Structural Damage Identification Based On Salp Swarm Optimization Algorithm And Sensitivity Analysis

With the rapid development of our national economy,civil engineering construction such as large-scale structures have emerged in recent years.As a result,more and more structural damage events are caused by the factors of overloading,fatigue and so on.Therefore,structural health monitoring(SHM)has been developed rapidly with the strong support of our nation.Structural damage identification(SDI)is an important part of SHM,which has been widely concerned by scholars all over the world.Swarm intelligence(SI)methods provide a new idea for SDI.SDI methods based on SI algorithms have better global optimization ability and faster iterative process.It can solve most high-dimensional and complex optimization problems that are difficult to be solved by traditional optimization techniques.As for this study,salp swarm algorithm(SSA)is applied in SDI based on its achievements in other fields.In order to improve accuracy and efficiency,an appropriate damage indicator and weighting strategies are selected as well as combined with sensitivity analysis at the same time.The main works are summarized as follows:(1)A literature review on SDI methods in recent years is summarized,which can be mainly divided into two categories: deterministic and probabilistic methods.According to this classification criterion,SDI methods can be classified as: model updating-based methods,timedomain signal processing-based method,natural frequency and mode shape-based static or dynamic fingerprints method;non-parametric system method,probability statistics-based method,etc.Each kind of the above methods is carried out on the brief elaboration respectively.(2)A SDI approach based on cosine-based indicator(CI)and SSA is proposed.The accuracy of different damage indicators are verified under the influence of noise.Compared with modal assurance criterion(MAC),CI is more sensitive to structural damage during this process.Meanwhile,weighting coefficients are introduced into the objective function to appropriately allocate the proportion of frequency and mode shape,which makes the results more reasonable.Then the basic theory of SSA is introduced and explained how to apply the algorithm to SDI.Numerical simulations on a simply-supported beam and a two-story rigid frame are carried out to verify the effectiveness of proposed method,and the proposed method is compared with the continuous colony optimization(CACO)method.(3)A SDI approach based on sensitivity analysis and SSA is proposed.The method firstly obtains the frequency and mode shape of a healthy structure by modal analysis,and constructs the sensitivity matrix of the frequency and mode shapes,respectively.Then SSA algorithm is used to obtain the distribution of damage reduction factors of the structure,that are,the positions of salps,and the corresponding sensitivity matrix is given to the specific salp’s position.Finally,the optimal solution is obtained by approaching the target value with Taylor’s formula.In order to improve the identification accuracy,noise robustness and reasonable results of the proposed method,on the basis of the new method,the high-order sensitivity matrix is deduced.Numerical studies on a twostory frame structure are conducted to investigate the effectiveness and efficiency of the proposed approach.(4)A cantilever beam structure is set up in laboratory to validate the SDI method based on CI indicator and SSA algorithm method.The experimental modal analysis is performed to get frequency information in both healthy and damage states of the structure.Besides,the parameters of the model are modified by SSA algorithm to obtain the finite element model which is closest to the experimental structure.Experimental data is used to analyze the damage cases and to verify that the proposed method is feasible and effective.