ANN based tensile force estimation for PSC girder using embedded EM sensor

The tensile force of PS tendons is most important factor for monitoring the structural health state of PSC girder bridges. The PS tendon located inside of PSC girder thus it is hard to apply the conventional nondestructive testing method. To measure the tensile force of PS tendons this study proposed an embedded EM sensor based tensile force estimation method using artificial neural network. The permeability of PS tendons is changed according to the induced tensile force due to its magneto-elastic effect. The embedded EM sensor can measure the permeability change of PS tendons and the tensile force can be estimated using the pattern of permeability. The embedded EM sensor consists of screw thread for connecting with the sheath, oblique end for insulting to the anchorage part, primary coil for generating magnetic field and secondary coil for measuring the magnetization of PS tendons. To verify the proposed method, the experimental study using three down-scaled PSC girder models was performed. The permeability of PS tendons was proportionally decreased according to increase of tensile forces. The artificial neural network was trained using test data to estimate the tensile force of PS tendons using permeability. As a result, the proposed ANN based tensile force estimation using embedded EM sensor could be one of the solution for evaluating the performance of PSC girder.