Cooperative monitoring and dynamic data evaluation technology for “borehole-pipe-pump” in coal mine gas drainage systems

Despite advancements in coal mine gas drainage technology, effective monitoring and evaluation of drainage systems remain challenging due to complex geological conditions and limitations in existing sensor-based methods. Traditional approaches rely heavily on manual analysis and fail to provide real-time, intelligent assessments of system performance. To address this gap, this study proposes a novel integrated framework combining physical simulation experiments with an intelligent data-driven evaluation model. Through simulated experiments on the “Borehole-Pipe-Pump” system, we identify characteristic pressure curve patterns under leakage and blockage conditions. Furthermore, we develop a convolutional neural network (CNN)-based model for intelligent classification of drainage effectiveness. Experimental results using real-world data from Wangzhuang Coal Mine show that the model achieves accuracies of 100% for Class I–III, 93% for Class IV, and 50% for Class V under optimal parameters (learning rate = 0.1, batch size = 256). This research provides a reliable and automated solution for real-time monitoring and evaluation of gas drainage systems, with significant implications for mine safety and sustainable energy practices.