Multi-objective optimization of carbon fiber–reinforced polymer drilling process based on grey fuzzy reasoning grade analysis

Abstract

With outstanding mechanical advantages such as high strength, light weight, high rigidity, low thermal expansion, good resistance corrosion, and high vibration resistance, carbon fiber-reinforced polymer (CFRP) is becoming increasingly popular in the automotive, aerospace, and medical spheres. For the CFRP machining, the biggest drawback is the effect of moisture on the mechanical properties caused by the cooling liquid and tool geometry. The appreciable body of study on the influence of wetness on the mechanical quality of carbon fiber-reinforced plastic has shown that there is no suitable use for lubrication fluids in the composite industry. However, there is a significant impact on the CFRP machining capabilities caused by the geometry of the cutting tool and cryogenic liquid cooling. This paper investigated the optimal choice of the parametric combinations in the CFRP drilling process. The optimization method was proposed using the Fuzzy interference system based on grey relation analysis. The experiment data was designed according to the Taguchi method. The twist angle of the drill, the coolant gas, and the feed rate were considered as the parameter inputs while observing several multi-responses such as thrust force and temperature generated during drilling process. The optimal result, as well as the case with highest grey fuzzy reasoning grade (GFRG), indicated that the low feed rate combined with the high twist angle drill in the hybrid cryogenic CO 2 -N 2 gas condition tended to minimize the value of the response output. The verification experiment results showed a good correlation with the optimal analytical result. The analysis of variance results also showed that the reliability of the data was satisfactory.

A new fuzzy rule based multi-objective optimization method for cross-scale injection molding of protein electrophoresis microfluidic chips

“…In order to verify the feasibility and performance improvement effect of the optimal scheme for protein electrophoresis microfluidic chip, the GFG of the optimal scheme must be predicted theoretically and tested in practice after determining the optimal scheme of protein electrophoresis microfluidic chip. The prediction formula of GFG is 38 : where is the total mean value of the AGFGs; denotes the mean value of the best parameter combination in the AGFGs; u is the amount for design variables.…”

“…Different and appropriate membership functions are selected to assign membership degrees for each language item. There are many membership functions that can be applied, for instance: triangular function, gaussian function, trapezoidal function, γ function and sigmoidal membership function, etc 38 . Among these membership functions, the triangular function was widely adopted due to its simplicity and high calculation efficiency in the existing study 42 .…”

“…Therefore, the triangular function and the gaussian function are adopted simultaneously to avoid the situation of the same fuzzy grade for output in this study, namely, the gaussian function is employed for input and the triangular function is employed for output. The triangular and gaussian functions are defined as follows 34 , 37 , 38 : where represents the variable; a, b, and c are the vertices of a triangle; denotes the coefficient of a gaussian function; m is the center of symmetry of a gaussian function.…”

“…In addition, fuzzy logic has been applied to the field of multi-objective optimization in recent years, because of its applicability, simple and flexible. Tran et al employed the mothed of grey fuzzy reasoning grade analysis to execute optimization on carbon fiber–reinforced polymer 38 . Yao et al optimized the beam-like structure by applying the fuzzy decision method 34 .…”

Hierarchical Multi-Objective Optimization of Automobile Seat Frame Based on Grey Fuzzy Logic System

“…Furthermore, FLS may exports continuous control signal based on the discrete membership functions and fuzzy rules. The general process of a FLS includes fuzzification, fuzzy logic inference and defuzzification [54], [55].…”