Method of Online Measurement and Assessment for the Deformation of Mining Ventilator Blade Based on Data Driven

In order to solve the problem of spatial geometric state analysis of mining ventilator blade, a method of online measurement and assessment for the blade deformation based on data driven is proposed, and the measurement principle and evaluation model of the method are discussed in detail. Firstly, based on the analysis of the geometric parameters and mechanical properties of ventilator blade, a complete index system for the characterization of blade geometric deformation is established, and the physical connotation and definition of each index quantity are clarified; Secondly, a measurement method of blade deformation based on two-dimensional laser technology is proposed, the geometric information of blade deformation is collected online using two-dimensional laser contour sensor, and the high-precision, high-efficiency and non-contact measurement of coordinate data is achieved by optimizing the position, attitude and measurement mode of the sensor; then, a data-driven method for blade deformation assessment is studied, the index quantities of blade deformation are extracted based on the raw coordinate data collected of blade profile using methodologies of the least squares fitting of quartic polynomial, curvature analysis and feature recognition, coordinate transformation and least squares matching algorithm, and the comprehensive characterization and evaluation of the amount of blade deformation are accomplished; finally, the experimental research is implemented, and the results show that the coordinate measurement accuracy reaches 2 μm level, the curve fitting error is controlled within ± 1%, which effectively solves the problem of high-precision extraction of measurement model for the blade profile. This method has the characteristics of non-contact, high efficiency and intelligence and provides a comprehensive technical solution for measuring and evaluating the deformation of mining ventilator blade.