Long-Term Cycling Costs In Short-Term Unit Commitment Models

Cycling of conventional power plants is becoming increasingly important in an electricity system with a large penetration of intermittent renewables. Power plant cycling entails short-term costs, e.g., additional fuel costs during start-up, and long-term costs, e.g., additional maintenance costs. Power plant operators should take long-term cycling costs into account when making short-term scheduling decisions, in order to reduce total generation costs. This paper presents a new approach to consider long-term start-up costs in a short-term unit commitment model. The approach is based on an iterative procedure, in which consecutively a unit commitment model is solved and the correct total start-up cost is recalculated. This new approach, referred to as the Cost Redistribution Unit Commitment (CRUC), is applied to a real-life case study based on the 2014 German electricity system. The performance of the CRUC model, in terms of generation costs and computational tractability, is compared with existing start-up cost formulations in the literature. The simulation results show that, for the considered case study, the CRUC model outperforms the existing formulations in terms of generation costs, but requires a longer run time.