Grinding Force Prediction Model by Discretizing Stochastic Grains

Abstract

Abstract Material removal process during grinding is the accumulative effect of numerous stochastic grains’ asperity interactions with workpiece, leading to extreme difficulty to predict grinding force. The existing analytical grinding force prediction models primarily focus on representing the stochastic interactions in contact zone using average effective grains number and undeformed chip thickness or assuming ideal distribution law for the stochastic characteristics of grains, which can result in relatively large error in predicting grinding force, whereas the experimentally empirical methods are usually time-consuming and laborious. To overcome these drawbacks, the research proposed a hybrid methodology to predict grinding force by means of discretizing stochastic grains, matrix calculation and two-dimension (2D) micro-grinding simulations. Consequently, grains’ stochastic geometry, size, position distribution, and corresponding undeformed chip size in contact region were considered to guarantee the authenticity and credibility of the predicted grinding forces. Firstly, the stochastic interaction between a random grain and workpiece was converted equivalently to a series of plane-cutting processes between micro-edges with different rake angles and micro-chip layers with corresponding thicknesses. Then, the rake angle matrix of micro edges and the corresponding cutting depth matrix in contact zone were identified through matrix calculation method based on single grain discretization analysis. Followed, both matrixes were incorporated into the plane-cutting force interpolant function derived from the finite element analysis (FEA) to achieve the grinding force matrix, and further the resultant grinding force along normal and tangential directions. The cross validation with experimental grinding results in literature showed the prediction error falls within 20% to verify the authenticity of proposed model. The model considers both influences of the stochastic grain distribution in grinding tool and their interactions with workpiece. Therefore, it bears significance in improving the prediction accuracy of grinding force, and promoting a better understanding of inherent logic from micro-level abrasive-workpiece interactions to macro-level grinding force prediction during the abrasive-based machining including grinding, honing, and polishing.