| Tunnel monitoring measurement is to obtain relevant parameters to surrounding rock state around tunnel face by the real-time filed survey, and then monitored data us processed to further predict and judge safety coefficients of surrounding rock and construction methods to maintain the stability of structural system. In this way, we can check the stability of surrounding rocks, and determine if the design and construction has reached standards, and provide corresponding technical support to the time selection of early supporting. At the meantime, it can provide reference for the time selection of secondary lining, which is the core part of new Austrian tunneling method, and provides corresponding guarantee to the safety and quality of construction. Based on corresponding standards and specifications, the tunnel monitoring measurement is divided into the essential and optional measuring projects; the former includes four aspects: inside and outside tunnel observations, horizontal clearance, rock bolt pulling resistance, and vault settlement volume measurement; vault settlement serves as an important basis for vault settlement research, and one of key indicators for the judgment of tunnel surrounding rock safety and stability, the measurement of which can bring an accurate pre-judgment on the deformation volume of surrounding rocks, and is of vital importance to tunnel construction. Therefore, in recent years surrounding rock deformation prediction has become one of hot topics in the study of tunnel construction safety.Over several decades of development, tunnel monitoring measurement technology has experienced considerable development in the field observation. Varied observation methods have emerged, and the accuracy of observation instruments and equipment is improved continuously. But because of the complexity of the operation environment inside tunnels and many interference factors, it is inevitable that some gross errors exist in observation data, so a series of high-precision data preprocessing of original data form the basis for the final prediction of tunnel deformation; Combining with construction experience, influence factors of arch crown subsidence is envisaged, and verified by the method of Grey Theory; In order to achieve the purpose of convenient calculation, calculation method of multivariable Kalman model is improved.And the main research content and achievements in this paper are listed as follows:1. Firstly, in combination with the specific example of "Pear Garden tunnel project of Chengdu-Chongqing passenger transport line", a brief introduction is given to filed monitoring process and work flow of monitoring measurement. And the concrete work is:the design procedure of monitoring measurement scheme; the specific implementation of monitoring measurement scheme; the frequency of monitoring and observation apparatus and method.2. Research mathematical method for prediction at home and abroad, tunnel surrounding rock deformation, although the methods to establish the forecast model have many kinds, because the Kalman filtering method and grey method in the pretreatment of tunnel data, correlation analysis and prediction have their respective advantages, which is the key of the research becoming the two most extensive method, we put their own advantages of comprehensive application in the final prediction model.3. To the question whether there are gross errors of the raw datas, we appreciate it from the classical regression model at the same time make efficiency and precise in the balance, it discrimination, the establishment of a self-dependent and independent variables as a function of relationship, the dependent variable is obtained analog value and error, confidence interval, select95%, which is the error when the difference between the measured value and the analog values of about2times, if we have a problem about the observation value, according to the circumstances of this observed interpolation to estimate the value or to be removed.4. In combination with actual working experience and relevant theories, it can be found that the deflection of tunnel vault is not only closely related to time, but also relevant to parameters of tunnel face and secondary lining mileage. So grey correlation analysis is adopted for verification, and grey absolute theory is applied to prove that the tunnel surrounding rock deformation quality is indeed related to excavation and secondary lining progress; and then the deflection and the correlation degree between two variables are attained. Taking the predicted accuracy of efficiency into comprehensive consideration, we combine two variables into one variable, together with time variable in the following prediction as independent variables.5. Kalman filtering is a recursion process of continuous prediction and modification. And data after treatment has the minimum variance and reliability, which cannot only apply to filtering with stationary sequence, but is also suitable for nonstationary Markov sequence; therefore, it is widely applied in data prediction. And Kalman filtering model are constructed in consideration of tunnel face and secondary lining parameters to predict change in the tunnel vault.6. Spss software is used to simulate the approximate function expression between settlement volume and time, and the approximate function expression between settlement volume and composite variable, and then above two expressions are combined into Kalman filtering Model, and their coefficient are determined by Kalman filtering method, and finally a parameter with relatively higher accuracy is obtained; therefore, the final function expression can predict settlement volume under the condition that the independent values are known.7. Comparing with the tunnel vault subsidence data in the difficulty with the operation. finally forecast precision and so on, the algorithm has advantages such as high fitting accuracy, not divergence etc.8. Put the content in the problem of safety distance and deformation prediction in to engineering practice. |
PDF Full Text Request