Manwan Hydropower Station Main Building Rock-mechanical Parameters Of The Study

Rock itself is a highly complex uncertain system. Its mechanical parameters, constitutive model, calculation of boundary conditions (stress) and measurements of displacement and stress can not be accurately determined. The anti-analytical method that determines each kind of computed parameters and model by monitor information has developed rapidly since its introduction . And now this method has become one of the important methods to solve complex geotechnical mechanics problems. This paper carries on a mechanical parameter anti-analysis for cave rocks of the main plant of Manwan Hydropower in the second phase project , using pro and con anti-analysis method to calculate cave rock mechanical parameters of the main plant. The main content of this paper is illustrated as follows:1. As the monitoring process of actual works is influenced by the surrounding environment, the data detected has errors to some extent. In order to reduce the impact of errors, at first the paper uses general ABAQUS software of FEM and Drucker-Prager model to simulate rock and provides a sample space for BP neural network. On this basis, we can make the parameter anti-analysis.2. Then the paper uses the BP neural network function of the MATLAB database to carry on BP neural network training for the sample space provided by ABAQUS. After training success, it substitutes monitor data and calculates the mechanical parameters.3. In order to verify the correctness of the model established by ABAQUS and the value calculated by BP neural network, the paper substitutes the anti-calculated data into ABAQUS once more. After that, it calculates the displacement value and makes comparison between the displacement value and the actual displacement value.