基于集合经验模态分解和支持向量机的溶解氧预测
|
摘要
应用集合经验模态分解(EEMD)和支持向量机(SVM)相结合的方法,建立一种天然水体溶解氧浓度预测模型。首先,利用EEMD方法将溶解氧时序分解成不同频段的分量,以降低序列的非平稳性;然后,根据各序列分量的自身特征建立合适的SVM预测模型,此过程通过相关分析确定各分量输入量;最后,将各子分量预测值合成得到最终的预测结果。使用该模型对嘉陵江北温泉段的溶解氧浓度进行预测,结果表明,与传统单一的SVM和BP神经网络模型相比,该模型能有效提高预测精密度,具有良好的应用前景。
A prediction model of dissolved oxygen( DO) concentration in natural water was established by a combination of ensemble empirical mode decomposition( EEMD) and support vector machine( SVM). Firstly,DO series were decomposed into several components with different frequency bands by EEMD to reduce the series instability. Secondly,an appropriate prediction model was built for each component of the sequence according to its own characteristics,and the input variables of each component were determined by correlation analysis method. Finally,the predicted value of each component was composed to get the final result. Taking the water quality monitoring sites in the north hot spring reach of Jialing River as an example for DO concentration prediction,results showed that the model had better generalization ability and a good application prospect compared with the traditional single SVM and back propagation neural network.
A prediction model of dissolved oxygen( DO) concentration in natural water was established by a combination of ensemble empirical mode decomposition( EEMD) and support vector machine( SVM). Firstly,DO series were decomposed into several components with different frequency bands by EEMD to reduce the series instability. Secondly,an appropriate prediction model was built for each component of the sequence according to its own characteristics,and the input variables of each component were determined by correlation analysis method. Finally,the predicted value of each component was composed to get the final result. Taking the water quality monitoring sites in the north hot spring reach of Jialing River as an example for DO concentration prediction,results showed that the model had better generalization ability and a good application prospect compared with the traditional single SVM and back propagation neural network.
引文
[1]ZHAI X Y,XIA J,ZHANG Y Y.Water quality variation in the highly disturbed Huai River Basin,China from 1994 to 2005 by multi-statistical analyses[J].Science of the Total Environment,2014,496:594-606.
[2]张思思.基于灰色理论的洱海流域水污染控制研究[D].武汉:华中师范大学,2011.
[3]卢金锁,黄廷林,韩宏大,等.基于决策树技术及在线监测的水质预测[J].环境监测管理与技术,2006,18(2):38-41.
[4]ILJIC'TOMIC'A N,ANTANASIJEVIC'D Z,RISTIC'M D-,et al.Modeling the BOD of Danube River in Serbia using spatial,temporal,and input variables optimized artificial neural network models[J].Environmental Monitoring and Assessment,2016,188(5):1-12.
[5]LI C,SANCHEZ R V,ZURITA G,et al.Multimodal deep support vector classification with homologous features and its application to gearbox fault diagnosis[J].Neurocomputing,2015,168:119-127.
[6]梁坚,何通能.基于小波变换和支持向量机的水质预测[J].计算机应用与软件,2011,28(2):83-86.
[7]张森,石为人,石欣,等.基于偏最小二乘回归和SVM的水质预测[J].计算机工程与应用,2015,51(15):249-254.
[8]KISI O,PARMAR K S.Application of least square support vector machine and multivariate adaptive regression spline models in long term prediction of river water pollution[J].Journal of Hydrology,2016,534:104-112.
[9]徐龙琴,李乾川,刘双印,等.基于集合经验模态分解和人工蜂群算法的工厂化养殖p H值预测[J].农业工程学报,2016,32(3):202-209.
[10]WANG W C,CHAU K W,XU D M,et al.Improving forecasting accuracy of annual runoff time series using ARIMA based on EEMD decomposition[J].Water Resources Management,2015,29(8):2655-2675.
[11]WANG T,ZHANG M C,YU Q H,et al.Comparing the application of EMD and EEMD on time-frequency analysis of seismic signal[J].Journal of Applied Geophysics,2012,83(6):29-34.
[12]SUYKENS J A K,GESTEL T V,BRABANTER J D,et al.Least squares support vector machines[M].Singapore:World Scientific,2002.
[13]BAI Y,WANG P,LI C,et al.A multi-scale relevance vector regression approach for daily urban water demand forecasting[J].Journal of Hydrology,2014,517:236-245.
[14]魏晶茹,马瑜,白冰,等.基于PSO-SVM算法的环境监测数据异常检测和缺失补全[J].环境监测管理与技术,2016,28(4):53-56.
[15]WANG S,YU L,TANG L,et al.A novel seasonal decomposition based least squares support vector regression ensemble learning approach for hydropower consumption forecasting in China[J].Energy,2011,36(11):6542-6554.
[16]李勇,白云,李川.基于小波分析与BP神经网络的PM10浓度预测模型[J].环境监测管理与技术,2016,28(5):24-28.
[17]WU Z,HUANG N E.Ensemble empirical mode decomposition:a noise-assisted data analysis method[J].Advances in Adaptive Data Analysis,2009,1(1):1-41.
[18]BAI Y,LI Y,WANG X,et al.Air pollutants concentrations forecasting using back propagation neural network based on wavelet decomposition with meteorological conditions[J].Atmospheric Pollution Research,2016,7(3):557-566.
[2]张思思.基于灰色理论的洱海流域水污染控制研究[D].武汉:华中师范大学,2011.
[3]卢金锁,黄廷林,韩宏大,等.基于决策树技术及在线监测的水质预测[J].环境监测管理与技术,2006,18(2):38-41.
[4]ILJIC'TOMIC'A N,ANTANASIJEVIC'D Z,RISTIC'M D-,et al.Modeling the BOD of Danube River in Serbia using spatial,temporal,and input variables optimized artificial neural network models[J].Environmental Monitoring and Assessment,2016,188(5):1-12.
[5]LI C,SANCHEZ R V,ZURITA G,et al.Multimodal deep support vector classification with homologous features and its application to gearbox fault diagnosis[J].Neurocomputing,2015,168:119-127.
[6]梁坚,何通能.基于小波变换和支持向量机的水质预测[J].计算机应用与软件,2011,28(2):83-86.
[7]张森,石为人,石欣,等.基于偏最小二乘回归和SVM的水质预测[J].计算机工程与应用,2015,51(15):249-254.
[8]KISI O,PARMAR K S.Application of least square support vector machine and multivariate adaptive regression spline models in long term prediction of river water pollution[J].Journal of Hydrology,2016,534:104-112.
[9]徐龙琴,李乾川,刘双印,等.基于集合经验模态分解和人工蜂群算法的工厂化养殖p H值预测[J].农业工程学报,2016,32(3):202-209.
[10]WANG W C,CHAU K W,XU D M,et al.Improving forecasting accuracy of annual runoff time series using ARIMA based on EEMD decomposition[J].Water Resources Management,2015,29(8):2655-2675.
[11]WANG T,ZHANG M C,YU Q H,et al.Comparing the application of EMD and EEMD on time-frequency analysis of seismic signal[J].Journal of Applied Geophysics,2012,83(6):29-34.
[12]SUYKENS J A K,GESTEL T V,BRABANTER J D,et al.Least squares support vector machines[M].Singapore:World Scientific,2002.
[13]BAI Y,WANG P,LI C,et al.A multi-scale relevance vector regression approach for daily urban water demand forecasting[J].Journal of Hydrology,2014,517:236-245.
[14]魏晶茹,马瑜,白冰,等.基于PSO-SVM算法的环境监测数据异常检测和缺失补全[J].环境监测管理与技术,2016,28(4):53-56.
[15]WANG S,YU L,TANG L,et al.A novel seasonal decomposition based least squares support vector regression ensemble learning approach for hydropower consumption forecasting in China[J].Energy,2011,36(11):6542-6554.
[16]李勇,白云,李川.基于小波分析与BP神经网络的PM10浓度预测模型[J].环境监测管理与技术,2016,28(5):24-28.
[17]WU Z,HUANG N E.Ensemble empirical mode decomposition:a noise-assisted data analysis method[J].Advances in Adaptive Data Analysis,2009,1(1):1-41.
[18]BAI Y,LI Y,WANG X,et al.Air pollutants concentrations forecasting using back propagation neural network based on wavelet decomposition with meteorological conditions[J].Atmospheric Pollution Research,2016,7(3):557-566.