Innovative Application of Intelligent Perception and Optimal Control in the Coal Washing Process
Abstract：To address issues such as low clean coal yield, high energy consumption, high equipment failure rate, and high safety risks in traditional coal washing plants, Shandong Sanhekou Mining Co., Ltd.'s coal washing plant has adopted intelligent perception and optimal control technologies. By using image recognition technology, a run-of-rough monitoring system was established; based on machine vision, a pulp surface gray-scale feature analysis system was developed; an MPC control algorithm was designed; a multi-source data fusion intelligent start-stop system was constructed; a equipment health prediction model was established using wavelet transform and LSTM neural networks; and precise identification of unsafe personnel behavior was achieved by combining UWB positioning with YOLOv5 algorithm. The application results show that the clean coal yield increased by 0.91%, the power consumption per ton of coal decreased by 1.8 kW·h, the equipment failure rate dropped by 13%, the fluctuation of heavy medium separation density was reduced to ±0.003 g/cm³, and continuous safe production reached 436 days, creating an annual economic benefit of 3.85 million yuan. This provides a technical model for the intelligent transformation of the industry.
Machine-generated Keywords：
Publication Date：2025-10-30
Online Publishing Date：2025-11-21(First online date of this platform, not the publication date of the document)
Pages：4( 68-71 )