A New Decomposition Ensemble Learning Approach with Intelligent Optimization for PM2.5 Concentration Forecasting
In this study, we focus our attention on the forecasting of daily PM2.5 concentrations. According to the principle of “divide and conquer,” we propose a novel decomposition ensemble learning approach by integrating ensemble empirical mode decomposition (EEMD), artificial neural networks (ANNs), and...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2020-01-01
|
| Series: | Discrete Dynamics in Nature and Society |
| Online Access: | http://dx.doi.org/10.1155/2020/6019826 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832561389798948864 |
|---|---|
| author | Guangyuan Xing Shaolong Sun Jue Guo |
| author_facet | Guangyuan Xing Shaolong Sun Jue Guo |
| author_sort | Guangyuan Xing |
| collection | DOAJ |
| description | In this study, we focus our attention on the forecasting of daily PM2.5 concentrations. According to the principle of “divide and conquer,” we propose a novel decomposition ensemble learning approach by integrating ensemble empirical mode decomposition (EEMD), artificial neural networks (ANNs), and adaptive particle swarm optimization (APSO) for forecasting PM2.5 concentrations. Our proposed decomposition ensemble learning approach is formulated exclusively to deal with difficulties in quantitating meteorological information with high volatility, irregularity, and complicacy. This decomposition ensemble learning approach mainly consists of three steps. First, we utilize EEMD to decompose original time series of PM2.5 concentrations into a specific amount of independent intrinsic mode functions (IMFs) and residual term. Second, the ANN, whose connection parameters are optimized by APSO algorithm, is employed to model IMFs and residual terms, respectively. Finally, another APSO-ANN is applied to aggregate the forecast IMFs and residual term into a collection as the final forecasting results. The empirical results show that the forecasting of our decomposition ensemble learning approach outperforms other benchmark models in terms of level accuracy and directional accuracy. |
| format | Article |
| id | doaj-art-d9a5157d958b4eaea715b659fed34ed6 |
| institution | Kabale University |
| issn | 1026-0226 1607-887X |
| language | English |
| publishDate | 2020-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Discrete Dynamics in Nature and Society |
| spelling | doaj-art-d9a5157d958b4eaea715b659fed34ed62025-02-03T01:25:17ZengWileyDiscrete Dynamics in Nature and Society1026-02261607-887X2020-01-01202010.1155/2020/60198266019826A New Decomposition Ensemble Learning Approach with Intelligent Optimization for PM2.5 Concentration ForecastingGuangyuan Xing0Shaolong Sun1Jue Guo2School of Economics and Finance, Xi’an Jiaotong University, Xi’an 710061, ChinaSchool of Management, Xi’an Jiaotong University, Xi’an 710049, ChinaSchool of Management, Xi’an Jiaotong University, Xi’an 710049, ChinaIn this study, we focus our attention on the forecasting of daily PM2.5 concentrations. According to the principle of “divide and conquer,” we propose a novel decomposition ensemble learning approach by integrating ensemble empirical mode decomposition (EEMD), artificial neural networks (ANNs), and adaptive particle swarm optimization (APSO) for forecasting PM2.5 concentrations. Our proposed decomposition ensemble learning approach is formulated exclusively to deal with difficulties in quantitating meteorological information with high volatility, irregularity, and complicacy. This decomposition ensemble learning approach mainly consists of three steps. First, we utilize EEMD to decompose original time series of PM2.5 concentrations into a specific amount of independent intrinsic mode functions (IMFs) and residual term. Second, the ANN, whose connection parameters are optimized by APSO algorithm, is employed to model IMFs and residual terms, respectively. Finally, another APSO-ANN is applied to aggregate the forecast IMFs and residual term into a collection as the final forecasting results. The empirical results show that the forecasting of our decomposition ensemble learning approach outperforms other benchmark models in terms of level accuracy and directional accuracy.http://dx.doi.org/10.1155/2020/6019826 |
| spellingShingle | Guangyuan Xing Shaolong Sun Jue Guo A New Decomposition Ensemble Learning Approach with Intelligent Optimization for PM2.5 Concentration Forecasting Discrete Dynamics in Nature and Society |
| title | A New Decomposition Ensemble Learning Approach with Intelligent Optimization for PM2.5 Concentration Forecasting |
| title_full | A New Decomposition Ensemble Learning Approach with Intelligent Optimization for PM2.5 Concentration Forecasting |
| title_fullStr | A New Decomposition Ensemble Learning Approach with Intelligent Optimization for PM2.5 Concentration Forecasting |
| title_full_unstemmed | A New Decomposition Ensemble Learning Approach with Intelligent Optimization for PM2.5 Concentration Forecasting |
| title_short | A New Decomposition Ensemble Learning Approach with Intelligent Optimization for PM2.5 Concentration Forecasting |
| title_sort | new decomposition ensemble learning approach with intelligent optimization for pm2 5 concentration forecasting |
| url | http://dx.doi.org/10.1155/2020/6019826 |
| work_keys_str_mv | AT guangyuanxing anewdecompositionensemblelearningapproachwithintelligentoptimizationforpm25concentrationforecasting AT shaolongsun anewdecompositionensemblelearningapproachwithintelligentoptimizationforpm25concentrationforecasting AT jueguo anewdecompositionensemblelearningapproachwithintelligentoptimizationforpm25concentrationforecasting AT guangyuanxing newdecompositionensemblelearningapproachwithintelligentoptimizationforpm25concentrationforecasting AT shaolongsun newdecompositionensemblelearningapproachwithintelligentoptimizationforpm25concentrationforecasting AT jueguo newdecompositionensemblelearningapproachwithintelligentoptimizationforpm25concentrationforecasting |