Estimating of cutting force and surface roughness in turning of GFRP composites with different orientation angles using artificial neural network

Abstract

Abstract Glass fiber-reinforced polymer (GFRP) composite materials are widely used in many manufacturing industries due to their low density and high strength properties, and consequently, the need for precision machining of such composites has significantly increased. Since composite materials have an anisotropic and heterogeneous structure, the machinability of composite materials is quite different from conventional materials. In the machining of GFRP composite pipes, tool wear, cracks or delamination, a rough surface, etc., many unwanted problems may occur. Therefore, GFRP composite pipes are difficult to process. To prevent such problems, it is very crucial to select suitable process parameters, thereby achieving the maximum performance for the desired dimensional integrity. In this study, through turning of GFRP composites with different orientation angles (30°, 60°, and 90°), the effects of cutting speed (50, 100, and 150 m·min−1), feed rate (0.1, 0.2, and 03 mm·rev−1), and depth of cut (1, 2, and 3 mm) on cutting force and surface roughness were determined. Then, with the use of these machining parameters, a model of the system for determining cutting force and surface roughness was established with artificial neural networks (ANNs). The ANN was trained using Levenberg–Marquardt backpropagation algorithm. It has been observed that the results obtained with the ANN model are very close to the data found in experimental studies. In both experimental and model-based analysis, minimum cutting force (44 N) and surface roughness (2.22 µm) were achieved at low fiber orientation angle (30°), low feed rate (0.1 mm·rev−1), and depth of cut (1 mm) at high cutting speeds (150 m·min−1).