摘要
在电动负载模拟器的控制系统设计中,其参数优化一般需要大量试验。根据负载模拟器的工作原理建立了伺服电机的数学模型,在分析CMAC神经网络控制结构的基础上,引入了更新系数和补偿环节,提出了一种基于CMAC神经网络的改进算法,对各控制参数进行了寻优,获得了参数的选择域。通过对系统的数字仿真表明,该方法可稳定地跟踪非线性输入,尤其在大角度高速下加载效果良好,满足项目精度性能要求,对负载梯度和频率变化具有自适应能力。
In the design of the control system for motor-driven load simulator,the optimal controller parameters are generally determined through lots of experiments.Based on analysis to the operating principle of the electric load simulator,the mathematic model of its servo motor was constructed.Then,control structure of CMAC neural network was studied,and renewed coefficients and compensatory segment were introduced.An improved arithmetic based on CMAC neural network was proposed to optimize each parameter,and the definable area was obtained.The result of digital simulation to the system shows that： 1） the method can track nonlinear input steadily,which performs especially well in case of large-angle and high frequency;2） it can satisfy the requirement to precision,and is adaptive to the change of load gradient and frequency.
出处
《电光与控制》
北大核心
2011年第4期72-76,共5页
Electronics Optics & Control
基金
航空科学基金资助(2008ZC13011)
