基于一维卷积神经网络的配电网无功优化

Reactive power optimization of distribution network based on one-dimensional convolutional neural network

深度学习和数据驱动技术的快速发展,为历史数据下的配电网无功优化提供了新的解决方法,提出了一种基于一维卷积神经网络的配电网无功优化方法。利用配电网节点的历史负荷数据,用优化算法得到对应的无功优化策略,并将无功优化策略进行二进制编码。通过训练一维卷积神经网络模型来映射配电网节点负荷和无功优化策略间的非线性关系,将训练好的模型用于配电网无功优化。在一个改造后的IEEE 33配电网节点系统进行仿真验证,结果表明相比九区图无功优化,所提方法的系统网损和节点电压偏移明显降低。

The rapid development of deep learning and data-driven technology provides a new solution for reactive power optimization of distribution network under historical data, and a reactive power optimization method of distribution network is proposed based on one-dimensional convolution neural network. Using the historical load data of distribution network nodes, the corresponding reactive power optimization strategy is obtained by optimization algorithm, and the reactive power optimization strategy is binary coded. By training one-dimensional convolutional neural network model to map the nonlinear relationship between node load and reactive power optimization strategy, the trained model is used for reactive power optimization of distribution network. The simulation verification is carried out in a modified IEEE 33 distribution network node system, and the results show that compared with the reactive power optimization in the nine-zone diagram, the system network loss and node voltage deviation of the proposed method are significantly reduced.