High-speed rail stratified foundation settlement early prediction confidence analysis method and high-speed rail stratified foundation settlement early prediction confidence analysis system

Abstract

The invention discloses a high-speed rail stratified foundation settlement early prediction confidence coefficient analysis method. The method comprises the steps that S1, a settlement prediction confidence coefficient change rule under the preloading stratified foundation condition based on the loading period and the standing period is determined; and S2, carrying out high-speed rail stratified foundation settlement early prediction confidence analysis on the basis of the settlement prediction confidence change rule under the preloading stratified foundation condition in the loading period and the standing period. The invention further discloses an analysis system, electronic equipment and a computer readable storage medium, a loading period and a standing period are used as research objects, a large amount of simulation settlement observation data under the multi-factor coupling effect is generated, and an adaptive curve fitting model is adopted to analyze the settlement observation data. A high-speed rail soft soil roadbed settlement prediction confidence coefficient change rule caused by observation condition randomness is statistically analyzed in combination with a PSO particle swarm algorithm; and constructing a settlement prediction confidence analysis model to carry out early prediction confidence analysis on the settlement of the stratified foundation of the high-speed rail.

Description

Technical Field

[0001] The present invention relates to the technical field of railway and foundation settlement prediction and analysis, and in particular to a confidence analysis method and system for early prediction of high-speed railway layered foundation settlement. Background Art

[0002] By the end of 2023, the operating mileage of my country's high-speed railways has exceeded 44,000 kilometers. At the same time, international projects such as the Jakarta-Bandung High-Speed ​​Railway, the Hungary-Serbia Railway, and the China-Thailand Railway have continued to take root. The rapid development of high-speed railways has become an important driving force for domestic economic growth and a beautiful business card for displaying an international image. Fast and safe high-speed railways have greatly met people's daily travel needs, but they have also put forward extremely strict requirements on line deformation control. As an important part of the high-speed railway line infrastructure, the post-construction settlement control of the roadbed is a key link to ensure construction quality and line smoothness.

[0003] As a strip-shaped project, high-speed railways pass through areas where soft soil is widely developed. Soft soil is a special soil that needs to be encountered in large quantities during high-speed railway construction. It is a fine-grained soil with a natural porosity greater than or equal to 1, a natural water content greater than the liquid limit, and an undrained shear strength less than 30kPa. Soft soil has low strength, poor permeability, high compressibility, poor uniformity, strong rheology, high thixotropy, complex layered distribution, and large differences in the physical and mechanical properties of each layer. The complex engineering characteristics and stratification make it extremely difficult to control the post-construction settlement of soft soil roadbed. Soft soil roadbeds can be treated with preloaded foundations and composite foundations. The preloading method combined with drainage shafts such as bagged sand wells and plastic drainage boards is widely used in the foundation reinforcement treatment of soft soil roadbeds. It can accelerate the settlement rate, ensure the long-term stability of the roadbed settlement, simplify the construction process, and ensure that the project cost is reasonable and controllable. However, the preloaded foundation has a larger settlement and a longer construction period than the composite foundation, which is not conducive to shortening the construction period, controlling post-construction settlement and estimating the track laying time. This is the main difficulty in its large-scale promotion and application in high-speed railway construction.

[0004] The settlement observation prediction and assessment method is based on limited observation data, and uses mathematical models to reasonably extrapolate the final settlement and post-construction settlement. Mathematical models generally include curve fitting models and fitting models based on neural networks and intelligent algorithms. The modeling and calculation process of fitting models based on neural networks and intelligent algorithms is relatively complex, and curve fitting models are still the main ones in engineering.

[0005] Curve fitting models generally include hyperbola method, three-point method, Asaoka method, exponential curve method, Hoshino method and other methods. The hyperbola method is a commonly used method in the prediction and evaluation of China's high-speed railway subgrade settlement. It was proposed by Nizipo Rović in 1955. The hyperbola method is suitable for fitting and extrapolating observation data during the static period. In order to effectively use the data during the filling period to achieve the purpose of early prediction, the German railway department proposed to expand the hyperbola method. Many scholars believe that the prediction results of the hyperbola method are conservative. Li Guowei and Feng Wenkai proposed corresponding correction methods; in 1959, Zeng Guoxi proposed the three-point method. Faced with the characteristics of small settlement magnitude and large data fluctuations encountered in the high-speed railway subgrade settlement, Chen Shan Xiong et al. integrated the exponential curve method, the hyperbola method and the three-point method, and proposed the improved three-point method, which achieved good results. In 1978, Japanese scholar Asaoka proposed the Asaoka method based on the one-dimensional vertical consolidation theory, also known as the Asaoka method or the graphical method. Some scholars believe that the deformation characteristics of the settlement curve after the inflection point conform to the characteristics of the exponential curve, and proposed the exponential curve method. Based on Terzaghi's one-dimensional consolidation theory, it is deduced that the degree of consolidation is proportional to the square root of time. The resulting Hoshino method is also widely used in the prediction and evaluation of high-speed railway subgrade settlement. Since then, the Verhulst model, Usher model, Logistic model, Compertz model, Weibul l model, settlement difference method model, etc. have been proposed one after another. These models can reflect the relationship between the settlement and time of the whole process, and are more often used for early prediction of subgrade settlement. The applicability of the curve fitting model is generally controlled by a correlation coefficient of not less than 0.92. Wang Xingyun et al., Liu Junfei et al., Zeng Juncheng et al., Liu Hong et al. compared and analyzed the applicability of various curve fitting models, and proposed indicators and methods for evaluating the applicability of curve fitting models.

[0006] In today's high-speed railway construction, post-construction settlement control is mainly achieved from two aspects. First, theoretical calculation of roadbed settlement under external load is carried out in the design stage. Due to the spatial variability of field sampling, the nonlinearity of soil stress-strain relationship, the limitations of theoretical calculation models and other issues, the calculation accuracy of roadbed settlement process and final settlement in the design stage is poor. Second, based on limited field observation data in the construction stage, the final settlement of the roadbed and post-construction settlement are reasonably extrapolated with the help of mathematical models. Since the degree of confidence of settlement prediction results is affected by soil properties, foundation treatment forms, and observation conditions, the settlement observation prediction and evaluation methods formed in the construction stage lack theoretical support. Surveys show that over-limit roadbed settlement is the main form of high-speed railway subgrade disease. At present, geotechnical engineering design and construction methods based on deterministic theory are difficult to accurately control foundation deformation, which brings great difficulties to high-speed railway construction and operation and maintenance.

[0007] Existing research has the following deficiencies. First, previous studies were mostly based on deterministic theories, focusing on the optimization and innovation of settlement prediction models, but there was little research on the possible distribution range of settlement prediction results. Second, the settlement prediction results are affected by the coupling of multiple factors such as soft soil conditions, foundation stratification, foundation treatment form, applicable scope of curve fitting models, observation conditions, etc., and the influence of multiple factor coupling on the confidence level of settlement prediction results is unclear. Third, when using observation data from the filling and loading process, the influence of construction technology on the confidence level of settlement prediction results has not been effectively discussed. Fourth, the current high-speed railway subgrade settlement prediction and evaluation methods lack theoretical support for their evaluation indicators. Therefore, there is a need to design a new technical solution for early prediction of high-speed railway stratified foundation settlement confidence analysis. Summary of the invention

[0008] The purpose of the present invention is to provide a confidence analysis method and system for early prediction of settlement of high-speed railway layered foundation to overcome the defects of the prior art and perform confidence analysis for early prediction of settlement of high-speed railway layered foundation.

[0009] The first aspect of the present invention is to provide a confidence analysis method for early prediction of high-speed railway layered foundation settlement, comprising:

[0010] S1, determine the changing rule of settlement prediction confidence under the condition of preloading layered foundation based on the loading period and static period;

[0011] S2, conducting an early prediction confidence analysis of high-speed railway layered foundation settlement based on the settlement prediction confidence variation rule under the conditions of preloaded layered foundation during the loading period and the static period.

[0012] Preferably, the S1 comprises:

[0013] S11, taking the loading period and the static period as the research objects, considering the changes of observation conditions based on the settlement theory model, processing the settlement theory curves under different soil properties, stratification and foundation treatment conditions, and generating a large number of simulated settlement observation data under the coupling of multiple factors; the observation conditions include changes in observation time, frequency and accuracy; the processing includes interception, discretization and superposition of random error processing;

[0014] S12, based on the optimal use conditions of different curve fitting models, adopts a settlement prediction model that can describe the graded loading conditions and the entire settlement time-varying process, and combines intelligent algorithms to quickly find and select the appropriate curve fitting model, combined with the PSO particle swarm algorithm, to predict the simulated settlement observation data;

[0015] S13, statistically analyzing the settlement prediction mean and standard deviation under different observation conditions to form a settlement prediction confidence analysis method;

[0016] S14, reveals the changing rules of confidence in the prediction of high-speed railway soft soil roadbed settlement caused by the randomness of observation conditions when using an appropriate curve fitting model under different soil properties, stratification and foundation treatment conditions.

[0017] Preferably, the step S11 generates a large amount of simulated settlement observation data under the coupling of multiple factors, including:

[0018] (1) Obtain theoretical data through the sedimentation theory model;

[0019] (2) The theoretical data is “intercepted” into an observation period and a settlement period at a certain observation period; the certain observation period is a filling period and a static period of 3 months, 6 months, 9 months or 12 months;

[0020] (3) The observation period data are discretized according to the observation frequency of roadbed settlement during the observation stages such as the roadbed filling period and the static period. The observation frequency is selected in accordance with the "Technical Specification for Railway Engineering Settlement Deformation and Evaluation" (Q / CR9230-2016) and the updated version of the Technical Specification for Railway Settlement Deformation and Evaluation;

[0021] (4) “Superimposing random noise” on the discrete observation period data, including: superimposing the observation error according to the deformation measurement level and accuracy requirements corresponding to the error in the elevation of the deformation point of different railway structure deformation measurement levels specified in the Railway Engineering Survey Specification (TB10101-2018), to generate discrete observation data with random noise;

[0022] (5) Repeat steps (2) to (4) to obtain the simulated settlement observation data under the coupling of a large number of multi-factors.

[0023] Preferably, the statistical analysis of the settlement prediction mean and standard deviation under different observation conditions in S13 to form a settlement prediction confidence analysis method includes statistically analyzing the settlement prediction results according to formulas (1)-(3).

[0024]

[0025]

[0026] In the formula, R is the correlation coefficient; are the sample means of the two sets of series respectively; is the covariance of the two series; is the variance of the two series; μ is the mean of settlement prediction; σ is the standard deviation of settlement prediction; N s is the number of samples to be calculated; S k The kth predicted value of settlement.

[0027] Preferably, the disclosure of the change rules of confidence in the prediction of high-speed railway layered foundation settlement caused by the randomness of observation conditions under different soil properties, layering properties, and foundation treatment conditions, when using an appropriate curve fitting model, includes:

[0028] Taking the loading period and static period as the research objects, a settlement prediction confidence analysis model is constructed;

[0029] Taking the loading period and static period as the research objects, the prediction results of various observation data models are compared and analyzed, and the prediction models are selected to obtain the optimal model;

[0030] Based on the changes in observation conditions, the optimal model is used to determine the impact of changes in the loading period and the static period on the mean and standard deviation of the settlement prediction, and a statistical analysis is performed on the changes in the confidence level of the high-speed railway stratified foundation settlement prediction caused by the randomness of the observation conditions.

[0031] Preferably, taking the loading period and the static period as the research objects, comparing and analyzing the prediction results of multiple observation data models and selecting the prediction model to obtain the optimal model includes:

[0032] Based on the correlation coefficient, relative error and coefficient of variation, a comprehensive evaluation index is formed, as shown in formula (4):

[0033] ω=0.3×R+0.4×(1-CV)+0.3×(1-10×δ)(4)

[0034] In the formula, ω∈(0,1) is a comprehensive evaluation index; CV∈(0,1) represents the coefficient of variation; R∈(0,1) represents the correlation coefficient; δ∈(0,1) represents the relative error;

[0035] Based on the comprehensive evaluation index, the prediction results of various observation data models are compared and analyzed, and the prediction models are selected to obtain the optimal model.

[0036] Preferably, S2 includes:

[0037] S21, choose different prediction models under different conditions;

[0038] S22, changing the value of the observation condition, so as to determine the variation law of the relative error and standard deviation with the observation condition, as the confidence analysis result of early prediction of the settlement of the high-speed railway layered foundation.

[0039] The second aspect of the present invention provides a high-speed railway layered foundation settlement early prediction confidence analysis system, which is used to implement the method of the first aspect, and is characterized by comprising:

[0040] A settlement prediction confidence change law determination module (101) is used to determine the settlement prediction confidence change law under the condition of preloading layered foundation based on the loading period and the static period;

[0041] The high-speed railway layered foundation settlement early prediction confidence analysis module (102) performs high-speed railway layered foundation settlement early prediction confidence analysis based on the settlement prediction confidence variation rule under the conditions of the preloading layered foundation during the loading period and the static period.

[0042] A third aspect of the present invention provides an electronic device, comprising a processor and a memory, wherein the memory stores a plurality of instructions, and the processor is configured to read the instructions and execute the method described in the first aspect.

[0043] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a plurality of instructions, and the plurality of instructions can be read by a processor and execute the method described in the first aspect.

[0044] Beneficial effects of the method and system of the present invention:

[0045] Settlement prediction is a method of extrapolating the final settlement of the roadbed and the post-construction settlement based on the field measured data and using mathematical means such as curve fitting methods. For the settlement prediction confidence analysis conducted by the present invention, it is very difficult to use field measured data for research. The reasons are that, first, the field conditions are extremely complex, and it is difficult to effectively and accurately control the changes in factors such as soil properties, stratification, foundation treatment conditions, and observation conditions; second, the settlement prediction confidence research requires a large number of repeated observations for the same working condition to obtain a large number of observation data with random errors. Based on the field measured data, a large amount of financial, material, and human resources are consumed; third, the measured roadbed settlement curve is not only affected by external loads, but also by climate, rainfall, temperature and other factors, resulting in the settlement curve obtained on site being unsuitable for scientific research. Therefore, in order to obtain a large number of reliable samples, the present invention adopts a settlement prediction confidence analysis method based on simulated observation data as the inventive concept. The settlement prediction confidence analysis method, which includes the generation of simulated settlement observation data, settlement prediction based on settlement observation data, and statistical analysis of prediction results, is one of the key technologies for achieving the analysis objectives of the present invention. Theoretical calculation, numerical simulation, intelligent algorithm, mathematical statistics and other methods are used to reveal the changing rules of confidence in the prediction of high-speed railway soft soil subgrade settlement caused by the randomness of observation conditions when using appropriate curve fitting models under different soil properties, stratification and foundation treatment conditions. The following technical results have been achieved:

[0046] (1) It can accurately control the main influencing factors of soft soil roadbed settlement under the condition of preloaded layered foundation, consider the influence mechanism of the randomness of observation conditions on the settlement prediction results, and reveal the change law of the confidence of the settlement prediction results.

[0047] (2) The target standard deviation of roadbed settlement prediction for different track structures can be used as the core to analyze the technical conditions such as observation period, observation frequency, and observation accuracy required for soft soil roadbed, and form a settlement assessment method based on confidence analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the related technologies, the drawings required for use in the specific embodiments or the related technical descriptions will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0049] Figure 1 A flow chart of a confidence analysis method for early prediction of high-speed railway layered foundation settlement according to an embodiment of the present invention;

[0050] FIG. 2( a ) is a schematic diagram of a method for analyzing the changing rule of settlement prediction confidence based on static period simulated observation data according to an embodiment of the present invention;

[0051] FIG2( b ) is a schematic diagram of a simulated observation data generation-cutout according to an embodiment of the present invention;

[0052] FIG2( c ) is a schematic diagram of simulated observation data generation — discreteness according to an embodiment of the present invention;

[0053] FIG2( d ) is a schematic diagram of simulated observation data generation—superposition of random errors according to an embodiment of the present invention;

[0054] FIG2( e ) is a schematic diagram of settlement prediction based on settlement observation data provided according to an embodiment of the present invention;

[0055] FIG2( f ) is a schematic diagram of statistical analysis of settlement prediction results provided according to an embodiment of the present invention;

[0056] Figure 3 This is a diagram of the architecture of a confidence analysis system for early prediction of settlement of layered foundation of high-speed railway provided according to an embodiment of the present invention;

[0057] Figure 4 A structural diagram of an electronic device provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0058] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0060] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0061] Embodiment 1

[0062] like Figure 1 As shown in FIG. 2( a ), this embodiment provides a confidence analysis method for early prediction of high-speed railway layered foundation settlement, comprising:

[0063] S1, determine the changing rule of settlement prediction confidence under the condition of preloading layered foundation based on the loading period and static period;

[0064] S2, conducting an early prediction confidence analysis of high-speed railway layered foundation settlement based on the settlement prediction confidence variation rule under the conditions of preloaded layered foundation during the loading period and the static period.

[0065] As a preferred embodiment, the S1 includes:

[0066] S11, taking the loading period and the static period as the research objects, considering the changes of observation conditions based on the settlement theory model, processing the settlement theory curves under different soil properties, stratification and foundation treatment conditions, and generating a large number of simulated settlement observation data under the coupling of multiple factors;

[0067] S12, based on the optimal use conditions of different curve fitting models, adopts a settlement prediction model that can describe the graded loading conditions and the entire settlement time-varying process, and combines intelligent algorithms to quickly find and select the appropriate curve fitting model, combined with the PSO particle swarm algorithm, to predict the simulated settlement observation data;

[0068] S13, statistically analyzing the settlement prediction mean and standard deviation under different observation conditions to form a settlement prediction confidence analysis method;

[0069] S14, reveals the changing rules of confidence in the prediction of high-speed railway soft soil roadbed settlement caused by the randomness of observation conditions when using an appropriate curve fitting model under different soil properties, stratification and foundation treatment conditions.

[0070] As a preferred implementation, the observation conditions include changes in observation duration, frequency and accuracy.

[0071] As a preferred embodiment, the processing includes truncation, discretization and superposition of random error processing.

[0072] As a preferred embodiment, the step S11 generates a large amount of simulated settlement observation data under the coupling of multiple factors, including:

[0073] (1) Obtain theoretical data through the sedimentation theory model;

[0074] (2) The theoretical data is "cut" into an observation period and a late settlement period at a certain observation period; the certain observation period includes a static period and a static period of 3 months, 6 months, 9 months or 12 months; as shown in Figure 2(b).

[0075] (3) The observation period data is discretized according to the observation frequency of the roadbed settlement during the observation stages such as the roadbed filling period and the static period, and the observation frequency is selected according to the "Technical Specification for Railway Engineering Settlement Deformation and Evaluation" (Q / CR9230-2016) and the updated version of the Technical Specification for Railway Settlement Deformation and Evaluation; as shown in Figure 2(c). In this embodiment, the observation frequency of the roadbed settlement during the observation stage of the preloading or roadbed filling completion is shown in the table of the observation frequency of the roadbed settlement during each observation stage in Table 1.

[0076] Table 1

[0077]

[0078] (4) “Superimposing random noise” on the discrete observation period data, as shown in Figure 2(d), includes: superimposing the observation error according to the deformation measurement level and accuracy requirements corresponding to the error in the elevation of the deformation point of different railway structure deformation measurement levels specified in the “Railway Engineering Survey Code” (TB 10101-2018), to generate discrete observation data with random noise;

[0079] The "Railway Engineering Surveying Specification" (TB 10101-2018) stipulates the mean error in the elevation of deformation points for different railway structure deformation measurement levels, as shown in Table 2 Deformation measurement levels and accuracy requirements.

[0080] Table 2

[0081]

[0082] (5) Repeat steps (2) to (4) to obtain the simulated settlement observation data under the coupling of a large number of multi-factors.

[0083] In this embodiment, Figure 2(a)-Figure 2(f) As shown as a preferred implementation, the observation and evaluation of railway subgrade deformation mostly use static load period observation data, and most curve fitting methods have stronger fitting ability for static load period observation data. The present invention uses static load period observation data for research, and the observation frequency is: once / week in the 1st to 3rd month after the completion of the subgrade filling; once / 2 weeks in the 4th to 6th month after the completion of the filling; once / month after 6 months after the completion of the filling. The observation period is 12 months. The accuracy of railway subgrade deformation observation should not be lower than the III level, that is, the absolute value of the error in the elevation of the deformation point is ≤1mm. The superimposed observation error in the calculation and analysis of the present invention is ±1mm. Therefore, the generation of a large number of simulated settlement observation data under the coupling of multiple factors includes:

[0084] (1) Obtain theoretical data through the sedimentation theory model;

[0085] (2) The theoretical data were “truncated” to the observation period and the late settlement period based on a 12-month observation period;

[0086] (3) The observation period data were discretized according to the frequency of once a week in the first to third months after the completion of the roadbed filling; once a week in the fourth to sixth months; and once a month after the sixth month.

[0087] (4) "Superimposing random noise" on the discrete observation period data, including: superimposing the observation error according to the III level, that is, the accuracy of the observation error ±1mm, to generate discrete observation data with random noise;

[0088] (5) Repeat steps (2) to (4) to obtain the simulated settlement observation data under the coupling of a large number of multi-factors.

[0089] As a preferred implementation, the S12 combined with the PSO particle swarm algorithm to carry out curve fitting settlement prediction and calculation on the simulated settlement observation data includes: based on the idea of ​​Monte-Carlo simulation, combined with the PSO particle swarm algorithm, a large number of repeated curve fitting settlement predictions and calculations are carried out using the hyperbola method, exponential curve method, Asaoka method or Hoshino method.

[0090] As a preferred embodiment, the statistical analysis of the settlement prediction mean and standard deviation under different observation conditions in S13 to form a settlement prediction confidence analysis method includes statistically analyzing the settlement prediction results according to formulas (1)-(3).

[0091]

[0092]

[0093] In the formula, R is the correlation coefficient; are the sample means of the two sets of series respectively; is the covariance of the two series; is the variance of the two series; μ is the mean of settlement prediction; σ is the standard deviation of settlement prediction; N s is the number of samples to be calculated; S k The kth predicted value of settlement.

[0094] like Figure 2(e) and 2(f) As shown, as a preferred embodiment, the disclosure of the change rules of confidence in the prediction of high-speed railway stratified foundation settlement caused by the randomness of observation conditions under different soil properties, stratification and foundation treatment conditions, when using an appropriate curve fitting model, includes:

[0095] Taking the loading period and static period as the research objects, a settlement prediction confidence analysis model is constructed;

[0096] Taking the loading period and static period as the research objects, the prediction results of various observation data models are compared and analyzed, and the prediction models are selected to obtain the optimal model;

[0097] Based on the changes in observation conditions, the optimal model is used to determine the impact of changes in the loading period and the static period on the mean and standard deviation of the settlement prediction, and a statistical analysis is performed on the changes in the confidence level of the high-speed railway stratified foundation settlement prediction caused by the randomness of the observation conditions.

[0098] As a preferred implementation, taking the loading period and the static period as the research objects, comparing and analyzing the prediction results of multiple observation data models and selecting the prediction model to obtain the optimal model includes:

[0099] Based on the correlation coefficient, relative error and coefficient of variation, a comprehensive evaluation index is formed, as shown in formula (4):

[0100] ω=0.3×R+0.4×(1-CV)+0.3×(1-10×δ)(4)

[0101] In the formula, ω∈(0,1) is a comprehensive evaluation index; CV∈(0,1) represents the coefficient of variation; R∈(0,1) represents the correlation coefficient; δ∈(0,1) represents the relative error;

[0102] Based on the comprehensive evaluation index, the prediction results of various observation data models are compared and analyzed, and the prediction models are selected to obtain the optimal model.

[0103] In this embodiment, the settlement prediction model should be able to better describe the time-varying characteristics of the settlement curve; for high-speed railway lines, especially for newly built 400km / h high-speed railway ballastless track lines, the subgrade settlement limit after construction is ≤15mm, and the prediction model is required to accurately calculate the final settlement; in the face of observation data containing random noise, the model prediction results should have high stability. The prediction model should have good description capabilities for the time-varying characteristics of the settlement curve, and the results should be accurate and stable.

[0104] The correlation coefficient is a commonly used indicator for evaluating the ability of the prediction model to describe the time-varying characteristics of the settlement curve, and R∈(0,1); the relative error can reflect the accuracy of the result, and in this invention, δ∈(0,1); the coefficient of variation can reflect the discrete degree of the prediction result and can eliminate the numerical differences in the prediction results of different models, and in this invention, CV∈(0,1). Based on the weighted average concept, a comprehensive evaluation index of the applicability of commonly used engineering models with correlation coefficient, relative error, and coefficient of variation as the core is constructed, ω∈(0,1); see formula (4).

[0105] ω=0.3×R+0.4×(1-CV)+0.3×(1-10×δ)(4)

[0106] Wherein, ω is the comprehensive evaluation index; CV is the coefficient of variation.

[0107] As a preferred embodiment, S2 includes:

[0108] S21, choose different prediction models under different conditions;

[0109] S22, changing the value of the observation condition, so as to determine the variation law of the relative error and standard deviation with the observation condition, as the confidence analysis result of early prediction of the settlement of the high-speed railway layered foundation.

[0110] In this embodiment, the process of establishing the settlement theory model includes:

[0111] (1) Basic assumptions for large deformation consolidation theory of ideal vertical drainage body:

[0112] ① The soil is completely saturated, and the soil particles and pore water are incompressible;

[0113] ② The radial and vertical drainage of the soil layer, and the radial and vertical seepage all obey Darcy's law;

[0114] ③ It meets the free strain condition, only considering the vertical compression deformation of the soil, and not considering the lateral deformation of the soil;

[0115] ④ The external load complies with the graded loading conditions;

[0116] ⑤ The top surface of the soil layer is completely permeable, the bottom surface is completely impermeable, well resistance and smear effect are not considered, and the radial outer boundary is impermeable.

[0117] (2) Determine the seepage continuity equation;

[0118] The large deformation consolidation theory needs to consider the relative speed of pore water seepage and soil particle movement when establishing the seepage continuity equation;

[0119] A cylindrical coordinate system (r,θ,ξ) is established, and the soil microelement with small time increment is taken as the research object.

[0120] The vertical outflow is:

[0121]

[0122] Where: e is the porosity ratio of soft soil; V s is the actual vertical movement speed of soil particles; V w1 is the actual vertical flow velocity of pore water.

[0123] The vertical inflow is:

[0124]

[0125] The vertical flow rate change is:

[0126]

[0127] The radial outflow is:

[0128]

[0129] Where: V w2 is the actual radial flow velocity of pore water;

[0130] The radial inflow is:

[0131]

[0132] The radial flow rate change is:

[0133]

[0134] The net flow rate can be obtained from the vertical flow rate change and the radial flow rate change:

[0135]

[0136] The change in volume of the soil microelement within a small time increment is the same as the change in flow within the microelement, so the continuity equation is:

[0137] (3) Establishing the stress-strain relationship of soft soil

[0138] The UH constitutive model can describe the stress-strain relationship of soft soil as shown in the following formula:

[0139]

[0140] Where: dε v is the volumetric strain increment; is the elastic volume strain increment, see the following formula:

[0141]

[0142] is the plastic volume strain increment caused by stress, see the following formula:

[0143]

[0144] Where: σ′ is the effective stress; C c is the compression index; C s is the rebound index; M is the stress ratio of the critical state, is the internal friction angle; M f is the stress ratio of the potential failure state, which can be obtained by the following three formulas.

[0145]

[0146] χ=M 2 / [12·(3-M)];

[0147]

[0148] Where: c ′0 is the initial consolidation stress;

[0149] (4) Establishing Darcy's law seepage equation

[0150] Darcy's law can describe the seepage of water in soil and the permeability coefficient is equal at all depths of the soil layer. The vertical and radial apparent flow velocities are as follows:

[0151]

[0152] Where: K ξ , K r are the vertical and radial permeability coefficients respectively; i ξ 、i r are the vertical and radial hydraulic gradients, respectively.

[0153] (5) Establishment of large deformation consolidation control equation

[0154] Substituting the soft soil stress-strain relationship and the vertical and radial apparent flow velocities into the continuity equation, the large deformation consolidation control equation is obtained as follows:

[0155]

[0156] Where: u is the pore water pressure; γ w The weight of water;

[0157] Based on the one-dimensional large deformation consolidation theory, the relationship between the vertical flow coordinate ξ and the solid phase coordinate z is shown as follows:

[0158]

[0159] The flow coordinate consolidation equation is converted into the solid phase coordinate consolidation equation, and the following equation is obtained:

[0160]

[0161] Where: G s is the specific gravity of soil particles;

[0162] The load form of graded loading conditions can be written as follows:

[0163]

[0164] The initial conditions of the theoretical model are as follows:

[0165] σ′(r,z,0)=σ z ′0+(G s -1)·γ w z;

[0166] Where: z '0 is the stress generated by the initial load.

[0167] The boundary conditions of the theoretical model are as follows:

[0168] σ′(r,0,t)=σ z ′0+p(t)

[0169]

[0170] σ′(r d ,z,t)=σ z ′0+(G s -1)·γ w z+p(t)

[0171]

[0172] The theoretical model is solved using the finite difference method;

[0173] (ii) Optimizing the fitting parameters of the sedimentation theory model to form the sedimentation theory model used in this embodiment.

[0174] In this embodiment, it includes:

[0175] (1) Soil parameter values

[0176] The empirical range of vertical permeability coefficient of soft soil of different origins is: 1×10 -4 ~1×10 -3 m / d. The vertical permeability coefficient is taken as 2.76×10 -4 ~1.36×10 -3 m / d expansion calculation.

[0177] In general, the radial permeability coefficient of soft soil is greater than the vertical permeability coefficient: the anisotropy ratio of Louiseville soil is determined to be 1.35-1.55; the anisotropy ratio of Boston red clay is 1.2-1.9; the anisotropy ratio of remolded kaolin is determined to be 1-1.6. The anisotropy ratio is taken as 1.1-1.9 for calculation.

[0178] (2) The calculation method of critical height of embankment based on Prandtl's solution to Fel lenius is shown in the following formula: γ w ·ρ·H crit =5.52·C u ;

[0179] In the formula, H crit The critical height of the roadbed is calculated to be 9m, which means the final load is 180kPa. Considering the surface sand cushion, the initial load is 20kPa.

[0180] As a preferred embodiment, S2 includes:

[0181] Statistically analyze the settlement prediction mean and standard deviation under different observation conditions to form a settlement prediction confidence analysis method;

[0182] The article reveals the changing rules of confidence in the prediction of high-speed railway layered foundation settlement caused by the randomness of observation conditions when using an appropriate curve fitting model under different soil properties, layering and foundation treatment conditions.

[0183] See also Figure 3 This embodiment provides a high-speed railway layered foundation settlement early prediction confidence analysis system, which is used to implement the confidence analysis method, and is characterized by comprising:

[0184] The settlement prediction confidence variation law determination module 101 is used to determine the settlement prediction confidence variation law under the condition of preloading layered foundation based on the loading period and the static period;

[0185] The high-speed railway layered foundation settlement early prediction confidence analysis module 102 performs high-speed railway layered foundation settlement early prediction confidence analysis based on the settlement prediction confidence variation rule under the conditions of the layered foundation preloaded by the loading period and the static period.

[0186] The present invention also provides a memory storing a plurality of instructions, wherein the instructions are used to implement the method as in the first embodiment.

[0187] like Figure 4 As shown, the present invention also provides an electronic device, including a processor 301 and a memory 302 connected to the processor 301, the memory 302 stores a plurality of instructions, and the instructions can be loaded and executed by the processor so that the processor can execute the method as in the first embodiment.

[0188] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A confidence analysis method for early prediction of high-speed railway layered foundation settlement, characterized in that: include: S1, determine the changing rule of settlement prediction confidence under the condition of preloading layered foundation based on the loading period and static period; S2, conducting an early prediction confidence analysis of high-speed railway layered foundation settlement based on the settlement prediction confidence variation rule under the conditions of preloaded layered foundation during the loading period and the static period.

2. The confidence analysis method for early prediction of high-speed railway layered foundation settlement according to claim 1 is characterized in that: The S1 includes: S11, taking the loading period and the static period as the research objects, considering the changes of observation conditions based on the settlement theory model, processing the settlement theory curves under different soil properties, stratification and foundation treatment conditions, and generating a large number of simulated settlement observation data under the coupling of multiple factors; the observation conditions include changes in observation time, frequency and accuracy; the processing includes interception, discretization and superposition of random error processing; S12, based on the optimal use conditions of different curve fitting models, adopts a settlement prediction model that can describe the graded loading conditions and the entire settlement time-varying process, and combines intelligent algorithms to quickly find and select the appropriate curve fitting model, combined with the PSO particle swarm algorithm, to predict the simulated settlement observation data; S13, statistically analyzing the settlement prediction mean and standard deviation under different observation conditions to form a settlement prediction confidence analysis method; S14, reveals the changing rules of confidence in the prediction of high-speed railway soft soil roadbed settlement caused by the randomness of observation conditions when using an appropriate curve fitting model under different soil properties, stratification and foundation treatment conditions.

3. The confidence analysis method for early prediction of high-speed railway layered foundation settlement according to claim 2 is characterized in that: The S11 generates a large amount of simulated settlement observation data under the coupling of multiple factors, including: (1) Obtain theoretical data through the sedimentation theory model; (2) "cutting" the theoretical data into an observation period and a settlement period at a certain observation period; the certain observation period is a filling period and a static period of 3 months, 6 months, 9 months or 12 months; (3) The observation period data are discretized according to the observation frequency of roadbed settlement during the observation stages such as the roadbed filling period and the static period. The observation frequency is selected in accordance with the "Technical Specification for Railway Engineering Settlement Deformation and Evaluation" (Q / CR9230-2016) and the updated version of the Technical Specification for Railway Settlement Deformation and Evaluation; (4) "Superimposing random noise" on the discrete observation period data, including: superimposing the observation error according to the deformation measurement level and accuracy requirements corresponding to the error in the elevation of the deformation point of different railway structure deformation measurement levels specified in the "Railway Engineering Survey Code" (TB 10101-2018), to generate discrete observation data with random noise; (5) Repeat steps (2) to (4) to obtain the simulated settlement observation data under the coupling of a large number of multi-factors.

4. The confidence analysis method for early prediction of high-speed railway layered foundation settlement according to claim 3 is characterized in that: The statistical analysis of the settlement prediction mean and standard deviation under different observation conditions in S13 to form a settlement prediction confidence analysis method includes statistically analyzing the settlement prediction results according to formulas (1)-(3). In the formula, R is the correlation coefficient; are the sample means of the two sets of series respectively; is the covariance of the two series; is the variance of the two series; μ is the mean of settlement prediction; σ is the standard deviation of settlement prediction; N s is the number of samples to be calculated; S k The kth predicted value of settlement.

5. The confidence analysis method for early prediction of high-speed railway layered foundation settlement according to claim 4 is characterized in that: The disclosure of the change rules of confidence in the prediction of high-speed railway layered foundation settlement caused by the randomness of observation conditions under different soil properties, layering properties and foundation treatment conditions when using an appropriate curve fitting model includes: Taking the loading period and static period as the research objects, a settlement prediction confidence analysis model is constructed; Taking the loading period and static period as the research objects, the prediction results of various observation data models are compared and analyzed, and the prediction models are selected to obtain the optimal model; Based on the changes in observation conditions, the optimal model is used to determine the impact of changes in the loading period and the static period on the mean and standard deviation of the settlement prediction, and a statistical analysis is performed on the changes in the confidence level of the high-speed railway stratified foundation settlement prediction caused by the randomness of the observation conditions.

6. The confidence analysis method for early prediction of high-speed railway layered foundation settlement according to claim 5 is characterized in that: The method of taking the loading period and the static period as the research objects, comparing and analyzing the prediction results of multiple observation data models and selecting the prediction model to obtain the optimal model includes: Based on the correlation coefficient, relative error and coefficient of variation, a comprehensive evaluation index is formed, as shown in formula (4): ω=0.3×R+0.4×(1-CV)+0.3×(1-10×δ)(4) In the formula, ω∈(0,1) is a comprehensive evaluation index; CV∈(0,1) represents the coefficient of variation; R∈(0,1) represents the correlation coefficient; δ∈(0,1) represents the relative error; Based on the comprehensive evaluation index, the prediction results of various observation data models are compared and analyzed, and the prediction models are selected to obtain the optimal model.

7. The confidence analysis method for early prediction of high-speed railway layered foundation settlement according to claim 6 is characterized in that: The S2 includes: S21, choose different prediction models under different conditions; S22, changing the value of the observation condition, so as to determine the variation law of the relative error and standard deviation with the observation condition, as the confidence analysis result of early prediction of the settlement of the high-speed railway layered foundation.

8. A confidence analysis system for early prediction of settlement of layered foundation of high-speed railway, used for implementing the method of any one of claims 1 to 7, characterized in that: include: A settlement prediction confidence change law determination module (101) is used to determine the settlement prediction confidence change law under the condition of preloading layered foundation based on the loading period and the static period; The high-speed railway layered foundation settlement early prediction confidence analysis module (102) performs high-speed railway layered foundation settlement early prediction confidence analysis based on the settlement prediction confidence variation rule under the conditions of the preloading layered foundation during the loading period and the static period.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a plurality of instructions, and the processor is used to read the instructions and execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a plurality of instructions, and the plurality of instructions can be read by a processor to execute the method according to any one of claims 1 to 7.

Images