Light robust co-optimization of energy and reserves in the day-ahead electricity market

To accommodate the stochasticity of variable renewable energy sources (VRES) while efficiently dispatching generation resources and procuring adequate reserves, previous research proposed co-optimizing energy and reserves in the day-ahead (DA) using various uncertainty-based mechanisms. However, the co-optimized mar-kets based on these mechanisms exhibit implementation limitations related to their high computational burden, complex customized solution algorithms, and over-conservative solutions. To address these shortcomings, this paper proposes a practical light robust optimization (LR) approach for the DA co-optimization of energy and reserves. The method results in a linear market clearing mechanism that easily enables the control of the robustness level of the solution through a tunable conservativeness parameter. In addition, the paper explores three different formulations for specifying the system reserve requirements considering, namely, fixed reserve requirements (LRF1), variable reserve requirements based on system uncertainty (LRF2), and a combined approach (LRF3). The formulations integrate the uncertainty from VRES in the market setting using a new bid format called uncertainty bid. The three formulations are then compared using a case study. The numerical results show the effects of the variation of the conservativeness parameter and the reserve requirements on the total socio-economic welfare (SEW), dispatched energy quantities, anticipated activation costs, and procured reserves. Moreover, the analyses showcase that sizing reserves based on system uncertainty (in LRF2) results in a 27%-61% decrease in reserve procurement costs when compared with LRF1, while the combined approach (in LRF3) results in a better performance than LRF2 in terms of reserve activation costs, with costs 61%-263% lower than in LRF2.

Automated summary

This paper proposes a practical light robust optimization (LR) approach for the day-ahead (DA) co-optimization of energy and reserves. The method enables control of the robustness level through a tunable conservativeness parameter and explores three different formulations for specifying the system reserve requirements. Numerical results demonstrate the effects of the conservativeness parameter and reserve requirements on various indicators.