Toolpath Optimization by Real-time Application of an Integrated Geometric/Mechanistic Model

点击翻译按钮获取中文摘要

DMI-9872575Jerard The goal of this research is to improve the productivity of Numerically Controlled (NC) machine tools. Models of the cutting process will be integrated with the NC machine controller. Sensor feedback will be used to monitor the process and continuously improve the model accuracy through real-time self-calibration. Machine tool controllers which use these methods will be capable of using high-level part descriptions for input as opposed to the low-level "G-codes" that are currently in common usage - an attribute that should greatly decrease the overall time required to produce parts on NC machines. Specific elements of the research include: (1) Development of an experimental test facility by retrofitting a 5-axis NC milling machine with an Open Architecture Controller (OAC). (2) Integrated geometric/mechanistic models of cutting on the OAC which are efficient enough to run in "real-time" mode. The models will be used to calculate optimal cutting conditions. Sensors on the machine tool will be used to perform continuous self-calibration of model parameters. (3) Continuous real-time comparison of actual and model-predicted cutting forces and the development of corrective actions to compensate for excessive differences. (4) Experimental comparisons between these methods and current "best practice" approaches. Complex sculptured surface programs developed by expert NC programmers at an industrial partner will be used. The research should lead to improved NC machining performance compared with the present NC control, in terms of machining productivity and product quality. The open-architecture controller on a 5-axis NC milling machine enables testing of the model in an environment close to real industrial setting.