In the heart of China’s coal mining industry, a technological breakthrough is poised to revolutionize safety and efficiency. Researchers at the College of Artificial Intelligence & Computer Science, Xi’an University of Science and Technology, have developed an intelligent method for identifying face pressure, a critical factor in preventing roof disasters in coal mines. Led by LUO Xiangyu, the team’s innovative approach promises to enhance the accuracy and adaptability of face pressure detection, potentially saving lives and reducing operational downtime.
Face pressure, the stress exerted on the coal face during mining, is a significant concern in coal extraction. Traditional methods of identifying face pressure rely on fixed rules and empirical data, which often fall short in complex mining environments. LUO Xiangyu and his team have addressed this challenge by transforming the problem into a binary classification task, leveraging the power of machine learning.
“The key was to extract the cycle-end resistance data from the vast amounts of working resistance data and then use this data to accurately determine face pressure,” LUO explained. The team employed a support vector machine (SVM) classifier to intelligently detect the end of coal cutting cycles, a process that previously depended on manual observation and rule-based systems.
The SVM classifier automatically identifies the end of each coal cutting cycle, allowing for the extraction of cycle-end resistance data. This data is then fused to create a single sequence that reflects the overall pressure state of the working face. The team developed a pressure judgment formula to determine face pressure based on this sequence.
The results are impressive. In experiments conducted on hydraulic support working resistance data from a non-contiguous coal mine, the method achieved precision, recall, and F1 scores of 85.91%, 81.84%, and 83.83%, respectively, for coal cutting cycle detection. For face pressure identification, the scores were 79.43%, 78.76%, and 79.09%, respectively. These figures outperform traditional methods, such as the sliding window extreme value method and threshold method, demonstrating significant advantages in cycle-end resistance identification and face pressure judgment.
The implications for the energy sector are substantial. Accurate and adaptive face pressure identification can lead to improved roof disaster prevention and control, enhancing miner safety and reducing the risk of operational disruptions. This technology could also optimize the use of hydraulic supports, extending their lifespan and reducing maintenance costs.
As the mining industry continues to embrace digital transformation, intelligent detection methods like this one are set to play a pivotal role. The research, published in Gong-kuang zidonghua, or “Mechanical Science and Technology,” marks a significant step forward in the application of artificial intelligence in coal mining. It opens the door to further innovations, such as real-time monitoring systems and predictive analytics, which could reshape the future of coal extraction.
The success of LUO Xiangyu’s research underscores the potential of interdisciplinary collaboration in addressing complex industrial challenges. As the energy sector navigates the demands of safety, efficiency, and sustainability, technologies like this one will be instrumental in driving progress. The future of coal mining is increasingly intelligent, and this breakthrough is a testament to that trend.