地震模拟振动台迭代学习控制算法对比研究  

Compare Study on Iterative Learning Control of Earthquake Simulation Shaking Table

作　　者：纪金豹[1] 杨森[1] 胡宗祥 JI Jinbao;YANG Sen;HU Zongxiang(Beijing Key Laboratory of Seismic Engineering and Structural Diagnosis and Treatment,Beijing University of Technology,Beijing 100124,China)

机构地区：[1]北京工业大学工程抗震与结构诊治北京重点实验室,北京100124

出　　处：《机床与液压》2023年第7期146-152,共7页Machine Tool & Hydraulics

基　　金：国家自然科学基金面上项目(51978015,51578024)。

摘　　要：为解决传统控制方法下电液伺服振动台复现波形精度不高的问题,对5种迭代学习控制算法进行了对比研究。采用的迭代学习控制算法包括时域迭代算法和频域迭代算法,时域迭代算法有P型学习律和PD型学习律,频域迭代算法有直接迭代、修正迭代和幂指数迭代,通过振动台仿真模型对比研究各种方法的控制效果。研究结果表明:在相同模型和相同波形条件下,所采用的5种算法中,频域迭代算法优于时域迭代算法,频域迭代算法下波形的相关系数可达98%以上,最大峰值误差在5%以内。In order to solve the problem of the low accuracy of the reproduced waveform of the electro-hydraulic servo shaking table under the traditional control method,five iterative learning control algorithms were studied.The used iterative learning control algorithm included time-domain iterative algorithms and frequency-domain iterative algorithms.The time-domain iterative algorithms have P-type learning law and PD-type learning law.Frequency-domain iterative algorithms have direct iteration,modified iteration and power exponential iteration.The simulation model was compared and studied with the control effect of various methods.The research results show that under the same model and the same waveform conditions,the frequency domain iterative algorithm is better than the time domain iterative algorithm in the five algorithms.The correlation coefficient of the waveform under the frequency domain iterative algorithm can reach more than 98%,and the maximum peak error is below 5%.

关 键 词：地震模拟振动台 迭代学习控制 时域迭代算法 频域迭代算法 

分 类 号：TU352[建筑科学—结构工程]