摘要
为了更有效合理地解决煤矿工作面中低氧问题,以神东某煤矿工作面监测数据为样本,考虑监测物理参数之间的相互影响关系,借助主成分分析法对广义回归神经网络(GRNN)进行改进,构建工作面氧气浓度预测模型,编制改进的GRNN模型程序。将预测氧气浓度结果与实测数据对比,证明改进后的GRNN模型具有良好的拟合准确度和泛化能力,比改进前GRNN模型和BP神经网络模型更适合于煤矿工作面低氧问题的预测;利用改进的GRNN模型分析了工作面进、回风压力及进风温度对工作面及回风平巷氧浓度的影响,为矿井工作面低氧预测及工作面低氧防治技术提供了参考。
In order to solve the problem of working face hypoxia in coal mine more effectively and reasonably,an improved general neural network(GRNN)model for prediction of oxygen concentration in coal mine was constructed,by taking the monitoring data of a working face in Shendong as samples and considering the interaction relationship between physical parameters,based on principal component analysis.Comparing the predicted oxygen concentration results with the measured data,it proves that the improved GRNN model has good fitting accuracy and generalization ability.By using the improved GRNN model,the original GRNN model and BP neural network model respectively in the comparative analysis of hypoxia problems,it found that the improved GRNN model has better effects and is more suitable for the prediction of hypoxia problems in coal mine face.The influence of inlet air pressure,outlet air pressure and inlet air temperature on the oxygen concentration were analyzed by the improved GRNN model.This improved GRNN model can give a reference to hypoxia prediction and hypoxia control technology of the working face.
作者
杨小彬
王逍遥
周世禄
张子鹏
Yang Xiaobin;Wang Xiaoyao;Zhou Shilu;Zhang Zipeng(School of Emergency Management and Safety Engineering,China University of Mining and Technology,Beijing 100083,China)
出处
《矿业科学学报》
2019年第5期434-440,共7页
Journal of Mining Science and Technology
基金
国家自然科学基金(51274207)
国家自然科学基金青年基金(50904071)
作者简介
杨小彬(1976—),男,重庆人,副教授,博士生导师,主要从事矿山动力灾害防治的研究工作。Tel:13522856957,E-mail:yangxiaobin02@126.com。
