Showing 110 results for Ica
Ayoub Hamidi, Ahmad Cheldavi, Asghar Habibnejad Korayem,
Volume 20, Issue 3 (9-2024)
Abstract
This paper proposes a structure for concrete composite materials that effectively attenuates transmitted power through the composite slab across a wide frequency range. The proposed structure is practical for electromagnetic interference shielding applications. To assess its effectiveness, the proposed structure has been compared with two other structures: a traditional wire mesh used in reinforced composites and an array of helices, a cutting-edge technique for manufacturing lightweight concretes with significant improvements in shielding properties. The comparison among full-wave simulation results indicates that the proposed method leverages the benefits of both techniques. It achieves a shielding effectiveness exceeding 30 dB from low frequencies up to 8.5 GHz and beyond 55 dB from low frequencies up to 4 GHz. Furthermore, an experimental measurement was conducted to validate the full-wave simulation results. An experimental sample was fabricated according to the simulated proposed structure, and the measured shielding effectiveness confirmed the composite's capability in wideband electromagnetic shielding. Theoretically, the proposed structure can enhance the concrete's mechanical characteristics while improving its shielding effectiveness, making it suitable for designing ultra-high-performance concretes.
Dalila Yessad,
Volume 20, Issue 4 (11-2024)
Abstract
This paper introduces the CTDRCepstrum, a novel feature extraction technique designed to differentiate various human activities using Doppler radar classification. Real data were collected from a Doppler radar system, capturing nine return echoes while monitoring three distinct human activities: walking, fast walking, and running. These activities were performed by three subjects, either individually or in pairs. We focus on analyzing the Doppler signatures using time-frequency reassignment, emphasizing its advantages such as improved component separability. The proposed CTDRCepstrum explores different window functions, transforming each echo signal into three forms of Short-Time Fourier Transform reassignments (RSTFT): time RSTFT (TSTFT), time derivative RSTFT (TDSTFT), and reassigned STFT (RSTFT). A convolutional neural network (CNN) model was then trained using the feature vector, which is generated by combining the cepstral analysis results of each RSTFT form. Experimental results demonstrate the effectiveness of the proposed method, achieving a remarkable classification accuracy of 99.83% by using the Bartlett-Hanning window to extract key features from real-time Doppler radar data of moving targets.
Arizadayana Zahalan, Samila Mat Zali, Ernie Che Mid, Noor Fazliana Fadzail,
Volume 21, Issue 2 (6-2025)
Abstract
Photovoltaic (PV) systems are vital in the global renewable energy landscape because of their capability to harness solar energy efficiently. Ensuring the continuous and efficient operation of PV systems is crucial in maximizing their energy contribution. However, these systems' reliability and safety remain critical because they are prone to various faults, mainly when operating in harsh environmental conditions. This study addresses these issues by exploring fault detection and classification in PV arrays using neural network (NN) -based techniques. A PV array model, consisting of 3x6 PV modules, was simulated using MATLAB Simulink to replicate real-world conditions and analyse various fault scenarios. An open circuit, a short circuit, and a degrading fault are the three types of faults considered in this study. The NN was trained on a dataset generated from the MATLAB Simulink model, encompassing normal operating and fault conditions. This training enables the network to learn the distinctive patterns associated with each fault type, enhancing its detection accuracy and classification capabilities. Simulation results demonstrate that the NN-based approach effectively identifies and classifies the three types of faults.
Hanim Suraya Mohd Mokhtar, Aimi Salihah Abdul Nasir, Mohammad Faridun Naim Tajuddin, Muhammad Hafeez Abdul Nasir, Kumuthawathe Ananda Rao,
Volume 21, Issue 2 (6-2025)
Abstract
The rapid growth of photovoltaic (PV) systems has highlighted the need for efficient and reliable defect detection to maintain system performance. Electroluminescence (EL) imaging has emerged as a promising technique for identifying defects in PV cells; however, challenges remain in accurately classifying defects due to the variability in image quality and the complex nature of the defects. Existing studies often focus on single image enhancement techniques or fail to comprehensively compare the performance of various image enhancement methods across different deep learning (DL) models. This research addresses these gaps by proposing an in-depth analysis of the impact of multiple image enhancement techniques on defect detection performance, using various deep learning models of low, medium, and high complexity. The results demonstrate that mid-complexity models, especially DarkNet-53, achieve the highest performance with an accuracy of 94.55% after MSR2 enhancement. DarkNet-53 consistently outperformed both lower-complexity models and higher-complexity models in terms of accuracy, precision, and F1-score. The findings highlight that medium-depth models, enhanced with MSR2, offer the most reliable results for photovoltaic defect detection, demonstrating a significant improvement over other models in terms of accuracy and efficiency. This research provides valuable insights for optimizing defect detection systems in photovoltaic applications, emphasizing the importance of both model complexity and image enhancement techniques for robust performance.
Edy Victor Haryanto S, Aimi Salihah Abdul Nasir, Mohd Yusoff Mashor, Bob Subhan Riza, Zeehaida Mohamed,
Volume 21, Issue 2 (6-2025)
Abstract
Malaria is a parasitic disease that causes significant morbidity and mortality worldwide. Early diagnosis and treatment are crucial for preventing complications and improving patient outcomes. Microscopic examination of blood smears remains the gold standard for malaria diagnosis, but it is time-consuming and requires skilled technicians. Deep learning has emerged as a promising tool for automated image analysis, including malaria diagnosis. In this study, we propose a novel approach for identifying malaria parasites in microscopic images using the GoogLeNet. Our method includes enhancement with the AGCS method, color transformation with grayscale, adaptive thresholding for segmentation, extraction, and GoogLeNet-based classification. We evaluated our method on a dataset of malaria blood smear images and achieved an accuracy of 95%, demonstrating the potential of GoogLeNet for automated malaria diagnosis.
Gholamreza Khademevatan, Ali Jalali,
Volume 21, Issue 3 (8-2025)
Abstract
A novel simplified EKV model base analog/RF CMOS design pre-SPICE tool is presented in this paper. Addition to facilitating the sizing process, this CAD tool can also optimize circuit characteristics. By having a web address, users can access it without installing any software. Using a graphical and a numerical view, the designer can select degrees of freedom and observe the MOS circuit performance. Through the use of charts versus IC, the graphical view can show tradeoffs in circuit performance in real-time. Charts can be displayed simultaneously in both linear and logarithmic scales. IC CRIT , is also available and can be displayed on the charts. This tool is not limited to one process and it is possible to select different processes. It is efficient for pre-SPICE designs, enhancing intuitive understanding and the designer's experience for future projects while eliminating the need for trial-and-error simulations. Furthermore, the predicted results align well with simulation outcomes, demonstrating the effectiveness of the design and optimization method presented. Two methodologies for selecting optimum ICs are presented by this tool. These are illustrated by the study of linearity indices, AIP3 and IIP3, in one-stage and two-stage differential amplifiers and the design of a single-ended OTA.
Somayeh Talebzadeh, Reza Radfar, Abbas Toloei Ashlaghi,
Volume 21, Issue 3 (8-2025)
Abstract
The accurate prediction of electricity demand is crucial for efficient energy management and grid operation. However, the complexities of demand patterns, weather variability, and socioeconomic factors make it challenging to forecast demand with high accuracy. To address this challenge, this research proposes a novel hybrid machine-learning approach for predicting electricity demand. In this research, first, different regression methods were investigated to solve the problem, the results showed that the multi-layer perceptron (MLP) regression model has the best performance in predicting electricity demand. Furthermore, the proposed system, BIMLP (Bagging-Improved MLP), is designed to iteratively improve its parameters using a binary search algorithm and reduce the learning error using bagging, a technique for ensemble learning. The proposed system was applied to the Electric Power Consumption data set and achieved a value of 0.9734 in the r2 criterion. The results of implementing and evaluating the proposed system demonstrate its satisfactory performance compared to existing techniques.
M.e. S.m Mehzabeen , Phd R Gayathri, Pattunnarajam Paramasaivam , Phd Ramya A,
Volume 21, Issue 4 (11-2025)
Abstract
Hepatitis C virus (HCV) detection is a critical aspect of early intervention and effective management of the disease. This paper presents a comprehensive study focused on enhancing the detection accuracy of HCV through the integration of advanced techniques - SMOTE, Optuna, and SHAP - alongside extensive exploratory data analysis (EDA). The study addresses class imbalance using Synthetic Minority Over-sampling Technique (SMOTE), optimizes model performance with Optuna for hyperparameter tuning, and provides model interpretability using SHAP (SHapley Additive exPlanations). EDA is leveraged to gain valuable insights into the dataset's characteristics, ensuring robust data preprocessing and feature engineering. The results show 97% improved HCV detection performance, highlighting the efficacy of the proposed methodology in medical diagnostics and aiding healthcare professionals in making informed clinical decisions.
Tara Sistani, Seyed Javad Kazemitabar,
Volume 21, Issue 4 (11-2025)
Abstract
Forests play several vital roles in our lives and provide various resources. However, in recent years, the increasing frequency of wildfires has led to the widespread burning and destruction of many forests and wildlands. Therefore, detecting forest fires and finding suitable solutions to address this issue has become one of the critical challenges for researchers. Today, with the advancement of artificial intelligence, forest fire detection using deep learning is an important method with the aim of increasing the efficiency of forest fire detection and monitoring systems. In this article, a method based on a type of convolutional neural network called Xception is proposed for classifying forest fire images. In this method, transfer learning technique is used on the proposed neural network and a new classifier is designed for the problem. Also, various hyperparameters have been used to optimize the performance of the proposed model. The proposed method is performed on the DeepFire dataset, which contains 1900 images equally divided between fire and no-fire classes. The results obtained from the implementation of the proposed method show that this method with an accuracy of 99.47% has achieved a favorable performance in classifying forest fire images.
Mohammad Reza Eesazadeh, Zahra Nasiri-Gheidari,
Volume 21, Issue 4 (11-2025)
Abstract
This research focuses on electromagnetic position sensors, particularly synchros, which play a crucial role in the closed-loop control systems of permanent magnet synchronous machines (PMSMs). Compared to two-phase resolvers, three-phase synchros provide enhanced reliability by ensuring continued operation even in the event of an open-circuit fault. One of the key challenges in designing such sensors lies in selecting optimal windings and configurations while also developing efficient modeling techniques to minimize computational complexity. To address this issue, the study introduces a matrix-based method for designing wound rotor (WR) synchros. This approach allows for flexible configurations depending on the number of pole pairs and stator tooth counts. The proposed design methodology ensures adaptability and precision, making it a valuable tool for engineers working on electromagnetic sensor development. To validate the effectiveness of the proposed method, the Field Reconstruction Method (FRM) is employed, providing a fast and accurate modeling technique that can be implemented using MATLAB. Additionally, a comparative analysis is conducted with finite element analysis (FEA) to confirm the accuracy and reliability of the approach. Results demonstrate that the matrix-based method is an efficient and effective solution for optimizing WR synchro designs, significantly improving performance and computational efficiency.
