Optimising the spinning reserve requirements considering failures to synchronise

System operators use spinning reserve (SR) to respond to unforeseen events such as generation outages and unforeseen load changes. The amount of SR that a system must carry is usually determined on the basis of deterministic criteria such as the capacity of the largest synchronised generating unit or a given percentage of the load. Several authors have argued that these deterministic criteria are sub-optimal. They have proposed various techniques to adjust the SR requirements considering the likelihood and the consequences of the failure of one or more of the generating units that are synchronised to the system. However, so far none of these approaches has considered explicitly in the optimisation the significant danger associated with the failure of generating units to synchronise with the system in a timely fashion. A three-state reliability model for generating units that takes into account not only the probability of failing while operating but also the probability of failing to synchronise is proposed. This model is then used in a cost/benefit analysis to estimate the SR requirements at each period of the optimisation horizon. Finally, a reserve-constrained unit commitment programme that enforces these requirements is used to find the optimal generation schedule. This optimal approach is compared with existing heuristic approaches on the basis of the IEEE-RTS system. © 2008 The Institution of Engineering and Technology.