论文部分内容阅读
随着计算机技术的发展,人工神经网络在结构损伤识别中的运用得于实现。对混凝土简支梁不同损伤的程度和位置进行模态分析,得到前5阶的频率数据作为神经网络所需要的训练和测试样本。运用不同的网络结构和传递函数建立起各种神经网络模型,通过比较可知神经网络能够很好地利用于混凝土结构的损伤识别,选择合适的网络结构和传递函数对神经网络模型的效果将产生关键的影响。
With the development of computer technology, the application of artificial neural network in structural damage identification has been realized. Modal analysis was conducted on the degree and location of different damage of concrete simply supported beams, and the first five steps of the frequency data were obtained as the training and test samples required by the neural network. By using different network structures and transfer functions, various neural network models are established. By comparison, it is known that neural networks can be applied to damage identification of concrete structures well. Selecting the appropriate network structure and transfer function will have a crucial effect on the neural network model Impact.