Quantification of Variability and Uncertainty in Hourly NO_x Emissions from Coal-Fired Power Plants

摘要

The purpose of this paper is to develop and demonstrate a method to quantify variability and uncertainty for NO_x emissions from coal-fired power plants. Data for hourly NO_x emissions, heat rate, gross load and capacity factor of 32 units from 9 different power plants were analyzed. Possible statistical dependencies among these parameters were evaluated and no substantial dependence was found except between the hourly NO_x emission factor and the hourly capacity factor for all units. The relationship between the hourly NO_x emission factor and hourly capacity factor was quantified using regression analysis. The residual error of the regression model was used to quantify the variability in hourly NO_x emissions that cannot be explained by the capacity factor. Regression models were developed and compared for individual units and for the combined data for groups of units. Three criteria were used to evaluate the adequacy of grouping data. The statistical representation of emissions for the groups of similar units was comparable to that for any individual unit contained within the group to within approximately plus or minus 10 percent. Therefore, it was concluded that it was reasonable to group the data in these cases. For a specific unit, the variability in the hourly NO_x emissions that cannot be explained by capacity factor was found to be -25 to +35% at a 95% confidence level. The implications of this range of variability for emissions inventory development, air quality modeling, and air quality management are discussed.