基于人工神经网络算法的配电网重构研究

Research of Distribution Network Reconfiguration Based on Artificial Neural Network Algorithm

针对传统配电网重构计算周期长和配电网负荷具有动态性的特征,为了充分挖掘电网配电潜能,论文提供了一套基于人工神经网络算法的配电网在线重构方案。首先将标准的Elman神经网络结构上增加输出层对隐含层节点的反馈,形成输出输入反馈型神经网络,提高网络的输出稳定性,然后在训练过程中,采用人类群体优化算法用于配电网重构计算,最后通过IEEE30电源电网进行训练分析,测试结果表明此方案在配电网重构方面优势明显。

Aiming at the long calculation period and dynamic load of traditional distribution network reconfiguration,an on-line reconfiguration scheme based on artificial neural network algorithm is proposed in this paper to fully tap the potential of dis⁃tribution network.Firstly,the output-input feedback neural network is formed by adding the output layer feedback to the hidden lay⁃er nodes in the standard Elman neural network structure to improve the output stability of the network.Then,in the training pro⁃cess,the human colony optimization algorithm is used to calculate the distribution network reconfiguration.Finally,the IEEE 30 power grid is trained and analyzed.The test results show that the scheme has obvious advantages in reconfiguration of distribution network.