The Research On Safety Assessment Technology Of The Existing Gravity Retaining Walls

The Gravity Retaining Walls,which serve as one of the common forms of road and rail slope support,whose security situation can chang greatly due to the combined effects of various factors after its operation for some time.Therefore,to determine the safety of the retaining walls,it is necessary to carry out a comprehensive assessment of the existing Gravity Retaining Walls.However,the current the assessment process of the assessment of the Gravity Retaining Walls often requires people to participate in it,so there are many factors influencing the objectivity of the evaluation results.Therefore,based on the evaluation methods proposed by scholars at home and abroad,this paper summarizes the state assessment method of Gravity Retaining Walls based on T-S fuzzy neural network and the evaluation method of Gravity Retaining Walls' bearing condition based on safety factor,which combines the qualitative assessment and quantitative assessment,so that to increases the accuracy,usability and objectivity of the assessment system.The main research contents are listed as follows:1.Research on Safety Evaluation Index System of Gravity Retaining Walls.By analyzing the common characteristics and damage forms of the Gravity Retaining Walls,the key factors influencing the damage of the Retaining Walls are summarized.Besides,Based on the selection method of the safety evaluation index of the existing subway structure and the concrete beam bridge,Evaluation Index System of Gravity Retaining Walls is established.2.Study on Scoring Criteria of Evaluation Indexes of Gravity Retaining Walls.Firstly,according to the importance of each index on Gravity Retaining Walls,the weight of each index is determined by AHP.Then,the scoring standard of each index is established.Finally,the weights of each index are multiplied by the operating state score,And the comprehensive score of the technical state of the Gravity Retaining Walls is obtained,which lays the foundation for the input of TS fuzzy neural network.3.Research on Technical State Evaluation of Existing Gravity Retaining Walls Based on T-S Fuzzy Neural Network.Firstly,a fuzzy neural network,which is suitable for multidimensional fuzzy reasoning is proposed,which is the T-S fuzzy neural network.T-S model can generate more complex nonlinear functions with a small number of fuzzy rules,which can effectively reduce the number of fuzzy rules when dealing with multivariable systems and is very suitable for the structure of Gravity Retaining Walls.Secondly,the structure of TS fuzzy neural network,the number of neurons in each layer and the learning algorithm of neural network are studied.Finally,the sample data of TS fuzzy neural network is introduced And the samples are divided into training samples and prediction samples.Training samples are used to train the program and the prediction samples are used for verification.The feasibility and accuracy of T-S fuzzy neural network assessment method are obtained through training and prediction.4.Study on Evaluation of Load Condition of the existing Gravity Retaining Walls.First of all,the five types of Gravity Retaining Walls,such as sliding damage,overturning damage,uneven foundation subsidence,wall damage and sliding damage along the weak layer,and reference to "highway bridge carrying capacity assessment procedures"(JTG / T J21-2011),the method of modifying the bearing capacity of Gravity Retaining Walls are put forward.Then,for the five kinds of failure forms,the safety factor method is used to evaluate the bearing capacity of the bridge as the final level of the assessment of the load condition of the Gravity Retaining Walls.Finally,the PKPM2012 foundation and the geotechnical toolbox model are used to calculate the mechanical analysis of the retaining wall,while keeping the method of maintaining the size of the single factor at the same sensitivity and changing only one factor to carry out the analysis of Gravity Retaining Walls Safety Assessment Parameters.5.Engineering application.Taking the gravity retaining wall detection and appraisal project in Laoshan scenic spot of Qingdao as an example.The T-S fuzzy neural network evaluation method and the safety factor method are used to evaluate the technical status and the bearing status of the existing gravity retaining wall in the case.Finally,the maintenance and reinforcement recommendations are given out according to the evaluation results.