Structural health monitoring of concrete bridge based predictive model on chloride ingress

• Main Author: Wong, Chee Lum
• Format: Undergraduates Project Papers
• Language: English
• Published: 2014
• Subjects: TG Bridge engineering
• Online Access: http://umpir.ump.edu.my/id/eprint/11606/1/WONG%20CHEE%20LUM.PDF; http://umpir.ump.edu.my/id/eprint/11606/

Concrete Bridge is a very useful structure nowadays to link the different community together by provided the path for human to across river, lake and sea. However, bridges possess high risk attack by corrosion by chloride ingress' if directly contact with sea water. The sea water contains high level of chloride compare to sulfate. Corrosion due to the chloride ingress becomes a serious issue nowadays. The concrete bridge should repair with an amount of cost to ensure the safety. Some of the concrete bridge even end of service life and cause the replacement of the old concrete bridge should carry out. The predictive model may use to determine the service life and the maintenance period. The engineer may assist by the model to make the decision on which method suitable use to maintenance in order to provide the safe structure and low cost maintenance fee. Thus, the objectives of this project are investigate the corrosion monitoring instruments, review the predictive model and estimate probability of rise of corrosion initiation phase at given time. The model will create by using Fick's second law. Monte Carlo simulation is widely used to obtain parameters by repeating sampling method. Bayes' Theorem- may use to interpret the parameters into Fick's second law with the experimental data to update the result. The analysis shown that the parameters will influence the result of the predictive model, good quality of parameter will give a good quality of result. The prior distribution may be updated to posterior distribution by using likelihood function. The likelihood function plays an important role in updating process because the likelihood strongly effect on the posterior distribution. So, it is very important to get the likelihood function right in order produce an accurate posterior distribution. The research shows the significance of providing the useful information to predict the condition of concrete bridge under the chloride ingress. An action may be performs with the help of the model in order to solve the problem by select the low cost and effective way. Keywords: Chloride ingress, predictive model, corrosion, health monitoring, concrete bridge