China Safety Science Journal ›› 2022, Vol. 32 ›› Issue (8): 45-51.doi: 10.16265/j.cnki.issn1003-3033.2022.08.2702

Previous Articles     Next Articles

Multi-objective optimization of surface settlement safety control during shield construction based on RF-NSGA-II

WU Xianguo1(), FENG Zongbao1, LIU Jun1, WANG Lei1,**(), CHEN Hongyu2, LI Xinyi1   

  1. 1 School of Civil and Hydraulic Engineering, Huazhong University of Science and Technology, Wuhan Hubei 430074, China
    2 School of Civil Engineering and Environment, Nanyang Technological University, Singapore 639798
  • Received:2022-02-12 Revised:2022-05-15 Online:2022-09-05 Published:2023-02-28
  • Contact: WANG Lei


In order to effectively adjust construction parameters of shields, and to achieve safe construction, a multi-objective optimization model based on RF and NSGA-II algorithm was established, in which control analysis of construction parameters were optimized with main shield parameters as research object and ground settlement as control target. Then, nine shield parameters controlling surface settlement were selected as input indexes of RF prediction model, and nonlinear relationship between the parameters and settlement was obtained as NSGA-II fitness function. Then, cutter wear was selected as the second optimization objective, and constraint range of construction parameters was set for multi-objective optimization. Finally, with a rail transit project in karst areas as an example, verification was conducted. The results show that using RF algorithm in training and simulating measured data of the project will result in high prediction accuracy. And the proposed model based on RF-NSGA-II features significant multi-objective optimization effect of ground settlement and cutter head wear, and it can obtain control range of shield construction parameters in karst areas.

Key words: random forest (RF), nondominated sorting genetic algorithm-II (NSGA-II), shield construction parameter, surface settlement, cutter head wear, multi-objective optimization