Research On Gross Error Detection Of CPⅢ Control Network

High-speed railway construction must meet the requirements of high ride comfort,high stability,and high precision.In the CPⅢ control network measurement process,the appearance of gross errors is inevitable,and gross errors will have a serious impact on the adjustment results.Therefore,it is very important to correctly detect gross errors in the CPⅢ control network data processing process.Most of the previous gross error detection methods use the traditional gross error detection method,but in the traditional gross error detection method,gross error detection and strict weighting affect each other.If strict weighting is performed first,the gross error in the observation value will be affected.Strictly fixed powers have an impact,resulting in inaccurate rigorous powers.If gross error detection is performed first,part of the gross error in the observation value will not be detected,and the optimal unbiased estimate will not be obtained,which will affect the subsequent adjustment calculation.Therefore,it is inaccurate to detect gross errors and rigorously determine the weights.Therefore,how to accurately detect gross errors is a problem worthy of study.First of all,in the data processing of CPⅢ control network,the calculation of approximate coordinates is a very critical link.If the calculation of approximate coordinates is inaccurate,and the deviation between the linear model and the nonlinear model is too large,model errors will occur and the approximate coordinates will deviate from the accurate value,Leading to misjudgment of gross errors,which will have a serious impact on the subsequent adjustment calculation and gross error detection,so it is very important to calculate the approximate coordinates correctly.Aiming at the traditional approximate coordinate calculation method,this paper proposes a linear model approximate coordinate calculation method,which improves the accuracy of approximate coordinates.Secondly,the traditional gross error detection method is due to the mutual influence between the detection gross error and the strict weighting.This paper proposes a new gross error detection method to solve the problems of the traditional gross error detection method,namely sub-item gross error detection.method.The sub-item gross error detection method uses Baarda data detection method to separate the horizontal observation value and distance observation value for gross error detection,that is,to detect the gross error of a single observation value,there is no problem of the interaction between gross error detection and strict weight determination,which solves the problems existing in the whole gross error detection method.Finally,this article uses MATLAB software to design and compile the program based on the sub-item gross error detection method and the traditional gross error detection method.The two gross error detection methods are verified through a large number of experiments,which proves the itemized gross error proposed in this paper.The detection method solves the problems in the traditional gross error detection method,and the sub-item gross error detection method is more accurate than the traditional gross error detection method.