Status : Verified
| Personal Name | Rahayel, Daniel Fadi D. |
|---|---|
| Resource Title | Hierarchical optimization of hybrid energy storage systems in islanded microgrids |
| Date Issued | June 2019 |
| Abstract | A recent improvement of storage systems for microgrids with renewable energy-based generation is the introduction of Hybrid Energy Storage Systems (HESS) that combines the advantages of different storage technologies to improve the lifetime and capacity of each storage. This thesis presents a hierarchical control strategy that employs two levels of Model Predictive Control(MPC): a fast controller controlling the dispatch minute-by-minute and a slower controller controlling the hourly dispatch. A simulated microgrid is built in order to compare the efficacy of Hierarchical Control with respect to a single-horizon MPC control (Flat MPC) and Hysteresis Band Control (HB). The simulated microgrid is set to a medium size microgrid of a nominal load of 60kW, which is a nominal load for a 30 household in an islanded rural community. Generation is served by a nominal solar generation of 30.5kW and a nominal wind generation of 30.5kW, resulting in a net load that is almost zero over the long-term. The fluctuations of this net load are solved by using a combination of three energy storage systems: a 1650 Ultracapcitor Bank(UC), a 24520AH Battery Bank(Batt), and a 50kW dispatchable energy combination of the Fuel Cell and Electrolysis cell (FC/EC). Each of the components of the microgrid are modeled in this study. The load demand is built from device probability simulation, solar and wind generation is computed from irradiance and speed values sourced from weather stations. The UC and Batt are modeled using equivalent circuits whereas the dispatch of the FC/EC is represented as a moving efficiency. Key performance indicators (KPIs) were identified from the cost equations of each storage utilization and state and are defined as the metrics for HESS dispatch control evaluation. Each of the models and the system states are used in MPC predictions and this is implemented using Template Matching where the match libraries are generated using K-means Clustering. A year's worth of micro |
| Degree Course | MS Electrical Engineering |
| Language | English |
| Keyword | optimization energy storage |
| Material Type | Thesis/Dissertation |
Preliminary Pages
121.78 Kb
Category : F - Regular work, i.e., it has no patentable invention or creation, the author does not wish for personal publication, there is no confidential information.
Access Permission : Open Access