Bulletin of Electrical Engineering and Informatics
Not a member yet
2885 research outputs found
Sort by
Development of a machine learning-based framework for predicting failures in heat supply networks
Full text linkThe increasing complexity and scale of heat supply systems leads to a higher risk of failures, which may cause significant economic and environmental consequences. This study develops a predictive mathematical framework for the early detection of emergency conditions in heat supply networks (HSNs) using machine learning (ML). The proposed approach is based on the LightGBM gradient boosting (GB) algorithm, chosen for its high accuracy and efficiency in handling large datasets. Real operational data (temperature, pressure, flow, and vibration) were considered. Data preprocessing, feature engineering (including SHAP analysis), and hyperparameter tuning with grid search and 5-fold cross-validation improved prediction quality. The model achieved accuracy of 85%, F1-score of 0.82, and receiver operating characteristic (ROC)-area under the curve (AUC) of 0.96, outperforming logistic regression (LR) and decision trees. The framework may be integrated into monitoring systems for predictive maintenance, reducing downtime and optimizing costs
Soil erosion analysis based on machine learning method
Full text linkSoil erosion poses a serious environmental and agricultural threat that undermines land productivity, sustainability, and ecosystem stability. This study develops a robust machine learning framework for predicting and analyzing soil erosion across diverse landscapes by integrating advanced remote sensing data, climate indicators, and soil characteristics. Spectral indices such as the normalized difference vegetation index (NDVI), moisture stress index (MSI), and surface albedo were employed to assess vegetation condition, moisture levels, and surface reflectance. The proposed model, based on the extreme gradient boosting (XGBoost) algorithm, classifies erosion stages with up to 99% accuracy, ranging from healthy land to severely degraded areas. The methodology includes comprehensive feature engineering, dataset preprocessing, and model evaluation. Furthermore, a comparative analysis with traditional models (USLE and RUSLE) highlights the superior predictive performance of the proposed approach. The findings offer valuable insights for sensor-based monitoring systems and cloud-based decision-support tools, supporting sustainable land use management, erosion risk mitigation, and effective soil conservation strategies
Load frequency control of multi-source power system using PID+DD controller based on chess algorithm
Full text linkThis article presents load frequency control for a nonlinear multi-source power system divided into three areas, consisting of thermal reheat power plants, hydropower, and wind generation, while considering generation rate constraints (GRC). A proportional–integral–derivative (PID) plus second-order derivative (PID+DD) controller optimized using the chess algorithm (CA) is proposed. The effectiveness of CA is validated against hippopotamus optimization (HO), grey wolf optimizer (GWO), and ant lion optimizer (ALO) under two scenarios: a 10% step load perturbation (SLP) and a random load pattern (RLP). Simulation results indicate that the proposed CA significantly improves dynamic performance. In scenario 1 (10% SLP), CA achieves a reduction of approximately 30.5% in integral weight time absolute error (ITSE) compared to GWO and 43.7% compared to HO, while also reducing frequency undershoot in Area 2 by 15.2% compared to HO. In scenario 2 RLP, CA maintains robustness, limiting tie-line power deviations to ±8 MW, whereas HO exhibits deviations exceeding ±12 MW. Overall, the CA-tuned PID+DD controller demonstrates superior damping, reduced overshoot and undershoot, and enhanced stability across multi-area interconnected renewable systems, making it a promising approach for future real-time load frequency control (LFC) applications with higher renewable penetration
Performance analysis of 3D assets in virtual reality simulations for climate change: a case study in sustainable energy systems
Full text linkThis study investigates the performance impact of 3D assets in a virtual reality (VR) simulation designed for climate change education, aiming to balance visual fidelity and system efficiency on standalone headsets. Using a case study modeled on a sustainable energy environment, key performance metrics frames per second (FPS), triangle count, and draw calls were measured to assess the effect of object density, material transparency, and batching strategies. Experimental results show that configurations with 20 trees and 20 characters maintained 101 FPS, while denser scenes with 30 trees and 30 characters dropped to 79 FPS approaching the minimum usability threshold for VR. Transparent tree foliage with alpha-cutout materials imposed higher graphics processing unit (GPU) loads than high-triangle opaque character models, highlighting the performance cost of material complexity. These findings offer practical guidelines for optimizing asset configurations in immersive educational VR content. Future work may explore integration of artificial intelligence (AI) behavior and user interaction to assess broader system performance
On-edge 2D-to-3D generative pipeline for seamless instance transformation
Full text linkDespite ongoing challenges with fragmented workflows, latency in device imports, and the main issue of limitations in object reconstruction functionality, relying on imperfect extraction networks remains an impractical solution for scalable object generation. To deal with these constraints, we proposed an end-to-end pipeline that leverages a re-designed self-consistency mechanism—aimed at reducing discrimination, along with the beneficial enhancement from level-set projection and gradient-surface orthogonality. In addition, our approach designs dynamic 3D object creation with minimal manual effort by unifying surface topology and optimizing data loading, enabling a streamlined reconstruction process and more flexible object projection. Our method supports rapid, resource-efficient mesh reconstruction and consistently demonstrates performance improvements across multiple instance benchmarks, covering virtual projection tasks. Improvements in mesh topology reconstruction, as measured by the L1 Chamfer distance (CD) metric, are consistently higher, while the system also achieves significant transmission speedups—up to 56.5×—near-instant importing—along with lowering latency in practical rendering on virtual reality (VR) devices. This result highlights that refining mesh binding improves re-creation fidelity. Our approach to scalability leads to faster user engagement and allows automated deployment without requiring human intervention during importing
Prediction of postpartum depression in Zacatecas Mexico using a machine learning approach
Full text linkPostpartum depression (PPD) is a silent disorder, difficult to detect by the mother who suffers from it. In this research project, we propose a classification model of PPD using machine learning (ML) techniques, following a supervised learning approach. This is model allows the prediction of PPD using sociodemographic and medical data through a dataset of 100 Zacatecan mothers previously classified with the result of Edinburgh Test. We use eight ML algorithms such as adaptative boosting classified (ABC), principal component analysis (PCA) boosting, decision trees (DT), k-nearest neighbors (KNN), support vector machines (SVM), random forests (RF), and boosting. Our results show that the proposed ML model based on ABC algorithm can outperform other classifiers yielding a precision of 90%, a recall of 90%, a F1-score of 78% and 74% for area under curve (AUC), illustrating a correct capability in the prediction of this disorder
A high-efficiency transformerless buck-boost inverter with fuzzy logic control for grid-connected solar PV systems
Full text linkTransformerless inverters are increasingly favored in grid-connected photovoltaic (PV) systems due to their higher efficiency, reduced size, and lower cost. This paper presents a novel transformerless inverter topology that integrates buck boost conversion with an advanced fuzzy logic controller (FLC) to enhance energy extraction and power quality under dynamically changing solar conditions. The proposed system employs a sine triangle pulse width modulation (PWM) scheme in conjunction with the FLC to improve waveform quality and system responsiveness. By dynamically adapting to variations in irradiance and load, the control strategy reduces the total harmonic distortion (THD) from 36.51% to 1.51%, significantly enhancing compliance with international grid standards. Additionally, a novel grounding technique is implemented to mitigate common mode leakage currents, a typical issue in transformerless systems, without the need for galvanic isolation. Comprehensive MATLAB/Simulink simulations validate the inverter’s performance, demonstrating superior dynamic behavior, harmonic suppression, and overall reliability. The proposed architecture offers a compact, cost effective, and high performance solution for next generation grid integrated solar PV systems
The effect of the number of NACA 4412 airfoil blades on the performance of a horizontal axis wind turbine
Full text linkThis work reports an experimental investigation of the effect of blade number on the performance of a small-scale horizontal-axis wind turbine (HAWT) using NACA 4412 airfoil blades. Two turbine prototypes (one with 7 blades and one with 9 blades) were fabricated and tested under controlled wind speeds (3.4–6.0 m/s). The turbine outputs were measured using INA219 current/voltage sensors and a TCRT5000 rotations per minute (RPM) sensor interfaced to an Arduino-based system for real-time data acquisition. Results show that the 9-blade turbine consistently generated higher electrical power and achieved a higher power coefficient than the 7-blade design. For example, at 3.4 m/s the 7-blade turbine produced about 0.0297 W versus 0.0471 W for the 9-blade turbine. The peak power coefficient reached ≈0.198 for the 9-blade rotor (vs. ≈0.195 for 7 blades) at the same wind speed. Sensor calibration indicated high accuracy (errors 1.2%), confirming the reliability of the measurements. These findings suggest that, for the tested design, increasing the number of blades improves small-HAWT performance. The developed wireless monitoring system and experimental results provide guidance for optimizing blade count in future small turbine designs
IoT-based real-time monitoring of agricultural wastewater using Raspberry Pi, Node-RED, and Grafana
Full text linkThis study introduces an internet of things-based agricultural wastewater monitoring system (IoT-AWMS) designed to enhance water management through real-time monitoring and advanced sensor integration. The system employs a Raspberry Pi for centralized control, node-RED for automation, InfluxDB for data storage, and Grafana for visualization. A key innovation is the integration of an alternative sensing approach for estimating electrical conductivity (EC), complementing conventional sensors for total dissolved solids (TDS), water temperature (DS18B20), and ambient conditions (DHT11). The system achieves over 85% accuracy in estimating EC across diverse water samples, including drinking water, agricultural runoff, and fertilizer-enriched solutions. Compared with conventional approaches, IoT-AWMS demonstrates superior accuracy, scalability, and cost-effectiveness. Its modular design supports applications in nutrient runoff detection, contamination monitoring, and optimized water resource utilization, with broader potential in precision farming and environmental monitoring. This work contributes a robust, adaptable IoT framework for sustainable agricultural water management
Intelligent building automation system using ESP32, Azure and internet of things technologies
Full text linkThe adoption of home automation systems in buildings faces limitations due to their cost, integration complexity, and protocol heterogeneity, which hinders the development of accessible solutions based on embedded devices to improve interaction in environments within buildings or homes. The literature review indicates that the selection of hardware and communication protocols in home automation systems considers factors such as cost, available infrastructure, and application context. In addition, approaches are identified that prioritize security, wired or wireless connectivity, and affordability. This paper presents the development of an affordable home automation system for building automation in Lima, using the ESP32 microcontroller and internet of things (IoT) technologies. The objectives focus on hardware design, implementation of control algorithms, remote monitoring interface, and validation in a simulated environment. The solution includes Wi-Fi connectivity, a cloud-based MySQL database, and a web interface. Key findings include the home automation system, integrated with Flask technology and web services, enabling monitoring and control via a responsive web interface, demonstrating its operability and ensuring lossless data transmission