Integrating Air Quality Considerations into Power System Optimization Modeling Development and Case Study

Emissions from fossil fuel-fired plants has been reported to lead to adverse air quality and human health impact, especially over densely populated urban area. Traditional power system model simulates the dispatch solutions by optimizing the system cost without considering the air quality and health impact associated with the changes in air pollutants emitted by power plants. Both power system and its air quality impact are highly complicated and affected by the power system configuration, meteorological variables, and chemical reactions. Therefore, this study focused on the investigation of the impact of dispatch solutions on the air quality and health effects associated with the emissions from power system using a Lagrangian puff model with detailed chemistry. The overarching goal of this study is to provide sufficient dataset of dispatch solutions and the corresponding air quality and health impact outcomes for development of a surrogate model to co-optimize the system cost and health impact burden for the power system. To reduce the complexity and computational burden but capture the essential characteristics of the system, representative days were selected using a machine learning approach with grid-related and meteorological variables. Then, we conducted power system modelling using National Renewable Energy (NREL)’s Sienna model to generate diverse electricity dispatch for the Los Angeles service area, specifically for thermal generators, to explore the impact of dispatch solutions on air quality impact. The emission inventories were generated for each solution using the Sienna outputs and the emission factor dataset we developed based on the US Environmental Protection Agency (EPA) Emissions & Generation Resource Integrated Database (eGRID) and the US EPA National Emissions Inventory (NEI) data. Then, air quality and health impact simulations were conducted using the Second-order Closure Integrated puff CHEMistry (SCICHEM) model and the Environmental Benefits Mapping and Analysis Program (BenMAP) model, respectively. Results shows total generation from thermal generators is highly positively correlated with the total PM2.5 concentration among all representative days. The distribution of air pollutants determined by the wind fields is important to the public health impact when population is considered. When the plume of air pollutants is accumulated over source area under weak wind condition or is transported by wind to the populated area, its health impact is stronger than that of plume of pollutants transported to rural area or ocean. The analysis and model outputs can provide meaningful insight and dataset for development of a power system optimization model that integrates air quality consideration to co-optimize the system cost and the health impact burden.