Prediction of bond strength between new and old concrete based on BP neural network
The prediction method of the interface bond strength between old and new concrete is examined.Based on hundreds of experimental data reported in the literature worldwide,a database is then built up.The bond strength between the old and new concrete calculated by the standard methods and the simplified methods are compared with the experimental results.Based on neural network technology,a BP neural network model is established for the prediction of the bond strength between new and old concrete.Eight factors(old concrete age,compressive strength of old concrete,compressive strength of new concrete,interfacial area,interface saturation degree,interface roughness,interface agent,test method)are taken into consideration.The predicted results obtained by the BP neural network achieve good agreement with the test results.
BP neural networkinterface between new and old concretebond strengthprediction
万方数据
