Research On Life Prediction Of Concrete Aqueduct Based On Carbonation-Freeze-Thaw Environment

Aqueduct is a kind of common canal system water delivery structure,which plays an irreplaceable role in the allocation of water resources in major irrigation areas.Most aqueducts in Guanzhong area of Shaanxi Province belong to concrete structure.The durability of materials is directly related to whether the concrete structure can reach the expected service life.The factors causing the deterioration of material durability are numerous and complex,which are mainly divided into two categories: the diseases caused by design and construction defects and the diseases caused by the erosion of harmful media in the environment.In order to ensure the safe operation of aqueduct in Guanzhong area,the durability evaluation and life prediction of concrete aqueduct has become an urgent problem.In this paper,combined with the scientific research project,the field detection and evaluation of several aqueducts in Guanzhong area are carried out.The prediction model of concrete carbonation depth is established from the perspective of material durability,and the life prediction of aqueduct concrete is carried out.The main research work and conclusions are as follows:(1)Through the analysis of design data and field data collection of aqueduct in Guanzhong area,the service characteristics and working state of concrete aqueduct in this area are studied.If the climate is warm in winter and cold in summer,the concrete is easy to be carbonized.The concrete of aqueduct is tested by rebound instrument,steel protection layer detector,crack depth detector and other instruments.It is found that the typical diseases of most aqueducts include concrete spalling,steel corrosion,concrete cavitation wear and so on.These diseases will directly affect the structural bearing capacity and material durability of aqueduct.Based on the analysis of field test data,it is considered that the main causes of durability deterioration of aqueduct concrete are carbonation and freeze-thaw.(2)The main characterization index(carbonation depth)of aqueduct concrete carbonation in freeze-thaw environment was studied.The influence factor α considering freeze-thaw cycle on carbonation was proposed.The calculation model of concrete carbonation depth considering concrete compressive strength is modified.The average error is 5.0%.(3)BP neural network and SOM-RBF neural network are used to predict the carbonation depth of concrete aqueduct.After training the relatively small data sets of different aqueduct concrete,the neural network prediction model can provide reliable prediction for the carbonation depth of aqueduct concrete.The BP neural network and som-rbf neural network are used to predict five groups of measured data.The prediction error of BP neural network is12.5%,6.6%,4.3%,13.1%,21.1%,and the average error is 11.5%;the prediction error of somrbf neural network is 0.7%,5.4%,3.0%,0.6%,6.6%,and the average error is 3.3%.It can be seen that the error of som-rbf neural network prediction model is small.(4)Based on the prediction model of concrete carbonation depth of aqueduct,the life prediction model of aqueduct concrete in the carbonation freeze-thaw environment in Guanzhong area is preliminarily established by taking the carbonation depth equal to the thickness of steel protection layer as the sign of concrete structure life termination,and the actual aqueduct life prediction is carried out,and the corresponding results are obtained.