On 17 December 2025, at 9:00 a.m., we held the master’s thesis defense of the student Iury Melo Américo, whose thesis was entitled: Prediction of Steel Flow Curves Using Artificial Intelligence Algorithms..

This work evaluates the performance of different machine learning algorithms for predicting metal flow curves using chemical composition, temperature, strain, and strain rate as input variables. The goal is to provide an alternative to laboratory experimentation, reducing the costs of generating these curves, which are essential in the analysis of the industrial metal forming process. The research is classified as applied technological, employing the implementation of a computational prototype divided into two stages: selection and evaluation of regression models, and the development of a web tool for use by engineers. Cross-validation methods, cleaning, and standardization were applied to the dataset extracted from QForm simulations, as well as hyperparameter optimization using libraries such as scikit-learn, TensorFlow, and Optuna. The main models evaluated were Random Forest, XGBoost, Gradient Boosting, SVR, and shallow and deep neural networks. The XGBoost model presented the best performance, surpassing even deep neural networks, with a mean absolute error of 7.53 MPa on the validation set, demonstrating robustness even for materials not present in the training data. It was found that both temperature and strain rate can influence performance, especially in high-alloy steels, suggesting potential topics for future studies. Finally, an API and a web application were developed to facilitate the practical application of the models in an industrial environment. The results confirm that machine learning techniques are capable of supporting the reliable prediction of flow curves, meeting the proposed objectives.

On 17 December 2025, at 9:00 a.m., we held the master’s thesis defense of the student Anderson Conti Soprana, whose thesis was entitled: Wireless Sensor Networks for Agriculture: Analysis of a Thread Network Based on IEEE 802.15.4.

Thesis abstract: This work aims to investigate the performance and reliability of the Thread network, based on the IEEE 802.15.4 standard, in wireless communication scenarios focused on precision agriculture. The research proposes the analysis of a Thread Wireless Sensor Network (WSN) as a viable alternative for monitoring environmental variables in locations with connectivity constraints, seeking greater energy efficiency, scalability, and security in data transmission. The methodology adopted combines an experimental approach and quantitative analysis, with packet capture and validation carried out using the Wireshark software. Payload tests were conducted under low and high load conditions, varying the distance and signal strength between devices, in order to measure metrics such as RSSI, LQI, jitter, and packet loss. The results obtained showed that the network maintains stability and a low error rate up to approximately 100 meters, with average RSSI values between -60 dBm and -80 dBm and LQI above 50. Beyond this distance, a progressive degradation of link quality and an increase in packet loss were observed, especially under higher loads, revealing the operational limits of the technology in long‑distance and high‑traffic scenarios. The study also implemented DHT11 sensors integrated into the Thread network, demonstrating the feasibility of collecting and transmitting temperature and humidity data to a web server, validating the integrity and stability of the communication. The analyses confirmed the robustness and applicability of the protocol in distributed monitoring and automation systems. It is concluded that Thread is a promising solution for low‑cost, low‑power mesh networks suitable for Internet of Things (IoT) applications in agricultural, residential, and small‑scale industrial environments, contributing to the advancement of communication technologies in intelligent systems.

Sorry, this entry is only available in Brazilian Portuguese.

As a result of the doctoral dissertation work of the researcher Peterson Duarte Diehl, the article entitled “Predictive modeling of hardness in bainitic steel forging processes using artificial intelligence” was published in The International Journal of Advanced Manufacturing Technology.

This work is the result of a strong partnership between UFSC and UFRGS, involving the Computer Engineering program (UFSC), the Graduate Program in ICT (UFSC), and the UFRGS Graduate Program PPG3M. The study comes from a project approved by CNPq, aimed at applying AI to mechanical manufacturing processes, especially mechanical forming.

Access link to the publication:
https://link.springer.com/article/10.1007/s00170-025-16888-8#citeas

On October 1, 2025, Professor Dr. Roderval Marcelino gave a lecture on the topic of Edge AI for the Artificial Intelligence vertical at Sigma Park in Tubarão. The lecture was followed by extensive exchanges of ideas and discussions about potential projects and partnerships among the participants.

The CPSlab research laboratory (https://cpslab.ufsc.br/) under the coordination of Professor Roderval has an open position for a scholarship student enrolled in the Computer Engineering or Information and Communication Technologies program, with a focus on web development of an application.

The selected scholarship recipient must dedicate 20 hours per week to the project, all mornings, at the laboratory located at UFSC Mato Alto campus, working on web development using React Native, basic database manipulation, and version control via GitHub.

Requirements:

• Currently enrolled in the Computer Engineering or ICT program.

• Basic knowledge of React Native for developing responsive interfaces.

• Basic understanding of databases (data modeling and simple queries).

• Use of Git and GitHub for code versioning.

Benefits:

• Scholarship of R$1,600.00

• Project duration: 18 months

• Dedication: 20 hours per week, all mornings

• Openings: One position

Interested candidates should contact: roderval.marcelino@ufsc.br

This Thursday, 08/14, at 10 AM, we will have a brilliant lecture with Prof. Marcelo Rovai about Edge AI. This lecture will be the inaugural class of my course Edge AI.
The lecture is open to the public, just access the link

https://lnkd.in/djyKcvYB