An enhanced least mean square(LMS)error identification algorithm integrated with Kalman filtering is proposed to resolve accuracy degradation induced by nonlinear dynamics and parameter uncertainties in continuous rot...An enhanced least mean square(LMS)error identification algorithm integrated with Kalman filtering is proposed to resolve accuracy degradation induced by nonlinear dynamics and parameter uncertainties in continuous rotary electro-hydraulic servo systems.This enhancement accelerates convergence and improves accuracy compared with traditional LMS.A fifth-order identification mod-el is developed based on valve-controlled hydraulic motors,with parameters identified using Kalman filter state estimation and gradient smoothing.The results indicate that the improved LMS effectively enhances parameter identification.An advanced disturbance rejection controller(ADRC)is de-signed,and its performance is compared with an optimal proportional integral derivative(PID)con-troller through Simulink simulations.The results show that the ADRC fulfills the control specifications and expands the system’s operational bandwidth.展开更多
代理模型由于可以有效地缩减学科分析时间,被广泛应用于优化领域。而构建高精度代理模型则取决于样本点在设计空间中的分布。为了建立拟合效率高的近似模型,在已有Kriging代理模型基础上,提出一种基于垂距和最大化点均方差取样(Integrat...代理模型由于可以有效地缩减学科分析时间,被广泛应用于优化领域。而构建高精度代理模型则取决于样本点在设计空间中的分布。为了建立拟合效率高的近似模型,在已有Kriging代理模型基础上,提出一种基于垂距和最大化点均方差取样(Integrated mean square error,IMSE)的多点取样算法,以保证预测精度的同时减少样本数量。该方法将垂距作为判定设计变量取值的标准,进行数据点的初步筛选。选取高斯函数作为设计点之间的相关函数,并且在边缘附近进一步修正。针对实际算例,将该取样算法与多点加点准则比较,结果表明使用的方法在保证全局精度的基础上,采用较少的采样点构建代理模型,具有较高的局部近似精度。展开更多
基金Supported by the National Natural Science Foundation of China(No.52375037)the Outstanding Youth of Pyramid Talent Training Project of Beijing University of Civil Engineering and Architecture(No.GDRC 20220801)+1 种基金the Graduate Innovation Fund Project of Beijing University of Civil Engineering and Architecture(No.PG2025160)the Special Fund for Cultivation Projects of Beijing University of Civil Engineering and Architecture(No.X24026).
文摘An enhanced least mean square(LMS)error identification algorithm integrated with Kalman filtering is proposed to resolve accuracy degradation induced by nonlinear dynamics and parameter uncertainties in continuous rotary electro-hydraulic servo systems.This enhancement accelerates convergence and improves accuracy compared with traditional LMS.A fifth-order identification mod-el is developed based on valve-controlled hydraulic motors,with parameters identified using Kalman filter state estimation and gradient smoothing.The results indicate that the improved LMS effectively enhances parameter identification.An advanced disturbance rejection controller(ADRC)is de-signed,and its performance is compared with an optimal proportional integral derivative(PID)con-troller through Simulink simulations.The results show that the ADRC fulfills the control specifications and expands the system’s operational bandwidth.
文摘代理模型由于可以有效地缩减学科分析时间,被广泛应用于优化领域。而构建高精度代理模型则取决于样本点在设计空间中的分布。为了建立拟合效率高的近似模型,在已有Kriging代理模型基础上,提出一种基于垂距和最大化点均方差取样(Integrated mean square error,IMSE)的多点取样算法,以保证预测精度的同时减少样本数量。该方法将垂距作为判定设计变量取值的标准,进行数据点的初步筛选。选取高斯函数作为设计点之间的相关函数,并且在边缘附近进一步修正。针对实际算例,将该取样算法与多点加点准则比较,结果表明使用的方法在保证全局精度的基础上,采用较少的采样点构建代理模型,具有较高的局部近似精度。
