Prediction of Coal Face Gas Concentration by Multi-Scale Selective Ensemble Hybrid Modeling
A selective ensemble hybrid modeling prediction method based on wavelet transformation is proposed to improve the fitting and generalization capability of the existing prediction models of the coal face gas concentration, which has a strong stochastic volatility. Mallat algorithm was employed for...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Eastern Macedonia and Thrace Institute of Technology
2014-06-01
|
| Series: | Journal of Engineering Science and Technology Review |
| Subjects: | |
| Online Access: | http://www.jestr.org/downloads/Volume7Issue2/fulltext87214.pdf |
| _version_ | 1819014539153244160 |
|---|---|
| author | WU Xiang YANG Zhong-ru Zhang Li Pilati Silvia |
| author_facet | WU Xiang YANG Zhong-ru Zhang Li Pilati Silvia |
| author_sort | WU Xiang |
| collection | DOAJ |
| description | A selective ensemble hybrid modeling prediction method based on wavelet transformation is proposed to improve the
fitting and generalization capability of the existing prediction models of the coal face gas concentration, which has a
strong stochastic volatility. Mallat algorithm was employed for the multi-scale decomposition and single-scale
reconstruction of the gas concentration time series. Then, it predicted every subsequence by sparsely weighted multi
unstable ELM(extreme learning machine) predictor within method SERELM(sparse ensemble regressors of ELM). At
last, it superimposed the predicted values of these models to obtain the predicted values of the original sequence. The
proposed method takes advantage of characteristics of multi scale analysis of wavelet transformation, accuracy and fast
characteristics of ELM prediction and the generalization ability of L1 regularized selective ensemble learning method.
The results show that the forecast accuracy has large increase by using the proposed method. The average relative error is
0.65%, the maximum relative error is 4.16% and the probability of relative error less than 1% reaches 0.785. |
| first_indexed | 2024-12-21T02:17:27Z |
| format | Article |
| id | doaj.art-499c324f724344bb9cd858b1c43e5461 |
| institution | Directory Open Access Journal |
| issn | 1791-2377 1791-2377 |
| language | English |
| last_indexed | 2024-12-21T02:17:27Z |
| publishDate | 2014-06-01 |
| publisher | Eastern Macedonia and Thrace Institute of Technology |
| record_format | Article |
| series | Journal of Engineering Science and Technology Review |
| spelling | doaj.art-499c324f724344bb9cd858b1c43e54612022-12-21T19:19:14ZengEastern Macedonia and Thrace Institute of TechnologyJournal of Engineering Science and Technology Review1791-23771791-23772014-06-01725359Prediction of Coal Face Gas Concentration by Multi-Scale Selective Ensemble Hybrid ModelingWU Xiang0YANG Zhong-ru 1Zhang Li 2Pilati Silvia 3School of Information and Electrical Engineering, China University Mining & Technology, Xuzhou, Jiangsu 221116, China / School of Medical Informatics,Xuzhou Medical College, Xuzhou, Jiangsu 221116, China2 China Huajin Coal Energy Co., Ltd. Hejin, Shanxi 043300, ChinaSchool of Medical Imaging, Xuzhou Medical College, Xuzhou, Jiangsu 221009, ChinaDepartment of Electrical and Computer Engineering, University of Massachusetts, Amherst, MA 01003, United StatesA selective ensemble hybrid modeling prediction method based on wavelet transformation is proposed to improve the fitting and generalization capability of the existing prediction models of the coal face gas concentration, which has a strong stochastic volatility. Mallat algorithm was employed for the multi-scale decomposition and single-scale reconstruction of the gas concentration time series. Then, it predicted every subsequence by sparsely weighted multi unstable ELM(extreme learning machine) predictor within method SERELM(sparse ensemble regressors of ELM). At last, it superimposed the predicted values of these models to obtain the predicted values of the original sequence. The proposed method takes advantage of characteristics of multi scale analysis of wavelet transformation, accuracy and fast characteristics of ELM prediction and the generalization ability of L1 regularized selective ensemble learning method. The results show that the forecast accuracy has large increase by using the proposed method. The average relative error is 0.65%, the maximum relative error is 4.16% and the probability of relative error less than 1% reaches 0.785.http://www.jestr.org/downloads/Volume7Issue2/fulltext87214.pdfGas ConcentrationMulti-scaleSelective Ensemble LearningHybrid Modeling |
| spellingShingle | WU Xiang YANG Zhong-ru Zhang Li Pilati Silvia Prediction of Coal Face Gas Concentration by Multi-Scale Selective Ensemble Hybrid Modeling Journal of Engineering Science and Technology Review Gas Concentration Multi-scale Selective Ensemble Learning Hybrid Modeling |
| title | Prediction of Coal Face Gas Concentration by Multi-Scale Selective Ensemble Hybrid Modeling |
| title_full | Prediction of Coal Face Gas Concentration by Multi-Scale Selective Ensemble Hybrid Modeling |
| title_fullStr | Prediction of Coal Face Gas Concentration by Multi-Scale Selective Ensemble Hybrid Modeling |
| title_full_unstemmed | Prediction of Coal Face Gas Concentration by Multi-Scale Selective Ensemble Hybrid Modeling |
| title_short | Prediction of Coal Face Gas Concentration by Multi-Scale Selective Ensemble Hybrid Modeling |
| title_sort | prediction of coal face gas concentration by multi scale selective ensemble hybrid modeling |
| topic | Gas Concentration Multi-scale Selective Ensemble Learning Hybrid Modeling |
| url | http://www.jestr.org/downloads/Volume7Issue2/fulltext87214.pdf |
| work_keys_str_mv | AT wuxiang predictionofcoalfacegasconcentrationbymultiscaleselectiveensemblehybridmodeling AT yangzhongru predictionofcoalfacegasconcentrationbymultiscaleselectiveensemblehybridmodeling AT zhangli predictionofcoalfacegasconcentrationbymultiscaleselectiveensemblehybridmodeling AT pilatisilvia predictionofcoalfacegasconcentrationbymultiscaleselectiveensemblehybridmodeling |