Eventos Anais de eventos
COBEM 2023
27th International Congress of Mechanical Engineering
AUTOMOTIVE BEARINGS ANALYSIS BASED ON REGRESSION MODELS
Submission Author:
Isabelle Therezinha Simão , PR , Brazil
Co-Authors:
Alan Lopes, Isabelle Therezinha Simão, LUIZ EDUARDO THOMAZ, Viviana Mariani, Leandro dos Santos Coelho
Presenter: Isabelle Therezinha Simão
doi://10.26678/ABCM.COBEM2023.COB2023-0822
Abstract
Using statistical models to predict new observations or events is one of the main goals of supervised statistical learning methods. In machine learning, one of the most important segments of Artificial Intelligence, a system or machine automatically learns to predict without being explicitly programmed to do so. Instead, it uses algorithms that allow the system to analyze data and recognize patterns based on certain prior examples, the data. The accuracy of the model depends on the data set used for training, the selection of the appropriate algorithms, and the choice of appropriate parameters. Machine learning is a constantly evolving field, with much research underway to develop new algorithms and techniques for regression, prediction, and classification. Currently, models have many applications in regression tasks in different fields of study, and the objective of this study is to apply such models to automotive bearing manufacturing processes. Regression models are widely used when it is desired to evaluate the impact of production factors on the quality of a product. The main objective of this study is to use statistical regression techniques, including linear and polynomial regression, and the neural network multi-layer perceptron (MLP), to understand and quantify the relationships between geometric variables of automotive bearings, as well as to compare the performance of the regression technique MLP through statistical metrics.
Keywords
Automotive bearing, Linear regression, polynomial regression, Multi-Layer Perceptron
