基于BP神经网络的建筑结构隔震初步设计系统

摘要

Based on BP neural network, as the seismic fortification type, fortification intensity, site type, seismic design group, heightwidth ratio,aspect ratio, stiffness, mass and of structure to main impact factor, as the maximum shear force ratio and the largest displacement of isolation bearing to output data, a base isolation structural initial design system is set up. After trained network by 25 training samples, it is used to test 15 test samples. It is show that accuracy is 96％ by comparing test with actual design results, which indicates that the analysis results of base isolation structure by base isolation structural initial design system based on BP neural network are more effective and accurate.%基于BP人工神经网络,以建筑结构的抗震设防类别、设防烈度、场地类别、地震分组、高宽比、长宽比、刚度,质量和面积为主要影响因子,以隔震后结构的最大层剪力比和支座最大位移作为输出结果,建立一个隔震初步设计系统.经25个训练样本对该网络进行训练后,利用15个测试样本对网络进行了测试.通过测试结果与实际设计结果的对比,网络的平均准确率达到96%,说明基于BP神经网络的隔震初步设计系统对隔震结构的减震效果分析具有高效性和准确性.