A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Development of hourly probabilistic utility NOx emission inventories using time series techniques: Part I—univariate approach
AbstractHistorical data regarding hourly variability in coal-fired power plant unit emissions based upon continuous emission monitoring enables estimation of the likely range of possible values in the near future for purposes of air quality modeling. Analyses were conducted for 32 units for a base case in 1995, an alternative 1998 case, and a 2007 future scenario case. Hourly inter-unit uncertainty was assumed to be independent. Univariate stochastic time series models were employed to quantify hourly uncertainty in capacity and emission factors. Ordinary least-squares regression models were used to quantify hourly uncertainty in heat rate. The models were used to develop an hourly probabilistic emission inventory for a 4-day period. There was significant autocorrelation for time lags 1, 2, 23, and 24 for the capacity and emission factor and a 24h cyclical pattern for the capacity factor. The uncertainty ranges for hourly emissions were found to vary for different hours of the day, with 95% probability ranges of typically ±20–40% of the mean. For the 1995 case, the 95% confidence interval for the daily inventory was 510–633t/d, representing approximately ±10% uncertainty with respect to the average value of 576t/d. Inter-annual changes in the mean and variability were assessed by comparison of 1998 data with 1995 data. The daily inventory for the 2007 scenario had an uncertainty range of ±8% of the average value of 175t/d. The substantial autocorrelation in capacity and emission factor, and the cyclic effect for capacity factor, indicate the importance of accounting for time series effects in estimation of uncertainty in hourly emissions. Additional work is recommended to account for inter-unit dependence, which is addressed in Part 2.
Development of hourly probabilistic utility NOx emission inventories using time series techniques: Part I—univariate approach
AbstractHistorical data regarding hourly variability in coal-fired power plant unit emissions based upon continuous emission monitoring enables estimation of the likely range of possible values in the near future for purposes of air quality modeling. Analyses were conducted for 32 units for a base case in 1995, an alternative 1998 case, and a 2007 future scenario case. Hourly inter-unit uncertainty was assumed to be independent. Univariate stochastic time series models were employed to quantify hourly uncertainty in capacity and emission factors. Ordinary least-squares regression models were used to quantify hourly uncertainty in heat rate. The models were used to develop an hourly probabilistic emission inventory for a 4-day period. There was significant autocorrelation for time lags 1, 2, 23, and 24 for the capacity and emission factor and a 24h cyclical pattern for the capacity factor. The uncertainty ranges for hourly emissions were found to vary for different hours of the day, with 95% probability ranges of typically ±20–40% of the mean. For the 1995 case, the 95% confidence interval for the daily inventory was 510–633t/d, representing approximately ±10% uncertainty with respect to the average value of 576t/d. Inter-annual changes in the mean and variability were assessed by comparison of 1998 data with 1995 data. The daily inventory for the 2007 scenario had an uncertainty range of ±8% of the average value of 175t/d. The substantial autocorrelation in capacity and emission factor, and the cyclic effect for capacity factor, indicate the importance of accounting for time series effects in estimation of uncertainty in hourly emissions. Additional work is recommended to account for inter-unit dependence, which is addressed in Part 2.
Development of hourly probabilistic utility NOx emission inventories using time series techniques: Part I—univariate approach
Abdel-Aziz, Amr (author) / Frey, H.Christopher (author)
Atmospheric Environment ; 37 ; 5379-5389
2003-09-04
11 pages
Article (Journal)
Electronic Resource
English
A probabilistic approach to screen and improve emission inventories
| Elsevier | 2020
Development of Probabilistic Emission Inventories of Air Toxics for Jacksonville, Florida
| Taylor & Francis Verlag | 2004
Construction Time Series Forecasting Using Univariate Time Series Models
| Springer Verlag | 2023
Univariate Time Series Modeling of Inflow
| British Library Conference Proceedings | 2001