综掘面风流调控下的瓦斯与粉尘浓度双目标预测模型研究  

Study on Dual-objective Prediction Mode of Gas and Dust Concentration under Regulation of Airflow in Fully Mechanized Excavation Face

作　　者：龚晓燕[1] 程傲 邹浩 孙育恒 杨富强[1] 张红兵[1] 李昊 孙康 GONG Xiaoyan;CHENG Ao;ZOU Hao;SUN Yuheng;YANG Fuqiang;ZHANG Hongbing;LI Hao;SUN Kang(College of Mechanical Engineering,Xi'an University of Science and Technology,Xi'an 710054,China;Weinan Shaanxi Coal Qichen Technology Co.,Ltd.,Weinan 714000,China)

机构地区：[1]西安科技大学机械工程学院,西安710054 [2]渭南陕煤启辰科技有限公司,陕西渭南714000

出　　处：《煤炭技术》2024年第1期153-157,共5页Coal Technology

基　　金：国家自然科学基金项目(51874235);陕西省自然科学基础研究计划-企业陕煤联合基金项目(2021JLM-01)。

摘　　要：针对综掘面瓦斯和粉尘浓度预测能力不足,导致瓦斯粉尘积聚难以提前解决,造成风筒出风口风流调控降尘排瓦效果不佳的问题,采用层次分析法确定了瓦斯和粉尘浓度分布的关键影响因素,建立了7-11-3结构的双目标预测神经网络模型,并进行出风口距端头5 m和10 m工况下的应用测试,结果表明:模型误差率最大9.85%,最小0.27%。瓦斯浓度最高降低了45%,粉尘浓度最高降低了40%,有效预防了瓦斯和粉尘浓度的积聚问题。In view of the insufficient prediction ability of gas and dust concentration in fully mechanized excavation face,it is difficult to solve the accumulation of gas and dust in advance,which leads to the poor effect of dynamic regulation and control of air flow at the outlet of air duct on dust reduction and tile removal.The key influencing factors of gas and dust concentration distribution are determined by analytic hierarchy process,and a dual-objective prediction mode with 7-11-3 structure is established.The application test is carried out under the working conditions of 5 m and 10 m from the outlet to the head.The results show that the maximum error rate of the model is 9.85%and the minimum is 0.27%.The maximum gas concentration is reduced by 45%,and the maximum dust concentration is reduced by 40%,which effectively prevents the accumulation of gas and dust concentration.

关 键 词：综掘面 风流调控 BP神经网络 瓦斯及粉尘浓度 双目标预测模型 

分 类 号：TD712[矿业工程—矿井通风与安全] TD724TD714