Research On Harmonic And Resonance Suppression Of Electro-Hydraulic Servo Shaking Table

As an important experimental equipment,electro-hydraulic servo shaking table is widely used in aerospace industry,automobile industry,bridge construction and other industrial fields.It is mainly used to simulate the vibration environment experienced by equipment under normal operation and verify its reliability level under specific working conditions.Compared with field test,shaking table test under laboratory condition has the advantages of repeatability,low cost and safety.The accurate loading of shaking table actuator is an important condition to ensure the accurate reproduction of reference signal.However,the inherent nonlinearity of the system and the characteristics of the experiment specimen itself make it very difficult to control the actuator accurately.In order to solve the problem of low accuracy of three-state control strategy of electrohydraulic servo shaking table,feedforward inverse compensation control strategy is used to improve the accuracy of time domain acceleration signal,but the accuracy is affected by dynamic characteristics and deviation of the model.The internal control strategy is proposed to improve the accuracy of time domain waveform reproduction of acceleration,which eliminates the influence of model characteristic on the tracking accuracy.For the non-minimum phase phenomenon caused by sampling and delay in inverse model identification,zero phase error tracking technique is used to design the non-minimum phase model.The proposed control strategy can effectively improve the accuracy of time domain acceleration waveform reproduction of the electro-hydraulic servo shaking table.In order to meet the needs of parameter identification and on-line controller adjustment,a VSS-NLMS adaptive algorithm is proposed based on LMS algorithm.Compared with LMS algorithm and NLMS algorithm,the proposed VSS-NLMS algorithm takes the third order energy of the error as the performance function,and obtains the optimal step size according to the current and the delayed instantaneous error.The proposed algorithm has faster convergence speed and lower steady-state error in steady-state and non-steady-state environments.In addition,the adaptive filtering algorithm provides theoretical basis and application guidance for harmonic suppression and resonance suppression of electro-hydraulic servo shaking table.Aiming at the harmonic distortion phenomenon caused by nonlinear factors of electrohydraulic servo shaking table,the time-domain and frequency-domain responses of acceleration response are analyzed,and the harmonic identification model is established.In view of the disadvantage of traditional optimization algorithm that it is easy to fall into the local optimal solution,WCA-VSS-NLMS hybrid algorithm is adopted to identify the high-order harmonics in sinusoidal acceleration test.The estimation method based on the WCA-VSS-NLMS can quickly and accurately identify higher harmonics information from the sinusoidal acceleration response.The harmonic suppression strategy can effectively suppress the dominant high order harmonics and realize the accurate reproduction of sinusoidal acceleration signal based on the identified harmonic information.During elastic load vibration test of electro-hydraulic shaking table,the resonance system formed by the coupling effect between the elastic load and the hydraulic system generates the resonance phenomenon.Aiming at the resonance system,an adaptive resonance suppression strategy based on resonance frequency identification filter is proposed.It not only identifies the time-varying resonant modes in frequency domain,but also realizes the on-line real-time adjustment of adaptive notch filter coefficients by using resonant frequency identification filter.The resonance frequency identification filter can quickly identify the central frequency of the resonance mode.Theoretical analysis and experimental results show that the proposed resonance suppression strategy can not only accurately identify the resonance frequency,but also effectively suppress the time-varying resonance mode of acceleration response.