Development of accurate estimation method of machining time in consideration of characteristics of machine tool

Abstract

In order to perform accurately scheduling of machining using a machine tool, it is necessary to estimate the actual machining time. The machining time is generally estimated by CAM systems. However, the error between the estimated and real cutting times is considerable because the systems do not consider the control and functional characteristics of the machine tool. In addition, control functions are installed in machine tools to achieve high precision and high speed motion while optimizing the tool paths and control parameters. The functions significantly affect the machining time. However, estimating machining time is difficult, thereby complicating optimization process. In this study, a method to identify the control characteristics and actual tool paths and a system to estimate the cutting time were developed. Furthermore, an estimation system using a deep neural network (DNN) was constructed to incorporate a control function. Finally, verification experiments were conducted wherein the estimation accuracy of the machining time was found to be within 5%.