Intelligent monitoring, analyzing and early warning system for underground coal mine safety and method thereof

Abstract

This invention relates to the field of coal mine dynamic disaster monitoring and early warning technology, specifically to an intelligent monitoring, analysis, and early warning system and method for underground coal mine safety. It includes: real-time acquisition of equipment operating parameters of intelligent fully mechanized mining equipment within the working face and actual monitoring signals from physical sensors; calculation of theoretical prediction signals corresponding to the physical sensors based on the equipment operating parameters; generation of residual data streams based on the actual monitoring signals and theoretical prediction signals; construction of a spatiotemporal data tensor for a preset spatial region within a preset time window based on the residual data stream; processing the spatiotemporal data tensor through a pre-trained convolutional neural network model to calculate the disaster risk index for the preset spatial region; and determination and output of the final early warning level based on the disaster risk index and a preset risk index threshold. This method significantly improves monitoring sensitivity and signal-to-noise ratio, reduces false alarms caused by changes in equipment operating conditions, and makes early warning results more accurate and reliable.

Images