Modeling start-up modes and corresponding cycling costs in the unit commitment problem

Cycling of thermal power plants is gaining importance with the increase of variable and partly unpredictable renewable energy sources (RES) such as wind and solar power. The long-term cycling costs, primarily maintenance costs, are difficult to determine and account for in the scheduling stage. Furthermore, the accrued damage and resulting maintenance costs are greatly affected by the required flexibility in terms of the start-up rate. In this paper, a unit commitment formulation that allows for multiple start-up modes and accounts for the appropriate long-term maintenance costs based on long-term service agreements (LTSAs) is presented. The impact of a slow and fast start-up mode on the scheduling results in the presence of renewables forecast errors and limited scheduling horizons is investigated. Results show that fast starting can be optimal despite the higher maintenance cost and a more cost-efficient operating strategy is obtained when considering fast start maintenance penalties.