Efficient identification of cutter axis offset in five-axis ball-end interrupted milling using twin data method free from cutting force model

Abstract

Although many methods for the identification of cutter axis offset have been proposed, almost all approaches are based on the computation models of cutting force. The mechanical behavior of the cutting tool cannot always be completely described by the existing force models. Once the cutting force is calculated inaccurately, the identification of cutter axis offset certainly is affected. In order to get rid of dependence on cutting force model, this paper presents a twin data driven model for the efficient identification of cutter axis offset in five-axis ball-end interrupted milling. The cutter coupled motion is divided into the two decouple standard movement units at first. The measured feature parameter of the axis offset is then extracted from cutting force signal by the geometric modeling technology. Subsequently, the theoretical and measured critical cutting positions are defined as a pair of twins. The axis offset parameters are identified by minimizing the distance of the twin data using the intelligent optimization algorithm. Lastly, the effectiveness of the proposed method is verified by the numerical examples and cutting experiments performed in five-axis ball-end milling.