Eventos Anais de eventos
COBEM 2023
27th International Congress of Mechanical Engineering
Statistical comparison of Taguchi method and full factorial on the quality of drilling on FRP composite
Submission Author:
Felipe Brum Rossato , SC
Co-Authors:
Felipe Brum Rossato, Jhonatan Acacio Silva, Gabriel Flores, Hector Perdoncini, Rodrigo Sant'Ana
Presenter: Felipe Brum Rossato
doi://10.26678/ABCM.COBEM2023.COB2023-1873
Abstract
Fiber-reinforced composite materials have gained immense popularity in various industries as a viable alternative to metallic materials due to their exceptional mechanical and physical properties. These materials offer a high strength-to-weight ratio, superior fracture resistance, and excellent chemical stability, making them highly desirable for applications that demand lightweight and durable materials. Among these composite materials, the FRP (Fiber reinforced plastic) has been used in large-scale in many industries and the necessity to machine them is a big concern due to the possibility of damaging the material and decrease its strength. Since these problems can render a part useless the normal approach is to test the products, but this attitude can generate a large financial cost and consume a lot of time because of that some statistical methods can be used to make this process more efficient. This research aims to verify the accuracy of the Taguchi method for experiment design, when compared to a full factorial analysis. The goal is to apply both methods to verify and optimize machining parameters based on their influences on the quality of holes; this quality is measure using both qualitative and quantitative factors, by the Davim delamination factor. The delamination is analyzed in the entry and exit of holes resulting from the drilling process in fiber-reinforced plastic, in an industrial setup. To carry this analysis a data set on the drilling of FRP previously done by the full factorial method was used as a base, the statistical method applied received a limited subset of the complete factorial analysis, and the results were compared. The full factorial set comprise of five types of drill bits (118°, 130°, 135°, 140° and E-90°) three feed rates (0,08mm/s ;0,10mm/s ;0,30mm/s) and three cutting speeds (10m/min, 28m/min 57m/min) in two materials ([0°,90°);[-45°,45°]). Both the entry and the exit of the hole were analyzed and a scored based on the eccentricity and delamination area and by sing Davim method a value was attributed to is quality. The full factorial analysis comprises of 45 tests for every material side and by using a L9 adapted orthogonal array a reduction of 5 times was achieved in the necessary performed experiments, witch reduced from 45 to only 9. For both methods 2 statistical analysis were performed the mean value analysis and the ANOVA. The results obtained by the Taguchi method were mixed, it correctly predicted all the mean values of cutting speed, feed rates and 3 out of the 5 drill bits (118°, 140° and E-90°), but failed to predict the correct values for 130° and 135°. Furthermore, in the ANOVA analysis Taguchi descanted cutting speed as a significant factor in the entry of both sheets. In conclusion, for the entries and the adapted setup the Taguchi method was not a valid method for the scientific studies but however it shown great potential in industrial setups were speed and cost is the upmost priority. The information obtained in this research intends to expand information and knowledge on this subject.
Keywords
Drilling, Taguchi's method, composites materials, Statistical Analysis
