基于自适应辅助变量的闭环运动对象辨识方法OA
A Closed-loop Motion Object Identification Method Based on Adaptive Instrumental Variables
精密运动控制系统中,高性能的前馈与反馈控制器的设计过程对被控对象模型精度的要求较高.然而,在闭环辨识过程中,由闭环引入的输入信号与噪声之间的相关性会导致对象辨识结果存在偏差,从而影响设计出的控制器的性能.为此,提出了一种基于自适应辅助变量的闭环运动对象辨识方法,在扩展辅助变量法的框架下,采用 BIC(Bayesian Information Criterion)准则确定模型与噪声阶次,通过自适应噪声辨识生成逆噪声滤波器,并结合扩展辅助变量法,在实现对象的无偏估计的同时提高了辨识精度.仿真与实验结果表明,提出的自适应辅助变量法在辨识精度上有所提升,且基本无需先验知识,对于未知噪声的鲁棒性较好.
In precision motion control systems,the design process of high-performance feedforward and feedback con-trollers requires high accuracy of the controlled object model.However,in the closed-loop identification process,the correla-tion between the input signal introduced by the closed-loop and the noise can lead to deviations in the object identification re-sults,thereby affecting the performance of the designed controller.To address this issue,this paper proposes a closed-loop motion object identification method based on adaptive instrumental variables.Within the framework of the extended instru-mental variable method,the BIC criterion is used to determine the model and noise order.Through adaptive noise identifica-tion,an inverse noise filter is generated,and combined with the extended instrumental variable method,the identification ac-curacy is improved while achieving unbiased estimation of the object.Simulation and experimental results show that the a-daptive instrumental variable method proposed in this paper has improved identification accuracy and requires little prior knowledge,with good robustness to unknown noise.
杨亮亮;王涵轩;鲁文其
浙江理工大学机械工程学院,浙江 杭州 310018浙江理工大学机械工程学院,浙江 杭州 310018浙江理工大学机械工程学院,浙江 杭州 310018
信息技术与安全科学
辅助变量法闭环辨识逆噪声滤波无偏估计
instrumental variable methodclosed-loop identificationinverse noise filteringunbiased estimation
《计算技术与自动化》 2026 (3)
36-43,8
国家自然科学基金资助项目(52277068)、浙江省科技厅重点研发计划资助项目(2024C01230,2023C01159、2022C01242)
评论