Purpose–The purpose of this paper is to apply a intelligent algorithm to conduct the force tracking control for electrohydraulic servo system(EHSS).Specifically,the adaptive neuro-fuzzy inference system(ANFIS)is sele...Purpose–The purpose of this paper is to apply a intelligent algorithm to conduct the force tracking control for electrohydraulic servo system(EHSS).Specifically,the adaptive neuro-fuzzy inference system(ANFIS)is selected to improve the control performance for EHSS.Design/methodology/approach–Two types of input–output data were chosen to train the ANFIS models.The inputs are the desired and actual forces,and the output is the current.The first type is to set a sinusoidal signal for the current to produce the actual driving force,and the desired force is chosen as same as the actual force.The other type is to give a sinusoidal signal for the desired force.Under the action of the PI controller,the actual force tracks the desired force,and the current is the output of the PI controller.Findings–The models built based on the two types of data are separately named as the ANFIS I controller and the ANFIS II controller.The results reveal that the ANFIS I controller possesses the best performance in terms of overshoot,rise time and mean absolute error and show adaptivity to different tracking conditions,including sinusoidal signal tracking and sudden change signal tracking.Originality/value–This paper is the first time to apply the ANFIS to optimize the force tracking control for EHSS.展开更多
The frequency response design method for PDFSV (Pseudo Derivative Feedback Subvariable) control of electrohydraulic servo system is introduced. Theoretical analysis and computer simulation show that PDFSV con...The frequency response design method for PDFSV (Pseudo Derivative Feedback Subvariable) control of electrohydraulic servo system is introduced. Theoretical analysis and computer simulation show that PDFSV control is a high robust system, and a very good performance can be obtained when this theory is employed in electrohydraulic servo system.展开更多
Combining the characteristics of servo systems , tracking variable structure control law is studied. Two kinds of new variable control law , the generalized exponential approaching vari- able structure control law and...Combining the characteristics of servo systems , tracking variable structure control law is studied. Two kinds of new variable control law , the generalized exponential approaching vari- able structure control law and the integral variable structure control law are put forward for dis- crete time domain. Taking pump-controlled-motor rotational speed servo system for example , the experiment investigation and digital simulation of integral variable structure control law for dis- crete time domain are performed , the rightness of conclusions are verified.展开更多
It is difficult to conduct shaking table tests that require large-displacement high-frequency seismic excitation due to the limited capacity of existing electrohydraulic servo systems.To address this problem,a double-...It is difficult to conduct shaking table tests that require large-displacement high-frequency seismic excitation due to the limited capacity of existing electrohydraulic servo systems.To address this problem,a double-layer shaking table(DLST)is proposed.The DLST has two layers of one table each(i.e.,an upper table and lower table)and aims at reproducing target seismic excitation on the upper table.The original signal is separated into two signals(i.e.,a high-frequency signal and low-frequency signal)through a fast Fourier transform/inverse fast Fourier transform process,and these signals are applied to the two tables separately.The actuators connected to different tables only need to generate large-displacement low-frequency or small-displacement high-frequency movements.The three-variable control method is used to generate large-displacement but low-frequency motion of the lower table and high-frequency but small-displacement motion of the upper table relative to the table beneath.A series of simulations are carried out using MATLAB/Simulink.The simulation results suggest that the DLST can successfully generate large-displacement high-frequency excitation.The control strategy in which the lower table tracks the low-frequency signal and the upper table tracks the original signal is recommended.展开更多
基金This work was supported by the National Key R&D Program of China“The study on Load-bearing and Moving Support Exoskeleton Robot Key Technology and Typical Application”(2017YFB1300502)This work is also supported by the National Natural Science Foundation of China“Research on gait detection and recognition technology of Parkinson’s disease based on all-fiber composite sensors”under Grant 61903280Hubei Key Laboratory of Digital Textile Equipment Open fund“Research on intelligent monitoring clothing based on micro-nano fiber composite sensor”under Grant DTL2019011.
文摘Purpose–The purpose of this paper is to apply a intelligent algorithm to conduct the force tracking control for electrohydraulic servo system(EHSS).Specifically,the adaptive neuro-fuzzy inference system(ANFIS)is selected to improve the control performance for EHSS.Design/methodology/approach–Two types of input–output data were chosen to train the ANFIS models.The inputs are the desired and actual forces,and the output is the current.The first type is to set a sinusoidal signal for the current to produce the actual driving force,and the desired force is chosen as same as the actual force.The other type is to give a sinusoidal signal for the desired force.Under the action of the PI controller,the actual force tracks the desired force,and the current is the output of the PI controller.Findings–The models built based on the two types of data are separately named as the ANFIS I controller and the ANFIS II controller.The results reveal that the ANFIS I controller possesses the best performance in terms of overshoot,rise time and mean absolute error and show adaptivity to different tracking conditions,including sinusoidal signal tracking and sudden change signal tracking.Originality/value–This paper is the first time to apply the ANFIS to optimize the force tracking control for EHSS.
文摘The frequency response design method for PDFSV (Pseudo Derivative Feedback Subvariable) control of electrohydraulic servo system is introduced. Theoretical analysis and computer simulation show that PDFSV control is a high robust system, and a very good performance can be obtained when this theory is employed in electrohydraulic servo system.
文摘Combining the characteristics of servo systems , tracking variable structure control law is studied. Two kinds of new variable control law , the generalized exponential approaching vari- able structure control law and the integral variable structure control law are put forward for dis- crete time domain. Taking pump-controlled-motor rotational speed servo system for example , the experiment investigation and digital simulation of integral variable structure control law for dis- crete time domain are performed , the rightness of conclusions are verified.
基金Scientific Research Fund of the Institute of Engineering Mechanics,China Earthquake Administration under Grant No.2019EEEVL0502Natural Science Foundation of China under Grant No.52078275+1 种基金the Institute for Guo Qiang,Tsinghua University under Grant No.2019GQC0001Beijing Natural Science Foundation under Grant No.JQ18029。
文摘It is difficult to conduct shaking table tests that require large-displacement high-frequency seismic excitation due to the limited capacity of existing electrohydraulic servo systems.To address this problem,a double-layer shaking table(DLST)is proposed.The DLST has two layers of one table each(i.e.,an upper table and lower table)and aims at reproducing target seismic excitation on the upper table.The original signal is separated into two signals(i.e.,a high-frequency signal and low-frequency signal)through a fast Fourier transform/inverse fast Fourier transform process,and these signals are applied to the two tables separately.The actuators connected to different tables only need to generate large-displacement low-frequency or small-displacement high-frequency movements.The three-variable control method is used to generate large-displacement but low-frequency motion of the lower table and high-frequency but small-displacement motion of the upper table relative to the table beneath.A series of simulations are carried out using MATLAB/Simulink.The simulation results suggest that the DLST can successfully generate large-displacement high-frequency excitation.The control strategy in which the lower table tracks the low-frequency signal and the upper table tracks the original signal is recommended.