Day-Ahead PM 2.5 Concentration Forecasting Using WT-VMD Based Decomposition Method and Back Propagation Neural Network Improved by Differential Evolution

Deyun Wang, Yanling Liu, Hongyuan Luo, Chenqiang Yue and Sheng Cheng
Additional contact information
Deyun Wang: School of Economics and Management, China University of Geosciences, Wuhan 430074, China
Yanling Liu: School of Economics and Management, China University of Geosciences, Wuhan 430074, China
Hongyuan Luo: School of Economics and Management, China University of Geosciences, Wuhan 430074, China
Chenqiang Yue: School of Economics and Management, China University of Geosciences, Wuhan 430074, China
Sheng Cheng: School of Economics and Management, China University of Geosciences, Wuhan 430074, China

IJERPH, 2017, vol. 14, issue 7, 1-22

Abstract: Accurate PM 2.5 concentration forecasting is crucial for protecting public health and atmospheric environment. However, the intermittent and unstable nature of PM 2.5 concentration series makes its forecasting become a very difficult task. In order to improve the forecast accuracy of PM 2.5 concentration, this paper proposes a hybrid model based on wavelet transform (WT), variational mode decomposition (VMD) and back propagation (BP) neural network optimized by differential evolution (DE) algorithm. Firstly, WT is employed to disassemble the PM 2.5 concentration series into a number of subsets with different frequencies. Secondly, VMD is applied to decompose each subset into a set of variational modes (VMs). Thirdly, DE-BP model is utilized to forecast all the VMs. Fourthly, the forecast value of each subset is obtained through aggregating the forecast results of all the VMs obtained from VMD decomposition of this subset. Finally, the final forecast series of PM 2.5 concentration is obtained by adding up the forecast values of all subsets. Two PM 2.5 concentration series collected from Wuhan and Tianjin, respectively, located in China are used to test the effectiveness of the proposed model. The results demonstrate that the proposed model outperforms all the other considered models in this paper.

Keywords: PM 2.5 concentration forecasting; wavelet transform; variational mode decomposition; differential evolution; back propagation neural network (search for similar items in EconPapers)
JEL-codes: I I1 I3 Q Q5 (search for similar items in EconPapers)
Date: 2017
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (3)

Downloads: (external link)
https://www.mdpi.com/1660-4601/14/7/764/pdf (application/pdf)
https://www.mdpi.com/1660-4601/14/7/764/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jijerp:v:14:y:2017:i:7:p:764-:d:104457

Access Statistics for this article

IJERPH is currently edited by Ms. Jenna Liu

More articles in IJERPH from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().