Full Life-cycle Optimal Battery Scheduling for Maximal Lifetime Value Considering Degradation
Zonglin Liu,Xin Wang,Feng Zhang
DOI: https://doi.org/10.1109/tec.2021.3131232
IF: 4.8767
2021-01-01
IEEE Transactions on Energy Conversion
Abstract:The dilemma between battery operating revenue and aging has not been addressed appropriately by previous studies. This paper presents a co-optimization method of battery full life-cycle scheduling and lifespan for its maximal lifetime value, involving usage limits, nonlinear aging characteristics, and time-value-of-money effects. The resulting dispatch model consists of numerous large-scale non-convex nonlinear programming (NLP) problems with the nonlinearity appearing in the equality constraint, which cannot be solved efficiently through conventional methods. Thus, we design an exact and fast solution approach including the equivalent problem transformation, a modified pegging algorithm, and a novel acceleration algorithm. The proposed methods are illustrated and validated by numerical tests using real historical data from Pecan St. Project. Test results suggest that it is not the case that the longer service life, the more battery lifetime profits can be generated. The proposed scheduling method can improve the lifetime profits than the conventional methods. For a battery with a 15-year design life, the optimal scheduling and lifespan with a 5-day resolution can be obtained within half an hour.
energy & fuels,engineering, electrical & electronic