Endogenous probabilistic reserve sizing and allocation in unit commitment models: Cost-effective, reliable and fast

Abstract

Abstract-In power systems with high shares of variable and limitedly predictable renewables, power system operators need to schedule flexible load, generation and storage to maintain the power system balance when forecast errors occur. To ensure a reliable and cost-effective power system operation, novel reserve sizing and allocation methods are needed. Although stochastic formulations of the unit commitment problem allow calculating an optimal trade-off between the cost of scheduling and activating reserves, load shedding and curtailment, these models may become computationally intractable for real-life power systems. Therefore, in this paper, we develop a novel set of probabilistic reserve constraints, which allows internalizing the reserve sizing and allocation problem in a deterministic unit commitment model, considering the full cost of reserve allocation and activation. Extensive numerical simulations show that this novel formulation yields UC schedules that are nearly as costeffective as the theoretical optimal solution of the stochastic model in calculation times similar to that of a deterministic equivalent.

RoCoF-Constrained Scheduling Incorporating Non-Synchronous Residential Demand Response

“…For example, the uncertainty associated with the provision of wind power emulated inertial response (which requires knowledge of the number of capable wind turbines online) is likely higher than that associated with provision from battery energy storage. More generally, representing the forecast error of RES within UCED is increasing in importance, with a range of implementations proposed [51]. (ii) If contingency reserve is sourced from synchronouslyconnected rotating loads, the (further) reduction in system inertia following the activation of a response.…”

Large-scale unit commitment under uncertainty: an updated literature survey

“…They thus neglect in the process any temporal correlations in the load; furthermore, since uncertainty is assumed Gaussian, they fail to exploit the fact that the set defined by the joint probability constraint is actually convex. A similar methodology, based on ICCO to compute accurate endogenous reserve levels is used in [90,322].…”

Joint generation and reserve scheduling of wind‐solar‐pumped storage power systems under multiple uncertainties

“…Generation and reserve are closely coupled in terms of generator capacity as well as transmission capacity. However, in UC model, the transmission constraints are often neglected due to the computation complexity involved . But in the operation procession where operating reserve is activated when uncertain events occur, generators have to adjust its output, and the transmission network must be able to accommodate these changes, so transmission constraints should be simultaneously considered in the scheduling stage.…”

“…8 In quantifying reserve stage, Chattopadhyay and Baldick 9 only consider the uncertainty of generator outages, and previous works [5][6][7][8] consider the uncertainties of generator outages as well as wind power and load forecast errors; generally, reserve requirement is determined by reliability index or cost-benefit analysis, and either way fails to give a concisely quantitative relationship between reserve and the system reliability. In reserve allocation stage, many works allocate reserve based on economic consideration 7,9,10 but ignore actual energy redistribution condition in post-uncertain events state; for those which do not ignore this factor, the consideration of uncertain events is not fully covered 2,3,[10][11][12][13][14][15][16] ; other studies only consider wind power/solar power/load forecast error, and Street et al 17 only consider generator outage. To the knowledge of the authors, previous works rarely consider multiple uncertainties (including wind and solar power output, load change, and generator failure) both in reserve quantifying and reserve allocation stage that is because these uncertainties have different characteristics and can be divided into two categories (continuous and discrete); different optimization methods are adopted to address these uncertainties: stochastic optimization, [18][19][20][21] robust optimization, 2,11,17,[22][23][24] and chance-constrained optimization.…”

“…However, in UC model, the transmission constraints are often neglected due to the computation complexity involved. 3,7,9,12 But in the operation procession where operating reserve is activated when uncertain events occur, generators have to adjust its output, and the transmission network must be able to accommodate these changes, 27 so transmission constraints should be simultaneously considered in the scheduling stage. To overcome the drawbacks of existing literature for lacking transmission constraints consideration, this paper presents a security-constrained unit commitment (SCUC) model for joint generation and reserve scheduling to facilitate transmission constraints consideration.…”