Continuous phase plate machining error identification method

Abstract

The invention discloses a continuous phase plate machining error identification method, and belongs to the field of phase plate machining error identification. The continuous phase plate machining error identification method adopts two times of matching, the first time of matching is carried out on a measured surface shape and a reference surface shape, roughly positioned translation position coordinates are obtained, then the second time of matching is carried out on the measured surface shape and the reference surface shape in a neighborhood range set according to the translation position coordinates, the error range is decreased, accurately positioned position parameters are obtained, and a machining error indicator is obtained according to an accurately positioned position function. The continuous phase plate machining error identification method overcomes the defect that error identification can not be carried out on a CPP requiring for high accuracy through one time of matching and improves the accuracy of error identification of the CPP through the two-time matching method with the progressively improved accuracy, the RMS value of error identification through the method can be below 30 nm, and the PV value can be below 1 wavelength.

Description

technical field [0001] The invention relates to the field of phase plate processing error identification, in particular to a continuous phase plate processing error identification method. Background technique [0002] In the process of magnetorheological processing of Continuous Phase Plate (CPP) components, generally CPP needs to be processed and inspected multiple times (depending on the surface accuracy requirements of CPP), and finally complete the phase accuracy index, while in the component During the detection process, due to factors such as clamping and positioning, ambient temperature and vibration, the components and the detection optical path are not properly matched, and various positioning errors exist. This kind of error displaces the position of the detected surface shape and the position of the reference surface shape, resulting in the misalignment of the measured surface shape data. If the misplaced surface shape data is imported into the NC software to gene...

AI Technical Summary

Problems solved by technology

However, the conventional image matching recognition method only performs one search, and the conventional algorithm (such as matching normalized product correlation method) for image matching not only has a large amount of calculation but also has a low matching probability.

However, CPP components are very sensitive to errors in the component plane, and a search and conventional algorithms are far from meeting the requirements of CPP high-precision recognition

Images