In this study, a hierarchical energy management system (EMS) is proposed, to coordinate different energy sources in an islanded multi-good microgrid. The first layer deals with the daily scheduling problem, while the second layer balances the generation in real-time. A novel second layer formulation, relying on model predictive control under a scenario-based stochastic approach (sMPC), is introduced and it is compared to a reference formulation, based on a central proportional-integral controller following the indications set by the first layer. The proposed sMPC explicitly accounts for uncertainty considering several scenarios of very-short term forecast errors, that act as disturbances for the system. The sMPC evaluates the control actions and the correction rules required to guarantee optimal operations through disturbance-feedback. The EMS is implemented in an experimental setup and tested for daily operations under a rolling horizon approach. The accuracy of the numerical system simulation is evaluated, resulting in an average discrepancy of 1.7%, in terms of operation cost, with respect to the experimental operations. Then, a test case comparing the proposed EMS with the reference approach shows that the adoption of sMPC allows to approach the lowest possible operation cost achievable by a second layer with an advantage of 2.7 % against the reference case. Finally, the developed sMPC leads to only 0.5% additional costs than an ideal controller working on the same control layer.

Development and experimental validation of hierarchical energy management system based on stochastic model predictive control for Off-grid Microgrids

Polimeni S.;Meraldi L.;Moretti L.;Leva S.;Manzolini G.
2021

Abstract

In this study, a hierarchical energy management system (EMS) is proposed, to coordinate different energy sources in an islanded multi-good microgrid. The first layer deals with the daily scheduling problem, while the second layer balances the generation in real-time. A novel second layer formulation, relying on model predictive control under a scenario-based stochastic approach (sMPC), is introduced and it is compared to a reference formulation, based on a central proportional-integral controller following the indications set by the first layer. The proposed sMPC explicitly accounts for uncertainty considering several scenarios of very-short term forecast errors, that act as disturbances for the system. The sMPC evaluates the control actions and the correction rules required to guarantee optimal operations through disturbance-feedback. The EMS is implemented in an experimental setup and tested for daily operations under a rolling horizon approach. The accuracy of the numerical system simulation is evaluated, resulting in an average discrepancy of 1.7%, in terms of operation cost, with respect to the experimental operations. Then, a test case comparing the proposed EMS with the reference approach shows that the adoption of sMPC allows to approach the lowest possible operation cost achievable by a second layer with an advantage of 2.7 % against the reference case. Finally, the developed sMPC leads to only 0.5% additional costs than an ideal controller working on the same control layer.
Energy management systems
MILP optimization
Off-grid Microgrid
Stochastic model predictive control
File in questo prodotto:
File Dimensione Formato  
1-s2.0-S2666792421000214-main.pdf

accesso aperto

: Publisher’s version
Dimensione 3.37 MB
Formato Adobe PDF
3.37 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11311/1200103
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 10
  • ???jsp.display-item.citation.isi??? ND
social impact