Economic evaluation of battery storage systems bidding on day-ahead and automatic frequency restoration reserves markets

Felix Nitsch,Marc Deissenroth-Uhrig,Christoph Schimeczek,Valentin Bertsch

doi:10.1016/j.apenergy.2021.117267

Felix Nitsch, Marc Deissenroth-Uhrig + Show 2 more

Open Access

https://doi.org/10.1016/j.apenergy.2021.117267

Copy DOI

Abstract

In future electricity systems, not only electricity generation but also frequency stabilization must be provided by low-carbon technologies. Battery systems are a promising solution to fill this gap. However, uncertainties regarding their revenue potential may hinder investments. Therefore, we apply the agent-based electricity market model AMIRIS to simulate a day-ahead market and an automatic frequency restoration reserves market. Demonstrating the model setup, we chose a scenario with high shares of renewable energies. First, we back-test our model with historic market data from Germany in 2019. The simulation results’ mean day-ahead prices of 39.20EUR/MWh are close to the historic ones of 38.70EUR/MWh. Second, we model both markets in a scenario for 2030. The simulated day-ahead market prices are higher on average than observed today, although, we find around 550 h/yr in which the load is fully covered by renewable energies. The variance in simulated prices is slightly higher compared to historic values. Bids on the reserve capacity market are derived from opportunity costs of not participating in the day-ahead market. This results in prices of up to 45EUR/MW for positive reserve while the prices for negative reserve are 0EUR/MW. Finally, we evaluate revenue potentials of battery storages. Compared to 2019, we see an improved economic potential and increased importance of the day-ahead market. High power battery storages perform best whereas improvements in round-trip efficiency only marginally improve revenues. Although demonstrated for Germany, the presented modular approach can be adapted to international markets enabling comprehensive battery storage assessments.

Highlights

Rising shares of fluctuating renewable energy (RE) lead to growing demand in flexibility options on various temporal scales
We present a novel approach for simulating the automatic frequency restoration reserves market alongside the day-ahead market in an agentbased electricity market model
The simulated electricity system features a significant share of renewable power plants supplying already 60% of the yearly electricity demand

Summary

Introduction

Rising shares of fluctuating renewable energy (RE) lead to growing demand in flexibility options on various temporal scales. One prominent example is maintaining the frequency of power grids which is very sensitive to changes either in electricity demand or generation. There are various different mechanisms to ensure a stable frequency. In Germany, for instance, the automatic Frequency Restoration Reserves (aFRR) – in combination with the Frequency Containment Reserves (FCR) and the manual Frequency Restoration Reserves (mFRR) – ensure frequency stabilization of the electricity system. At the moment, these markets are mostly supplied by conventional power plants and pumped hydro stor ages [1]. This may become a challenge, as Nomenclature

Objectives

Methods

Discussion

Conclusion