Impact of the Extreme Operating Gusts on Power Converter Connected to PMSG based Wind Turbine for Reliability Analysis

A new reliability analysis explores the effects of Extreme Operating Gusts (EOG) on back-to-back (BTB) power converters in PMSG-based wind turbines. Using a 4^2 factorial design and data from La Ventosa, Oaxaca, researchers evaluated the impact of EOG’s amplitude and intensity on electrical variables like current, voltage, and active power. The study, simulated with PSIM® software, identifies the most critical factor affecting BTB converter reliability.

Read the full article in Latam Transactions, October Issue (2024).

 

Implementation of an adaptive data logging algorithm in low-cost IoT nodes for supply chain transport monitoring

In IoT-powered supply chain transportation, minimizing data loss, reducing energy use, and optimizing travel times are critical. This study presents a cutting-edge adaptive data logging algorithm designed for cost-effective IoT nodes, ensuring real-time data capture and display through a web interface. Tested in both indoor and outdoor environments, this innovation shows promising results in enhancing transportation efficiency by reducing data by 74%, improving power use and processing needs. A step forward in using IoT to revolutionize supply chain systems!

Read the full article in Latam Transactions, October Issue (2024).

Real-time Object Detection Performance Analysis Using YOLOv7 on Edge Devices

Real-time object detection is key in computer vision for applications like security, autonomous vehicles, and robotics! This study tested the YOLOv7-tiny model on three edge hardware platforms: Raspberry Pi 4B, Jetson Nano, and Jetson Xavier NX. Results?
– Raspberry Pi 4B: 0.9 FPS
– Jetson Nano: 7.4 FPS (max)
– Jetson Xavier NX: 30 FPS (max), showing CPU speed impacts FPS more than GPU!

Read the full article in Latam Transactions, October Issue (2024).

 

Addressing Class Imbalance in Healthcare Data: Machine Learning Solutions for Age-Related Macular Degeneration and Preeclampsia

Exciting news in healthcare! A recent study highlights how machine learning is revolutionizing disease diagnosis and treatment optimization.

Researchers tackled the challenge of imbalanced medical data, often due to privacy regulations, by proposing a hybrid approach combining SMOTE and undersampling techniques.

Their findings show that methods like Balanced Bagging and Balanced Random Forest significantly improve performance, ensuring better patient outcomes.

This innovative solution could transform how we handle healthcare data, paving the way for more effective interventions!

Read the full article in Latam Transactions, October Issue (2024).