<<
[an error occurred while processing this directive] | [an error occurred while processing this directive] >>
On the Optimization of Deformation Prediction Method for Karst Tunnels
in Complex Geological Conditions
(1 School of Civil Engineering, Beijing Jiao Tong University, Beijing 10044; 2 The No.6 Engineering Ltd. of China Railway 20th Bureau Group, Xianyang 711120)
Abstract It is of great importance to monitor the deformation of karst tunnels in order to evaluate the tunnel stability. To improve the accuracy of prediction models, an IOWHA operator-based combined model is set up. Firstly the grey theory and time series method are optimized to fit predicted tunnel deformation. Secondly an individual method based on the predicted results is used to establish a combined prediction model. Using original data obtained from a karst tunnel, the corresponding results of this combined prediction model are given, and the results are compared to those from former models. Results show that the IOWHA operator-based combined model combines both the advantages of grey theory and the time series method, improving prediction accuracy greatly.
Abstract:
It is of great importance to monitor the deformation of karst tunnels in order to evaluate the tunnel stability. To improve the accuracy of prediction models, an IOWHA operator-based combined model is set up. Firstly the grey theory and time series method are optimized to fit predicted tunnel deformation. Secondly an individual method based on the predicted results is used to establish a combined prediction model. Using original data obtained from a karst tunnel, the corresponding results of this combined prediction model are given, and the results are compared to those from former models. Results show that the IOWHA operator-based combined model combines both the advantages of grey theory and the time series method, improving prediction accuracy greatly.
.On the Optimization of Deformation Prediction Method for Karst Tunnels
in Complex Geological Conditions[J] MODERN TUNNELLING TECHNOLOGY, 2013,V50(5): 87-91