人工神经网络在高温后静置混凝土抗压强度预报中的应用

APPLICATION OF ARTIFICIAL NEURAL NETWORK IN THE PREDICTION OF COMPRESSIVE STRENGTH OF LONG STANDING CONCRETE AFTER EXPOSURE TO HIGH TEMPERATURE

根据试验结果,应用人工神经网络(ANN)方法对火灾高温后静置混凝土的抗压强度进行了预报,预报值与试验值吻合良好。探讨了受火温度、高温后静置时间及冷却和养护方式对火灾高温后混凝土抗压强度的影响。此外,利用该网络模型分析了火灾高温下混凝土的抗火性能。表明该方法是可行的。

Based on test results, the compressive strength of standing concrete after exposure to high temperature is predicted by artificial neural network (ANN). Good agreement is reached between the predicted results and the test data. The effects of high temperature, standing time after cooling, methods of cooling and curing on the compressive strength of concrete are discussed. The fire resistance performance of concrete exposed to high temperature fire is analyzed by the ANN. It is concluded that the ANN method is feasible to estimation of the compressive strength of standing concrete after exposure to high temperature in practice.