000009343 001__ 9343
000009343 005__ 20141205150157.0
000009343 04107 $$aeng
000009343 046__ $$k2008-10-12
000009343 100__ $$aAdnan, Azlan
000009343 24500 $$aThe Application of Artificial Neural Network in Predicting Bridge Condition Based on Seismic Zonation

000009343 24630 $$n14.$$pProceedings of the 14th World Conference on Earthquake Engineering
000009343 260__ $$b
000009343 506__ $$arestricted
000009343 520__ $$2eng$$aIn this study, artificial intelligent methodology is applied to bridge inspection system. Artificial neural network (ANN) is developed to predict bridge condition rating based on different intensity of seismic zonation. Inspection results from nondestructive evaluation are used as an indicator to the structural condition. Numbers of systems are developed to determine the effective parameters and neural network structure in order to build the most predictive ANN system. Backpropagation algorithm with one hidden layer is used to develop the neural network and Borland C++ is used as the programming language. 75 concrete bridges under the supervision of Public Works Department, PWD (Malaysia) have been selected for further inspection using nondestructive evaluation technique which includes the rebound hammer test, Ultrasonic Pulse Velocity, and electromagnetic cover meter. These tests were conducted to determine the bridge strength, structural damages, and level of the damages. Results from this inspection are then applied to the ANN together with the seismic zonation parameter and other bridge parameters in order to develop the intelligent system. Generally, this study showed that the ANN has a potential to be used to predict the condition rating based on different seismic intensity. Prediction values are up to 90% correct. Linear correlation coefficient between the prediction and actual value ranges from 0.5 to 0.8, which shows a strong relationship between these two values. This intelligent system can help the authority to forecast bridge condition after tremors. Critical bridges can be short listed and prioritized for the allocation of maintenance budget. The intelligent system has large potential to be used as inspection aided tool in bridge monitoring and thus any attempt to enhance the system are very much recommended.

000009343 540__ $$aText je chráněný podle autorského zákona č. 121/2000 Sb.
000009343 653__ $$aSeismic zonation, earthquake monitoring system, artificial neural network, nondestructive evaluation.

000009343 7112_ $$a14th World Conference on Earthquake Engineering$$cBejing (CN)$$d2008-10-12 / 2008-10-17$$gWCEE15
000009343 720__ $$aAdnan, Azlan$$iAlih, Sophia.C.$$iIsmail, Rozaina
000009343 8560_ $$ffischerc@itam.cas.cz
000009343 8564_ $$s170906$$uhttps://invenio.itam.cas.cz/record/9343/files/05-02-0052.pdf$$yOriginal version of the author's contribution as presented on CD, Paper ID: 05-02-0052.
000009343 962__ $$r9324
000009343 980__ $$aPAPER