Study on the relationship between machining errors and transmission accuracy of planetary roller screw mechanism using analytical calculations and machine-learning model

Abstract Correlation between machining errors and transmission accuracy of planetary roller screw mechanism (PRSM) plays an important role in tolerance design. In this study, analytical calculations, machine learning, experimental verification are utilized for exploring the internal correlation PRSM. A multi-roller meshing error model is established, which comprehensively considers eccentric error, nominal diameter flank angle cumulative pitch screw, roller, nut. The importance coefficients various on determined using random forest algorithm. genetic algorithm-back propagation neural network algorithm-based method training dataset generated via calculations. results show that proposed calculation reflects alternate characteristics rollers during PRSM motion, providing a more accurate prediction than existing methods. For actual mean travel deviation, most significant whereas bandwidth useful travel, formulae considering essential illustrate reasonable accuracy, with average 10.63% deviation 14.27% compared experiments, can effectively support direct design engineering practice.