基于深度学习的主动配电网短路计算方法

Deep Learning-Based Short-Circuit Calculation Method for Active Distribution Networks

在低电压穿越(LVRT)控制策略下,具有逆变器接口的分布式电源(IIDG)需要根据连接节点的电压改变逆变器的输出模式。短路电流与系统额定容量、网络短路阻抗和分布式功率输出有关。因此,基于深度学习算法,提出了一种预测电压降的方法。通过预测连接节点的电压,可以根据LVRT控制确定IIDG的输出模式,从而准确建立了IIDG的故障计算模型。最后,在IEEE13节点测试系统中进行编程分析技术来验证所提出方法的有效性。结果表明,与传统的故障电流计算方法相比,所提出的方法可以简化过程,提高故障计算的速度,为主动配电网的保护设置和故障分析奠定了坚实基础。

Under the low voltage ride through(LVRT)control strategy,distributed generators with inverter interfaces(IIDGs)adjust their output modes based on the voltage changes at the connecting nodes.The short-circuit current of the system is related to the rated capacity of the IIDG,network short-circuit impedance,and distributed power output.Therefore,a deep learning algorithm is employed to propose a method for predicting voltage drops.By forecasting the voltage at the connecting nodes,the output mode of the IIDG can be adjusted according to LVRT control requirements,thus accurately establishing a fault calculation model for IIDGs.Finally,this method is verified through programming analysis techniques in the IEEE 13-node test system,demonstrating its effectiveness.The results show that compared to traditional fault current calculation methods,the proposed approach simplifies the process and enhances fault calculation speed,providing a solid foundation for protection settings and fault analysis in active distribution networks.