Holds a postdoctoral degree from ULSTER University in Northern Ireland. Holds a Ph.D. and a master's degree in Engineering from the Federal University of Rio Grande do Sul (UFRGS) (CAPES grade 7). Specialized in Industrial Automation at the Federal University of Santa Catarina (UFSC) in 1998. Earned a bachelor's degree in Computer Science from the University of Southern Santa Catarina (UNISUL) in 1998. Currently, serves as a full-time professor at UFSC - Federal University of Santa Catarina. Has experience in the field of Computer Science, with an emphasis on Embedded Systems, working primarily on the following topics: microprocessors, automation, renewable energy, software, and education technology. Leads the LPA - Applied Research Laboratory, a CNPq research group. Also a member and researcher at the LABTEL research laboratory.
Machine learning prediction of the forming limit curve of dual phase steels André Rosiak, Peterson Duarte Diehl, Roderval Marcelino, Lirio Schaeffer International Journal of Material Forming, 2026 Accurate prediction of the Forming Limit Curve (FLC) is essential for the design of sheet metal stamping processes; however, its experimental determination is costly and limited by data availability. This work investigates the use of Machine Learning techniques to predict the FLC of Dual Phase (DP) steels based on mechanical properties obtained from uniaxial tensile tests. To overcome the scarcity of experimental data, a synthetic database was developed based on statistical consistency and physical constraints, using Kernel Density Estimation, PCA projections, and controlled probabilistic interpolation, followed by the application of physicometallurgical plausibility criteria. The models use physics-based descriptors as input variables, which reflect known metallurgical mechanisms associated with plastic instability, without explicitly incorporating differential equations into the training process. The results show that all models were able to reproduce the characteristic geometry of the FLC, with errors on the order of 10⁻³–10⁻². Among the investigated techniques, Random Forest exhibited the best performance (MAE = 0.0052; MSE = 0.00011; R² = 0.943), followed by XGBoost, while the Neural Network showed greater variability and a tendency toward overfitting. The results demonstrate that the combination of physics-based descriptors, statistically validated synthetic expansion, and ensemble machine learning methods constitutes a robust and efficient strategy for modeling FLCs of DP steels.
Blockchain solutions for scientific paper peer review: a systematic mapping of the literature Allan Farias Fávaro, Roderval Marcelino, Cristian Cechinel Data Technologies and Applications, 2024 PurposeThis paper presents a review of the state of the art on the application of blockchain and smart contracts to the peer-review process of scientific papers. The paper seeks to analyse how the main characteristics of the existing blockchain solutions in this field to detect opportunities for the improvement of future applications.Design/methodology/approachA systematic review of the literature on the subject was carried out in three databases recognized by the research community (IEEE Xplore, Scopus and Web of Science) and the Frontiers in Blockchain journal. A total of 1,967 articles were initially found, and after the exclusion process, the 26 remaining articles were classified according to the following dimensions: System Type, Open Access, Review Type, Reviewer Incentive, Token Economy, Blockchain Access, Blockchain Identification, Blockchain Used, Paper Storage, Anonymity and Maturity of the solution.FindingsResults show that the solutions are normally concerned on offering incentives to the reviewers' work (often monetary). Other common general preferences among the solutions are the adoption of open reviews, the use of Ethereum, the implementation of publishing ecosystems and the use of InterPlanetary File System to the storage of the papers.Originality/valueThere are currently no studies covering the main aspects of blockchain solutions in the field of scientific peer review. The present study provides an overall review of the topic, summarizing important information on the current research and helping new adopters to develop solutions grounded on the existing literature.
IoT sensors integrated with the distributed protocol IOTA/Tangle: Bosch XDK110 use case Wellington Fernandes Silvano, Daniel De Michele, Daniel Trauth, Roderval Marcelino Brazilian Symposium on Computing System Engineering Sbesc, 2020 Decentralized systems can provide privacy, security, and immutability without relying on a central authority. In the context of the Internet of Things, Distributed Ledger Technologies (DLT) can be important facilitators for implementing IoT economies. IOTA is a public DLT, which was created to provide support for the Internet of Things. The IOTA protocol enables high transaction throughput at zero cost, while also being highly scalable. These characteristics make it possible to collect sensor data with transparency and security, using a distributed public network. Despite the great potential of IOTA as a distributed protocol for sensor data, there are still no studies that demonstrate how the connection between sensors and IOTA can occur, and what the characteristics and limitations of the system are. We began exploratory technological research which includes the modeling of a system that allows the integration of IOTA with sensors, using IOTA as the data layer. An implementation of the model has been developed, based on data collection from the Bosch XDK110 multisensor and a storage/visualization application. Because the IOTA protocol requires Proof of Work to send and broadcast data over the network, which takes a few seconds, it is not possible to have granular live data from sensors directly interacting with the DLT. However, we will study the case of an application that allows for this by buffering data and inserting it to the IOTA network at a rate that nodes can handle.
LABCONM: A remote lab for metal forming area Lucas B. Michels, Luan C. Casagrande, Vilson Gruber, Lirio Schaeffer, Roderval Marcelino Lecture Notes in Networks and Systems, 2018
Remote experimentation model based on digital TV Ranieri Alves dos Santos, Eliane Pozzebon, Luciana Bolan Frigo, Roderval Marcelino Exp at 2015 3rd Experiment International Conference Online Experimentation, 2016
Remote experiments and 3D virtual world in education Caroline Porto Antonio, Joao Paulo Cardoso De Lima, Joao Bosco da MotaAlves, Roderval Marcelino, Juarez Bento da Silva, et al. Exp at 2015 3rd Experiment International Conference Online Experimentation, 2016
A remote experimentation and 3D virtual world for basic education Caroline Porto Antonio, Joao Paulo Cardoso De Lima, Joao Bosco da Mota Alves, Roderval Marcelino, Juarez Bento da Silva, et al. Exp at 2015 3rd Experiment International Conference Online Experimentation, 2016
Studying in 3D environments Roderval Marcelino, Juarez Bento da Silva, Vilson Gruber, Simone Meister Bilessimo, Janaina Oliveria, et al. International Journal of Online Engineering, 2014
