Study On The Stability Analysis Of Real-Time Dynamic Substructuring System

Test is the main method of civil engineering structure researches.However,the capacity of the existing facilities have seriously restricted the development of the engineering structure researches.In Real-time dynamic substructure testing(RTDS),in order to promote the scale of the specimens,the structural parts or components which are the experiment focus on are tested physically and the rest part is built for numerical model and simulated in the computer.The stability of RTDS system is the basic of a successful experimental implementation.Therefore,it is significant to predict the stability accurately.The experimental performance of RTDS testing is determined by numerical integration algorithm,the control algorithm,the loading system,the test specimen and the sensor.However,the existing research on the stability predictions are inadequate for considering the above coupling various factors.In this paper,in order to consider the coupling effects of factors on the stability,a novel analysis method considering the coupling effect of the loading system,the dynamic characteristics of the specimens and the numerical integration algorithm was developed.Firstly,based on the concept of gain margin,a stability analysis method which can consider the influence of the comprehensive factors of RTDS system is proposed and verified experimentally.This method combines the numerical integral algorithm,the loading system,sensors and the physical substructures together through frequency domain transfer functions perfectly.At the same time,the stability performance of RTDS system can be predicted onlybased on the measured data,while the accurate mathematical models of the loading system,sensors and the physical substructure are unnecessary,which greatly reduces the difficulty of stability prediction.Using the developed stability analysis methods given above,a dynamic coupling model which considering the transfer-system-specimen interaction was established for actuator and shaking table.Besides,the influence of transfer-system-specimen interaction on the stability of actuator and shaking table RTDS system was discussed respectively.The analysis results show that the transfer-system-specimen interaction is sometimes beneficial and sometimes detrimental to RTDS stability.In multi degree of freedom(MDOF)system,the parameters required in the stability analysis are often complex and the index of stability assessment has unspecific physical meaning.In order to solve this problem,this paper developed an analytical method for MDOF RTDS system incorporating the mode superposition method and the concept of gain margin.This method can be used to predict the stability of the system only by obtaining the physical substructure modal parameters and the loading system models.That is simple and easy.Besides,the effectiveness of this method was verified experimentally.Through the z transform,the numerical integration algorithm is translated into discrete transfer function.Combining with the stability prediction method of multi degree of freedom system based on gain margin,a stability method of considering all factors in RTDS systems was achieved.Besides,a comparative analysis of coupling effects of different numerical integration methods are considered.Existing studies believe that with the increasing of the frequency and the time step,the stability will definitely be reduced.However,the results of our study show that there is a strong coupling effect between the numerical integration algorithm and other factors of the test system,and the numerical integration algorithm may enhance or reduce the stability of RTDS system in difference cases.At last,the method is extended to the nonlinear stage,and the method to predict the stability of nonlinear RTDS system is developed by linear means.Besides,a comparative analysis of the stability of actuator and shaking table RTDS system was analyzed.The existing studies believe that the nonlinearity is always beneficial to RTDS stability.Whereas,we find that the stability of RTDS system is not definitely improved after the structures get into the nonlinear part.