An optimization model of autonomous decentralized energy network and energy routing algorithms based on machine learning techniques

我们研究了一个由多个集群组成的分散式太阳能网络，这些集群对应着配备了光伏单元、充电电池、电器和电力路由器的家庭。路由器管理集群上的电力传输。为了评价整个能源网络的效率和稳定性，建立了一个数学规划模型来优化整个能源网络的设计和利用。通过一些仿真结果，对所提出的模型的有效性和潜力进行了研究，例如，对于澄清电池的影响。此外，我们还建立了分散太阳能网络中电池退化控制的数学规划模型，以明确系统效率与退化控制程度之间的关系。

We study a decentralized solar energy network composed of multiple clusters, which correspond households equipped with PV units, rechargeable batteries, electrical appliances, and an electric power router. The routers manage the power transfer overthe clusters. A mathematical programming model is formulated for optimizing the design and the utilization of this overall energy network in order to evaluate efficiency and stability of the whole energy network. Through some simulation results, the effectiveness and the potential, e.g. for clarifying the effect of the batteries, of the proposed model are investigated. Moreover, we also develop a mathematical programming model which controls degradation of batteries in the decentralized solar energy networks in order to clarify the relationship between the system efficiency and the degree of degradation control.