POST-EARTHQUAKE EVALUATION OF STRUCTURESTHROUGH THE MODEL UPDATING DAMAGE DETECTIONMETHOD USING PSO OPTIMIZATION ALGORITHM ANDDIFFERENT FITNESS FUNCTIONS

The assessment of structures after an earthquake is a critical aspect of ensuring the safety andreliability of structural engineering. This study presents an evaluation of the efficacy of four fitnessfunctions based on modal characteristics of structures and the Particle Swarm Optimization (PSO)Algorithm in the context of structural damage detection. The primary aim is to assess the capabilitiesof this algorithm with different fitness functions in accurately identifying the location and severity ofdamage within the structures. The findings of this study underscore the varying degrees of accuracyand computational efficiency exhibited by PSO optimization algorithms in detecting damage withinthe truss-like structures. By assessing their effectiveness across a damage scenario, this researchprovides valuable insights into selecting the most suitable fitness function for use in optimizationalgorithms in model updating techniques based on the specific characteristics of the structuralproblem. The obtained results show that all four fitness functions almost reached the correct answer onthe other hand the location and severity of damage introduced in the target model were fulfilled by allfitness functions after ۶۰۲۰ function evaluations. Among these results, Fitness_۱ has the best resultsdue to the minimum Root Mean Square Error (RMSE) of results and simultaneously solving theproblem in lesser function evaluation. After that, the Fitness_۴, Fitness_۲, and Fitness_۳ withminimum RMSE and function evaluations find the solution result. Therefore, the proposed fitnessfunctions to find the location and severity of damage in truss structures are Fitness_۱, Fitness_۴,Fitness_۲, and Fitness_۳ respectively.

کلیدواژه ها:

Natural Frequency ، Damage Detection ، Model Updating ، Fitness Function ، PSO Algorithm