Energy Storage Location and Capacity Planning Based on Distribution Network Partition and ResNet DNN

Abstract

With the large-scale grid connection of renewable energy, a large amount of energy may be wasted. The energy storage system with strong regulation ability provides a solution. Reasonable planning of energy storage system is the key point. Therefore, aiming at the problem of renewable energy consumption in the new power system, this paper proposes an energy storage location and capacity planning based on distribution network partition and ResNet DNN. Firstly, the mathematical model is established according to the optimization objective of the energy storage system; Secondly, according to the output and consumption of nodes, the Lssvm is used to classify the data of load and renewable energy output, so as to partition the distribution network and determine the distribution area of the energy storage system; Finally, a deep neural network is established in each region and a residual block is introduced to determine the location and capacity of energy storage with the objective of the best consumptive ability. It decomposes the large-scale energy storage optimization problem into several sub problems, so as to meet the configuration requirements of large-scale energy storage system under the high proportion of renewable energy access. We use the typical monthly real data set in the standard IEEE 33 node for experimental analysis. Under different energy storage configurations, the proposed method can effectively promote the accommodation of renewable energy generation.