https://repositorio.ufu.br/handle/123456789/19216 2026-04-23T14:28:19Z 2026-04-23T14:28:19Z https://repositorio.ufu.br/handle/123456789/47964 2026-01-06T06:19:55Z 2025-12-01T00:00:00Z

Title: Inovação sustentável: sistema de quebra de coco com solução mecânica segura e ergonômica utilizando materiais reaproveitados Abstract: This work presents the development of a babassu coconut cutting machine aimed at improving the working conditions of coconut breakers, an activity traditionally characterized by high physical effort and accident risk. The project was conceived based on the principles of mechanical engineering, ergonomics, and sustainability, using recycled materials easily found in local workshops and scrapyards, such as automotive and metallic components, to ensure low cost and community replicability. The mechanical system of the machine consists of a rack-and-pinion mechanism coupled to an automotive steering sector, which converts the rotary motion of the lever into a linear displacement of the cutting blade, significantly reducing the operator’s physical effort. Poka-Yoke safety systems and adaptations to NR-12 and NR-17 standards were implemented to ensure safe and ergonomic operation. Performance tests indicated that the equipment can cut each fruit in approximately 10 seconds, representing a productivity increase of up to 85% compared to the manual method, while also providing greater comfort and operational safety. It is concluded that the developed prototype is technically viable, sustainable, and socially relevant, contributing to the enhancement of the babassu extractivist communities’ work and the modernization of the babassu production chain.

2025-12-01T00:00:00Z https://repositorio.ufu.br/handle/123456789/47963 2026-01-06T06:19:54Z 2025-12-30T00:00:00Z

Title: Máquina semiautomatizada e sustentável para quebra do Coco Babaçu: uma solução tecnológica de baixo custo para redução de esforço e risco nas comunidades extrativistas Abstract: This Final Paper presents the development of a low-cost electric machine for cracking babassu coconuts, designed as an ergonomic, safe, and sustainable alternative to the traditional manual process carried out by babassu coconut breakers. The project was based on principles of applied mechanical engineering and social technology, aiming to reduce physical effort and accident risks while increasing productivity and improving working conditions. The development included conceptual and experimental stages, evolving from early prototypes using washing machine motors and gear reduction systems to the final configuration, which employs a sliding gate motor with internal reduction, chain transmission, and a connecting rod–crank–piston system. Three-dimensional modeling was performed using Autodesk Inventor software, allowing for kinematic analysis, component sizing, and geometric validation of the assembly. The results indicate that the project meets the criteria of simplicity, low cost, and safety, demonstrating technical feasibility and strong potential for application in extractivist communities. It is concluded that the proposed machine represents an efficient social technology, capable of combining mechanical innovation, sustainability, and cultural appreciation of babassu coconut breakers.

2025-12-30T00:00:00Z https://repositorio.ufu.br/handle/123456789/47962 2026-01-06T06:19:44Z 2025-09-24T00:00:00Z

Title: A aplicação de deep learning na previsão de propriedades mecânicas de materiais — revisão sistemática e aprofundada Abstract: The application of deep learning in materials science represents a new frontier for accelerating the discovery and optimization of emerging technologies. This study aims to conduct a systematic literature review to map the state of the art, methodologies, challenges, and future trends in the use of deep neural networks for predicting mechanical properties. The methodology involved the analysis of 73 scientific articles published between 2017 and 2025. The results indicate an exponential growth of the field since 2020, with a predominant focus on predicting properties such as tensile strength and fatigue. Convolutional Neural Networks (CNNs) dominate the landscape, driven by the use of microstructure images as input data, alongside hybrid approaches that combine experimental and synthetic datasets for database construction. Despite the high predictive accuracy of the models, the main challenges lie in their "black-box" nature, which limits interpretability, and in ensuring generalization to real-world scenarios. Future trends point to the consolidation of Explainable Artificial Intelligence (XAI) techniques and the integration of such models into Inverse Materials Design platforms, marking the transition from a purely predictive data-driven science to a prescriptive approach.

2025-09-24T00:00:00Z https://repositorio.ufu.br/handle/123456789/47961 2026-01-06T06:19:42Z 2025-09-16T00:00:00Z

Title: Aprimorando processos de usinagem com inteligência artificial : uma revisão abrangente Abstract: This work consists of a comprehensive bibliometric review on the application of artificial intelligence techniques in machining processes. The central objective of the study was to map the scientific research scenario from 2021 to 2025, aiming to identify the main trends and research gaps, quantifying the impact of various artificial intelligence techniques on processes such as milling, turning, drilling, and grinding. To enable this research, 85 articles were statistically analyzed to generate data on the topic. As a result, milling and turning were the most investigated processes, while machine learning and artificial neural networks are the most dominant techniques. It was revealed that the main focus of the research lies in optimizing surface roughness and tool wear, although a notable growth in the evaluation of environmental and economic impacts is observed. It was also possible to identify persistent challenges, such as big data management and the difficulty of generalizing artificial intelligence models for different industrial scenarios. It was concluded that the research field is mature in quality and wear optimization but points to new promising directions, such as the use of hybrid approaches and digital twins to overcome existing limitations. It is suggested for future work to focus on more accessible and standardized solutions that allow the transition to intelligent manufacturing, ensuring maximum efficiency, quality, and sustainability.

2025-09-16T00:00:00Z