摘要
为准确预测瓦斯涌出量,选取某煤矿的开采煤层、临近煤层、采空区3个瓦斯涌出源作为实例研究,将BP神经网络、粒子群算法(PSO)、Ada Boost迭代提升算法和瓦斯涌出分源预测法相结合,建立基于PSOBP-Ada Boost算法的瓦斯涌出量分源预测模型,并将其与BP神经网络算法进行比较分析。结果表明,PSOBP-Ada Boost算法预测的3个瓦斯涌出源平均相对误差分别为3.24%,2.11%,3.21%;BP神经网络的平均相对误差分别为6.73%,3.19%,4.27%,基于PSOBP-Ada Boost模型的预测精度明显优于BP神经网络模型。
In order to predict the quantity of gas emission from different sources in a mine,the coal seam being mined,the adjacent coal seam and the gob of a certain coal mine in China were taken as the gas emission sources,and a PSOBP-Ada Boost based model was built for predicting the quantity. Effectiveness of the model was verified with the data acquired in the mine.It was found that accuracy of prediction of the model is better than that of the BP neural network model.
引文
[1]任晓奎,侯洪涛,江海朋.基于模糊粗糙集与支持向量机的瓦斯涌出量预测研究[J].计算机测量与控制,2012,20(9):2 369-2 371.REN Xiaokui,HOU Hongtao,JIANG Haipeng.Research on forecasting of gas emission quantity[J].Computer Measurement and Control,2012,20(9):2 369-2 371.
[2]杨武艳,郁钟铭.GM(1,1)灰色预测模型在矿井瓦斯涌出量预测中的应用[J].矿业工程研究,2012,27(4):46-49.YANG Wuyan,YU Zhongming.Application of GM(1,1)grey forecasting model in mine gas emission prediction[J].Mineral Engineering Research,2012,27(4):46-49.
[3]王巍,刘德胜.基于支持向量机理论的煤矿瓦斯涌出量预测研究[J].煤矿机械,2011,32(2):78-80.WANG Wei,LIU Desheng.Prediction of emission of coal mine gas based on support vector machine theory[J].Coal Mine Machinery,2011,32(2):78-80.
[4]王生全,刘柏根,张召召,等.基于遗传算法BP网络模型进行瓦斯涌出量预测[J].西安科技大学学报,2012,32(1):51-56.WANG Shengquan,LIU Baigen,ZHANG Zhaozhao,et al.Prediction of gas emission quantity of mining faces based on genetic BP neutral network optimal model[J].Journal of Xi'an University of Science and Technology,2012,32(1):51-56.
[5]魏春荣,李艳霞,孙建华,等.灰色-分源预测法对煤矿瓦斯涌出量的应用研究[J].采矿与安全工程学报,2013,30(4):628-632.WEI Chunrong,LI Yanxia,SUN Jianhua,et al.Gas emission rate prediction in coal mine by grey and separated sources prediction method[J].Journal of Mining and Safety Engineering,2013,30(4):628-632.
[6]李松,解永乐,王文旭.Ada Boost_BP神经网络在铁路货运量预测中的应用[J].计算机工程与应用,2012,48(6):233-234.LI Song,XIE Yongle,WANG Wenxu.Application of Ada Boost neural network in prediction of railway freight volumes[J].Computer Engineering and Applications,2012,48(6):233-234.
[7]柳玉,郭虎全.基于Ada Boost与BP神经网络的风速预测研究[J].电网与清洁能源,2012,28(2):80-83.LIU Yu,GUO Huquan.Wind speed prediction based on Ada Boost and BP neural networks[J].Power System and Clean Energy,2012,28(2):80-83.
[8]王改革,郭立红,段红,等.基于Elman_Ada Boost强预测器的目标威胁评估模型及算法[J].电子学报,2012,40(5):901-906.WANG Gaige,GUO Lihong,DUAN Hong,et al.The model and algorithm for the target threat assessment based on Elman-Adaboost strong predictor[J].ACTA Electronic Sinca,2012,40(5):901-906.
[9]颜爱华.煤层瓦斯含量多源数据分析及其预测研究[D].北京:中国矿业大学(北京),2010.YAN Aihua.Study on multisource data analysis and prediction of gas content[D].Beijing:China University of Mining&Technology(Beijing),2010.
[10]付华,谢森,徐耀松,等.基于ACC-ENN算法的煤矿瓦斯涌出量动态预测模型研究[J].煤炭学报,2014,39(7):1 296-1 301.FU Hua,XIE Sen,XU Yaosong,et al.Gas emission dynamic prediction model of coal mine based on ACC-ENN algorithm[J].Journal of China Coal Society,2014,39(7):1 296-1 301.
[11]张强,贾宝山,董晓蕾,等.PCA-GA-SVM的回采工作面瓦斯涌出量预测[J].辽宁工程技术大学学报:自然科学版,2015,34(5):572-577.ZHANG Qiang,JIA Baoshan,DONG Xiaolei,et al.Working face gas emission prediction based on PCA-GA-SVM[J].Journal of Liaoning Technical University:Natural Science,2015,34(5):572-577.
[12]薛鹏骞,吴立锋,李海军.基于小波神经网络的瓦斯涌出量预测研究[J].中国安全科学学报,2006,16(2):22-25.XUE Pengqian,WU Lifeng,LI Haijun.Predicting the amount of gas emitted based on wavelet neural network[J].China Safety Science Journal,2006,16(2):22-25.
[13]崔洪庆,樊帅帅,关金锋,等.采煤工作面瓦斯涌出量分源计算[J].中国安全科学学报,2015,25(10):78-83.CUI Hongqing,FAN Shuaishuai,GUAN Jinfeng,et al.Calculation of methane emission from different sources at coal mining face based on monitoring data[J].China Safety Science Journal,2015,25(10):78-83.
[14]温廷新,张波,邵良杉.矿井突水水源识别预测研究:以新庄孜矿为例[J].中国安全科学学报,2014,24(2):100-106.WEN Tingxin,ZHANG Bo,SHAO Liangshan.Research on prediction of mine water inrush source identification:Xinzhuangzi coalfield as an example[J].China Safety Science Journal,2014,24(2):100-106.
[15]温廷新,张波,邵良杉.矿井突水水源识别的QGA-LSSVM模型[J].中国安全科学学报,2014,24(7):111-115.WEN Tingxin,ZHANG Bo,SHAO Liangshan.QGA-LSSVM model for mine water inrush source identification[J].China Safety Science Journal,2014,24(7):111-115.