Stent torsion prediction method and storage medium

Abstract

This application provides a method and storage medium for predicting the torque of a support structure, belonging to the field of mechanical manufacturing technology. The method includes: acquiring target sample data, which includes multiple sets of torque target sample data composed of target sample torque values ​​and factor sampling parameters corresponding to multiple key assembly factors; performing regression analysis on the multiple sets of torque target sample data using a first random forest model to obtain the feature importance degree and basic predicted value of each key assembly factor, whereby the feature importance degree indicates the contribution of the corresponding key assembly factor to the torque variable corresponding to the target sample torque value; determining the target linear logistic regression equation by using each feature importance degree as a coefficient and the basic predicted value as the intercept; predicting the torque of the target support structure according to the target linear logistic regression equation and outputting the torque prediction result. This application embodiment can improve the torque pass-through rate after assembly.

Images