应用人工神经网络预测锈蚀钢筋与混凝土粘结性能

钢筋混凝土粘结强度受多种因素 (如混凝土强度、混凝土保护层厚度、钢筋直径、钢筋类型、钢筋锈蚀率等 )的共同作用 ,建立计算模型比较困难。在试验研究的基础上 ,利用人工神经网络技术 ,分别在考虑单一因素和多种因素的情况下建立BP网络模型预测锈蚀钢筋与混凝土之间的极限粘结力 ,从而不需建立具体的数学模型就可以得到较满意的预测结果。为受腐蚀钢筋混凝土结构力学性能的研究提供一种新方法和新思路 ,为工程实际应用提供简便的预测方法

APPLICATION OF ARTIFICIAL NEURAL NETWORKS IN PREDICTION OF BOND PROPERTY BETWEEN CORRODED REINFORCEMENT AND CONCRETE

Bond property between the reinforcing steel and concrete is influenced by many factors such as concrete strength, depth of concretecover, reinforcement diameter, reinforcement type, corrosion ratio etc., which make it difficult to build any calculation model. Based on the test results, Artificial Neural Network(ANN) was introduced into the prediction of bond property between the corroded reinforcement and concrete in this paper. Considering single factor and multi factors, BP Neural Network models were built respectively. It is shown that satisfactory results can be achieved by the given BP models, whereas need no mathematical model. Therefore, ANN will break a new approach for the study on mechanical property of corroded reinforced concrete structure, which will provide a simple calculation method for engineering in practice as well.