A Methodology for Dynamic Procurement of Secondary Reserve Capacity in Power Systems with Significant vRES Penetrations

The European pathway to carbon neutrality indicates strong investments in variable renewable energy sources. Demand-side and variable renewable players rely on forecasts to participate in day-ahead markets closing between 12 and 37 hours ahead of real-time operation. As such, the further from the real-time operation those forecasts are provided, the higher their errors and the uncertainty. Deviations from the market schedules are balanced using real-time reserves. Traditionally, transmission system operators (TSOs) use a symmetrical procurement of up and down reserves based on the expected demand. This work considers the computation of a dynamic up and down procurement of secondary capacities by considering the expected deviations, using the day-ahead programmed and expected dispatches of variable renewables, demand, other technologies, and the cross-border capacities. The study uses operational open data from the Spanish TSO from 2019 to 2022. The proposed methodology allows increasing the usage of the up and down secondary capacities by almost 13% and 8%, respectively, freeing up 11% of the allocated resources, on average.

I. Introduction

The 2030 European National Energy and Climate Plans indicate sector coupling and a high investment in variable renewable energy sources (vRES), such as wind and solar photovoltaic (PV) technologies to achieve the goal of a carbon-neutral society by 2050 [1],[2]. Increasing penetrations of vRES are raising the usage and costs of reserves across Europe due to their stochastic production profile [3],[4]. The duck curve effect originated from high penetration of solar PV is already a reality in several power systems with significant up and down net load ramps (load minus vRES generation). This situation leads to the so-called merit-order effect, which reduces market prices in the case of negative or nearly zero net loads and increases balancing costs in the case of net load ramps [5],[6]. Market participants pay the balancing costs of their imbalances as Balance Responsible Parties (BRPs) [7],[8]. The self-cannibalization effect and the increase in balancing costs originated by increasing penetrations of vRES are obstacles for their market integration without externalities like support schemes [9],[10]. Demand-side and vRES players rely on forecasts to participate in electricity markets. These forecasts tend to have higher uncertainty, especially in time horizons farther from the real-time operation [8],[11].