| Abstract
|
:
|
Particulate matter (PM) pollution is currently receiving a great environmentalconcern due to its harmful effect on human health. The sources of PM pollution maycome from direct local emissions, secondary atmospheric formation or long-rangetransport from other regions.In this study, an attempt was made to develop a neural network model to correlatehourly PM concentration with several selected input parameters in the Golden Trianglearea. The considered input parameters included hourly data of nitric oxide, nitrogendioxide, sulfur dioxide, wind speed, temperature, local wind direction, regional winddirection, hour of day, solar radiation, pressure, relative humidity, mixing height andrainfall. These hourly data were obtained from the TCEQ's (Texas Commission onEnvironmental Quality) eaumont monitoring stations (C02 & C54) during the periodfrom January 1,2001 through December 31,2001 .The neural network analysis wasperformed using the Brain aker Professional from California Scientific Software. Theresults have indicated that the R~ values can be as high as 0.85 depending on the month.In general, the higher the wind speed and rainfall values, the lower the observed PM2 5concentrations; however, the precursor concentrations including NO2 and SO2 areobserved to have positive effect on PM2,5 concentrations..............-....,....-...,..tested for theQ1 PC1 bus cardBoth these projects mere sofixare des elopment efforts tonards contributing to dlfferentaspects of Roboucs and lZ1echatronics projects m the Controls and Roboucs Group..
|