TY - JOUR
T1 - Degradation and Operation-Aware Framework for the Optimal Siting, Sizing, and Technology Selection of Battery Storage
AU - Sayfutdinov, Timur
AU - Patsios, Charalampos
AU - Vorobev, Petr
AU - Gryazina, Elena
AU - Greenwood, David M.
AU - Bialek, Janusz W.
AU - Taylor, Philip C.
PY - 2020/10
Y1 - 2020/10
N2 - This paper addresses the problem of optimal siting, sizing, and technology selection of Energy Storage System (ESS) considering degradation arising from state of charge and Depth of Discharge (DoD). The capacity lost irreversibly due to degradation provides the optimizer with a more accurate and realistic view of the capacity available throughout the asset's entire lifetime as it depends on the actual operating profiles and particular degradation mechanisms. When taking into account the ESS's degradation, the optimization problem becomes nonconvex, therefore no standard solver can guarantee the globally optimal solution. To overcome this, the optimization problem has been reformulated to a Mixed Integer Convex Programming (MICP) problem by substituting continuous variables that cause nonconvexity with discrete ones. The resulting MICP problem has been solved using the Branch-And-Bound algorithm along with convex programming, which performs an efficient search and guarantees the globally optimal solution. We found that the optimal battery use does not necessarily correspond to it reaching its End of Life state at the end of the service lifetime, which is the result of nonlinear degradation mechanicms from both idling and cycling. Finally, the proposed methodology allows formulating computationally tractable stochastic optimization problem to account for future network scenarios.
AB - This paper addresses the problem of optimal siting, sizing, and technology selection of Energy Storage System (ESS) considering degradation arising from state of charge and Depth of Discharge (DoD). The capacity lost irreversibly due to degradation provides the optimizer with a more accurate and realistic view of the capacity available throughout the asset's entire lifetime as it depends on the actual operating profiles and particular degradation mechanisms. When taking into account the ESS's degradation, the optimization problem becomes nonconvex, therefore no standard solver can guarantee the globally optimal solution. To overcome this, the optimization problem has been reformulated to a Mixed Integer Convex Programming (MICP) problem by substituting continuous variables that cause nonconvexity with discrete ones. The resulting MICP problem has been solved using the Branch-And-Bound algorithm along with convex programming, which performs an efficient search and guarantees the globally optimal solution. We found that the optimal battery use does not necessarily correspond to it reaching its End of Life state at the end of the service lifetime, which is the result of nonlinear degradation mechanicms from both idling and cycling. Finally, the proposed methodology allows formulating computationally tractable stochastic optimization problem to account for future network scenarios.
KW - Battery degradation
KW - convex optimization
KW - energy storage
UR - http://www.scopus.com/inward/record.url?scp=85084940404&partnerID=8YFLogxK
U2 - 10.1109/TSTE.2019.2950723
DO - 10.1109/TSTE.2019.2950723
M3 - Article (Academic Journal)
AN - SCOPUS:85084940404
SN - 1949-3029
VL - 11
SP - 2130
EP - 2140
JO - IEEE Transactions on Sustainable Energy
JF - IEEE Transactions on Sustainable Energy
IS - 4
M1 - 8889420
ER -