Application Of Inequality Constrained Adjustment In Data Processing Of High Speed Railway

With the rapid development of surveying and mapping technology,measurement modes are becoming more and more abundant,and the acquiring methods and quantities of measurement data are also increasing.There are a lot of priori information hidden in observation data.Classical surveying adjustment can’t fully utilize these information,which result in loss of data resources.In view of the current situation that prior information can’t be effectively used in the processing of engineering survey data,the algorithm and accuracy evaluation method of the inequality constraint adjustment model are researched,and the data of the deformation monitoring control network are processed by inequality constraint adjustment,which improves the accuracy of parameter estimation.The main research work of this thesis is as follows:1.In order to extract valid prior information from engineering measurement data,the ways of obtaining the prior information in the literature are sorted out.and the prior information was divided into the random prior information and the prior information in the function model according to the way the prior information was used in the mathematical model.The inequality constraint adjustment model is studied deeply,the optimality condition of the model is given,and the advantages and disadvantages of various inequality constraint algorithms are compared by practical examples.2.After analyzing the deficiencies of common inequality constrained adjustment algorithms,a genetic algorithm based on adaptive penalty function is proposed.By comparing the advantages and disadvantages of four penalty functions and summarizing the construction principle of penalty function,a simple and efficient adaptive penalty function is constructed,and the complexity and convergence of the new algorithm are proved.The influence of inequality constraint on parameter estimation and accuracy is deduced.The quality evaluation methods of three types of inequality constraint adjustment parameter estimation are compared.The accuracy of inequality constraint adjustment solution is calculated by Monte Carlo simulation,which shows that this method can obtain more accurate accuracy information.3.The problem of inequality constraint adjustment for deformation monitoring data is that it is difficult to obtain effective and accurate prior information.According to the periodicity of deformation monitoring,the prediction value calculated by the combined prediction model is studied as prior information.In order to obtain accurate prior information,it is proposed to use Kalman filtering to solve the model parameters.And L-M method is used to optimize the state covariance matrix of each iteration to ensure that the observation information is fully utilized.Through the calculation of the measured data,which is proved that the Kalman combined prediction model based on L-M can obtain accurate and effective prior constraint information.4.The inequality constraint conditions are established by the predictive value and its RMSE,the genetic algorithm based on the adaptive penalty function is used to solve the settlement data of the high-speed railway,and the adjustment results under different constraint scales are compared.The inequality constraint conditions are established by the horizontal cumulative displacement of the high-speed rail.The new algorithm is used to solve the plane data of high-speed railway.Finally,the Monte Carlo simulation method is used to evaluate the quality of the unconstrained and constrained adjustment results,which proves that the inequality constrained adjustment can significantly improve the accuracy of parameter estimation.