| Resumo: | In this paper, an optimization model is presented for the solving a multi period operation problem for battery energy storage systems, and an investment analysis. The batteries are installed in a distribution medium voltage grid environment, in which stochastic consumption and dispersed generation profiles were considered, as well as penalty factors for power flow and technical grid constrains. Since the problem is characterized to have a large dimension, with dependences between time periods, the optimization requires a model with high efficiency and low computational weight. The solution presented is an alternative variant of Benders decomposition, with connected multi time stages called dual dynamic programming, that allows the division of the main problem in smaller subproblems, that are easier to solve separately. An integral linear programing model was also implemented to compare the results between both approaches. The results suggest that the utilization of the dual dynamic programming, in this case study, increased the computational performance, as well as good battery operation solutions.
|
|---|