Future daily PM10 concentrations prediction by combining regression models and feedforward backpropagation models with principle component analysis (PCA)

Ul-Saufie, Ahmad Zia; Yahaya, Ahmad Shukri; Ramli, Nor Azam; Rosaida, Norrimi; Hamid, Haslinda Abdul

doi:10.1016/j.atmosenv.2013.05.017

Cited by 121 publications

(77 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In practice the assumption of linearity cannot be conirmed, and linear regression models are considered to be acceptable provided there are only minor deviations from this assumption. PM 10 and its explanatory variables typically do not meet the assumption of linearity [22]. Often the variables do not have a normal distribution due to the presence of outliers.…”

Section: Regression Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Particulate Matter Sampling Techniques and Data Modelling Methods

Whalley¹,

Zandi²

2016

Air Quality - Measurement and Modeling

View full text Add to dashboard Cite

Particulate mater with 10 μm or less in diameter PM 10 ) is known to have adverse efects on human health and the environment. For countries commited to reducing PM 10 emissions, it is essential to have models that accurately estimate and predict PM 10 concentrations for reporting and monitoring purposes. In this chapter, a broad overview of recent empirical statistical and machine learning techniques for modelling PM 10 is presented. This includes the instrumentation used to measure particulate mater, data preprocessing, the selection of explanatory variables and modelling methods. Key features of some PM 10 prediction models developed in the last 10 years are described, and current work modelling and predicting PM 10 trends in New Zealand a remote country of islands in the South Paciic Ocean are examined. In conclusion, the issues and challenges faced when modelling PM 10 are discussed and suggestions for future avenues of investigation, which could improve the precision of PM 10 prediction and estimation models are presented.

show abstract

Section: Regression Methodsmentioning

confidence: 99%

“…If high humidity is not accompanied by rainfall but is accompanied by high temperatures, humidity has been found to contribute to higher PM 10 concentrations. It has been suggested that when the relative humidity is over 55%, then PM 10 concentrations are afected [22].…”

Section: Study Reference [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ 9] [ ] [ ]mentioning

confidence: 99%

Particulate Matter Sampling Techniques and Data Modelling Methods

Whalley¹,

Zandi²

2016

Air Quality - Measurement and Modeling

View full text Add to dashboard Cite

show abstract

“…In this study, only a factor loading that is greater than 0.4 is considered significant (Ul-Saufie et al, 2013). The sufficiency of the monitoring data for PCA was assessed using Kaiser-Meyer-Olkin (KMO) and Bartlett's tests.…”

Section: Principal Component Analysis (Pca)mentioning

confidence: 99%

The Influence of Spatial Variability of Critical Conversion Point (CCP) in Production of Ground Level Ozone in the Context of Tropical Climate

Awang

Elbayoumi

Ramli

et al. 2016

Aerosol Air Qual. Res.

Self Cite

View full text Add to dashboard Cite

Critical conversion point (CCP) is a very crucial step in production of the ground level O 3 chemistry. Thus, a multivariate analysis was applied on the dataset of nine selected locations in Malaysia from 1999 to 2010. It incorporated hierarchical agglomerative cluster analysis (HACA) to explore the spatial variability of CCP and principal component analysis (PCA) to determine the major sources of the air pollutants that influence ozone CCP. High variability in CCP was observed between the monitoring stations that occurred during critical conversion time (CCT) from 8:00 a.m. to 11:00 a.m. The HACA results grouped the nine monitoring stations into three different clusters, based on the characteristics of ozone concentrations during CCT period. Results of PCA for the three clusters showed that the contributions to O 3 level variation during CCT by meteorological variables (UVB, temperature, relative humidity, and wind speed) are higher at 51.6%, 48.5%, and 33.3% than that of primary air pollutants (NO 2 , SO 2 , PM 10 ) at 19.2%, 21.4%, and 15.2% for cluster 1, cluster 2, and cluster 3, respectively. Therefore, applying a targeted spatial control strategy for ground level O 3 precursors during the CCT period is a crucial step.

show abstract

“…The FFBP model has been applied in many previous studies of renewable energy (e.g., Ravinesh C Deo & Sahin, 2017;Ghorbani, Khatibi, Hosseini, & Bilgili, 2013). The FFBP model offers a competent learning environment that minimizes error between the target and the obtained values (Ul-Saufie, Yahaya, Ramli, Rosaida, & Hamid, 2013). As shown in Figure 1 the network of FFBP usually consists of an input layer (X 1 , X 2 , X 3 ,..X n ), several hidden layers, and an output layer (Y).…”

Section: Artificial Neural Networkmentioning

confidence: 99%

“…According to Kriesel (Kriesel, 2007), the transfer function in the hidden layers must use a nonlinear transfer function (otherwise the result end up with only linear separable solutions). Therefore, the optimal activation functions are obtained using sigmoid transfer function (Ul-Saufie, Yahaya, Ramli, Rosaida, & Hamid, 2013). The most common sigmoid functions are linear (purelin), log-sigmoid (logsig) and hyperbolic tangent sigmoid (tansig).…”

Section: Artificial Neural Networkmentioning

confidence: 99%