Research On Risk Assessment And Prediction Of Ground Collapse Caused By Pipeline Leakage

In recent years,urban ground collapse accidents have occurred frequently,causing huge property losses and posing a huge threat to people's lives.Through investigation and analysis of the causes of ground collapse accidents,it is found that hydraulic pipeline damage is the main cause of ground collapse.The hydraulic pipeline is damaged and leaked,and the surrounding soil is eroded to form underground concealed cavities,endangering public safety.Therefore,analyzing and researching the risk of ground collapse caused by the damage and leakage of hydraulic pipelines is of great significance to protecting people's lives and property.In this paper,with the aid of the dual-pipeline hydraulic collapse model test system,based on the mechanism of the formation of underground cavities by the leakage of erosion soil after the hydraulic pipeline is damaged,indoor model tests are conducted to study the development process of underground cavities and different influencing factors after the pipeline is damaged,and support vectors are used.The computer performs data prediction and simulation,and develops a collapse risk evaluation system based on MATLAB GUI.In view of the risk assessment and prediction of road surface collapse caused by pipeline leakage,the main work content and research conclusions of this paper are as follows:Dangerous Analysis of Ground Collapse Caused by Pipeline Leakage This double pipeline hydraulic subsidence model test analyzes the main forms of ground collapse caused by pipeline leakage under the influence of dual factors(velocity-compactness),selects the working conditions with representative results,records the pictures of the working conditions,and analyzes the law of collapse.The collapse features are summarized as collapse time and collapse range,and the analysis results are obtained.(1)The density is fixed and the flow velocity changes(working condition one and working condition two)when the water velocity increases,the collapse time of the model is shortened accordingly.The top surface collapse area of condition 1 is twice that of condition 2;the amount of sand loss on the side of condition 1 is greater than that of condition 2,and the area of the cavity in the front view of condition 1 is about that of condition 2.The area of the cavity in the front view is 4 times.(2)When the flow velocity is fixed and the compactness changes(working condition one and working condition three),when the soil compactness increases,the collapse time of the model also increases.The collapse area at the top of the working condition 1 is three times of the collapsed area of the working condition 3.The area of the cavity in the front view of Case 1 is roughly similar to that of the front view of Case 3,and the radius of the collapsed cavity is similar,both from top to bottom,decreasing as the height decreases.Prediction and analysis of ground collapse caused by pipeline leakage Through the analysis software MATLAB,the support vector machine SVM is used to complete the predictive analysis of the collapse time scale.And input the processed data into SVM,and explore the performance of the two sets of data in the support vector machine SVR simulation.Subsequent experiments show that the processed collapse time data can not only reflect the characteristics of the initial collapse time,but also be suitable SVM performs the recognition and prediction of the two-factor collapse pattern.Analysis of Ground Collapse and Damage Caused by Pipeline Leakage According to the scale of the collapse range,the coordinates of the characteristic points of the representative working conditions are extracted,and the damage range definition analysis is completed.Based on the previous theoretical algorithms and experimental results,this study constructed a system that can learn SVM parameters offline,train models,and identify faults online for test samples.Based on this algorithm flow,a support vector machine collapse risk evaluation of MATLAB GUI was developed.In the system,subsystem one can bring in the collapse time data,and perform parameter fitting to obtain the best prediction result under specific working conditions.Subsystem two can bring in soil compactness parameters,water velocity parameters,and position coordinate parameters to obtain the risk level of the actual collapse location,and provide corresponding risk prevention countermeasures.