Status | 已發表Published |
Statistical Modeling of PM10 concentrations in Macau by using Kalman filter | |
Hoi, K. I.; Yuen, K. V.; Mok, K. M. | |
2007-12-01 | |
Source Publication | Proceedings of the APCOM’07 in conjunction with EPMESC XI (CD-ROM) |
Abstract | In the present study, Kalman filter was applied to forecast the daily average of PM10 concentrations in Macau during the summer period of 2004. The Kalman filter was implemented on two different model classes, namely an empirical model and a semi-empirical model. It was found that there was a time-delay problem associated with the predictions of the empirical model, i.e., the predicted PM10 concentrations generally lagged behind the measured PM10 concentrations. The root mean square error (RMSE) and the correlation coefficient (r) for the empirical model are 27.20μg/m3 and 0.68, respectively. For the semi-empirical model, the time-delay problem was not substantial and the prediction accuracy was increased. The RMSE and the value of r for the semi-empirical model are 20.63μg/m3 and 0.81, respectively. The semi-empirical model outperforms the empirical model since the meteorological conditions which can influence the PM10 concentrations on the day of prediction are considered. Therefore, the process noise of the system is reduced. It is concluded that the Kalman filter was a promising tool in air quality prediction. However, caution should be made in the selection of the model classes. |
Keyword | air quality prediction Kalman filter PM10 Macau |
Language | 英語English |
The Source to Article | PB_Publication |
PUB ID | 48433 |
Document Type | Conference paper |
Collection | HENRY FOK PEARL JUBILEE COLLEGE |
Recommended Citation GB/T 7714 | Hoi, K. I.,Yuen, K. V.,Mok, K. M.. Statistical Modeling of PM10 concentrations in Macau by using Kalman filter[C], 2007. |
APA | Hoi, K. I.., Yuen, K. V.., & Mok, K. M. (2007). Statistical Modeling of PM10 concentrations in Macau by using Kalman filter. Proceedings of the APCOM’07 in conjunction with EPMESC XI (CD-ROM). |
Files in This Item: | There are no files associated with this item. |
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Edit Comment