Ancient Timber-structured Buildings Reliability Assessment And Residual Life Prediction

The ancient Chinese timber-structured buildings are dramatically unique in terms of the design andconstruction which contain profound cultural and artistic value, making them the precious historical andcultural heritage in the world. However, due to the influences of external environment factors and thematerial structure of the timber building itself, there are possibilities to trigger sudden invalid of thestructure, undermining danger in the structure of the building. Therefore, the study of reliability estimationand residual life prediction of the historic wooden structure buildings has important guiding significance.The essay will study and research the reliability estimation and residual life prediction of thetimber-structure buildings, which is based on the existing domestic and international researches and thereliability theory. Moreover, in order to solve the problem of the complex and massive calculation of usingthe multi-factorial variability resistance model, it will be a positive trying to apply the BP nerve network tothe residual life prediction under the capability of its nonlinear mapping. The essay is consist of thefollowing aspects:Firstly, analyzing the main influencing factors in consideration of the features of the resistancestatistics, synthetically applying the basic theories: the variable reliability theory and theory of strength, tobuild up a timber-structured architecture attenuation model functioning under the multi-factorial variabilityresistance model, then revising the resistance statistical features in Bayes’ method. Finally predicting theresidual life of the timber-structure building according to the measured data of the wooden truss based onthe reliability theory. Secondly, combining the additional momentum method and study rate increasing method to improvethe BP algorithm. Establishing a BP neural network prediction model to predict the residual life accordingto the measured data of each wood trusses parts and predicted sampled data getting from the reliabilitytheory.The study explores an innovative method to provide residual life predictions for the preservation andmaintenance of ancient timber-structured buildings, by using this method, we can assure when and how tomaintain the timber-structure building, and also we can figure out the exact part which is in need ofmaintenance. It has a significant and sustainable meaning for study and extending of residual life oftimber-structured buildings.