Research on mill distance monitoring system based on machine vision

Abstract

Objective： Real-time monitoring of the grinding roller distance of the pulverizer. Methods： CCD industrial camera was used to collect the rolling distance image, threshold segmentation and morphology were used to preprocess the gray image, and then the edge detection algorithm combined with MATLAB software and mathematical operation were used to eliminate the irrelevant edge points, and finally the rolling distance of the mill was calculated. The feasibility and veracity of the system were verified by comparing the value of gauge and the value of system monitoring. Results： The data fluctuation range of the online monitoring system was 0.001~0.002 mm in the steady state, and the fluctuation range was larger in the speeding state, which was 0.001~0.005 mm. Conclusion： The system can measure rolling distance in real time.

Keywords

• roller flour mill
• rolling distance
• machine vision
• non-contact feeler
• image processing
• monitoring system