Алгоритм регистрации характеристик качества по параметрам вибраций технологического оборудования

Abstract

The paper presents an algorithm for recording the quality characteristics of the milling process based on the vibration parameters of technological equipment. Multilayer discrete wavelet decomposition and calculation of statistical features in a sliding window were used to analyse the signals, on the basis of which the classification of cutting modes was carried out using machine learning methods. Experimental studies have shown high accuracy of cutting speed identification by vibration features, in particular, when using Random Forest classifier.