An energy allocation model for air pollution episode control

This paper reports the results of an exploratory study conducted at Argonne National Laboratory in which atmospheric dispersion analysis was linked with modern control theory. Its initial objective was to introduce comprehensive air pollution episode control planning into the electric utility operations. The planning activity was based on the assumption that the ability of the utility plants to switch from coal fuel to natural gas could be utilized for air resource management purposes. This study resulted in the formulation of a computerized optimal episode control system for a subset of the power plant network. The control system, which inputs standard meteorological parameters and power demand forecasts, generates a schedule of power load shifting and natural gas utilization for the duration of a pollution episode. This optimal control schedule represents a minimum cost, minimum gas use strategy which restricts the pollution levels produced by power plants to a maximum acceptable threshold level within the limits of the real-time system operational and economic constraints and the capacity of the power plant network to react to control measures. The strategy can be updated as rapidly as altered meteorological conditions, a changing power demand schedule or unforeseen equipment failures warrant. An actual episode was used to test this control model. Recorded meteorological data and operating conditions supplied by the local Chicago utility were employed in the experiment. The control model was used to computer an optimal operating schedule. An indication of the air quality resulting from the use of this schedule was based on an atmospheric diffusion model.