Publication: Aplicación de técnicas de IA en mecanizado
Date
2022
Editor
Director
Advisor
Coordinator
Commentator
Reviewer
Illustrator
Access rights
info:eu-repo/semantics/openAccess
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad Nacional de Educación a Distancia (España), Universidad Politécnica de Madrid. Departamento de Ingeniería Mecánica
Abstract
El boom de la digitalización permite disponer de un volumen de datos que provienen de máquinas herramienta en su etapa operativa. El análisis de estos datos puede ayudar a obtener mejoras en el ámbito del mantenimiento de las máquinas, pero también mejoras en los procesos de fabricación que realizan las máquinas herramienta. El objetivo del proyecto consiste en desarrollar algoritmos o técnicas basadas en Inteligencia Artificial, que permitan analizar datos procedentes de centros de mecanizado, que faciliten procesos más productivos, desatendidos o robustos. Se parte de los datos históricos de dos centros de mecanizado en pleno rendimiento productivo que fabrican una variedad muy amplia de piezas diferentes. Se busca a través de los datos encontrar patrones que permitan controlar el funcionamiento adecuado de cada herramienta. El control se basará en el funcionamiento de la herramienta para las mismas condiciones en un histórico de datos. El núcleo del trabajo es el pre-procesamiento de los datos (eliminación de pausas y momentos no productivos, división de datos por operaciones, identificación del tipo de operaciones) y el estudio de las alternativas existentes de Machine-Learning que pueden ayudar a modelizar el funcionamiento de la herramienta utilizando las variables apropiadas
The digitization boom makes available a volume of data coming from machine tools in their operational stage. The analysis of this data can help to obtain improvements in the field of machine maintenance, but also improvements in the manufacturing processes performed by machine tools. The objective of the project is to develop algorithms or techniques based on Artificial Intelligence, which allow the analysis of data from machining centers, to facilitate more productive, unattended or robust processes. The starting point is the historical data of two machining centers in full productive performance that manufacture a wide variety of different parts. The data is used to find patterns that allow to control the proper operation of each tool. The control will be based on the performance of the tool for the same conditions in a data history. The core of the work is the pre-processing of the data (elimination of non-productive moments, division of data by operations, identification of the type of operations) and the study of the existing Machine-Learning alternatives that can help to model the operation of the tool using the appropriate variables.
The digitization boom makes available a volume of data coming from machine tools in their operational stage. The analysis of this data can help to obtain improvements in the field of machine maintenance, but also improvements in the manufacturing processes performed by machine tools. The objective of the project is to develop algorithms or techniques based on Artificial Intelligence, which allow the analysis of data from machining centers, to facilitate more productive, unattended or robust processes. The starting point is the historical data of two machining centers in full productive performance that manufacture a wide variety of different parts. The data is used to find patterns that allow to control the proper operation of each tool. The control will be based on the performance of the tool for the same conditions in a data history. The core of the work is the pre-processing of the data (elimination of non-productive moments, division of data by operations, identification of the type of operations) and the study of the existing Machine-Learning alternatives that can help to model the operation of the tool using the appropriate variables.
Description
UNESCO Categories
Keywords
inteligencia artificial, mantenimiento predictivo, máquina-herramienta,, ánalisis de datos
Citation
Center
E.T.S. de Ingenieros Industriales
Department
Mecánica