基于改进粒子群算法优化长短时记忆神经网络的孤网电压控制方法

Isolation Network Voltage Control Method Based on Improved Particle Swarm Optimization of Long Short Term Memory Neural Networks

随着风力发电机、光伏发电机、储能装置等分布式电源的大量接入,电网的组成和运行模式越来越复杂,更多的不确定因素影响着孤网电压的恢复。针对不确定因素下孤网电压优化控制的问题,提出一种基于改进粒子群算法优化长短时记忆神经网络的孤网电压控制方法。首先,分析分布式电源控制器增益与孤网控制器性能的关系;然后,使用粒子群算法优化长短时记忆神经网络的学习率和神经元数量等超参数,防止超参数选择不当影响网络的训练效果;最后,将优化后的长短时记忆神经网络映射分布式电源电压偏差和控制器增益之间的关系,构建孤网电压优化控制模型。仿真结果表明,分布式电源电压经过0.92 s时使电压稳定于1,说明所提的优化控制方法提高了孤网电压水平,验证了该方法的有效性。

With the extensive integration of distributed power sources such as wind turbines,photovoltaic generators,and energy storage devices,the composition and operation mode of the power grid are becoming increasingly complex,and more uncertain factors affect the recovery of isolated grid voltage.Aiming at the problem of optimizing the control of isolated network voltage under uncertain factors,a method for optimizing the control of isolated network voltage based on particle swarm optimization of long and short term memory neural networks is proposed.Firstly,analyze the relationship between the gain of distributed power controller and the performance of isolated network controller;Then,particle swarm optimization algorithm is used to optimize the hyperparameter such as learning rate and number of neurons of the short-term and short-term memory neural network,so as to prevent improper selection of hyperparameter from affecting the training effect of the network;Finally,the optimized long and short term memory neural network is mapped to the relationship between distributed power supply voltage deviation and controller gain,and an optimized control model for isolated network voltage is constructed.The simulation results show that the distributed power supply voltage stabilizes at 1 after 0.92 seconds,indicating that the proposed optimization control method improves the voltage level of the isolated network and verifies the effectiveness of the method.