Assessing the influence of solar forecast accuracy on the revenue optimization of photovoltaic + battery power plants in day-ahead energy markets

Abstract

Energy storage has become a subject of great interest in the last years due to the increasing penetration of non-dispatchable renewable energy power plants, especially solar photovoltaics (PV) and wind. Energy storage systems can add value to the grid in many ways: energy schedule optimization, ancillary services, renewable energy integration at high penetration levels, and so on. However, from the perspective of the asset's owner, it is important that the system regulator provides a regulatory framework and energy prices that stimulate the private plant operator to dispatch its energy storage system to assist the grid and get rewarded by these operations. An optimization problem is presented from the perspective of an asset owner with a PV plant and battery energy storage system. From this perspective, the agent must know the day-ahead spot-market energy prices and predict the day-ahead PV generation to decide on how to dispatch the battery system. We have evaluated 11 forecast accuracies to assess how the forecast accuracy can impact on the agent's total revenue.The findings showed that, if the agent's objective is to use the battery system only to guarantee the contracted energy, the battery system does not add as much value as it could; and higher forecast uncertainties lead to higher revenues, because the battery will then operate closer to the tariff arbitrage strategy, which is the optimal strategy for battery dispatch, independent of PV generation, relying only on the spot-market energy prices.