A behavioral model of solar/diesel/Li-ion hybrid power systems for off-grid applications: Simulation over a lifetime of 10 years in constant use
Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Remote hybrid power systems (RHPS) serve local off-grid loads or various island grids when no grid extension is possible. They combine renewable resources, conventional generators and energy storage systems in order to balance the load at any moment, while ensuring power quality and energy security similar to large centralized grids. Modelling such a complex system is crucial for carrying out proper simulations for predicting the system’s behavior and for optimal sizing of components. The model should include an estimation of the renewable resource availability over the period of the simulation, a prediction of the load consumption and time variation, the choice of technologies, a prior dimensioning approach, an energy dispatching strategy between the components and, finally, the behavioral model of all components.
This study limits its scope to the simulation of a RHPS composed of solar PV panels, a diesel generator set, and a Li-ion battery bank supplying a dynamic isolated load with a daily demand variation between 10 kW and 80 kW. Methods for building load scenarii are explained first and then, rules of thumbs for selecting the technologies and pre-sizing the components are reviewed. Commonly used dispatching strategies are described before detailing the algorithm of a Matlab behavioral model for the system’s components with an emphasis on the proper prediction of performance and aging for the Li-ion battery model. Finally, a 10-year simulation is carried out over a case study and the results are analyzed.
Place, publisher, year, edition, pages
2015. , 98 p.
Hybrid power plant, off-grid, renewable, Li-ion battery, dispatching strategy, battery aging, modelling, Matlab, Simulink
IdentifiersURN: urn:nbn:se:kth:diva-160967OAI: oai:DiVA.org:kth-160967DiVA: diva2:793129
2015-02-24, 10:51 (English)