Virtual Power Plants (VPPs) and Multi-Energy Systems (MESs) are aggregated energy systems comprising renewable energy sources, energy storage systems and dispatchable units. The presence of such diverse systems unlocks the possibility of a near-zero carbon emission energy generation while overtaking the main drawback of renewables sources that is their lack of control. Among the different energy storage systems, large scale (seasonal) H2 storages (e.g. salt cavern or depleted oil field) would allow shifting the excess solar energy from the hot to the cold season. High round-trip efficiency (electricity to electricity) and unpaired operational flexibility could be achieved using H2 in state-of-the-art combined cycles. This work investigates the optimal design and operation of a fully renewable VPPs integrating PV panels, batteries for short-term storage, electrolyzers, H2 seasonal storage and H2-fired combined cycles. The optimal design and optimal yearly operation of such complex VPP are formulated as Mixed Integer Linear Programs (MILP) and solved to global optimality imposing to meet the highest possible fraction of the electricity demand profile. Results indicate that the optimal VPP design features a 490 MWhel of battery, 687 MWel of PV panels, 392 MWel of electrolyzer and requires a minimum H2 storage size of 168 GWhH2,LHV to power a combined cycle of 58 MWel. In case of a geological H2 seasonal storage availability, the resulting cost of generated electricity is above 430 $/MWhel, considerably higher with respect to the average electricity prices in Italy (in the range 50–80 $/MWhel) underlining the need of achieving better power-to-gas efficiencies and lower specific investment costs of conversion technologies in the next years. Furthermore, if the H2 storage needs to be built on purpose, the resulting cost of electricity would be even higher.
This work presents a two-stage stochastic Mixed Integer Linear Programming model for the optimization of the design of an aggregated energy system (AES) (i.e., multi-energy systems, microgrids, energy districts, etc.) serving a university campus featuring electricity and heating demands. The off-grid system design is obtained by considering a set of representative periods for both demands by means of a carefully modified k-medoids algorithm. N-1 reliability is also considered in the model, by introducing the concept of "break-down scenarios" that allows the solution of the problem to be able to meet the user demands for every possible contingency in which one of the AES's units fails. The effect of including N-1 reliability in the model is then showed by comparing the optimal design obtained by considering such approach against one with no break-down scenarios.
This work proposes two ad hoc part-load control strategies for steam cycles adopted in concentrated solar power plants. The control strategies are designed to keep the molten salt temperature above the minimum allowed value set by solidification issues in the 30–100 % load range. Particularly critical is the temperature of molten salts in contact with the heat exchanger tubes, the so called skin temperature. The first control strategy adopts a turbine with controlled extraction and readmission valve while the second strategy employs a throttling valve and a feedwater preheating loop. Off-design simulations show that both strategies are capable of avoiding the molten salts solidification issue but at the cost of a non negligible penalty (up to −1.9 percentage points) in power block efficiency at low loads (30–50%). The off-design analysis considers also the effect of ambient temperature variations and the optimization of the cooling fan rotational speed. The results are used to derive best-fit polymonials relating the power block efficiency to the ambient temperature and load.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.