The presence of cracks in a concrete structure reduces its performance and increases in the size of cracks result in the failure of the structure. Therefore, the accurate determination of crack characteristics, such as location and depth, is one of the key engineering issues for assessment of the reliability of structures. This paper deals with the inverse analysis of the crack detection problems using triple hybrid algorithms based on Particle Swarm Optimization (PSO); these hybrids are Particle Swarm Optimization-Genetic Algorithm-Firefly Algorithm (PSO-GA-FA), Particle Swarm Optimization-Grey Wolf Optimization-Firefly Algorithm (PSO-GWO-FA), and Particle Swarm Optimization-Genetic Algorithm-Grey Wolf Optimization (PSO-GA-GWO). A strong correlation exists between the changes in the natural frequency of a concrete beam and the crack parameters. Thus, the location and depth of a crack in a beam can be predicted by measuring its natural frequency. Hence, the measured natural frequency can be used as the input parameter of the algorithm. In this paper, this is applied to identify crack location and depth in a cantilever beam using the new hybrid algorithms. The results show that among the proposed triple hybrid algorithms, the PSO-GA-FA and PSO-GWO-FA algorithms are much more effective than PSO-GA-GWO algorithm for the crack detection.