Оценка параметров текстуры прецизионных поверхностей с применением квазиоптимальных корреляционных алгоритмов

Abstract

The authors considered a new method of texture analysis of machined precision surfaces based on using computer optics and the autocorrelation method of processing the images of micro-relief textures under the study. This method is based on a probabilistic comparative evaluation of the unknown texture of the micro-relief under the investigation with the available textures of reference micro-patterns, in which microrelief parameters are determined. The paper proposes an approach to identify the profile surface roughness of a gas turbine engine (GTE) blade after vibro-contact polishing according to the parameters of correlation surface texture. The authors studied the surface micro-geometry of the blade back and pressure side using the optoelectronic complex based on the calculation of the average amplitude of the variable component of an autocorrelation function resulting from computer processing of a surface video image. The application of the electrooptic method for evaluating the surface texture of compressor and turbine blades allows building the surface roughness fields and more deeply analyzing the technology of final processing of the GTE blade feather profile. The relevance and novelty of the study lie in the promising technique to evaluate the surface quality parameters using the electrooptic method. A special feature of this method is the measurement of surface area roughness, while the stylus methods measure the roughness of the surface profile. An important advantage of the proposed method is its application to measure the roughness parameters of a curved surface by a non-contact method, which is advanced since there are surfaces of parts that do not imply being scratched with a diamond needle.