Application Research Of Fuzzy C-Means And Cloud Model Comprehensive Evalution In CPⅠ Stability Analysis

During the long construction period of the high-speed railway construction phase,the stability analysis of the control points in the re-test of the CPⅠ precision survey network is a prerequisite and an important link to ensure the safety of the subsequent projects.Since the stability of the control point is a fuzzy problem,in order to evaluate the stability of the CPI point scientifically and reasonably,fuzzy clustering and fuzzy comprehensive evaluation methods are introduced to analyze the stability of the CPI point.This thesis takes practical engineering as an example,and the main research contents are as follows:1.Aiming at the shortcomings of the existing FCM algorithm,the improved FCM algorithm is used: the optimal classification is determined by using the fuzzy equivalent fuzzy clustering method and F statistic,and then the clustering results are further corrected by the FCM algorithm to obtain the optimal classification number for each classification.The classification of control points and the degree to which each control point belongs to each stability class.The feasibility and effectiveness of the improved FCM algorithm in the stability analysis of control points are proved by the actual data.2.Considering that each attribute index of measured data has different effects on clustering in practical problems,weights are introduced and discussed for 9 methods of constructing fuzzy similarity matrix.Through the calculation of two sets of measured data,the appropriate method for constructing the fuzzy similarity matrix is screened out according to the actual situation.Taking the second re-measurement of the CPI precision measurement network in the new stage of a high-speed railway as an example,the compatibility analysis of the starting point based on fuzzy clustering based on the fuzzy equivalence matrix is used to verify the applicability of this method;The fuzzy similarity matrix is established by the change of the coordinates of the CP Ⅰ point and the relative relationship between the adjacent points,and then the improved FCM algorithm is analyzed to obtain the classification of the stability of each control point in the CPⅠ precision survey network re-test.3.Establish a fuzzy comprehensive evaluation model for the stability of CPI points in the new stage,and analyze the stability of each control point from three aspects: the quality of observation data,the accuracy of baseline solution and adjustment,and the difference of coordinate results;the cloud model is used to determine the degree of membership to make up for the shortcomings of the traditional membership determination method,the improved AHP method is used to determine the subjective weight of the criterion layer and the measure layer,and it is proposed to use the anti-entropy weight method to calculate the objective weight of each measure layer based on the measured data,and then combine the subjective and objective weights based on the game theory.The weights are comprehensively empowered,which solves the problem of inconsistency between subjective and objective empowerment.The validity degree of the maximum membership degree is tested.When the maximum membership degree principle is ineffective or invalid,the comprehensive evaluation vector is quantitatively processed,and then the stability of each control point is comprehensively and reasonably evaluated.