Resonance Peaks Suppression For Electro-hydraulic Servo Shaking Table Based On Adaptive Notch Filter

There exists a wide range of vibration forms inevitably when the products are in the process of actual application.Therefore vibration test equipment is needed for product testing to verify the reliability and other performance requirements.As a typical simulation of vibration test equipment,electro-hydraulic servo shaking table has been widely used in the vibration test.In the test specimen,load not only can be regarded as inertia,also can be considered as elastic.When the load is the elastic load,the interaction between elastic load and the shaking table is coupled into a resonant system,thus weakening the dynamic characteristics of the system and producing resonance peak in the bandwidth requirements.Supported by the National Natural Science Foundation of China,this thesis studied and analyzed the coupling phenomenon,and proposed the method of using adaptive notch filter to suppress the resonance peak.First of all,the development of vibration and related research were described,the purpose and significance of the study were briefly introduced.The thesis introduced the working principle and hardware composition of the shaking table.The mathematical model of the valve controlled hydraulic cylinder was established.The frequency characteristic of the system was obtained according to the transfer function,and further to test the existence of the resonance peak.Secondly,based on the basic principle of the three-variable control strategy,the thesis designed the three-variable controller of feedforward and feedback parameters used pole assignment method,making the performance of the corrected system greatly improved.By comparing the amplitude frequency characteristic of the system before and after the correction,it can be seen that the bandwidth of the system can meet the test requirement.Thirdly,through the study of the principle of least mean square algorithm,the structure of the algorithm has been analyzed.Then the search mode of the gradient descent method was analyzed and the step size factor was improved.Based on the study of the algorithm,the filter was analyzed,the basic principle of the filter and the influence factors of the performance index were described in detail.The performance of the filter is simulated and modeled.Finally,the resonant peak under the elastic load was studied and an adaptive notch filterbased on LMS algorithm has been proposed.The principle block diagram of the resonance peak suppression was obtained based on the analysis of the notch filter and the suppression principle.The resonant peak model was established by Simulink.Through the analysis of the resonance peak model,the resonance peak suppression power spectrum was obtained.The suppression of the resonance peak can be seen in the graph.