A method for real-time trajectory compensation based on measurement data

Abstract

The present invention discloses a method for implementation of real-time track compensation based on measurement data. The method comprises the steps of: measuring and calculating deformation of the position of a measurement point, employing the obtained deformation and the position of the measurement point to train a neural network model, taking the path point position on a theory track into theneural network model to obtain an output function value, namely deformation of the path point position, and overlapping the deformation and a theory path to obtain an actual path. The method can effectively avoid path inflection points, and is simple to obtain a compensation value and more accurate in the commemoration value.

Description

Technical field [0001] The invention belongs to the field of numerical control processing, and particularly relates to a tool path deviation compensation method for contour processing of a workpiece. Background technique [0002] At present, for contour and chamfering processing, due to the influence of external force, such as uneven force in the clamping state, it is easy to cause contour deformation. The processing requirements must meet the requirements of each cutting amount, otherwise uneven chamfer width will occur. The existing numerical control system generally adds macro variables to the original program based on the measured data, predicts the product change trend with a linear model, and uses the filtering function to achieve precise chamfering, but this method often ignores the large clamping deformation At the same time, when there is a large deviation between the actual measured position and the reference position, a tiny shape path inflection point may appear at t...

AI Technical Summary

Problems solved by technology

Existing CNC systems generally add macro variables to the original program based on measured data, predict product change trends with a linear model, and use filtering functions to achieve accurate edge processing, but this method often ignores large clamping deformation. At the same time, when there is a large deviation between the measured position and the reference position, there may be a small shape of the path inflection point at the measurement node, resulting in ribs on the processing surface during processing and affecting the surface quality; and the other through the spline curve The compensation method of the model is used to improve the above existing problems, but because the spline curve has no definite model calculation, it is more troublesome to obtain the compensation value of each point on the workpiece through graphic mapping

Images