Optimization of surface finish in fused deposition modeled components using hybrid metaheuristic techniques and abrasive flow machining

Fused deposition Modelling (FDM) technique is extensively used in industrial applications. The main advantage of this technique is to provide flexibility and ease of operation, but at the same time, surface roughness is a significant drawback to FDM, caused by layer-by-layer deposition resulting in staircase effects and surface irregularities. For this purpose, the present study used Abrasive Flow Machining (AFM) to enhance the surface quality of PLA/Cu composite specimens produced by FDM with a biodegradable abrasive medium made from paper pulp, peanut oil, and stainless-steel particles. Experimental investigations were carried out to analyze the effect of the number of strokes, abrasive concentration, and extrusion pressure on surface roughness. Taguchi L9 experimental data were used to construct a predictive surface roughness model and then combined with a hybrid optimization algorithm, namely Cuckoo Search-Differential Evolution (CS-DE), to determine the optimum parameters of the AFM process. The optimum parameter combination was determined as 100 strokes, 0.00981 MPa extrusion pressure, and 35 wt% abrasive concentrations from the optimization results. In such a situation, the surface roughness predicted was 6.193624 μm, and the confirmation experiment result was 6.30 μm with a prediction error of about 1.71%. The SEM-based surface characterization results confirmed a significant decrease in layer-induced asperities and ridge height, which confirmed an effective surface smoothing process. The results show that the proposed hybrid CS-DE optimization approach offers an effective and reliable methodology to improve the surface finish of the FDM-fabricated parts treated with AFM, and the agreement between the proposed and experimental results is excellent.