Title: Optimising the supporting structure of a bridge's foundation pit based on hybrid neural network
Authors: Rong Deng; Lei Zhou
Addresses: Big Data and Internet of Things School, Chongqing Vocational Institute of Engineering, Chongqing, 402260, China ' No. 6 Supervision Consulting Company of Chongqing Construction Science Research Institute, Chongqing, 400016, China
Abstract: Aiming at the problem of large settlement of nearby buildings caused by displacement deformation of supporting structure of a bridge foundation pit, an optimisation design method of supporting structure of a bridge foundation pit based on hybrid neural network is proposed. The structure model of hybrid neural network is setup and the variables are inputted. The minimum deformation prediction results of the bridge foundation pit support structure are obtained through training. The earth pressure value is calculated. The minimum soil depth and pile length of the supporting pile are calculated to complete the optimisation of the supporting structure of the bridge foundation pit. The actual engineering test shows that the average values of vertical displacement, horizontal displacement and peripheral settlement of the proposed method are 6 cm, 15 cm and 13 cm respectively, which are all within the allowable range, which can effectively guarantee the safety of excavation and support of foundation pit engineering.
Keywords: hybrid neural network; bridge; supporting structure of foundation pit; optimal design.
International Journal of Critical Infrastructures, 2022 Vol.18 No.2, pp.159 - 171
Received: 29 May 2020
Accepted: 03 Sep 2020
Published online: 20 Jun 2022 *