EM - Escola de Minas

Agora exibindo 1 - 3 de 3

A statistical analysis of the relationship of civil construction GDP to cement production in Brazil.
(2022) Souza, Ana Carolina Rodrigues da Rocha; Gomes, Helton Cristiano; Guimarães, Irce Fernandes Gomes
The ICC plays an important role in the Brazilian economy. This participation in the country's GDP remains, on average, above 5% per year. Cement, one of the main resources in this context, is used in almost all types of constructions in the country. The Brazil are among the 10 largest producers in the world and cement be the main component of concrete, makes widely used. The generation of different data is the starting point for decisions, optimization and forecasting of the activities of this communication network. To transform this data into information, many institutions use with tools such as Data Science. In this sense, this article presents the result of analysis of the behavior of cement production in Brazil, based on results generated through Machine Learning. Trends and seasonality periods were identified, as well as prediction models for future periods were proposed. Verified the existence of a strong positive correlation between cement production and the ICC GDP in Brazil. Machine Learning models were proposed and compared to predict the ICC GDP based on the annual cement production in Brazil, which showed high accuracy. It was concluded that the Ensemble Learning methods adapted better to the data, especially Random Forest.
Sensitivity analysis of coating mortars according to their specific heat, specific gravity, thermal conductivity, and thickness in contribution to the global thermal performance of buildings.
(2022) Mendes, Vítor Freitas; Fardin, Welington; Barreto, Rodrigo Rony; Caetano, Lucas Fonseca; Mendes, Júlia Castro
Although coating (plastering) mortars are an important element of masonry systems, their impact on the building's overall thermal performance is still unclear. In this sense, the present work performed a sensitivity analysis on the influence of the thermophysical properties of coating mortars on the internal temperature and thermal load of two buildings. The authors aimed to fill the gap between the mortars' properties, their manufacturing specifications, and the actual effect of their application on the building's total energy perfor- mance. The methodology included energy simulations on EnergyPlus considering all Brazilian bioclimatic zones. We varied the mortars' specific heat, specific gravity, thermal conductivity, and thickness from 25% to 200% from baseline values. We also analysed the results through Decision Tree technique (XGBoost). The thermal conductivity (proportional to the specific gravity) was the less significant property, whereas the thickness and the specific heat were the most influential ones. The differences between the best and worst mortars reached 356 ◦C and 224 kWh/year for the house, and 736 ◦C and 45 kWh/year for the commercial building. The results showed that the optimal combination of the tested properties is a function of the bioclimatic characteristics of the region, the building layout, and the existence and schedule of the HVAC system. The simulations also evidenced that the strategy of solely decreasing the thermal conductivity without considerations for the thermal capacity, which is often used in the manufacturing of conventional insulating mortars, is ineffective. Therefore, assertively adjusting the mortars' thermophysical properties can be a promising complementary strategy for improving the thermal performance of buildings.
Automatic system for visual detection of dirt buildup on conveyor belts using convolutional neural networks.
(2020) Santos, André Almeida; Rocha, Filipe Augusto Santos; Reis, Agnaldo José da Rocha; Guimarães, Frederico Gadelha
Conveyor belts are the most widespread means of transportation for large quantities of materials in the mining sector. Therefore, autonomous methods that can help human beings to perform the inspection of the belt conveyor system is a major concern for companies. In this context, we present in this work a novel and automatic visual detector that recognizes dirt buildup on the structures of conveyor belts, which is one of the tasks of the maintenance inspectors. This visual detector can be embedded as sensors in autonomous robots for the inspection activity. The proposed system involves training a convolutional neural network from RGB images. The use of the transfer learning technique, i.e., retraining consolidated networks for image classification with our collected images has shown very effective. Two different approaches for transfer learning have been analyzed. The best one presented an average accuracy of 0.8975 with an F-1 Score of 0.8773 for the dirt recognition. A field validation experiment served to evaluate the performance of the proposed system in a real time classification task.

EM - Escola de Minas

Notícias

Navegar

Filtros

Configurações

Ordenar por

Resultados por página

Resultados da Pesquisa