This paper presents chaos synchronization between two different chaotic systems by using a nonlinear controller, in which the nonlinear functions of the system are used as a nonlinear feedback term. The feedback contr...This paper presents chaos synchronization between two different chaotic systems by using a nonlinear controller, in which the nonlinear functions of the system are used as a nonlinear feedback term. The feedback controller is designed on the basis of stability theory, and the area of feedback gain is determined. The artificial simulation results show that this control method is commendably effective and feasible.展开更多
高渗透率新能源波动下系统动态频率预测是实现受端网络频率安全态势感知的基础。该文提出一种基于混合量测和物理状态方程联合驱动的新能源电力系统双向树状长短期记忆网络(combined equation-of-state-driven and data-driven bi-direc...高渗透率新能源波动下系统动态频率预测是实现受端网络频率安全态势感知的基础。该文提出一种基于混合量测和物理状态方程联合驱动的新能源电力系统双向树状长短期记忆网络(combined equation-of-state-driven and data-driven bi-directional tree-struct long short term memory,CEOSD-BITREE-LSTM)动态频率预测方法。首先,引入双层多头注意力图神经网络,提出考虑同步相量测量单元(synchronous phasor measurement unit,PMU)和数据采集与监视控制系统装置(supervisory control and data acquisition,SCADA)量测差异性和时序同步性的混合量测融合策略;其次,依据PMU密集采样特性,建立计及源网荷物理联系的线性时变状态方程,刻画物理-数据空间的频率特征交互关系;然后,考虑新能源出力、负荷波动等不确定因素,结合以PMU并行搜索调频资源形成的拓扑结构,构建CEOSD-BITREE-LSTM动态频率预测模型,实现系统频率态势的高精度预测。最后,以改进新英格兰10机39节点、三区互联系统为算例,验证该文所提方法的可行性和有效性。展开更多
基金Project Supported by the National Natural Science Foundation of China (Grant No 20373021) and Natural Science Foundation of Liaoning Province, China (Grant No 20052151).
文摘This paper presents chaos synchronization between two different chaotic systems by using a nonlinear controller, in which the nonlinear functions of the system are used as a nonlinear feedback term. The feedback controller is designed on the basis of stability theory, and the area of feedback gain is determined. The artificial simulation results show that this control method is commendably effective and feasible.
文摘高渗透率新能源波动下系统动态频率预测是实现受端网络频率安全态势感知的基础。该文提出一种基于混合量测和物理状态方程联合驱动的新能源电力系统双向树状长短期记忆网络(combined equation-of-state-driven and data-driven bi-directional tree-struct long short term memory,CEOSD-BITREE-LSTM)动态频率预测方法。首先,引入双层多头注意力图神经网络,提出考虑同步相量测量单元(synchronous phasor measurement unit,PMU)和数据采集与监视控制系统装置(supervisory control and data acquisition,SCADA)量测差异性和时序同步性的混合量测融合策略;其次,依据PMU密集采样特性,建立计及源网荷物理联系的线性时变状态方程,刻画物理-数据空间的频率特征交互关系;然后,考虑新能源出力、负荷波动等不确定因素,结合以PMU并行搜索调频资源形成的拓扑结构,构建CEOSD-BITREE-LSTM动态频率预测模型,实现系统频率态势的高精度预测。最后,以改进新英格兰10机39节点、三区互联系统为算例,验证该文所提方法的可行性和有效性。