Advancing viscoelastic material characterization through computer vision and robotics: MIRANDA and RELAPP
This study introduces MIRANDA, a computer vision system, and RELAPP, a complementary force measurement system, developed for characterizing viscoelastic materials. Our aim was to evaluate their combined ability to predict key rheological parameters and demonstrate their utility in material analysis,...
| Autores: | , , , , , , , , , |
|---|---|
| Tipo de documento: | artigo |
| Data de publicação: | 2025 |
| País: | España |
| Recursos: | Universitat Politècnica de Catalunya (UPC) |
| Repositório: | UPCommons. Portal del coneixement obert de la UPC |
| Idioma: | inglês |
| OAI Identifier: | oai:upcommons.upc.edu:2117/452431 |
| Acesso em linha: | https://hdl.handle.net/2117/452431 https://dx.doi.org/10.3390/ma18214827 |
| Access Level: | Acceso aberto |
| Palavra-chave: | Robotics Computer vision Viscoelastic material Chopin alveograph Material characterization Viscoelasticity Viscosity Àrees temàtiques de la UPC::Enginyeria electrònica::Microelectrònica |
| Resumo: | This study introduces MIRANDA, a computer vision system, and RELAPP, a complementary force measurement system, developed for characterizing viscoelastic materials. Our aim was to evaluate their combined ability to predict key rheological parameters and demonstrate their utility in material analysis, offering an alternative to traditional methods. We analyzed five distinct flour dough samples, correlating MIRANDA and RELAPP variables with established rheological reference values. Support Vector Machine (SVM) regression models were trained using MIRANDA’s stable TR and elasticity data to predict industrially relevant parameters: baking strength (W), tenacity (P), extensibility (L), and final viscosity (RVU) from Chopin alveograph and viscosimeter. The predictive models showed promising results, with R2 values of 0.594 (p = 0) for W, 0.575 (p = 0) for P, and 0.612 (p = 0.03763) for viscosity, all statistically significant. While these findings are promising, it is important to note that the small sample size may limit the generalizability of these models. The synergy between the systems was evident, exemplified by strong positive correlations, such as between MIRANDA’s Elasticity and RELAPP’s c_exp (parameter ‘c’ of its mathematical model m1, r = 0.858) and final resistive force (r = 0.839). Despite the limited sample size, these findings highlight MIRANDA’s versatility and speed for efficient material characterization. MIRANDA and RELAPP offer significant industrial implications for viscoelastic materials, including accelerating development cycles and enhancing continuous quality control. This approach has strong potential to reduce reliance on slower, traditional methods, warranting further validation with larger datasets. |
|---|