Electrification and distributed energy sources are central strategies for accelerating the green energy transition and still pose significant challenges to the existing energy systems, requiring a deep rethinking of the energy infrastructures, both under the point of view of resilience and economic sustainability. Indeed, smart grids and battery energy storage systems (BESS) are rising in importance both for technical and economical aspects as they enable a series of services such as active demand response, load peak reduction or mitigation, improved voltage stability etc. However, the economic viability of distributed generation coupled with storage systems is still being debated, as hidden costs associated to their use are complex to include in planning tools, especially when considering aging, and remain difficult to evaluate ex-ante, limiting the large scale penetration of BESS technologies. Here we present BRAINS (Balancing Responsive Artificial Intelligence Storage), a tool for planning and real time management of electrochemical storage systems based on a cross-disciplinary approach bridging Artificial Intelligence (AI), data science, and economics. BRAINS is used in the operation of a microgrid acting as a virtual power plant: the optimal sizing of the storage is determined according to the specific (and configurable) load characteristics, while the realtime management of the State of Charge (SoC) takes into account degradation costs while minimizing the total costs for the whole system. We also show that by including the aging and cycle costs in the BESS management, the expected return of investment (ROI) is drastically reduced, and in some cases economic losses are observed. Furthermore, under the planning point of view we also show that storage sizing is critical, as we observe that the larger is not the better, while an optimal size must be computed for each specific application case.
ASSESSING BATTERY MANAGEMENT FOR ENERGY COMMUNITIES: ECONOMIC EVALUATION OF A ARTIFICIAL INTELLIGENCE (AI) LED SYSTEM
Alessandro Rubino
Writing – Original Draft Preparation
;
2019-01-01
Abstract
Electrification and distributed energy sources are central strategies for accelerating the green energy transition and still pose significant challenges to the existing energy systems, requiring a deep rethinking of the energy infrastructures, both under the point of view of resilience and economic sustainability. Indeed, smart grids and battery energy storage systems (BESS) are rising in importance both for technical and economical aspects as they enable a series of services such as active demand response, load peak reduction or mitigation, improved voltage stability etc. However, the economic viability of distributed generation coupled with storage systems is still being debated, as hidden costs associated to their use are complex to include in planning tools, especially when considering aging, and remain difficult to evaluate ex-ante, limiting the large scale penetration of BESS technologies. Here we present BRAINS (Balancing Responsive Artificial Intelligence Storage), a tool for planning and real time management of electrochemical storage systems based on a cross-disciplinary approach bridging Artificial Intelligence (AI), data science, and economics. BRAINS is used in the operation of a microgrid acting as a virtual power plant: the optimal sizing of the storage is determined according to the specific (and configurable) load characteristics, while the realtime management of the State of Charge (SoC) takes into account degradation costs while minimizing the total costs for the whole system. We also show that by including the aging and cycle costs in the BESS management, the expected return of investment (ROI) is drastically reduced, and in some cases economic losses are observed. Furthermore, under the planning point of view we also show that storage sizing is critical, as we observe that the larger is not the better, while an optimal size must be computed for each specific application case.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
