论文部分内容阅读
建立了飞机泵源负载模拟系统的数学模型,针对系统的非线性和各种不确定因素,提出了基于小脑神经网络的复合控制方法,控制器由比例-积分-微分(PID)和小脑模型神经网络控制器(CMAC)构成,该方法在传统的PID前馈控制方法上加入了CMAC神经网络快速学习算法,保证了快速实时跟进,进一步提高了控制精度。仿真结果表明,CMAC-PID能够较好解决PID在快速性和控制精度(稳定性)之间的矛盾,对抑制系统的非线性时变性具有一定效果。
A mathematic model of aircraft pump load simulation system is established. Aiming at the nonlinearity and various uncertainties of the system, a hybrid control method based on cerebellar neural network is proposed. The controller is composed of proportional-integral-derivative (PID) Network controller (CMAC). This method adds the CMAC neural network fast learning algorithm to the traditional PID feedforward control method to ensure rapid real-time follow-up and further improve the control accuracy. The simulation results show that the CMAC-PID can solve the conflict between the fastness and the control accuracy (stability) of the PID, and has certain effect on suppressing the nonlinear time-varying of the system.