using cylinder A (2 cm diameter) and cylinder B (3 cm diameter). Results indicate that while cylinder A results in reduced energy harvesting efficiency, cylinder B significantly enhances it. Overall, this study identifies the optimal conditions for maximizing energy harvesting from aeroelastic flutter, considering the combined effects of flag positioning, flow characteristics, bluff body shape, and wind speed. To the best of our knowledge, a hand in hand analysis of these parameters has not been explored in the hitherto literature
Hand-Crafted Features With A Simple Deep Learning Architecture For Sensor-Based Human Activity Recognition
No full textWith the growth in the wearable device market, wearable sensor-based human activity recognition systems have been gaining increasing interest in research because of their rising demands in many areas. This research presents a novel sensor-based human activity recognition system that utilizes a unique feature extraction technique associated with a deep learning method for classification. One of the main contributions of this work is dividing the sensor sequences time-wise into non-overlapping 2D segments. Then, statistical features are computed from each 2D segment using two approaches; the first approach computes features from the raw sensor readings, while the second approach applies time-series differencing to sensor readings prior to feature calculations. Applying time-series differencing to 2D segments helps in identifying the underlying structure and dynamics of the sensor reading across time. This work experiments with different numbers of 2D segments of sensor reading sequences. Also, it reports results with and without the use of different components of the proposed system. Additionally, it analyses the best-performing models’ complexity, comparing them with other models trained by integrating the proposed method with an existing transformer network. All of these arrangements are tested with different deep-learning architectures supported by an attention layer to enhance the model. Four benchmark datasets are used to perform several experiments, namely, mHealth, USC-HAD, UCI-HAR, and DSA. The experimental results revealed that the proposed system outperforms human activity recognition rates reported in the most recent studies. Specifically, this work reports recognition rates of 99.17%, 81.07%, 99.44%, and 94.03% for the four datasets, respectively.American University of Sharja
Design and Verification of High Efficiency Energy Harvesting System at RF Frequencies
No full textA Master of Science thesis in Electrical Engineering by Hebah Rabah entitled, “Design and Verification of High Efficiency Energy Harvesting System at RF Frequencies”, submitted in November 2024. Thesis advisor is Dr. Lutfi Albasha and thesis co-advisor is Dr. Hasan Mir. Soft copy is available (Thesis, Completion Certificate, Approval Signatures, and AUS Archives Consent Form).Energy harvesting systems stand as a promising method to power Internet of Things (IoT) devices efficiently. This thesis presents the novel design of an energy harvesting system operating at 5.8GHz for IoT applications. The work was divided into two main phases. The first phase involved designing and simulating a rectifier circuit at 5.8GHz using the TSMC 65nm process. Challenges associated with high-frequency operation, such as parasitic capacitances, frequency-dependent leakage currents, and impedance mismatches, were addressed. These challenges were carefully managed to establish good performance at a higher frequency. For example, layout techniques were applied to reduce parasitic capacitance and leakage current, thereby minimizing unwanted energy loss and improving efficiency. A high-pass impedance matching network was also designed to reduce reflection loss and facilitate optimal power transfer to the rectifier. The second phase focused on creating a physical layout of the rectifier circuit, followed by a performance evaluation of the extracted circuit. After designing the layout, critical testing was conducted to compare the simulated and layout-extracted circuits. Key performance metrics, such as output voltage and power conversion efficiency (PCE), were analyzed. The post-layout extracted, closest to fabricated results, achieved an output voltage of 2.88V and a PCE of 82.94%, demonstrating high efficiency despite the presence of parasitic elements introduced during the layout process. The novelty of this thesis lay in being the first to design, simulate, and create a layout for a Dickson charge pump rectifier at 5.8GHz using TSMC 65nm process. The design successfully passed Design Rule Checks (DRC) and Layout Versus Schematic (LVS) tests, and the extracted circuit accurately represented real-world chip performance, with an efficiency of 82.94%. This research contributed to the advancement of self-powered IoT systems by optimizing energy harvesting at high frequencies, reducing reliance on traditional batteries, and enhancing the longevity of IoT networks.College of EngineeringDepartment of Electrical EngineeringMaster of Science in Electrical Engineering (MSEE
Assessment of Brain Function After 240 Days Confinement Using Functional Near Infrared Spectroscopy
No full textFuture space exploration missions will expose astronauts to various stressors, making the early detection of mental stress crucial for prolonged missions. Our study proposes using functional near infrared spectroscopy (fNIRS) combined with multiple machine learning models to assess the level of mental stress. Objective: The objective is to identify and quantify stress levels during 240 days confinement scenario. In this study, we utilize a diverse set of stress indicators including salivary alpha amylase (sAA) levels, reaction time (RT) to stimuli, accuracy of target detection, and power spectral density (PSD), in conjunction with functional connectivity networks (FCN). We estimate the PSD using Fast Fourier Transform (FFT) and the FCN using partial directed coherence. Results: Our findings reveal several intriguing insights. The sAA levels increased from the first 30 days in confinement to the culmination of the lengthy 240-day mission, suggesting a cumulative impact of stress. Conversely, RT and the accuracy of target detection exhibit significant fluctuations over the course of the mission. The power spectral density shows a significant increase with time-in-mission across all participants in most of the frontal area. The FCN shows a significant decrease in most of the right frontal areas. Five different machine learning classifiers are employed to differentiate between two levels of stress resulting in impressive classification accuracy rates: 96.44% with-nearest neighbor (KNN), 95.52% with linear discriminant analysis (LDA), 88.71% with Naïve Bayes (NB), 87.41 with decision trees (DT) and 96.48% with Support Vector Machine (SVM). In conclusion, this study demonstrates the effectiveness of combining functional near infrared spectroscopy (fNIRS) with multiple machine learning models to accurately assess and quantify mental stress levels during prolonged space missions, providing a promising approach for early stress detection in astronauts.Mohammed Bin Rashid Space CenterAmerican University of SharjahCollege of EngineeringDepartment of Electrical Engineerin
The Scalability of Third Generation Photovoltaics: Deposition Techniques and Modularity
No full textThe development of third-generation photovoltaic (TGPV) technologies promises to address some of the limitations of conventional silicon-based solar cells. However, the scalability of these technologies is a critical factor in their practical application. TGPV include Perovskite Solar Cells (PSC), Organic Solar Cells (OSC), and Dye-Sensitized Solar Cell (DSSC). This paper reviews and summarizes the recent trends and research on the deposition techniques and modularity of TGPV. Various deposition techniques such as slot-die coating, thermal evaporation, and spray pyrolysis are discussed, along with their advantages and limitations. Modularity, which allows for the integration of TGPV cells into different structures, is also examined as a critical factor in scalability. The paper concludes that the scalability of TGPV technologies depends on the development of efficient and cost-effective deposition techniques and modularity, which will facilitate the integration of the TGPV cells into various structures and enable the widespread use of these promising technologies
Elastic Buckling Behavior of Functionally Graded Material Thin Skew Plates with Circular Openings
No full textThis study investigates the elastic buckling behavior of Functionally Graded Material (FGM) thin skew plates featuring a circular opening. FGMs, known for their unique property gradients, have gained prominence in structural engineering due to their mechanical performance and durability. Including a circular opening introduces a critical geometric consideration, influencing the structural stability and load-carrying capacity of FGM plates. The study examines the effects of the skew angle, plate’s aspect ratio, opening position, and size on the critical buckling load, normalized buckling load, and various buckling failure modes through computer modeling and finite element analysis. The results offer valuable insights into the interplay between material heterogeneity, geometric configuration, and structural stability. For instance, the critical buckling load increases by 29%, 82%, and 194% with an increment in skew angle from 0° to 30°, 45°, and 60°, respectively. Moreover, as the opening shifts from the plate’s edge closer to the center, the critical buckling load decreases by 26%. The critical buckling load is also dependent on the power index, as an increase in the power index from 0.2 to 5 reduced the buckling load by 1698 kN. This research contributes to the advancement of our understanding of FGM thin plates’ behavior under skew loading conditions, with implications for the design and optimization of innovative structures. The findings presented provide a foundation for further exploration of advanced composite materials and their applications in structural engineering.Open Access Program in the American University of SharjahFaculty Professional Development Plan Program in the American University of Sharja
Behavior of normal and recycled aggregates beams strengthened with different types of externally bonded shear reinforcement
No full textThis study aims at investigating the effect of strengthening shear-deficient recycled aggregate concrete (RAC) beams with carbon fiber-reinforced polymer (CFRP) laminates. Five RAC beams were cast, four of which were strengthened with different CFRP shear strengthening configurations: U-wraps bonded at 45°, vertical U-wraps, continuous U-wraps along the shear span, and side-boned laminates. In addition, one RAC specimen was left unstrengthened to act as a benchmark specimen. For comparison purposes, an additional five normal aggregate concrete (NAC) beams were cast, three of which are strengthened with similar CFRP schemes as that of the RAC, and one was left unstrengthened. All beams are loaded under four-point bending tests, and the results in terms of shear force-deflection graphs and failure modes are analyzed and compared. Experimental results indicated that the shear force values obtained in NAC and RAC beams are comparable. In fact, the percentage increase in the shear strength compared to the respective control beam was higher for RAC beams than that of NAC beams. This proves the effectiveness of using different shear strengthening configurations and the viability of using CFRP shear strengthened RAC beams compared to CFRP shear strengthened NAC beams