In this paper we propose two optimization-based algorithms for coordinating residential battery storage to balance increases in daily operational savings that accrue to residential customers with the management of bi-directional power flows in the distribution grid. Bi-directional power flows are managed to improve the supply voltage for residential customers with rooftop solar PV, in addition to alleviating (potentially infrequent) congestion that occurs in the evening when PV production is unavailable. Our objectives are threefold: (1) to reduce reverse power flow leading to significant voltage rise; (2) to reduce peak loads creating sustained under-voltages and/or approaching a network thermal capacity; and (3) to increase operational savings for the residential customer. To achieve our objectives we present a Distributed-Receding Horizon Optimization (D-RHO) algorithm, wherein charge and discharge rates of residential battery storage are coordinated so as to directly influence power flows along a distribution feeder. We also present an Adaptive-Receding Horizon Optimization (A-RHO) algorithm, in which charge and discharge rates of residential battery storage are coordinated to more directly manage supply voltages. To assess the distributor benefit, both RHO-based algorithms are applied to a publicly available model of an Australian distribution region located in Elermore Vale. The results of this case study confirm that the A-RHO algorithm improves supply voltages in a low voltage network, and that the D-RHO algorithm offers a peak load reduction of 32% along the Elermore Vale medium voltage feeder.

在本文中，我们提出了两种基于优化的算法，用于协调住宅电池存储，以平衡住宅用户通过配电网中双向功率流管理而产生的日常运营节省。双向功率流得到管理，以改善屋顶光伏太阳能住宅用户的供电电压，此外还缓解了（有时不常发生的）晚上无法使用光伏发电的交通拥堵情况。我们的目标有三方面：（1）减少导致大幅度升压的反向功率流；（2）减少峰值负载，从而产生持续的欠压和/或接近电网的热容量；（3）增加居民用户的运营成本。为了实现我们的目标，我们提出了分布式后备水平优化（D-RHO）算法，其中协调住宅电池存储的充放电速率，以便直接影响沿着配电馈线的电流。我们还提出了一种自适应渐进式地平线优化（A-RHO）算法，该算法可协调住宅电池存储的充放电速率，以更直接地管理电源电压。为了评估分销商的利益，将两种基于RHO的算法应用于位于Elermore Vale的澳大利亚分销地区的公开可用模型。此案例研究的结果证实，A-RHO算法可改善低压网络中的电源电压，并且D-RHO算法可沿Elermore Vale中压馈线将峰值负载降低32％。