中文摘要：

为提高发电机励磁汽门系统机端电压和转速的控制性能,提出了基于在线学习和扩张状态观测器（ESO）的改进神经网络逆控制方法.首先,将神经网络逆与被控励磁汽门系统组成的复合伪线性系统等效为含有扰动的线性系统.然后,基于自抗扰控制方法,设计了用于在线估计复合伪线性系统状态和扰动的ESO,并在设计的伪控制量中对扰动进行补偿,利用线性系统理论证明了ESO的收敛性和闭环系统的稳定性;同时,在离线训练的基础上设计了基于在线梯度方法的神经网络逆在线学习算法,利用李雅普诺夫稳定性理论证明了神经网络逆在线学习的收敛性.最后,以典型的两区域四机系统为例进行数值仿真,结果表明,与传统的AVR/PSS方法和基于离线训练的神经网络逆控制方法相比,所提方法明显提升了电力系统的暂态性能.

英文摘要：

To improve the performance of the terminal voltage and rotor speed of a turbogenerator,an improved artificial neural network-inversion（ANNI） control scheme is proposed based on online learning and extended state observer（ESO）.First,the composite pseudo linear system,which is composed of the ANNI system and the controlled excitation and valve system,is equivalent to a linear system with disturbance.Then,an ESO is designed based on the auto-disturbance-rejection-control（ADRC） method to estimate the states and the disturbance of the composite pseudo linear system online.The pseudo control input with disturbance compensation is designed for the composite pseudo-linear system,and the convergence of the ESO and the stability of the closed-loop system are proved by the linear system theory.Meanwhile,an online learning algorithm of the ANNI is proposed with online gradient descent method based on offline training,and the convergence of the online learning algorithm of the ANNI is proved according to the Lyapunov stability principles.Finally,a case study is fulfilled on a typical two-area four-machine power system.The results show that compared with the conventional AVR/PSS and the offline trained based ANNI control scheme,the proposed control scheme can greatly improve the transient performance.