Articles

Access the latest knowledge in applied science, electrical engineering, computer science and information technology, education, and health.

Advanced Search
Filter Icon

Filters article

Years

FAQ Arrow
Single Year
Range Year
0
0

Source Title

FAQ Arrow

Authors

FAQ Arrow

30,411 Article Results

Optimal sizing and performance evaluation of hybrid photovoltaic-wind-battery system for reliable electricity supply

10.11591/ijece.v15i5.pp4341-4354
Youssef El Baqqal , Mohammed Ferfra , Reda Rabeh
Given the advantages of hybrid renewable energy systems over single-source systems, this study proposes the optimal sizing and performance evaluation of a hybrid photovoltaic-wind battery system to meet the electricity demand of an isolated community in Dakhla, Morocco. The objective is to achieve an economical approach to electricity generation. Particle swarm optimization (PSO) and grey wolf optimizer (GWO) techniques were used to determine the optimal configuration of system components, including photovoltaic (PV) panels, wind turbines, and battery storage. The annual system cost (ACS) is minimized as the optimization objective, and the levelized cost of electricity (LCOE) is used for economic comparison. MATLAB serves as the platform for implementation and evaluation. Results demonstrate the convergence and effectiveness of PSO and GWO in delivering high-quality solutions. PSO, however, achieves superior system reliability with a lower loss of power supply probability (LPSP) during peak demand. The optimal configuration achieves a minimal LCOE of 0.1065 USD/kWh, representing a 33.44% reduction compared to the applicable rate. These findings highlight the potential of advanced optimization techniques to improve the economic and operational performance of hybrid renewable energy systems, making them a viable solution for rural electrification in regions with limited grid access.
Volume: 15
Issue: 5
Page: 4341-4354
Publish at: 2025-10-01

Optimized fractional-order direct torque control with space vector modulation strategy for two-wheel-drive electric vehicles

10.11591/ijece.v15i5.pp4409-4420
Touhami Nawal , Ouled-Ali Omar , Mansouri Smail , Benhammou Aissa
Electric vehicles (EVs) are a sustainable and efficient transportation choice, offering zero emissions, lower operating costs, and advanced performance features like instant torque and regenerative braking. They promote energy independence, improve urban livability, and support the global shift toward cleaner, renewable energy-powered mobility, making them a future-proof investment. The electric motor is a critical component in electric vehicles (EVs), the importance of which lies in its high efficiency, instant torque delivery, and smooth operation, which enhances performance and energy use. This paper focuses on a two-wheel drive electric vehicle (TWD EV) configuration powered by an energy storage battery system (ESBS), driven by two permanent magnet synchronous motors (PMSMs), and controlled using direct torque control with space vector modulation (DTC-SVM). fractional-order proportional integral derivative (FOPID) controllers, optimized via the grey wolf optimizer (GWO) algorithm, are implemented for precise speed control of the PMSMs. An electronic differential (ED) is incorporated to ensure vehicle stability, safety, and performance. The simulation results show that the proposed GWO-FOPID controller gave super results by reducing electromagnetic torque overshoot by 33%, improves torque settling time by 55%, and achieves the lowest electromagnetic torque ripple of approximately ±1 Nm compared to conventional DTC-SVM and GWO-PID approaches. Additionally, it optimized speed overshoot and undershoot by 44%, significantly enhancing system performance, responsiveness, and drive smoothness. This novel combination of fractional-order control, metaheuristic optimization, and electronic differential integration marks a meaningful advancement in high-precision and efficient control for 2WD EVs.
Volume: 15
Issue: 5
Page: 4409-4420
Publish at: 2025-10-01

Comparative analysis of metaheuristic algorithms (genetic algorithm, artificial bee colony, differential evolution) in the design of substrate integrated waveguide dual bandpass filter

10.11591/ijece.v15i5.pp4682-4691
Souad Akkader , Abdennaceur Baghdad , Hamid Bouyghf , Aziz Dkiouak , Yassine Gmih
A well-optimized substrate integrated waveguide (SIW) filter can significantly enhance the performance of modern technologies, including wireless communication systems, radar, and sensors. The frequencies of 5 and 6 GHz play a crucial role in these applications. Metaheuristic algorithms such as genetic algorithm (GA), artificial bee colony (ABC), and differential evolution (DE) are effective for designing SIW filters specifically tailored to these needs. This paper evaluates the performance of evolutionary optimization techniques in the design of substrate integrated waveguide filters. The optimization focuses on achieving optimal impedance matching within the frequency range of 4 to 8 GHz. The attenuation constant serves as the cost function, guiding the optimization process to ensure reliable and accurate results from each algorithm. The filter parameters derived from the most efficient algorithm are verified using ANSYS HFSS, resulting in two bands with S11=-45 dB and S21=-0.2 dB in the first band, and S11=-28 dB and S21=-0.5 dB in the second band. Additionally, two transmission zeros with rejections of -23 and -12 dB are achieved at 6.4 and 7.08 GHz, respectively. These results highlight the practicality of SIW technologies in designing microwave circuits, particularly for internet of things (IoT) applications.
Volume: 15
Issue: 5
Page: 4682-4691
Publish at: 2025-10-01

A non-destructive approach for estimation of Hb, HCT and red blood cells using reflectance spectroscopic technique

10.11591/ijece.v15i5.pp4569-4580
P. Divyabharathi , Neelamegam Devarasu
Paediatric haematology involves the use of non-invasive methods and technologies to evaluate haematological parameters in children. These techniques attempt to offer precise measurements of blood constituents without the necessity of intrusive procedures such as venipuncture or blood draws, which can be difficult and unpleasant for paediatric patients. The data gathered from the elbow will be given priority for further investigations to find haematological profiles. Estimates of haemoglobin, haematocrit, and red blood cell count were done and compared against the values obtained using conventional methods. This method achieves an accuracy of 75.56% with high precision and specificity which makes the method particularly beneficial for paediatric applications, potentially due to physiological differences or enhanced calibration for younger populations. The sensitivity varies with red blood cells (RBC) showing the lowest true positive detection rate. Future work could focus on improving the sensitivity of these parameters to enhance the accuracy. Conventional techniques cannot monitor continuously and remotely, which is crucial for a point-of-care screening device in the current era. The proposed non-destructive technique offers the benefits of infection control, pain reduction, and minimal operational cum maintenance expenses, all while being portable and child friendly.
Volume: 15
Issue: 5
Page: 4569-4580
Publish at: 2025-10-01

Enhancing facial landmark detection with ControlNet-based data augmentation

10.11591/ijece.v15i5.pp4907-4915
Kritaphat Songsri-in , Munlika Rattaphun , Sopee Kaewchada , Sunisa Kidjaideaw , Sangjun Ruang-On , Wichit Sookkhathon , Patompong Chabplan
Facial landmark detection plays a pivotal role in various computer vision applications, including face recognition, expression analysis, and augmented reality. However, existing approaches often struggle with accuracy due to the variations in lighting, poses, and occlusion. To address these challenges, this study explores the integration of ControlNet with Stable Diffusion to enhance facial landmark detection via data augmentation. ControlNet, an advanced extension of diffusion models, improves image generation by conditioning outputs on structured inputs such as landmark coordinates, enabling precise control over image attributes. By leveraging annotated landmark data from the 300W dataset, ControlNet synthesizes diverse facial images that supplement traditional training datasets. Experimental results demonstrate that ControlNet-based augmentation reduces the interocular normalized mean error (INME) in landmark detection from a baseline of 4.67 to a range of 4.63 to 4.74, with optimal parameter tuning yielding further accuracy gains. These findings highlight the potential of generative models in complementing discriminative approaches and improving robustness and precision in facial landmark detection. The proposed method offers a scalable solution for enhancing model generalization, particularly in applications requiring high-fidelity facial analysis. Future research can extend this framework to broader computer vision tasks that demand detailed feature localization and structured data augmentation.
Volume: 15
Issue: 5
Page: 4907-4915
Publish at: 2025-10-01

Performance evaluation of a high-gain 50 W DC-DC flyback boost converter for variable input voltage low-power photovoltaic applications

10.11591/ijece.v15i5.pp4520-4530
Muhammad Hafeez Mohamed Hariri , Lim Kean Boon , Tole Sutikno , Nor Azizah Mohd Yusoff
DC-DC boost converters are essential for stabilizing the voltage output of photovoltaic (PV) modules. This paper analyzes a unique 50 W high-gain DC-DC flyback boost converter for various input voltage PV applications. Scientific analysis was employed to determine suitable parameters for critical circuit components. Simulations were conducted to evaluate the proposed high-gain DC-DC boost converter's performance. Subsequently, a prototype of the high-gain DC boost converter was fabricated with a printed circuit board (PCB) size of 100×100 mm. The proposed prototype's performance is compared to that of conventional boost converters based on criteria such as input voltage, output voltage, component count, voltage stress, voltage gain, efficiency, and rated power. The results indicate that the proposed converter can achieve a 300 V output voltage with a 50 W power rating from variable input voltages ranging between 12 V and 36 V. The highest gain achieved was 25 with a 12 V input voltage, though at a lower power rating of 15 W. A peak efficiency of 84.30% was measured with a 24 V DC input voltage. The proposed converter's features, particularly its high step-up voltage gain, make it suitable for industrial and renewable energy applications.
Volume: 15
Issue: 5
Page: 4520-4530
Publish at: 2025-10-01

A novel approach for recommendation using optimized bidirectional gated recurrent unit

10.11591/ijece.v15i5.pp5019-5030
Prakash Pandharinath Rokade , Swati Babasaheb Bhonde , Prashant Laxmanrao Paikrao , Umesh Baburao Pawar
In today's world, every one of us refreshes our mood and gets energy through entertainment and enjoyment. Human nature is to provide feedback through ratings or comments for products used, services received, or films viewed. The recommendation system serves the user with recommendations based on historical stored information of user preferences. These systems amass information about the user in order to provide personalized experiences. These systems put efforts into delivering personalized experiences by accumulating information about the user. Hybrid algorithms are necessary to address the issues recommendation systems confront, which include low prediction accuracy, output that exceeds range, and inadequate convergence speed. This study suggests building a movie recommendation system using the remora optimization algorithm (ROA) and the bidirectional gated recurrent unit (BiGRU), the most recent version of the recursive neural network (RNN). The proposed method's results are compared with those of the genetic algorithm (GA), feed forward neural network (FFNN), and multimodal deep learning (MMDL). In terms of movie recommendation, BiGRU with ROA performs better than GA, MMDL, and FFNN.
Volume: 15
Issue: 5
Page: 5019-5030
Publish at: 2025-10-01

Efficient mask region-based convolutional neural network-based architecture for COVID-19 detection from computed tomography data

10.11591/ijece.v15i5.pp4751-4761
Nader Mahmoud , Ashraf B. El-Sisi
The worldwide effect of the coronavirus disease (COVID-19) pandemic has been catastrophic, leading to a significant number of fatalities worldwide. In response to the outbreak, health care institutions have proposed the use of chest computed tomography (CT) as an important diagnosis tool for rapid diagnosis, leveraging deep learning approaches for disease detection. This paper aims to progress a robust methodology towards accurate diagnosis of COVID-19 based on deep learning approaches with chest CT images. We propose a mask region-based convolutional neural network (Mask R-CNN) model architecture that is well-trained and used to discriminate between COVID-19-infected and uninfected cases. In order to improve feature extraction, the proposed model incorporates a fuzzy color enhancement preprocessing technique that reduces image fuzziness and increases contrast. A publicly available chest CT dataset is considered for quantitative evaluation of the proposed architecture model, which includes various frontal image views of COVID-19 and non-COVID-19 cases. The proposed approach yielded an accuracy of 98.8% with 98.4% precision and 98.5% recall. Additionally, the proposed model architecture has been quantitatively evaluated in comparison with benchmark approaches, yielding superior performance in terms of conventional evaluation metrics.
Volume: 15
Issue: 5
Page: 4751-4761
Publish at: 2025-10-01

An efficient direction oriented block-based video inpainting using morphological operations and adaptively dimensioned search region with direction-oriented block-based inpainting

10.11591/ijece.v15i5.pp4705-4713
Shyni Shajahan , Y. Jacob Vetha Raj
Video inpainting is a technique in computer vision used to remove unwanted objects from video sequences while preserving visual consistency, so that modifications remain unnoticeable to the human eye. This paper presents an accurate video inpainting model based on the adaptively dimensioned search region with direction-oriented block-based inpainting (ADSR-DOBI) algorithm. The model operates in five main phases: preprocessing, background separation, morphological operations, object removal, and video inpainting. Initially, the input video is converted into frames, followed by preprocessing steps such as deionizing and resizing. These frames are then processed using a background subtraction module, where object localization and foreground detection are performed using the binomially distributed foreground segmentation network (BDFgSegNet) and morphological techniques. This results in segmented foreground objects tracked across frames. The object removal phase eliminates the identified foreground objects and defines the missing regions (holes) to be filled. The ADSR-DOBI algorithm is then applied to inpaint these regions seamlessly. Experimental results demonstrate that this approach outperforms existing state-of-the-art methods in both accuracy and efficiency.
Volume: 15
Issue: 5
Page: 4705-4713
Publish at: 2025-10-01

On design of a small-sized arrays for direction-of-arrival-estimation taking into account antenna gains

10.11591/ijece.v15i5.pp4642-4652
Ilia Peshkov , Natalia Fortunova , Irina Zaitseva
In the paper a technique for designing antenna arrays composed of directional elements for direction-of-arrival (DOA) estimation is proposed. Especially this approach is applied for developing hybrid antenna arrays with increased accuracy which features digital spatial spectral estimation after preliminary analog beamforming. The earlier obtained explicit formula for calculating the Cramér–Rao lower bound (CRLB) which determines the relationship between the variance of the DOA-estimation and antenna elements' radiation patterns, array geometry, has been used. Main idea of the proposed technique is that it takes into account spatial pattern and gain of the antenna elements. The high gain unlike the number of the antenna elements or interelement distance is the most important factor which allows reducing the value of the DOA-estimation errors. A couple of the examples of calculating radiation patterns of antenna elements improving accuracy of DOA-estimation with super-resolution are provided in the paper. Proposed antenna arrays are modeled according to the method of moments (MoM). The values of the root mean square error after the DOA-estimation are obtained. It is shown that the resulting hybrid systems can reduce the error value in DOA-estimation with super-resolution.
Volume: 15
Issue: 5
Page: 4642-4652
Publish at: 2025-10-01

Bone-Net: a parallel deep convolutional neural network-based bone fracture recognition

10.11591/ijece.v15i5.pp4692-4704
Md. Hasan Imam Bijoy , Nusrat Islam Kohinoor , Syeda Zarin Tasnim , Md Saidur Rahman Kohinoor
Many people suffer from bone fractures, which can result from minor accidents, forceful blows, or even diseases like osteoporosis or bone cancer. In the medical realm, accurately identifying bone fractures from X-ray images is paramount for effective diagnosis and treatment. To address this, a comparative study is conducted utilizing three distinct models: a traditional convolutional neural network (CNN), MobileNet-V2, and a newly developed parallel deep convolutional neural network (PDCNN). The primary aim is to evaluate and contrast these models in terms of precision, sensitivity, and specificity for diagnosing bone fractures. X-ray images of fractured and non-fractured bones are sourced from Kaggle and subjected to various image processing techniques to rectify anomalies. Techniques such as cropping, resizing, contrast enhancement, filtering, and augmentation are applied, culminating in canny edge detection. These processed images are then used to train and test models. The results showcased the superior performance of the newly developed PDCNN model, achieving an impressive accuracy of 92.89%, surpassing both the traditional CNN and pretrained MobileNet-V2 models. A series of ablation studies are conducted to fine-tune the hyperparameters of the PDCNN model, further validating its efficacy. Throughout the investigation, PDCNN consistently outperformed MobileNet-V2 and traditional CNN, underscoring its potential as an advanced tool for streamlining bone fracture identification.
Volume: 15
Issue: 5
Page: 4692-4704
Publish at: 2025-10-01

A hybrid approach to phishing email detection: leveraging machine learning and explainable artificial intelligence

10.11591/ijece.v15i5.pp4865-4874
Tarek Zidan , Fadi Abu-Amara , Ahmad Hasasneh , Muath Sawaftah , Seth Griner
With the increasing use of emails in our daily lives, they have become a prime target of phishing attacks, posing a significant threat to users. Attackers pretend to be trusted sources and use email phishing attacks to trick people into clicking malicious links or opening attachments. The aim of these attacks is to obtain sensitive information, such as financial information, login credentials, and personally identifiable information. Emails have attributes including the URL, sender, subject, receiver(s), and body. This paper proposes a hybrid intelligence model that integrates machine learning algorithms (ML) and natural language processing (NLP) techniques for email phishing detection. Three ML algorithms are employed: logistic regression, decision tree, and random forest. In addition, a customized ChatGPT model has been developed to receive email classification results from the hybrid model. This model educates users on recognizing phishing emails by explaining email classifications, highlighting keywords, and offering security tips. The proposed approach to detecting phishing emails raises awareness and educates users on recognizing and reporting email phishing attacks.
Volume: 15
Issue: 5
Page: 4865-4874
Publish at: 2025-10-01

Anomaly-based intrusion detection leveraging optimized firewall log analysis: a real-time machine learning solution

10.11591/ijece.v15i5.pp4785-4802
Tran Cong Hung , Dam Minh Linh , Han Minh Chau , Ngo Xuan Thoai , Thai Duc Phuong , Huynh De Thu
Firewall logs play a vital role in cybersecurity by recording network traffic and flagging potential threats. This study evaluates five machine learning algorithms-decision tree (DT), random forest (RF), extra trees (ET), CatBoost (CB), and AdaBoost (AB)-on a dataset of 65,532 firewall log entries. Models were assessed using accuracy, precision, recall, training/prediction time, and Pearson correlation for feature selection, across multiple train-test splits. The DT model achieved the best performance, reaching 99.45% test accuracy, 97.457% precision, and 93.389% recall at a 7:3 split, along with the fastest training time (0.20642s). We propose real-time flow-level intrusion detection (RT-FLID), novel, lightweight, real-time intrusion detection system that leverages multithreaded processing and flow-level analysis to boost detection speed and scalability. Unlike existing approaches that rely heavily on deep packet inspection or computationally intensive processing, RT-FLID requires minimal resources while maintaining high detection accuracy. The architecture efficiently handles large traffic volumes and dynamically identifies anomalies such as distributed denial-of-service (DDoS) and port scans. Validated on real-world logs, the system maintained high accuracy in critical classes like “deny” and “reset-both.” These findings highlight RT-FLID’s novelty and practical advantages, demonstrating its potential for deployment in high-throughput, low-latency network environments.
Volume: 15
Issue: 5
Page: 4785-4802
Publish at: 2025-10-01

Decomposition and multi-scale analysis of surface electromyographic signal for finger movements

10.11591/ijece.v15i5.pp4593-4604
Afroza Sultana , Md. Tawhid Islam Opu , Md. Shafiul Alam , Farruk Ahmed
Decomposition of the surface electromyography (sEMG) signal is vital for separating the composite, complex, noisy signals recorded from muscles into their integral motor unit action potentials (MUAPs). By precisely identifying each motor unit’s activity, this method offers greater insights into the functioning of the neuromuscular system, which helps isolate each motor unit's contribution, making it essential for understanding muscle coordination and diagnosing neuromuscular disorders. In this study, we employ the maximal overlapping discrete wavelet transform (MODWT), which is well-suited for analyzing signals in the time-frequency domain. The study decomposed the sEMG signal into six levels to identify the neural activity of finger movements and analyzed the motor unit action potential (MUAP). In the frequency range of 30.2 and 64.6 Hz, the signal exhibits the highest MUAP which is independent of movement. Using inverse MODWT, it was rebuilt from the decomposed levels. With 95.8% accuracy, the similarity between the reassembled signal and the original signal was determined using correlation analysis to assess the efficacy of the method.
Volume: 15
Issue: 5
Page: 4593-4604
Publish at: 2025-10-01

Optimizing vehicle selection in supply chain management with data-driven strategies

10.11591/ijece.v15i5.pp4899-4906
Imane Zeroual , Jaber El Bouhdidi
Logistics has undergone significant transformation to address the complex economic, social, and environmental challenges of the modern era. To maintain competitiveness, logistics providers have been compelled to optimize operations, meet increasing customer expectations, and improve satisfaction. Critical issues impacting logistics performance include traffic congestion, infrastructure limitations, rising demand, and the complexities of vehicle scheduling, coordination, and management. These challenges frequently disrupt delivery operations, undermining efficiency and overall system performance. This paper proposes the application of three machine learning models aimed at optimizing delivery processes, with a focus on improving vehicle assignment for order deliveries. By leveraging these models, logistics providers can enhance decision-making and operational efficiency. The study defines the core problem and evaluates several machine learning approaches to bolster logistics delivery systems.
Volume: 15
Issue: 5
Page: 4899-4906
Publish at: 2025-10-01
1 2 3
...
2027 2028
Show 149 of 2028

Discover Our Library

Embark on a journey through our expansive collection of articles and let curiosity lead your path to innovation.

Explore Now
Library 3D Ilustration