Tracking And Synchronization Control For Multi-Motor Driving Servo Systems

With the increasing demand for high power drive,the multi-motor driving servo systems?MDSS?have been paid more attentions and widely applied in industry and military.However,the multi-motor drive and the transmission parts exist in the MDSS,which may result in the coupling issue of load tracking and motors synchronization,friction nonlinearity,backlash nonlinearity and so on,such that the system performances and control accuracy are degraded.Therefore,the researches of load output tracking and motors synchronization control are of great significance for the MDSS with uncertain nonlinearities.This thesis studies the MDSS with uncertain system states,friction nonlinearity,backlash nonlinearity and input saturation.By incorporating the neural network?NN?state observer,adaptive robust control?ARC?,switching scheme and cascade control,the load tracking and motors synchronization control are studied under different conditions.The main contents of this thesis are summarized as follows:?1?By incorporating the variable gain NN state observers,the adaptive controllers are designed for the dual-motor driving servo systems with uncertain states and nonlinearities,which can guarantee the load position tracking and motors speed synchronization.By applying the variable gain NN observer,the states and uncertain nonlinearities of servo systems are online estimated with the fast convergence rate and small steady-state error,where the effects from external disturbance are suppressed as well.Based on the estimated states,the adaptive recursive fast terminal sliding mode tracking scheme and cross-coupling synchronization strategy are designed to achieve the precise load tracking and rapid speed synchronization.In particular,a novel nonlinear synchronization factor characterizing the degree of speed synchronization is proposed and applied in the synchronization control,which is proven to reduce the coupling effect of tracking and synchronization.Finally,the simulation results are included to verify the effectiveness of the proposed algorithms.?2?The minimum learning parameters?MLP?-based ARC with dynamic gain is presented for four-motor driving servomechanism with uncertain nonlinearities,such that the coupling problem of load position tracking and motors synchronization is addressed.The L₂gain observer is proposed based on the time-varying gain K-filter to estimate uncertain state,where the robustness of observer is enhanced under the external disturbance.Then,an ARC is designed to address this coupling issue,which consists of tracking and synchronization parts.In the tracking part,an integral sliding mode controller is proposed based on the optimal time scheme to achieve the little overshoot load tracking with the short settling time.Particularly,the proposed controller can eliminate the singularity of the classical sliding mode and ensure the initial error on the sliding mode,which improves the convergence rate.In the synchronization part,the adaptive cross-coupling synchronization controller with dynamic gain is investigated to maintain the fast speed synchronization of four motors,which can also weaken its coupling effect on the tracking performance.Besides,the MLP-based adaptive robust compensation scheme is presented in the two controllers,which can compensate for the uncertain friction nonlinearities with the MLP.?3?A novel switching scheme is presented for the MDSS with backlash,which is utilized to achieve the backlash compensation and guarantee the transient and steady-state performances of the load position tracking and motors speed synchronization.According to the backlash nonlinearity,the control process is divided into two parts:contact mode and backlash mode.In contact mode,the generalized sliding surface?GSS?is proposed to deal with the coupling issue of tracking and synchronization,where the synchronization control and tracking control are described by the reaching and sliding phases,respectively.Based on the GSS,the modified prescribed performance function is introduced to attain load tracking with prescribed performances?e.g.,the maximum overshoot,the maximum steady-state error and the convergence rate?,and L_?performance of speed synchronization is guaranteed by initialization method,maintaining the transient performance of synchronization behavior.In backlash mode,a robust control is adopted to compensate for the backlash and simultaneously eliminate the low speed crawling phenomenon of friction,such that the motors can traverse backlash at an exponential rate.Finally,the comparative simulations on four-motor driving servo system verify the effectiveness and reliability.?4?The generalized cascade systems with multi-driving systems and a driven system are considered in this paper,where the MDSS is a special case of the cascade systems.In the generalized cascade systems,the single driven system is actuated by the bounded inputs through the multi-driving subsystems,resulting in the complicated coupling problem of output tracking and subsystems synchronization.To address this issue,the optimal algorithm is firstly incorporated with the data filtering technology to design the adaptive parameter law,which maintains the finite-time parameter estimation of the driven system with the optimized performances.Based on the estimation results,the desired position is proposed by the guaranteed cost scheme,where the precise load tracking is attained as the driving subsystems converge to the desired position.By incorporating the desired signal and graph theory,the generalized coupling error?GCE?is designed to convert the complicated coupling issue into the GCE convergence,such that the complexity controller design is simplified and the coupling effect of tracking and synchronization is minimized.By applying the echo neural network,a novel time-varying sliding mode controller is presented to guarantee the finite-time GCE convergence under the input saturation,which can successfully eliminate the reaching phase and singularity of the traditional sliding mode to improve error convergence rate and weaken the chattering.Finally,the comparative simulation results illustrate the effectiveness of the proposed algorithms.?5?Considering the influences of backlash and friction,the experiments are implemented on a four-motor driving test rig to verify the tracking and synchronization control schemes.According to the MDSS with backlash and friction nonlinearities,the effectiveness of integral sliding mode tracking control and adaptive synchronization scheme is verified.By comparing with the classical sliding mode control and robust synchronization method respectively,the experiential results illustrate the benefits of the proposed algorithms.Moreover,for the MDSS with input saturation and friction,the novel time-varying sliding mode is proposed based on the GCE.From the comparative results with the conventional nonsingular sliding mode schemes,it is verified that the proposed algorithms provide better transient and steady-state performances.