In a significant advancement for the mining sector, researchers have tackled a persistent issue in ventilation and dust removal systems that has long plagued comprehensive excavation workfaces. Led by Liu Dandan from the School of Electrical and Control Engineering at Heilongjiang University of Science and Technology, this innovative study, published in Gong-kuang zidonghua (translated as “Mining Automation”), presents a groundbreaking approach to optimizing airflow and reducing dust concentrations, which are critical for both worker safety and operational efficiency.
Traditional long-pressure short suction systems often struggle with vortex formation and dead air zones, leading to inefficiencies and increased dust dispersion. Liu and her team leveraged the Coanda effect—a phenomenon where fluid adheres to surfaces—to redesign the system’s structure. By nesting the exhaust and pressure ducts, they significantly enhanced airflow within the negative pressure duct, ensuring that dust is more effectively contained and removed from the environment.
“Our research demonstrates that by optimizing the structure and pressure-extraction ratio, we can achieve substantial reductions in dust levels, which is a vital concern in mining operations,” Liu stated. The findings revealed an optimal pressure-extraction ratio of 2:3, resulting in a remarkable decrease in dust concentrations—5.56% at the driver’s position and an astonishing 55.41% downwind, compared to traditional systems.
The implications of this research extend beyond mere dust reduction. With the mining industry continuously seeking ways to enhance productivity while ensuring the health and safety of its workforce, the introduction of intelligent parameter control using convolutional neural networks (CNN) marks a pivotal shift. By adjusting key parameters such as the distance between the duct and the dust-producing surface and the positioning of the ducts, the system can adapt to varying conditions, further improving dust removal efficiency.
Liu emphasized the importance of this adaptability, noting, “Our experiments demonstrate that the CNN model significantly outperforms traditional neural networks in predicting dust concentrations, which allows for real-time adjustments and greater stability in operations.” With initial dust concentrations ranging from 300 to 900 mg/m3, the optimized system achieved an impressive average reduction of 51.49% to 83.88%.
As mining companies face increasing pressure to comply with stricter environmental regulations and improve workplace safety, this research not only offers a solution to a pressing problem but also positions companies to enhance their operational efficiency and sustainability. The ability to integrate advanced technologies like CNNs into existing systems can lead to significant cost savings and improved worker conditions, making this a transformative development for the industry.
The study’s findings hold promise for the future of mining operations, paving the way for more intelligent and efficient systems that prioritize both productivity and safety. As the sector continues to evolve, Liu’s work exemplifies how innovative approaches can address longstanding challenges, ultimately reshaping the landscape of mining technology.
For more information on Liu Dandan’s research, visit lead_author_affiliation.
