Study of condition monitoring of bridges using genetic algorithms

Monitoring the condition and performance of structures is an essential issue of structural maintenance in civil engineering to ensure the service quality of structures. Periodic monitoring of the condition of a structure provides a procedure to keep track of the variations of a structure under normal service loads. Non-Destructive Evaluation (NDE) techniques provide feasible methods of directly identifying the structural irregularity through examining the entire structure.; A new methodology is developed to monitor the structural condition through non-destructive evaluation techniques using genetic algorithms. System identification techniques are used to establish the mathematical model of the real structural system through parametric methods. The problem of structural condition monitoring is formulated as an optimization problem. The remotely sensed static measurements, such as displacements and strains, under normal service loads are used to estimate the state of the structure. Genetic algorithm (GA), one of the computational intelligence methods, employs a highly parallel, massive, and adaptive computational environment to accomplish the mathematically difficult tasks in the structural condition monitoring. The feasibility of the proposed methodology is verified by the application to the detection and identification of structural damage. The location and magnitude of the damage in the element level defined as the stiffness change in structural members; and the location and size of the damage in the detailed level defined as the fracture damage in beam structures are successfully determined.