Volume 19, number 1, 2025. Print version and published papers:
ORIENTATION OF PHOTOVOLTAIC PANELS USING A PYRAMIDAL SENSOR
Copy citation: P. Nistor, I. Orha, Orientation of Photovoltaic Panels Using a Pyramidal Sensor, Carpathian Journal of Electrical Engineering, vol. 19, no. 1, pp. 7-20, 2025, https://doi.org/10.34302/CJEE/DACO1267.
Paul NISTOR, Ioan ORHA, Technical University of Cluj-Napoca, Romania, paul.nistor@ieec.utcluj.ro, ioan.orha@ieec.utcluj.ro
Keywords: sensors, solar panels, renewable energy sources, solar power generation
Abstract: The work presents the realization of a solar cell orientation system using a pyramidal sensor. At the base of this system is the pyramidal sensor that captures the luminous flux and transmits it in the form of information to the block responsible for signal processing. The main component within this block is the operational amplifier AO 741, which amplifies the signal before reaching the motor drive block. Within this block we will meet an H-bridge that has the role of changing the direction of the current in the motors, thus resulting in the change of the direction of rotation. In this way, going through all the blocks described, the signal resulting from the pyramid sensor leads to the finality of the movement, namely, the top of the pyramid is oriented towards the sun. When the light flux will fall from a different angle on the sensor, the process will start again, reorienting the pyramidal sensor.
SELECTING SWITCHES FOR SWITCHING INDUCTIVE LOADS: TECHNICAL ASPECTS AND PRACTICAL CONSIDERATIONS
Copy citation: C. M. Ruano González, Selecting Switches for Switching Inductive Loads: Technical Aspects and Practical Considerations, Carpathian Journal of Electrical Engineering, vol. 19, no. 1, pp. 21-29, 2025, https://doi.org/10.34302/CJEE/ECCK7473.
Carlos Manuel RUANO GONZÁLEZ, Department of Electrical Engineering, Electromechanical Faculty, University of Camaguey, Cuba, cmruanog@gmail.com
Keywords: Switches of half tension, inductive loads, surges, electric arches
Abstract: In power systems, reactive compensation is a common and necessary practice to maintain the quality of electrical service. Disconnection of reactors or reactor banks is frequent, sometimes up to two or three times a day. Therefore, circuit breakers for this application must operate satisfactorily under the energization and de-energization processes of “inductive currents” that can cause the chopping interruption phenomenon, as well as in cases of magnetizing currents from unloaded transformers and load currents from induction motors. The proper selection of circuit breakers to operate inductive loads in medium-voltage systems is a critical aspect in the design and operation of electrical networks. Inductive loads, such as motors, transformers, and reactors, present unique challenges due to the voltage and current transients associated with their connection and disconnection. This article addresses the essential technical criteria for breaker selection, including interrupting capacity, surge resistance, and arc flash management. In addition, relevant international regulations are discussed, and case studies illustrating industry’s best practices are presented. The aim is to provide comprehensive guidance for engineers and designers seeking to optimize the reliability and safety of their electrical systems.
REGRESSION-BASED PREDICTION OF ANXIETY SEVERITY
Copy citation: B. Chiș, D. Dulf, M. Morar, E. Pop, Regression-Based Prediction of Anxiety Severity, Carpathian Journal of Electrical Engineering, vol. 19, no. 1, pp. 30-39, 2025, https://doi.org/10.34302/CJEE/BUEP2247.
Bogdan CHIȘ, Diana DULF, Mădălina MORAR, Eleonora POP, Technical University of Cluj-Napoca, chis.do.paul@student.utcluj.ro, dulf.da.diana@student.utcluj.ro, jula.an.madalina@student.utcluj.ro, eleonora.pop@ieec.utcluj.ro
Keywords: Mental health, Anxiety disorders, Machine learning, Regression analysis
Abstract: The present study explores the use of machine learning to predict self-reported anxiety levels based on demographic, behavioral, and physiological data. To this end, we used a dataset comprising 11,000 survey responses and applied multiple regression models, including Linear Regression, Gradient Boosting Regression, Extreme Gradient Boosting Regression Light Gradient-Boosting Machine Regression after data preprocessing. The performance of the models was evaluated using the mean absolute error, the root mean square error, and the coefficient of determination. Among the models evaluated, Gradient Boosting Regression achieved the best results, with a mean cross-validated R² of 0.759 after five-fold cross-validation.
SIMULATION OF AN IOT AIR QUALITY MONITORING SYSTEM
Copy citation: B. Chiș, C. Costea, E. Pop, Simulation of an IOT Air Quality Monitoring System, Carpathian Journal of Electrical Engineering, vol. 19, no. 1, pp. 40-47, 2025, https://doi.org/10.34302/CJEE/LDGV2056.
Bogdan CHIȘ, Cristinel COSTEA, Eleonora POP, Technical University of Cluj-Napoca, chis.do.paul@student.utcluj.ro, cristinel.costea@ieec.utcluj.ro, eleonora.pop@ieec.utcluj.ro
Keywords: Air quality, Air pollution, Anomaly detection, Internet of Things
Abstract: Air pollution remains a major global health and environmental concern, driving the need for efficient and scalable monitoring systems. This paper presents a simulation-based Internet of Things architecture for air quality monitoring, focusing on real-time data transmission, anomaly detection, and visual analysis. The system simulates 25 monitoring stations, each with six virtual sensors representing key pollutants, totaling 150 sensor processes. These sensors communicate with dedicated gateway processes via MQTT and transmit data to a cloud-hosted web server for aggregation, air quality index calculation, visualization, and database storage. The simulation uses real-world data collected in Seoul to emulate realistic pollution conditions. An interactive web interface displays live air quality index values through maps and time series charts, allowing the detection of anomalies.
A DATA-DRIVEN ANALYSIS OF REMOTE WORK SALARIES AND SATISFACTION
Copy citation: I. A. Vlad, D. Tepfenhart, A Data-Driven Analysis of Remote Work Salaries and Satisfaction, Carpathian Journal of Electrical Engineering, vol. 19, no. 1, pp. 48-55, 2025, https://doi.org/10.34302/CJEE/MEIZ3377.
Ioan Alexandru VLAD, Dacian TEPFENHART, Technical University of Cluj-Napoca, Romania, al3xro@gmail.com, dacianXpaul@gmail.com
Keywords: remote work, salary analysis, data science, job satisfaction
Abstract: This paper provides a data-driven analysis of remote work salaries, leveraging a real-world dataset processed with Python. The study investigates how industry, experience, employment type, and remote flexibility impact salary and job satisfaction. Insights are visualized using statistical plots and support a broader understanding of global salary trends in remote work environments.
Copy citation: S. Kujabi, E. A. Frimpong, F. B. Effah, Enhancing Islanding Detection in PV-Based Distributed Systems Using CEEMDAN and Pattern Recognition Neural Network, Carpathian Journal of Electrical Engineering, vol. 19, no. 1, pp. 56-74, 2025, https://doi.org/10.34302/CJEE/LRBT6157.
Sulayman KUJABI, Emmanuel Asuming FRIMPONG, Francis Boafo EFFAH, Department of Electrical and Electronic Engineering, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana, kujabisaul@yahoo.com
Keywords: Distributed generation, islanding detection, empirical mode decomposition, pattern recognition neural network, CEEMDAN
Abstract: This paper presents an enhanced and practically validated islanding detection framework for grid-connected solar PV systems, integrating Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) and a Pattern Recognition Neural Network (PANN). The method processes negative sequence voltage signals at the Point of Common Coupling (PCC) to extract intrinsic mode functions (IMFs), mitigating mode mixing and improving signal fidelity. Significant IMFs are selected based on their power percentile, and three statistical features—maximum value, standard deviation, and entropy—are extracted and normalized before classification by the PANN. Unlike prior studies, this work extends evaluation to zero-power mismatch scenarios, noisy environments, and load-switching conditions, providing practical validation of real-time detection performance. Simulation results demonstrate a classification accuracy of 98.6% with a detection time of 0.1806 seconds, complying with IEEE 1547 standards. The proposed approach exhibits robust and reliable islanding detection across diverse operating conditions, significantly reducing the non-detection zone (NDZ) and enhancing the safety and reliability of modern distribution systems.