The estimation of the surface ozone concentration promotes the creation of data useful for planning the air quality forecast, which is a key element for the management of public health. The aim of this study is to dev...The estimation of the surface ozone concentration promotes the creation of data useful for planning the air quality forecast, which is a key element for the management of public health. The aim of this study is to develop an Artificial Neural Network (ANN) to estimate the concentration of surface ozone from daily climate data. ANN is an equivalent form of Feedforward Multilayer Perceptron whose data has been inserted from the daily concentration of measured ozone. In the intermediate and output layers activation functions like tan-sigmoid and linear have been used, respectively. The performance of the developed ANN is actually very good and it can be considered like part of the set of indirect methods to estimate the concentration of surface ozone. The proposed model may be used by governmental agencies as a tool to enable the public interventional actions during the period of atmospheric stagnation, when ozone levels in the atmosphere represent risks to the public health.展开更多
This study aims to develop a statistical modelling framework of urban water consumption forecast for the city of Aquidauana, Brazil from year 2005 to 2014, monthly data, using multiple linear regression, cluster analy...This study aims to develop a statistical modelling framework of urban water consumption forecast for the city of Aquidauana, Brazil from year 2005 to 2014, monthly data, using multiple linear regression, cluster analysis, and principal component analysis. These forecasts were based on historical data collected through SANESUL System (Water Systems of South Mato Grosso). The meteorological data were provided by the Water Resources Monitoring Center of South Mato Grosso—CEMTEC. The statistical model developed explains 71.5% of the variance with three factors: number of consumers (19.3%), seasonality (37.8%), and climate regression (14.3%). The model was further validated using an independent set of data from January 2005 to November 2014, with an R2 of 86% and error of 1.7%. The results indicated no intervention of climate variables in the phenomenon. This tool, combined with the perception of the potential and limitations of managers of water resources and public policy makers, can be used in the regulation of per capita consumption, and thereby achieve the optimization of available resources and also contribute to the sustainable perspective of water resources.展开更多
文摘The estimation of the surface ozone concentration promotes the creation of data useful for planning the air quality forecast, which is a key element for the management of public health. The aim of this study is to develop an Artificial Neural Network (ANN) to estimate the concentration of surface ozone from daily climate data. ANN is an equivalent form of Feedforward Multilayer Perceptron whose data has been inserted from the daily concentration of measured ozone. In the intermediate and output layers activation functions like tan-sigmoid and linear have been used, respectively. The performance of the developed ANN is actually very good and it can be considered like part of the set of indirect methods to estimate the concentration of surface ozone. The proposed model may be used by governmental agencies as a tool to enable the public interventional actions during the period of atmospheric stagnation, when ozone levels in the atmosphere represent risks to the public health.
文摘This study aims to develop a statistical modelling framework of urban water consumption forecast for the city of Aquidauana, Brazil from year 2005 to 2014, monthly data, using multiple linear regression, cluster analysis, and principal component analysis. These forecasts were based on historical data collected through SANESUL System (Water Systems of South Mato Grosso). The meteorological data were provided by the Water Resources Monitoring Center of South Mato Grosso—CEMTEC. The statistical model developed explains 71.5% of the variance with three factors: number of consumers (19.3%), seasonality (37.8%), and climate regression (14.3%). The model was further validated using an independent set of data from January 2005 to November 2014, with an R2 of 86% and error of 1.7%. The results indicated no intervention of climate variables in the phenomenon. This tool, combined with the perception of the potential and limitations of managers of water resources and public policy makers, can be used in the regulation of per capita consumption, and thereby achieve the optimization of available resources and also contribute to the sustainable perspective of water resources.
