 |
Center for Nonlinear Studies |
The amount of natural gas used as a primary energy source in power system operations has increased dramatically in recent years. Existing natural gas-fired generation accounted for about 42% of the total installed capacity in the United States in 2015. Because many natural gas-fired units choose interruptible natural gas-supply contracts, the availability of natural gas supply can threaten the secure operation of electricity systems. We propose a two-stage stochastic unit commitment model that integrates natural gas-supply constraints into the commitment and dispatch processes. The scenarios in the stochastic optimization model capture uncertainties in the hourly natural gas-pipeline capacities. This problem has day-ahead unit commitment decisions in the first stage, and real-time dispatch and market-balancing decisions in the second stage. We use our proposed model to study the effects of natural gas-supply constraints on power system operations under two system paradigms. The first is a case in which natural gas prices are relatively low but potential natural gas-pipeline congestion limits the extent to which the system can rely on natural gas-fired units. The second case has relatively high natural gas prices, but the flexibility of natural gas-fired units must be used to accommodate steep ramps in the load profile. The effects of these two types of natural gas supply conditions on unit commitment, dispatch, and day ahead energy prices are examined.
Host: Chris Neale