Eventos Anais de eventos
COBEM 2023
27th International Congress of Mechanical Engineering
Multi-Objective Genetic Algorithm Optimization to Examine the Durability of Cutting Tools with NASA Milling Dataset
Submission Author:
Danilo dos Santos Oliveira , DF , Brazil
Co-Authors:
Danilo dos Santos Oliveira, jackson paz bizerra de souza, Priscila Morais, Rhander Viana, Alberto Alvares
Presenter: Danilo dos Santos Oliveira
doi://10.26678/ABCM.COBEM2023.COB2023-0234
Abstract
This work presents an application of multi-objective genetic algorithm (MOGA) optimization to examine the durability of cutting tools using the NASA milling dataset. This dataset encapsulates a variety of cutting tool conditions, including cutting speeds and depths, which were monitored via three sensor types - Acoustic Emission Sensor, Vibration Sensor, and Current Sensor. The primary objective of this research was to develop an optimized method capable of accurately evaluating the durability of cutting tools under different usage conditions. To accomplish this, multi-objective genetic algorithm optimization was utilized, circumventing the drawbacks inherent in traditional optimization techniques, such as direct optimization and response surface optimization. Accordingly, a non-dominated genetic classification algorithm (NSGA-II) was implemented. The data processed through this method presented optimal conditions for various factors. These factors included machining time, spindle vibration, table vibration, spindle current, and spindle acoustic emission, which were all examined as functions of the depth of cut and feed variables. As a result, the study was able to reveal all pairs of optimal combinations for the aforementioned objective functions. The study concludes that MOGA is a promising tool for optimizing tool wear and extending tool life by precisely adjusting milling environment variables. By providing a novel approach for optimizing cutting tools in a range of industrial applications, this research could make significant contributions to the enhancement of cutting tool durability across various commercial sectors.
Keywords
Artificial Intelligence, Multi-Objective Genetic Algorithm, cutting tool wear, milling process, Dataset Analysis
