Dadingshan Tunnel Rock Parameters Back Analysis And Construction Optimization

Geotechnical engineering calculations, geotechnical parameters is very important. In this paper evolutionary neural network method for inversion to obtain, for the future for further research and practical engineering application provides a new means and methods. With feedback to the designer in order to achieve "just in time to modify design, reduce costs, shorten the construction period" effect. And can further validation evolutionary neural network method has the advantage, but also more accurately predict the displacement of tunnel construction provides a theoretical basis.In this paper, Dadingshan tunnel engineering background, according to the obtained tunnel deformation monitoring measurement field measured data, measured values for data fitting analysis, resulting in deformation of the distribution of measured values. In addition, the selection of FLAC3D software as a platform for establishing the tunnel displacement and mechanical parameters and provide the mapping between the sample data, as evolutionary neural network analysis of samples. Through a network of training, learning and other technical means, build parameters and the displacement between the non-linear mathematical model. After distribution of the deformation has been built into a good nonlinear mapping model inversion rock mechanical parameters. Finally, based on the inversion of this paper is analyzing the results of the same cross section surrounding rock deformation category value projections and comparative analysis to optimize the construction program.Having statistical analysis of field data, combined with theoretical analysis and numerical simulation, using decimal genetic algorithm BP neural network training process automatically search for the optimal training effect parameters of the network topology, the formation of neurons evolutionary algorithm, neural network training in order to achieve the required accuracy parameters back analysis. Furthermore neurons using evolutionary algorithms Dadingshan Tunnel Deformation mechanical parameters of rock physics, inversion results will be used for the construction of the tunnel surrounding rock deformation subsequent short-term forecasting and optimization comparative analysis of the construction, the use of the results show that the method has higher prediction accuracy of short-rock deformation.These research results Dadingshan soft rock tunnel section construction of deformation has been applied, the results show that this method is in good agreement with the actual excavation displacement, effectively ensure that paragraph rock tunnel construction safety and engineering quality, and to design and construction of similar projects provide some guidance.