欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
通过对61根型钢混凝土柱试验数据的整理，利用神经网络原理建立5-6-1型反向传播（BP）神经网络模型，分析不同参数对型钢混凝土柱位移延性系数的影响．分析及预测结果表明，学习样本和测试样本的预测值与实验值之比的均值分别为1．0006，0．9980；标准差分别为0．0203，0．0596，预测值与试验值吻合良好．当轴压力系数增加到0．42以后，位移延性的变化较小；体积配箍率增加到1．9％后，位移延性的增长减缓；当剪跨比小于1．5时，型钢混凝土柱的延性系数随剪跨比的增加而提高；但当剪跨比大于1．5时，随着剪跨比的增加，型钢混凝土柱位移延性系数有所降低．
英文摘要：
Based on the test data of 61 steel reinforced concrete columns, a model with 5 input layers, 6 implicit layers and 1 output layer （5-6-1） is developed to analyze the influence of various parameters on displacement ductility by the principle of back propagation （BP） neural network. The ratio of prediction value to experimental value is 1. 000 6 for the learning samples, 0. 998 0 for the testing samples; the standard deviation is 0. 020 0 and 0. 059 6 respectively, indicating that the prediction results conform well to the test results. When the axial compression ratio increases more than 0.42, the variation of the displacement ductility ratios is small; when the stirrup ratio per unit volume increases more than 1.9%, the displacement ductility increases slowly; with increasing the shear span ratio 2 , the displacement ductility increases for λ〈1.5, decreases for ,λ〉1.5.
同期刊论文项目
同项目期刊论文
期刊信息
位置：