基于RBF神经网络滑模控制的互联电力系统混沌控制研究  

RBF neural network based sliding mode controller and its application to chaos control in interconnected power system

作　　者：王年[1,2] 陈辉 丁大为[1] 胡永兵 程志友[1,2] Wang Nian;Chen Hui;Ding Dawei;Hu Yongbing;Cheng Zhiyou(School of Electronics and information engineering,Anhui University,Hefei,230601,China;Power Quality Engineering Research Center,Ministry of Education Anhui University,Hefei,230601,China)

机构地区：[1]安徽大学电子信息工程学院,合肥230601 [2]安徽大学教育部电能质量工程研究中心,合肥230601

出　　处：《南京大学学报（自然科学版）》2018年第5期911-920,共10页Journal of Nanjing University（Natural Science）

基　　金：国家自然科学基金(61201227);安徽省科技重大专项(16030901058)

摘　　要：电力系统是一种具有多自由度、强耦合特性的非线性动力学系统,混沌振荡会对电力系统的安全性造成极大危害.通过研究互联电力系统的时序图、最大Lyapunov指数图、相平面图来讨论电磁功率扰动幅值对系统动力学特性的影响.提出了一种基于继电特性函数的改进型径向基函数(Radial Basis Function,RBF)神经网络自适应滑模控制方法,快速平滑地使系统达到控制目标.实验表明,相比自适应滑模控制方法和基于继电特性函数的滑模控制方法,提出的改进型RBF神经网络滑模控制方法不仅可以使系统有较快的收敛速度,而且对抖震现象有很好的抑制作用,对外部扰动具有很强的鲁棒性.Power system is a nonlinear dynamical system with multiple degrees of freedom and strong coupling.When chaotic oscillation occurs in a power system,it has a bad effect on the stability of whole system.In this paper,by using of timing diagram,maximum Lyapunov exponent,and phase diagram,we discuss the relationship of the electromagnetic power disturbance and the stability of the interconnect power system.Then,an improved adaptive sliding mode control method based on RBF neural network and relay characteristic function has been proposed in order to restrain the chaotic oscillation in such interconnect power system.By combining these two methods,the interconnect power system not only attains the expected target,but also has strong robustness when there is an external disturbance.Comparing with traditional adaptive sliding mode control,the RBF neural network based sliding mode control method can make the convergence speed faster.Meanwhile this method can reduce the buffetingeffectively.Finally,some numerical simulations are given to demonstrate the effectiveness of the proposed methods.

关 键 词：互联电力系统 混沌控制 RBF神经网络 滑模控制 

分 类 号：N93[自然科学总论]