摘要
提出了1种基于频域特性分析的在线模型频域特性辨识方法,该方法算法简单,易于实现,且具有较强的抗干扰性和较高的辩识精度。在PID调节器的输出信号和过程输出信号中均包含与辩识频谱信号相同的频谱成分,当频谱的幅值和相位特性发生变化时,该变化与PID调节器、控制对象和闭环控制系统的特性相关。通过对PID调节器输出信号中频谱成分的幅值和相位与过程输出信号中频谱成分的幅值和相位比较,可辩识得到控制对象的频域特性,仿真试验及应用研究验证了该方法的有效性。当在参考滤波器模块组输入中加入带宽为80.0Hz,平均幅值为方波辨识信号幅值100%的高频高斯干扰信号后,对二阶惯性环节相位频率特性的仿真辩识精度大于0.64°,二阶惯性环节幅值频率特性仿真辩识精度大于1.1%。
A novel online model identification method based on frequency domain analysis has been proposed,which has simple algorithm and is easy to realize.Moreover,this method has good noise immunity and identification precision.Input and output signal of the PID regulator both contain spectrum composition which is the same as the identification spectrum signal.The change of amplitude characteristic and phase property of spectrum is related to the properties of PID regulator,control object and the closed loop control systems.The control model can be identified through amplitude and phase comparison for spectrum of input and output signals.The simulation and application results verified the effectiveness of this method.When the Gauss interference signal with 80 Hz bandwidth and whose average amplitude is 100% of square wave identification signal was added to the input of reference filter module group,the simulation accuracy of phase-frequency characteristics of second order inertia link identification was higher than 0.64°,and the simulation accuracy of amplitude-frequency characteristics of that was better than 1.1%.
出处
《热力发电》
CAS
北大核心
2012年第9期44-49,共6页
Thermal Power Generation
关键词
频域特性
在线辨识
模型
频谱
PID调节器输出信号
过程输出信号
frequency domain characteristic
online identification
model
spectrum
PID regulator
output signal
process output signal
