|
||||||||
|
||||||||
|
||||||||
Measurement of Physical Quantities |
||||||||
|
|
|||||||
|
||||||||
|
||||||||
Pages: |
|
|||||||
1-9 |
Neven Kanchev, Nikolay Stoyanov, Georgi Milushev: Prediction of the Natural Gas Compressibility Factor by using MLP and RBF Artificial Neural Networks Abstract: The compressibility factor indicates the deviation of the real natural gas from the ideal behavior. It is one of the most important parameters in the natural gas industry. In the present study, two different types of neural networks – multi-layer perceptron (MLP) and radial basis functions (RBF) – were used to predict the compressibility factor Z of natural gas. The pressure, temperature, and speed of sound (SoS) were chosen as input parameters for the artificial neural network (ANN) models. The training and testing of the MLP-ANN and RBF-ANN were carried out on the basis of 151 days of continuous measurements. Different variants of both types of neural networks were implemented and a comparative analysis of their modeling capabilities was performed. The models developed show a very high prediction accuracy, with the results obtained showing a certain advantage of the RBF-ANN. The comparative analysis was performed on the basis of standard performance indicators such as R2, root mean square error (RMSE), mean square error (MSE), mean absolute error (MAE). The present study shows an intelligent method implemented in two different variants to determine the compressibility factor of natural gas without the need to use the equation of state.
|
|||||||
10-14 |
Mohamed Yacin Sikkandar, S. Sabarunisha Begum, Ahmad Alassaf, Ibrahim AlMohimeed, Khalid Alhussaini, Adham Aleid, Abdulrahman Khalid Alhaidar: Abstract: This study introduces an innovative method for minimizing artifacts in electroencephalography (EEG) signals by integrating brainstorm optimization (BSO) with a variational autoencoder generative adversarial network (VAE-GAN), resulting in the BrOpt_VAGAN model. EEG signals are critical for the diagnosis of neurological disorders, for brain-computer interface (BCI) applications, and for the monitoring of neurological disabilities. However, EEG data often contains artifacts from physiological sources — such as electro- oculographic (EOG), electromyographic (EMG), and electrocardiographic (ECG) signals — which can distort the accuracy of brain activity readings. Our proposed BrOpt_VAGAN model combines BSO with a VAE-GAN framework to more effectively remove these artifacts, thus improving the clarity and accuracy of EEG signals. In this model, the VAE first reduces the raw EEG signals into a lower-dimensional representation that captures the essential signal patterns while filtering out the noise. The GAN component then refines this representation via adversarial training, effectively minimizing artifacts and improving the quality of the processed EEG data. BSO optimally adjusts the encoding and decoding parameters within the VAE-GAN structure, enabling the model to handle different noise levels and helps to find different neurological disorders. Preliminary results show that BrOpt_VAGAN performs significantly better with an accuracy of 98.5 % and an error rate of 11.23 %, enabling a clearer and more precise EEG signal reconstruction. |
|||||||
15-21 |
Hsiung-Cheng Lin, Yan-Hao Peng: Elevator Operation Health Diagnosis using Vibration Region Segmentation Algorithm via Internet Abstract: The safety of elevator operation is an indispensable issue in the elevator industry. The most important factor affecting the elevator performance is car vibration. For this reason, vibration analysis is considered an important topic for elevator maintenance as it can be used to detect potential problems before breakdown. Currently, vibration measurement is typically performed using vibration analyzers operated by personnel, resulting in a time-consuming process and experience-dependent interpretation. While there are some machine learning algorithms that are used to diagnose elevator condition, their computational complexity still makes it difficult to translate into a real-world application. Therefore, in this study, the elevator condition diagnosis model is developed using a simple vibration region segmentation method. Based on the elevator operation characteristics, the cut-off point is determined by the abrupt acceleration variation condition to define the acceleration segment region. With the binarization process, the digital array is used at each time of acceleration variation to evaluate the state of elevator operation. Normally, the elevator operation can be divided into five segments, e.g., start up, acceleration, steady state, deceleration, and stop. The rule for determining the critical point for segmentation is thus formulated based on an abrupt acceleration change. If the number of segmented areas exceeds five, it can be considered as an abnormal case. To develop the system, the elevator vibration data is collected by a 3D accelerometer and then processed in the PC using the proposed algorithm. The results are then transferred to the cloud for online monitoring. The experimental results show that the proposed model is quite simple but effective for elevator diagnosis and maintenance. |
|||||||
22-29 |
Pavle Stepanić, Nedeljko Dučić, Jelena Vidaković, Jelena Baralić, Marko Popović: Implementation and Evaluation of Machine Learning Algorithms in Ball Bearing Fault Detection Abstract: The subject of this research is the development of a classifier based on machine learning (ML) that is able to recognize defective and healthy ball bearings. For this purpose, vibration measurements were performed on the bearings, on a total of 196 samples. For each recorded vibration signal, a feature extraction was performed by digital processing in the time domain. The following ML algorithms were used to develop the classifier: K-nearest neighbor (KNN) and support vector machine (SVM) as well as improved versions of the aforementioned algorithms. Improved versions of the mentioned algorithms were obtained by optimizing their hyperparameters. The corresponding models of the KNN and SVM algorithms showed a high percentage of success in classification, 98.5 % and 99.5 %, respectively. By optimizing the hyperparameters, models with a maximum classification success of 100 % were achieved.
|
|||||||
30–39 |
G. Puvaneswari: Abstract: This work proposes an optimized support vector model and a variable ranking-based test node selection approach for identifying parametric faults in analog circuits using a fault dictionary. Test node selection is essential for fault dictionary-based fault detection to reduce the dimensionality and test process complexity. To determine an appropriate set of test nodes, a feature selection technique based on variable ranking is used, as it is computationally efficient and involves sorting and score estimation. In the proposed method, test nodes are ranked using a score function based on data variability, where the nodes with the highest data variability are assigned the highest rank. This ranking ensures that the most informative test nodes are prioritized for fault detection. An optimized support vector model is used for fault diagnosis to improve classification accuracy. The results show the effectiveness of this approach. The performance of the proposed method is validated by measuring the fault detection accuracy on benchmark circuits.
|
|||||||
40–47 |
Juan Wang, Wei Liu, Yong Zhang, Zhi Liu, Xiaolei Zheng, Yuxin Wang, Jianshu Hao, Xuanding Dai: Abstract: In the context of the increasing spread of electric vehicle (EV) charging stations, the accuracy and reliability of electric energy measurement is becoming increasingly important for consumers. Degradation in the performance of smart meters at these stations is often due to factors such as aging and malfunctions. Traditional approaches to solving this problem usually involve manual on-site inspections, which require significant investment in manpower and materials. To overcome this challenge, this study proposes an error estimation method that integrates highway convolutional neural networks with bidirectional long short-term memory (LSTM) networks, which enables realtime prediction of measurement performance at charging piles. First, the convolutional module is combined with the highway network to extract spatial features from smart meter data for charging facilities while retaining some original information to improve model prediction performance. The features are then fed into a bidirectional LSTM network to obtain temporal characteristics, which improves the accuracy of relative error predictions. Empirical validation of this method at a charging station in the region has shown that it has higher efficiency compared to existing advanced models.
|
|||||||
|
Top ↑ |
|||||||
|
|
|||||||
|
||||||||
|
||||||||
48–52 |
A. Velliangiri, Madhavi Damle, Peter Soosai Anandaraj, Jampani Satish Babu: Dynamic Trust-based Access Control with Hybrid Encryption for Secure IoT Applications Abstract: The rapid growth of internet of things (IoT) applications, especially in wireless sensor networks (WSNs), has led to the generation of large amounts of real-time data from interconnected devices. This growth leads to challenges in securing data access and managing resources efficiently. To address these challenges, we propose a dynamic trust-based access control (DTAC) model for IoT and WSN applications. The DTAC model integrates behavioral trust evaluation and context-aware decision making to dynamically adapt access permissions to network conditions in real-time. the trust scores are calculated using fuzzy logic and machine learning techniques, which enable adaptive decision-making. To increase security, the model uses a hybrid encryption scheme that combines elliptic curve cryptography (ECC) with lightweight symmetric encryption, ensuring data confidentiality with minimal computational overhead. In addition, access control decisions are refined by contextual factors such as user roles, device locations, and data sensitivity. The model includes a collaborative re-evaluation mechanism that periodically updates trust scores and isolates malicious nodes without compromising network stability. The DTAC model is evaluated on key metrics such as security resilience, energy efficiency, and latency and demonstrates better performance than existing solutions. This model provides a scalable, energy-efficient, and secure framework for IoT and WSN applications that ensures reliable data access and privacy in diverse environments.
|
|||||||
53–66 |
Aleksandar Djuric, Sreten Peric, Momir Drakulic, Aleksandar Bukvic, Snezana Jovanovic: Abstract: The aim of this paper is to contribute to the possibility of using vibration diagnostics as a modern, non-destructive tool for assessing the condition of gears in the gearbox of off-road vehicles when the vehicles are used in real operating conditions, despite the non-stationary change in load during work. The method of vibration analysis in a vehicle gearbox under real operating conditions was determined and the adaptability of vibration analysis tools in determining the conditions of gears in off-road vehicles was considered. The data collected in the research shows that the vibration diagnostic method as a non-destructive tool for determining the condition of gears in off-road vehicle gearboxes under real operating conditions provides data on the condition of gear elements at different stages of vehicle operation. In this way, research time is shortened, there is no need to develop special test-tables for gearbox tests, the research method is non-destructive and it can identify processes that are characteristic of both early and later operating stages of the vehicle. Finally, the decision as to whether certain forms of gearbox maintenance are required can be made based on the assessed condition of the gears in the gearbox.
|
|||||||
64–71 |
Haoran Ma, Bei Peng, Guochuan Zhao, Shuang Wang, Yun Rong, Yibo Li: Grain Truck Compartment Localization Method based on Point Cloud Projection Abstract: Quality control is an essential step before grain storage. It requires the localization of grain truck compartments and guiding robotic arms to automatically sample grains. However, the diverse types of grain trucks and the variability in parking lead to difficulties in compartment localization and inaccurate measurements. To solve this problem, a rotating 3D laser scanner is proposed to scan grain trucks. After ground calibration, the XOY plane of the rotating scanned point cloud is aligned parallel to the ground. To avoid complex point cloud segmentation, grain truck point clouds are clipped using pre-defined regions of interest (ROI). Since only 2D corner points are required, this paper presents a projection-based point cloud processing method. Here, the points of the grain truck are projected onto the XOY plane and then the points of the rear and side panels of the projected compartment are extracted for line fitting. To robustly extract compartment corners, a region growing method based on density variations is proposed. Along the fitted line, the 2D corners of the rear and side panels are extracted to obtain the length and width dimensions of the compartment. Extensive tests have shown that the proposed method can accommodate various grain truck models with a corner extraction accuracy of less than 9.8 cm, making it suitable for the automated grain truck localization and measurement tasks.
|
|||||||
72–82 |
Yingkuang Zhu, Zhenhua Pan, Huajun Li, Jianyang Chen, Yihao Sheng: Abstract: Optical tomography, a critical component of process tomography, is an important tool for determining the absorption coefficient of cross-sectional media, with significant applications in multi-phase flow analysis, chemical processing, and combustion studies. However, the precise reconstruction of these media distributions is a major challenge. In this work, a sophisticated optical tomography (OT) system coupled with an innovative reconstruction algorithm is presented. The system architecture includes 25 light sources and 25 strategically placed fan-beam receivers. We present a convolutional neural network (CNN) with an encode-decode configuration, augmented by residual connections and a squeeze-and-excitation (SE) attention mechanism. Initial evaluations performed using MATLAB simulations showed the algorithm's superior performance compared to existing methods, with notable improvements in relative error (RE) and correlation coefficient metrics. Subsequent practical experiments validated these findings and emphasized the efficiency of the residual and SE components in improving reconstruction accuracy. While this study focuses on high-contrast binary scenarios, the proposed RESE-CNN framework provides a basic architecture for future extensions to weakly absorbing and scattering media where nonlinear reconstruction problems dominate.
|
|||||||
83–92 |
Hoang Vu Viet, Lubomír Kremnický, Martin Bereta, Michal Teplan: Experimental Platform for Investigation of Low-Frequency Magnetic Field Effects on Cells Abstract: This study presents a novel experimental platform designed to systematically investigate the effects of low-frequency (LF) magnetic fields (MFs) on biological systems. By overcoming key methodological challenges, including variability in environmental conditions and poor reproducibility, this platform sets a new standard for experimental reliability. It integrates precise temperature regulation, high MF homogeneity, and a modular structure, that ensures adaptability to diverse experimental setups. The platform was validated using Saccharomyces cerevisiae CCY 21-4-99 as a model organism, cultivated under controlled conditions with and without MF exposure. The minimal growth variations observed between chambers confirm the ability of the platform to maintain reproducible conditions and support statistically robust experimental designs. The platform can be applied to diverse biological systems and can be technically adapted to different experimental requirements, including different electromagnetic field sources. It provides a highly controlled environment for investigating the cellular and subcellular effects of LF MF, creating a solid foundation for future research into its biological mechanisms and potential applications.
|
|||||||
93–99 |
Rajalakshmi K, Palanivel Rajan S: An Enhanced Measurement of Epicardial Fat Segmentation and Severity Classification using Modified U-Net and FOA-guided XGBoost Abstract: The amount of epicardial fat around the heart has a significant impact on cardiovascular function and requires precise measurement for timely treatment. In this work, an improved U-Net architecture is proposed for accurate segmentation of epicardial fat in computer tomography (CT) images. The proposed method integrates a modified squeeze-and-excitation (MSE) block and a multi-scale dense (MS-D) convolutional neural network (CNN) to improve feature extraction. In addition, a metaheuristic optimization algorithm from falcon optimization algorithm (FOA) is used for efficient feature selection. The selected features are then classified using the XGBoost algorithm to determine the fat severity. Experimental evaluations on a CT image dataset show the superior segmentation performance of the proposed U-Net compared to existing architectures. It achieves a mean intersection over union (Mean IOU) of 89.5 %, a mean Dice score (MDS) of 94.3 %, and a Pearson correlation coefficient (PCC) of 0.973. FOA-guided feature selection further increases the accuracy of severity classification. The overall classification accuracy of the model is 98 %. These results highlight the technological advancements and measurement accuracy of the proposed U-Net architecture. They also demonstrate the suitability of the model to improve cardiovascular risk assessment and management.
|
|||||||
|
|
|||||||
|
Top ↑ |
|||||||
|
|
|||||||
|
||||||||
|
||||||||
|
||||||||
|