不确定分数阶同步发电机混沌系统的滑模自适应控制及参数辨识

Adaptive Sliding Mode Control and System Identification in Fractional-order Chaotic Synchronous Generator System with Uncertainty

考虑了一种带有周期性负荷扰动和电磁功率扰动的同步发电机系统模型.基于分数阶稳定性原理,对分数阶系统的动力学特性做了仿真分析.针对系统内部参数和扰动幅值的不确定,利用径向基函数神经网络的万能逼近特性改进滑模控制方法,设计了一种基于径向基函数的自适应滑模控制器,实现参数辨识,使系统输出能渐近跟踪目标轨迹,进而抑制分数阶系统混沌振荡.研究表明,该方法控制时间短、逼近误差小,而且有效地消除了抖振,具有实时控制、鲁棒性高等特点.

Chaos oscillation of a new fractional order synchronous generator system was studied,and its dynamic behaviors under two disturbances via bifurcation diagram and phase plane were discussed.The result proves that the uncertain fractional-order chaotic system can be asymptotically stable.The radial basis function（RBF）neural network has been applied in identification of nonlinear part in fractional-order system.The identification model has been joined in system to compensate nonlinearity,and a no-chattering sliding mode control strategy to control chaos is designed.Simulation results show the effectiveness of the method.