Reliability Assessment And Residual Service Life Prediction Of Existing Building Structural

With the rapid development economic development rapid and the improvement of urbanization, the number of urban houses and total construction area is increasingly rising dramatically. The expansion and application of the scale of construction improved the people’s living conditions, make life more convenient, but corresponding building safety issues are also increasing. Due to the construction service time is constantly increasing, so old buildings structure will corresponding increase, The safety and reliability of the structure has been paid more and more attention. So existing building structure reliability assessment and residual service life prediction is not only can reduce the occurrence of building structure’s safe hidden trouble, but also can provide scientific basis for decision-making for maintenance and reinforcement.This paper introduces BP neural network to existing building structure reliability assessment and residual service life prediction. Firstly, through the existing structure reliability evaluation standard and relevant research material, built on the existing building structure reliability assessment index system. In the basis of the theory of reliability, based on BP neural network existing building structure reliability assessment model for reliability evaluation. On the basis of reliability evaluation, and introduction the structural damage index, use BP neural network prediction model for existing building structure residual service life prediction. By combining the neural network theory, and use the neural network toolbox of MATLAB, to assessment the reliability and predict residual service life of existing building structure. Finally, with the engineering case analysis, the application of the model performance is verified. Using BP neural network method for reliability assessment and residual service life prediction, can improve the work efficiency, and provide support for the related decision makers.