CAJ | 학술논문

为了提高二级倒立摆系统实时控制的响应速度和稳定性,在设计Mamdani型模糊推理规则控制器控制倒立摆系统稳定的基础上,设计了一种更有效率的基于Sugeno型模糊推理规则的模糊神经网络控制器.该控制器使用BP神经网络和最小二乘法的混合算法进行参数训练.能够准确归纳输入输出量的模糊隶属度函数和模糊逻辑规则.通过与Mamdani型控制器的仿真对比及实际控制实验结果,表明该Sugeno型模糊神经网络控制器时二级倒立摆实验装置的控制具有良好的稳定性、快速性和较高的控制精度.
위료제고이급도립파계통실시공제적향응속도화은정성,재설계Mamdani형모호추리규칙공제기공제도립파계통은정적기출상,설계료일충경유효솔적기우Sugeno형모호추리규칙적모호신경망락공제기.해공제기사용BP신경망락화최소이승법적혼합산법진행삼수훈련.능구준학귀납수입수출량적모호대속도함수화모호라집규칙.통과여Mamdani형공제기적방진대비급실제공제실험결과,표명해Sugeno형모호신경망락공제기시이급도립파실험장치적공제구유량호적은정성、쾌속성화교고적공제정도.
To improve the response speed and the stability of double inverted pendulum system in real-time control,a Sugeno fuzzy neural network controller is designed based on a Mamdani fuzzy controller which can stably control the double inverted pendulum system.A BP network with the least square method is used to train the Sugeno fuzzy neural network controller.The membership functions and fuzzy logic ndes of the input and output variables are modified.The simulation results and real-time control experiment show that the Sugeno fuzzy neural network controller has better stability,quicker response and higher control precision as compared with the Mumdani fuzzy controller.