Storage requirements for PV power ramp-rate control in a PV fleet

• Published in: Solar energy Vol. 118; pp. 426 - 440
• Main Authors: de la Parra, I., Marcos, J., García, M., Marroyo, L.
• Format: Journal Article
• Language: English
• Published: New York Elsevier Ltd 01.08.2015; Pergamon Press Inc
• Subjects: Aggregated PV plants; Alternative energy; Electricity distribution; Electricity generation; Energy storage sizing; Photovoltaic cells; Power fluctuation smoothing; Ramp-rate control; Solar energy; Navarre Spain; Energy storage sizing; Ramp-rate control; Power fluctuation smoothing; Aggregated PV plants
• ISSN: 0038-092X; 1471-1257
• DOI: 10.1016/j.solener.2015.05.046

•Ramp-rate control is simulated for smoothing PV power fluctuations.•An analytical model to calculate energy storage requirements in a PV fleet is presented.•Model is validated with 1s real operational data.•Energy storage requirements for centralised and distributed energy storage systems are compared.•The centralised energy storage system results in considerably cost savings. Considering the rapid development of PV power generation, one of the greatest challenges to the integration of this solar resource into the electricity grid, is its intermittent nature and, therefore, the variability of the power produced. As a result, some transmission system operators have issued new grid codes to limit the maximum allowable ramp-rate and make it easier for them to react appropriately to harmful short-term power fluctuations. In order to comply with the established grid codes, there is a need to include some kind of energy storage systems, thereby increasing the cost of the renewable energy generation system. This paper presents a methodology to size the storage requirements of a centralised energy storage system connected at the distribution level, serving to smooth out the power variability of a PV fleet in either a grid sector or in the main grid in the case of stand-alone power islands. Then, a study is conducted of the storage requirement savings by taking advantage of the smoothing effect of geographical dispersion on PV power output using a centralised instead of a distributed energy storage system. The general validity of the model was confirmed through simulations carried out with real operational PV power output data taken every 1s in the course of one year at five PV power plants located in the south of Navarre (Spain).