摘要
针对电液振动台加速度时域波形复现精度低的问题,提出了一种基于内环前馈逆补偿和外环非线性滤波的在线自适应控制策略。首先,构建了电液振动台伺服驱动机构模型,采用三状态控制策略获得了初步加速度控制性能;其次,基于多新息遗忘梯度算法辨识了三状态控制下系统参数化模型,利用零幅值跟踪技术构建了内环前馈逆补偿策略,拓展了系统频宽,降低了相位滞后误差;然后,引入了外环非线性Volterra自适应滤波算法,抑制了电液振动台模型参数偏差、非线性动态特性和不确定性扰动的影响;最后,在搭建的水平单轴电液振动台上开展了时域波形复现对比实验。结果表明,提出的自适应控制策略与传统策略相比具有更优越的加速度时域波形复现精度。
Aiming at solving the problem of low-accuracy acceleration waveform replication for electrohydraulic shake tables,an online adaptive control strategy combing inner loop feed-forward inverse compensation and outer loop nonlinear filtering is proposed.Firstly,the theoretical model of the actuating mechanism of an electro-hydraulic shake table is constructed,and the three variable controller is adopted to obtain a preliminary control performance.Secondly,the system model with three variable controller is identified by multiinformation forgetting gradient algorithm,and the inner loop feed-forward inverse compensation strategy is constructed using zero magnitude error tracking technology,which effectively extends system's bandwidth and reduces phase lag error.Then,the outer loop nonlinear Volterra adaptive filtering algorithm is introduced to effectively suppress the influences of parameter deviation,nonlinear dynamic characteristics and uncertainty disturbance for the electro-hydraulic shake table.Finally,comparative experiments are carried out on a horizontal uniaxial electro-hydraulic shake table.The experimental results indicate that the proposed adaptive control strategy has better time domain acceleration waveform tracking accuracy compared with traditional strategies.
作者
汤裕
朱真才
沈刚
刘韩
凌疆语
TANG Yu;ZHU Zhencai;SHEN Gang;LIU Han;LING Jiangyu(School of Mechatronic Engineering,China University of Mining and Technology Xuzhou,222116,China)
出处
《振动．测试与诊断》
EI
CSCD
北大核心
2024年第3期456-465,616,共11页
Journal of Vibration,Measurement & Diagnosis
基金
国家自然科学基金资助项目(52175068,51805532)
江苏高校优势学科建设工程资助项目(PAPD)。
关键词
电液振动台
三状态控制
前馈逆控制
自适应控制
波形复现
electro-hydraulic shaking table
three variable control
feedforward inverse control
adaptive control
waveform replication
