Optimization Design And Parameter Identification Of Multi-ribbed Composite Wall Structure Based On Subset Simulation Optimization

As a new stochastic optimization algorithm,the subset simulation optimization algorithm(SSO)has the advantages of fast convergence and avoiding falling into local optimal solutions when solving high-dimensional complex optimization problems.It is suitable for solving continuous variable unconstrained and constrained optimization problems and discrete variable optimization problems.Multi-ribbed composite wall structure,as a new architectural structure,is mainly composed of prefabricated multiribbed composite walls,invisible outer frame and floor slab.The multi-ribbed composite wall is the main force-bearing member of the structure,which is composed of components of different materials,and its force state is complex.Optimal design is an important part of its seismic design theory,and there are both discrete variables and continuous variables in the optimization design.In addition,the influence of structural dynamic effects on structural optimization design has not yet been considered.Therefore,in this paper,the SSO is respectively extendedly used to solve the optimization design problem of multiribbed composite wall structures that include continuous and discrete variables and the optimization design problem considering the dynamic effects of the structure.Furthermore,the restoring force model reflects the seismic performance of the multiribbed composite wall structure,and is the basis for structural response prediction.The identification of its parameters is very important.In this paper,the SSO is combined with Bayesian theory to identify the parameters of the interlayer restoring force model of the multi-ribbed composite wall structure.The research content and conclusions of this paper are as follows:(1)Considering both discrete variables and continuous variables,the SSO algorithm is applied to the cost optimization design of multi-ribbed composite wall structures under the constraints of component bearing capacity.Based on the idea of segment optimization,the mathematical model for the optimal design of the multilayer multi-ribbed composite wall structure was established,which is constrained by the bearing capacity of walls,and aims at minimizing the cost of the structure,and then optimize the design of a 7-layer multi-ribbed composite wall structure.Based on the multi-level optimization idea,the mathematical model for the optimal design of the mid-high-rise multi-ribbed composite wall structure was established,and then optimize the design of a 12-layer multi-ribbed composite wall structure.After optimization,the costs of structures are reduced,and both structures meet the lateral displacement requirement,and walls meet the requirements of bearing capacity,which proves the feasibility and effectiveness of the proposed improved subset simulation optimization algorithm.(2)In order to consider the structural integrity and the dynamic elastoplastic effect of the structure under massive seism,the multi-ribbed composite wall structure is further optimized based on the seismic performance.In this paper,the optimal design method and process of the multi-ribbed composite wall structure based on the SSO is proposed,which is constrained by the he overall material consumption of the structure and aims at minimizing the maximum interstory displacement angle of the structure.The optimized design of a 4-story and a 12-story multi-ribbed composite wall structure was performed.After optimization,the displacement angles between the layers of the two structures were significantly reduced,and meanwhile the seismic vulnerabilities of the structures were both reduced.(3)Considering the uncertainty in the parameter identification process of the structural hysteresis model,and in order to effectively avoid the local optimal solution and improve the calculation efficiency,a method of hysteresis model parameter identification based on Bayesian theorem and SSO is proposed.Taking the measured interlayer displacement response of the multi-ribbed composite wall structure as the observed measurement,basing on Bayesian theory,the proposed method derive the posterior probability distribution function of the restoring force model parameters,and the SSO is used to identify the most likely structural interlayer restoring force model parameters value.The proposed method is applied to the identification of interlayer restoring force parameters of a 4-layer multi-ribbed composite wall structure.The Clough biline model is used to simulate the interlayer dynamic characteristics of the structure,and the SSO is used to calculate the interlayer restoring force parameters.And compared with the recognition results of the TMCMC algorithm,it shows that the recognition accuracy and calculation efficiency of the subset simulation optimization algorithm are higher.