A study on the extreme learning machine based prediction of machining times of the cycloidal gears in CNC milling machines

Abstract

In this study, the machining times of the cycloidal gears manufactured in a CNC milling machine according to the radial machining method are investigated by considering the design and manufacturing parameters of the gear. According to these parameters, the machining times of the cycloidal gears manufactured in Dyna 4M CNC milling machine are determined. It is observed from the experimental manufacturing process that the CAD/CAM parameters of these gears significantly affect the machining times of the gear. In addition, obtained machining times are modeled using extreme learning machine (ELM) which is one of the computational intelligence (CI) algorithms. The proposed ELM model is also compared with the feed-forward and back-propagation based artificial neural network (ANN) algorithm. When both CI methods are compared in terms of modeling performances for training and test phases, it is found that ELM prediction method is extremely fast, accurate and gives higher performance according to ANN method. As a result, ELM method is verified to be able to be used safely to obtain a prediction model for the manufacturing process in a variety of CNC machines where many experiments are required.