节点文献

两轴转台的非线性多模型建模与自适应鲁棒控制

Nonlinear Multi-model Modeling and Adaptive Robust Control of Two-axis Turntable

【作者】 刘明;

【导师】 甘明刚;

【作者基本信息】 北京理工大学 , 控制工程(专业学位), 2018, 硕士

【摘要】 如今,高精度转台应用较广,尤其是在目标跟踪的应用发展较快。本文以实际工程项目中的两轴转台为研究对象,根据转台的特性建立模型,设计控制器,有效抑制外界扰动,提升转台精度,具有重要的现实意义。本文首先对转台进行建模。根据Lagrange法应用第二类Lagrange方程建立转台运动方程,然后综合分析了经典的摩擦模型以及摩擦的静态特性和动态特性后建立基于速度不同情况下的摩擦非线性多模型,最后建立含转台运动特性以及摩擦非线性多模型的伺服系统非线性多模型。然后对非线性多模型系统进行参数辨识。针对模型的结构已知,参数未知的情况对辨识问题进行讨论研究。考虑到存在高频噪声、参数非线性等问题,辨识前对模型进行预处理、参数线性化。在不同的角速度条件下,采用递推最小二乘法分别辨识出了多模型参数的实际值,并通过仿真实验与转台实际实验进行对比,验证了模型结构与参数的准确性。然后在得到的转台伺服系统非线性多模型下,设计自适应鲁棒控制器。对于多模型中每一个具体模型根据参数区间再划分为多个固定模型。其中,针对多模型结构提出了一种速度直接切换和性能指标间接切换相结合的切换策略。对于多固定模型采用性能指标间接切换的策略。根据多个固定模型的参数选取,设计一种参数区间变化的自适应率,然后采用了模型参数辨识器得到模型参数辨识值,经过卡尔曼滤波的方法,将参数的自适应估计值和辨识器下的辨识值进行参数数据融合,得到最优化的参数估计值。同样,在设计鲁棒控制器部分中,参数区间也将根据当前多固定模型的参数选取得到。最后本文在实际应用中对上述内容进行验证。在构建的基于x PC Target半实物仿真实验系统中,分别进行了参数辨识实验、模型验证实验以及控制效果验证实验,得出了实际效果与理论分析相一致的结果。

【Abstract】 Nowadays,the high-precision turntable is widely used,especially in the application of target tracking.This paper takes the two axis turntable in the actual project as the research object.The model of the two axis turntable is established according to the characteristics of the turntable,and the controller is designed to effectively suppress the external disturbance and improve the accuracy of the turntable,which has important practical significance.First,this paper carries out the modeling of the turntable.Based on the Lagrange method,the second kind of Lagrange equation is used to establish the motion equation of the turntable.Then the classical friction model and the static and dynamic characteristics of the friction are synthetically analyzed.The nonlinear multi-model of friction is established on the basis of different velocity conditions.Finally,the servo system nonlinear multi-model with the motion specificity of the turntable and the multi-model of the friction nonlinear model is established.Then,after obtaining the nonlinear multi-model structure of the system,the multiple models are identified.For the structure of the models are known and the parameters are unknown,the identification problem is discussed.Considering the high frequency noise and other problems,the model is preprocessed and the parameters are linearized before identification.Under the different angular velocity conditions,the actual values of multi-model parameters are identified by recursive least square method,and the accuracy of the model structure and parameters is verified by comparing the simulation experiment with the actual experiment of the turntable.Then,the adaptive robust controller is designed under the nonlinear multi-model of the turntable servo system.For a multi-model,each specific model is further divided into several fixed models according to the parameter interval.In allusion to the multi-model structure,a switching strategy combining speed direct switching and performance index indirect switching is proposed.In allusion to multiple fixed models,performance index indirect switching strategy is adopted.According to the parameter selection of multiple fixed models,the adaptive rate of parameter interval change is designed.Then model parameter identifier is used to get the model parameter identification value.After Calman filtering,the adaptive estimation value of parameters and identification value under the identifier are fused,and the optimal parameter estimation is obtained.Similarly,in the design of robust controller,the parameter interval will be selected according to the parameters of the current multiple fixed models.Finally,the above contents are verified in practical application.In the hardware in the loop simulation experiment system based on x PC Target,the parameters identification experiment,model verification experiment and control effect verification experiment are carried out respectively,and the results are consistent with the theoretical analysis.

  • 【分类号】TP273
  • 【被引频次】1
  • 【下载频次】75
节点文献中: 

本文链接的文献网络图示:

本文的引文网络