Analysis and forecasting of the particulate matter (PM) concentration levels over four major cities of China using hybrid models

Qin, Shanshan; Liu, Feng; Wang, Jianzhou; Sun, Baoshan

doi:10.1016/j.atmosenv.2014.09.046

Cited by 100 publications

(41 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The proximity is described by the gray relational degree, which is a measure of the similarities of discrete data that can be arranged in sequential order (Jia et al 2010;Qin et al 2014). The gray correlation degree quantitatively represents the correlation between different driving factors.…”

Section: Methods and Available Data Brief Introduction To Methodsmentioning

confidence: 99%

See 1 more Smart Citation

A hybrid study of multiple contributors to per capita household CO2 emissions (HCEs) in China

Qin

Liu

et al. 2015

Environ Sci Pollut Res

Self Cite

View full text Add to dashboard Cite

Given the large expenditures by households on goods and services that contribute a large proportion of global CO 2 emissions, increasing attention has been paid to household CO 2 emissions (HCEs). However, compared with industrial CO 2 emissions, efforts devoted to mitigating HCEs are relatively small. A good understanding of the effects of some driving factors (i.e., urbanization rate, per capita GDP, per capita income/disposable income, Engel coefficient, new energy ratio, carbon intensity, and household size) is urgently needed prior to considering policies for reducing HCEs. Given this, in the study, the direct and indirect per capita HCEs were quantified in rural and urban areas of China over the period 2000-2012. Correlation analysis and gray correlation analysis were initially used to identify the prime drivers of per capita HCEs. Our results showed that per capita income/disposable income, per capita GDP, urbanization rate, and household size were the most significantly correlated with per capita HCEs in rural areas. Moreover, the conjoint effects of the potential driving factors on per capita HCEs were determined by performing principal component regression analysis for all cases. Based on the combined analysis strategies, alternative polices were also examined for controlling and mitigating HCEs growth in China.

show abstract

Section: Methods and Available Data Brief Introduction To Methodsmentioning

confidence: 99%

“…Otherwise, it will weaken the major factor (Zhang and Zhang 2007). Detailed information about GCA can be referenced to (Jia et al 2010;Qin et al 2014).…”

Section: Methods and Available Data Brief Introduction To Methodsmentioning

confidence: 99%

A hybrid study of multiple contributors to per capita household CO2 emissions (HCEs) in China

Qin

Liu

et al. 2015

Environ Sci Pollut Res

Self Cite

View full text Add to dashboard Cite

show abstract

“…Therefore, a reasonable and effective reasonable assessment of these relationships is of great importance. Traditional statistical methods, such as the Spearman's rank correlation and the gray correlation analysis can result in poor assessment quality, and may cause a large error [25,28,31,36]. To overcome these hurdles, new methodologies have been developed for classification and regression, such as neural network and boosted regression tree (BRT), which offer advantages over interaction and prediction [37,38].…”

Section: The Role Of Meteorological Elements In Relation To Pmmentioning

confidence: 99%

“…For example, Li, Qian, Ou, Zhou, Guo, and Guo [24], Zhang et al [28] and Tian, Qiao, and Xu [25] have analyzed the impacts of meteorological conditions on PM pollution in different season. Jian et al [29], Li et al [30], and Qin et al [31] have predicted PM concentrations using meteorological data. However, most studies have some limitations, due to their analyses focusing on a certain city or region, which is hard to reflect back to the whole state of PM pollution [24,25,28].…”

Section: Introductionmentioning

confidence: 99%

Characteristics of Particulate Pollution (PM2.5 and PM10) and Their Spacescale-Dependent Relationships with Meteorological Elements in China

Chen

Zeng

et al. 2017

Sustainability

View full text Add to dashboard Cite

Particulate matter (PM) pollution in China has an obvious characteristic of spatial distribution. It is well known that intensive anthropogenic activities, such as fossil fuel combustion and biomass burning, have great influence on the spatial distribution of PM pollution. However, the spacescale-dependent relationships between PM concentrations and weather conditions remain unclear. Here, we investigated the characteristics of two types of particulate pollution, including PM 2.5 and PM 10 , and their spatial relationships with meteorological elements in 173 cities throughout China from March 2014 to February 2015. Results: (1) High PM 2.5 concentrations were distinctly located southeast of the Hu Line, and high PM 10 concentrations were distinctly situated north of the Yangtze River; (2) Spacescale-dependent relationships were found between PM pollution and meteorological elements. The influence of temperature had similar inverted V-shaped characteristics, namely, there was serious PM pollution when temperature was about 15 • C, and there was slight PM pollution when temperature was less or more than 15 • C. Annual precipitation, wind speed, and relative humidity were negatively correlated with PM, while annual atmospheric pressure was positively correlated with PM; (3) The ideal meteorological regions were identified according to the quantified spatial relationships between PM and meteorological elements, which could be defined by a combination of the following conditions: (a) temperature <10 • C or >21 • C; (b) precipitation >1500 mm; (c) atmospheric pressure <900 hPa; (d) wind speed >3 m/s; and (e) relative humidity >65%, where air pollutants can easily be scavenged. The success of this research provides a meteorological explanation to the spatial distribution characteristics of PM pollution in China.

show abstract

“…Moreover, the forecasting difficulty is principally due to the complex characteristics of data. The works of Antanasijević et al (2013) and Qin et al (2014) are among the few studies that address the variable importance issue. In the first work, the authors constructed a nonlinear predictive model based on gray correlation analysis (GCA), Ensemble Empirical Mode Decomposition (EEMD), Cuckoo search (CS) and Back-propagation artificial neutral networks (BPANN) to identify the pertinent predictors giving rise to an accurate model to forecast the PM10 concentrations in different climatic and environmental areas.…”

Section: Related Workmentioning

confidence: 99%

Variable Selection Based on Statistical Learning Approaches to Improve PM10 Concentration Forecasting

Ishak¹

2016

J ENV INFORM

View full text Add to dashboard Cite

ABSTRACT. In this work, the problem of variable selection for regression is investigated in order to improve the forecasting accuracy. To that effect, the support vector regression (SVR) and the random forests (RF) are used to assess the variable importance. Then, a stepwise algorithm is built to select the best subset of predictors. An intensive comparative study is conducted on simulated and real datasets. The real datasets expose the problem of particulate matter concentration forecasting in two monitoring stations from Tunisia. We have proposed a combined approach using SVR and RF for variable importance assessment and for variable selection. We have achieved a significant improvement in forecasts accuracy for the two stations when using only a reduced number of selected predictors.

show abstract

Analysis and forecasting of the particulate matter (PM) concentration levels over four major cities of China using hybrid models

Cited by 100 publications

References 21 publications

A hybrid study of multiple contributors to per capita household CO2 emissions (HCEs) in China

A hybrid study of multiple contributors to per capita household CO2 emissions (HCEs) in China

Characteristics of Particulate Pollution (PM2.5 and PM10) and Their Spacescale-Dependent Relationships with Meteorological Elements in China

Variable Selection Based on Statistical Learning Approaches to Improve PM10 Concentration Forecasting

Contact Info

Product

Resources

About