Controller-dynamic-linearization-based data-driven load frequency control for interconnected power systems
ZHANG Yan
BU Xu-hui
CHEN Zong-yao
Abstract：To the problem that complex power systems are difficult to model accurately,system parameter perturba-tions and nonlinear physical constraints lead to the degradation of frequency modulation performance,a data-driven load frequency control(LFC)algorithm based on controller dynamic linearization is proposed.Firstly,the nonlinear inter-connected power system is equivalent to a dynamic linear function model and the estimated value of the pseudo partial derivative(PPD)is obtained by using an adaptive observer.Assuming the existence of an ideal controller,the equivalent parameterized realizable controller form is given.Secondly,a long short-term memory(LSTM)neural network is construct-ed to adjust the controller parameters online.The stability of closed-loop power system and the convergence of observer estimation method are proved strictly in theory.Finally,the effectiveness of the proposed load frequency control algorithm in realizing frequency regulation is verified by simulation on the interconnected power system,which is independent of the model information of the power system and does not measure the system status signal.
Keywords：interconnected power systemsload frequency controldynamic linearizationlong short-term memory neural networks
Publication Date：2025-09-30
Online Publishing Date：2025-10-28(First online date of this platform, not the publication date of the document)
Pages：9( 1818-1826 )