基于改进神经网络的预应力锚杆布置间距

Range interval distance of preforced grouting anchors using update neural network

以山东省境内的 1 0 4国道界河立交桥加筋土挡土墙的失稳加固为实例 ,根据预应力锚杆在拉拔过程中 ,不同的距离范围内应力增量的变化试验 ,利用改进的 BP神经元网络对这一试验值进行了学习 ,经过检验发现网络的输出值与期望值之间的误差较小 ,所以网络具有比较强的推广能力。利用这一网络对试验数据作了进一步的推广得到了另一组试验数据。根据这一组数据 ,在给锚杆施加一定预应力的条件下 ,可以求出锚杆的作用范围。根据这一范围和挡土墙破坏状况以及外部载荷的分布情况 ,利用极限平衡理论得出了比较合理的锚杆布置间距。

As a kind of common reinforcement technique for the instability rock soil structures, the reinforcement grouted anchors was a very effective and reliable technics. For size of the complicated properties of rock soil, it is very difficulty to choose and design reasonable engineering parameters. Taking the practical reinforced engineering of a reinforced soil retaining wall as an example, which located in Shandong province and set on No.104 national highway,the stress spread behaviors of the reinforcement grouted in the preforced proceeding was tested. Accounting to the test data, and by use of the update backpropagation algorithm neural network,the test method and it's mechanism were studied by the network, then the learning results show the MSE and output data were less than the goal values(the main aim, in this paper, is to build a neural network directly from the in situ test results(the learning phase)). The learning and adjustment abilities of the neural network permit us to develop the test data, a new group of test data were acquired from the neural network. By use of the provided data, as well as the failure situation and carried loading capacity of the retaining wall, the reasonable range interval distance of reinforcement grouted anchors were carried out.