Shiploader Conveyor Belt No-Load Anomaly Detection

Shiploader conveyor belts consume a large amount of electrical energy when running for a long time in the state of no load, foreign matter in the cargo and less cargo. To enhance the efficiency and energy utilization of the ship loader, this paper proposes a method that combines a contrast algorithm with a coal quantity discrimination module to detect real-time coal quantity for adjusting conveyor belt speed. Initially, a fusion algorithm based on Fourier transform is employed to enhance material images on the conveyor belt. The contrast algorithm compares the similarity of images every 15 adjacent frames to determine the operational state of the conveyor belt. Subsequently, images with coal and without impurities are inputted into the coal quantity discrimination module. After processing such as hole filling and edge feature extraction on coal images, the coal quantity on the conveyor belt is detected. Finally, the conveyor belt speed is adjusted based on the estimated coal quantity feedback. Experimental results demonstrate the effectiveness of this method in accurately discriminating between three different states, achieving an accuracy of 87.72%.