基于WT与GALSSVM的瓦斯涌出量预测
|
摘要
为了准确预测工作面瓦斯涌出量,加强煤矿安全生产,基于小波变换(WT)和进化最小二乘支持向量机(GALSSVM),建立了瓦斯涌出量的新型预测模型.首先,通过小波分解将瓦斯涌出量时间序列分解成具有不同频率特征的信号;然后利用互信息法和伪近邻法得到的时间延迟和嵌入维数对各信号进行相空间重构;之后根据各个相空间的特点建立相应的GALSSVM预测模型;最后把各信号的预测结果进行小波重构,作为最终的瓦斯涌出量的预测结果.以晋城市成庄矿2315综放工作面瓦斯涌出量为例,进行了预测研究.实例表明,该方法具有很高的预测精度和较强的泛化能力.
In order to forecast the gas emission volume from working face and ensure the coalmine safety production,a novel model based on wavelet transform(WT) and genetic algorithm-least square support vector machine(GALSSVM) for gas emission forcast is presented.Firstly,gas emission time series is decomposed into different frequency signals through the wavelet transform;Secondly,phase space of each signals is reconstructed,and time delay and embeding dimension are determined by mutual information method and false nearest neighbor mehod respectively.Then,the respective forcasting model of GALSSVM is constructed according to different characteristics of each phase space.Lastly,the predicted results of the signals are reconstructed to be used as the final prediction result of gas emission.Gas emission from working face No.2315 of Cheng zhuang mine in Jing cheng city was predicted by using this model.The calculation result shows that this model has a higher forecasting precision and greater generality ability.
In order to forecast the gas emission volume from working face and ensure the coalmine safety production,a novel model based on wavelet transform(WT) and genetic algorithm-least square support vector machine(GALSSVM) for gas emission forcast is presented.Firstly,gas emission time series is decomposed into different frequency signals through the wavelet transform;Secondly,phase space of each signals is reconstructed,and time delay and embeding dimension are determined by mutual information method and false nearest neighbor mehod respectively.Then,the respective forcasting model of GALSSVM is constructed according to different characteristics of each phase space.Lastly,the predicted results of the signals are reconstructed to be used as the final prediction result of gas emission.Gas emission from working face No.2315 of Cheng zhuang mine in Jing cheng city was predicted by using this model.The calculation result shows that this model has a higher forecasting precision and greater generality ability.
引文
[1]梁华珍.工作面瓦斯涌出量预测的研究与应用[D].淮南:安徽理工大学能源与安全学院,2007.
[2]朱红青,常文杰,张彬.回采工作面瓦斯涌出BP神经网络分源预测模型及应用[J].煤炭学报,2007,32(5):504-508.ZHU Hong-qing,CHANG Wen-jie,ZHANG Bin.Different-source gas emission prediction model ofworking face based on BP artificial neural networkand its application[J].Journal of China Coal Society,2007,32(5):504-508.
[3]施式亮,伍爱友.GM(1,1)模型与线性回归组合方法在矿井瓦斯涌出量预测中的应用[J].煤炭学报,2008,33(4):415-418.SHI Shi-liang,WU Ai-you.Application of GM(1,1)and line regression for predicting amount of mine gasemission in coal mine[J].Journal of China Coal Soci-ety,2008,33(4):415-418.
[4]张剑英,程健,侯玉华,等.煤矿瓦斯浓度预测的ANFIS方法研究[J].中国矿业大学学报,2007,36(4):494-498.ZHANG Jian-ying,CHENG Jian,HOU Yu-hua,etal.Forecasting coalmine gas concentration based onadaptive neuro-fuzzy inference system[J].Journal ofChina University of Mining&Technology,2007,36(4):494-498.
[5]赵洪波.岩土力学与工程中的支持向量机分析[M].北京:煤炭工业出版社,2008:1-10.
[6]廖建朝,谢正文.灰色支持向量机在瓦斯流量预测中的应用[J].工业安全与环保,2008,34(8):27-29.LIAO Jian-chao,XIE Zheng-wen.The application ofgray support vector machine in gas flow forecast[J].Industrial Safety and Environmental Protection,2008,34(8):27-29.
[7]郝仲熙.基于分形理论的煤矿瓦斯涌出规律预测研究[D].太原:太原理工大学矿业工程学院,2006.
[8]郑治真,沈萍,杨选辉,等.小波变换及其MAT-LAB工具的应用[M].北京:地震出版社,2001:35-106.
[9]SUYKENS J A K,VANDEWALLE J.Least squares support vector machines[J].Neural Pro-ceeding Letters,1999,9(3):293-300.
[10]陈帅,朱建宁,潘俊,等.最小二乘支持向量机的参数优化及其应用[J].华东理工大学学报:自然科学版,2008,34(2):278-282.CHEN Shuai,ZHU Jian-ning,PAN Jun,et al.Pa-rameters optimization of LS-SVM and its application[J].Journal of East China University of Science andTechnology:Natural Science Edition,2008,34(2):278-282.
[2]朱红青,常文杰,张彬.回采工作面瓦斯涌出BP神经网络分源预测模型及应用[J].煤炭学报,2007,32(5):504-508.ZHU Hong-qing,CHANG Wen-jie,ZHANG Bin.Different-source gas emission prediction model ofworking face based on BP artificial neural networkand its application[J].Journal of China Coal Society,2007,32(5):504-508.
[3]施式亮,伍爱友.GM(1,1)模型与线性回归组合方法在矿井瓦斯涌出量预测中的应用[J].煤炭学报,2008,33(4):415-418.SHI Shi-liang,WU Ai-you.Application of GM(1,1)and line regression for predicting amount of mine gasemission in coal mine[J].Journal of China Coal Soci-ety,2008,33(4):415-418.
[4]张剑英,程健,侯玉华,等.煤矿瓦斯浓度预测的ANFIS方法研究[J].中国矿业大学学报,2007,36(4):494-498.ZHANG Jian-ying,CHENG Jian,HOU Yu-hua,etal.Forecasting coalmine gas concentration based onadaptive neuro-fuzzy inference system[J].Journal ofChina University of Mining&Technology,2007,36(4):494-498.
[5]赵洪波.岩土力学与工程中的支持向量机分析[M].北京:煤炭工业出版社,2008:1-10.
[6]廖建朝,谢正文.灰色支持向量机在瓦斯流量预测中的应用[J].工业安全与环保,2008,34(8):27-29.LIAO Jian-chao,XIE Zheng-wen.The application ofgray support vector machine in gas flow forecast[J].Industrial Safety and Environmental Protection,2008,34(8):27-29.
[7]郝仲熙.基于分形理论的煤矿瓦斯涌出规律预测研究[D].太原:太原理工大学矿业工程学院,2006.
[8]郑治真,沈萍,杨选辉,等.小波变换及其MAT-LAB工具的应用[M].北京:地震出版社,2001:35-106.
[9]SUYKENS J A K,VANDEWALLE J.Least squares support vector machines[J].Neural Pro-ceeding Letters,1999,9(3):293-300.
[10]陈帅,朱建宁,潘俊,等.最小二乘支持向量机的参数优化及其应用[J].华东理工大学学报:自然科学版,2008,34(2):278-282.CHEN Shuai,ZHU Jian-ning,PAN Jun,et al.Pa-rameters optimization of LS-SVM and its application[J].Journal of East China University of Science andTechnology:Natural Science Edition,2008,34(2):278-282.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心