| Bridges occupy a large proportion of the existing railway lines in China,and their structural safety has also attracted the attention of many railway bridge scholars.As an important index of bridge dynamic performance,modal parameters of bridge structure play an important role in updating numerical model,damage detection and structural health monitoring(SHM).The traditional bridge detection usually adopts the way of manually arranging acquisition instruments on the bridge,which consumes a lot of manpower and material resources,and has a certain degree of risk.In view of this situation,based on the vehicle bridge interaction theory and stochastic subspace method,this paper deduces the indirect identification method of bridge modal parameters based on stochastic subspace method.The effectiveness of the method is verified by numerical simulation and indoor model test.The crack element is introduced to simulate the bridge damage.According to the first mode shape and regional mode shape curvature(RMSC),an indirect identification method of bridge damage based on stochastic subspace method is proposed.The feasibility of the method is verified by numerical simulation.The main research work and conclusions are as follows(1)Derivation and numerical verification of indirect identification method for bridge modal parameters based on stochastic subspace method1)Based on the vehicle bridge interaction theory,the vehicle is assumed to be a quarter car model,and the bridge is regarded as a simply supported Euler-Bernoulli beam.The finite element calculation model of vehicle bridge system is compiled by MATLAB,and the dynamic response of vehicle bridge system is solved.By introducing the state space theory,the state space model of vehicle bridge interaction system is established.2)Based on the Cov-SSI method,an indirect identification method of bridge frequencies is derived by using the acceleration output vector of moving point(wheel or vehicle body).Based on the Ref-SSI method,the indirect identification method of bridge mode shapes is derived by using the output vector composed of acceleration response of moving point and reference point(any point on the bridge).3)The indirect identification algorithm program of bridge modal parameters(frequency and mode shape)based on stochastic subspace method is compiled by MATLAB.The acceleration response of moving point(wheel or vehicle body)and reference point(24m position of bridge)calculated by vehicle bridge model is taken as input.It is found that the first three order frequencies and mode shapes of bridge can be identified through the acceleration response of wheel or vehicle body,the correctness of the proposed indirect identification method is verified.On this basis,the influence of vehicle speed,vehicle suspension system parameters(stiffness and damping),track irregularity and environmental noise on the results of bridge modal parameters(frequency and mode shape)are studied.The results show that:(1)lower vehicle speed can ensure that the obtained results of bridge modal parameters have higher accuracy.(2)The change of the stiffness of the vehicle suspension system will only affect the mode shapes obtained from the acceleration of the vehicle body.(3)The damping of vehicle suspension system can improve the accuracy of bridge mode sahpes based on vehicle body acceleration.The larger the damping of suspension system is,the more accurate the mode shapes based on vehicle body acceleration are.(4)Compared with the vehicle body acceleration,under the influence of track irregularity and environmental noise,the accuracy of bridge modal parameters identified by wheel acceleration is higher.(2)Indoor model test of indirect identification of bridge modal parameters based on stochastic subspace methodThrough the indoor vehicle bridge model test,according to the derived indirect identification method of bridge modal parameters,the acceleration of vehicle body and reference point is calculated and analyzed,and the first two frequencies and mode shapes of the bridge are identified,which verifies the feasibility of the indirect identification method of bridge modal parameters proposed in this paper.(3)Derivation and numerical verification of bridge damage indirect identification method based on stochastic subspace methodThe crack element is introduced to simulate the bridge damage.According to the first mode shape identified by the acceleration of wheel or vehcicle body and regional mode curvature(RMSC),an indirect identification method of bridge damage based on stochastic subspace method is proposed.The results show that:(1)the wheel acceleration can identify the damage at any position(single position and multi position)on the bridge;However,the bridge damage can not be identified by the acceleration of vehicle body.(2)The degree of bridge damage can be identified by the wheel acceleration.(3)Track irregularity will cause the undamaged position of the bridge to be misjudged as damaged. |