| The current research on earthquake collisions usually uses finite element methods to model and study earthquake collisions.The calculation results of this method are relatively accurate,but the calculation efficiency is low.However,the requirements for the calculation efficiency of earthquake collisions are getting higher and higher.Therefore,it is necessary to seek high-precision and high-efficiency earthquake collision calculation methods.This paper proposes a new method for studying earthquake collision: based on the principle of equivalent dynamic characteristics of the structure,a chain topology system mechanics suitable for efficient calculation of the multi-body system transfer matrix method(MS-TMM)is established for adjacent building structures.Model;In order to improve the accuracy of parameter identification of the standard quantum swarm algorithm,a parameter identification method that mixes quantum particle swarm with standard particle swarm is proposed,and based on this method,the modal and structural modes calculated by the finite element method are used as the basis.Based on the displacement,the relevant physical parameters of this mechanical model are determined;in order to improve the calculation accuracy of the classic collision model,an improved Hertz-damp model with a variable recovery coefficient is proposed.Based on this model,the dynamic simulation calculation of the collision of adjacent structures is realized through the MS-TMM method;finally,an engineering example of a ten-story frame structure and a five-story frame structure is used to calculate the structural collision through MS-TMM and ANSYS methods,and The calculation results of the two methods are compared and analyzed,and the effects of the period ratio,spacing and input peak ground acceleration of adjacent structures on the seismic collision response of adjacent structures are studied.The main conclusions obtained are as follows:(1)The improved Hertz-damp collision analysis model with variable restitution coefficient proposed in this paper calculates the restitution coefficient of the model at different initial collision velocities.The calculated results have smaller errors compared with the experimental results of previous scholars.Therefore,the model proposed in this paper can Determine the coefficient of restitution in the collision model more accurately.(2)Based on the structural modes calculated by the finite element method,when using the improved quantum particle swarm algorithm to identify the elastic model parameters of the frame structure chain topology dynamic model,the errors of the first two frequencies are 6.3% and 0.1%,this error is significantly lower than the other two parameter identification methods.Therefore,compared with ordinary quantum particle swarm optimization and adaptive particle swarm optimization,the method proposed in this paper has higher recognition accuracy.(3)Based on the structural displacement calculated by the finite element method,the improved quantum particle swarm algorithm is used to identify the elastoplastic model parameters of the frame structure chain topology dynamic model,and substitute the recognition result into the MS-TMM method to calculate the structural displacement,The maximum error between the calculation result and the ANSYS method calculation result is27.3%,and the change trend of the displacement of each layer is basically the same.Therefore,this method can determine the unknown parameters of the mechanical model of the frame structure.(4)The collision force and top displacement of the two frame structures are calculated by the ANSYS method and the MS-TMM method.The results of the calculation example show that the collision force and top displacement between adjacent frame structures are calculated by the MS-TMM method.The maximum error between the results and the calculation results of the ANSYS method is 9.82% and 9.04%,respectively.The calculation results of the two methods are not much different,but the calculation time of the MS-TMM method is about 1/20 of the ANSYS method. |