Aquesta és una còpia de la versió author’s final draft d'un article publicat a la revista [International journal of energy research]. URL d'aquest document a UPCommons E-prints: http://hdl.handle.net/2117/127058

Primary control reserve and maximising self-consumption are currently two of the main applications for large-scale battery storage systems. Although being currently the most profitable application for large-scale batteries in Germany, storage systems applying primary control reserve have not been implemented in a grid supportive manner in distribution grids yet. Despite a current unfavourable regulatory framework and reimbursement scheme for community electricity storages in Germany, they are potentially more profitable than residential storages, which is mainly due to their economy of scale, and thus they may become the major large scale battery application in the future. The two applications: primary control reserve and maximising self-consumption, are combined with a grid supportive behaviour by providing reactive power control and/or peak shaving and are fitted to a vanadium redox flow battery prototype, which is installed in a distribution grid in southern Germany. Based on measured data from the prototype, two battery models for two different time resolutions (1s, 1min) are presented in detail along with their respective operation models. The operation strategy model for primary control reserve comprises the so-called degrees of freedom used to reduce the energy needed to recharge the battery. The operation strategy to maximise self-consumption is based on a persistence forecast. The model for the operation strategy for a grid supportive primary control reserve was validated in a field test revealing a relative error of 2.5 % between the simulated and measured state of charge of the battery for a multi-week time period. The technical assessment of both applications shows that the use of the degrees of freedom can reduce the energy to recharge the battery by 20 %; and in the case of self-consumption, the curtailment losses can be kept under 1 %. The economic assessment, however, indicates that even for the most promising primary control reserve case, the investment costs of vanadium redox flow batteries must be reduced by at least 30 % in order to break even. Finally, the encouraging key finding is that the negative impact of a grid supportive behaviour, additionally to its primary purpose, is less than 1 % of the revenues. This may encourage distribution grid and battery operators to consider the integration of large scale batteries in distribution grids as part of the solution of a rising share of a decentralised renewable energy generation.