Analysis Method for the Reliability of Highway Subgrade Slope Collapse Monitoring

Abstract

The present invention discloses a reliability analysis method for monitoring the collapse of highway subgrade slopes, belonging to the technical field of subgrade slopes, and comprising the following steps: S1: collecting and processing the historical data of highway subgrade slopes; S2: determining the optimal prediction model for the collapse of highway subgrade slopes; S3: analyzing and predicting the real-time data of highway subgrade slopes, judging the stability of highway subgrade slopes, and determining the prediction result of the collapse of highway subgrade slopes; S4: analyzing the reliability of the prediction result of the collapse of highway subgrade slopes, and determining the analysis result of the reliability of the monitoring of the collapse of highway subgrade slopes. The present invention solves the problem that the existing method cannot perform a reliability analysis on the monitoring of the collapse of highway subgrade slopes, resulting in poor monitoring effects of slope collapses. The present invention can perform a reliability analysis on the monitoring of the collapse of highway subgrade slopes, can clarify whether the monitoring of the collapse of highway subgrade slopes is accurate, can provide an effective prediction for the monitoring of the collapse of highway subgrade slopes, and can improve the monitoring effect of the collapse of highway subgrade slopes.

Description

Technical Field

[0001] The present invention relates to the technical field of subgrade slopes, and particularly to a reliability analysis method for monitoring the collapse of highway subgrade slopes. Background Art

[0002] With the rapid development of expressways in China, the problem of highway subgrade slope collapse has received increasing attention. Among them, the subgrade slope refers to the inclined plane connecting the two sides of the subgrade cross-section with the ground, which is divided into embankment slopes and cutting slopes and is an important factor affecting the stability of the subgrade. In actual engineering, due to the influence of factors such as geological environment, hydrological conditions, and construction quality, highway subgrade slope collapse accidents occur from time to time, posing a serious threat to road traffic safety.

[0003] Existing technical means fail to timely conduct reliability analysis on the monitoring of highway subgrade slope collapse, fail to timely clarify whether the monitoring of highway subgrade slope collapse is accurate, and fail to timely provide effective prediction for the monitoring of highway subgrade slope collapse, resulting in the frequent occurrence of highway subgrade slope collapse or potential safety hazards. Summary of the Invention

[0004] The purpose of the present invention is to provide a reliability analysis method for monitoring the collapse of highway subgrade slopes, which can conduct reliability analysis on the monitoring of highway subgrade slope collapse, can clarify whether the monitoring of highway subgrade slope collapse is accurate, can provide effective prediction for the monitoring of highway subgrade slope collapse, and can improve the monitoring effect of highway subgrade slope collapse, thus solving the problems raised in the above background art.

[0005] To achieve the above purpose, the present invention provides the following technical solutions:

[0006] A reliability analysis method for monitoring the collapse of highway subgrade slopes, comprising the following steps:

[0007] S1: Collect historical data of highway subgrade slopes based on big data, process the historical data of highway subgrade slopes based on big data, and determine the characteristic data of highway subgrade slopes based on big data;

[0008] S2: According to the reliability analysis requirements for monitoring the collapse of highway subgrade slopes, construct a prediction model for the collapse of highway subgrade slopes based on machine learning, test and optimize the prediction model for the collapse of highway subgrade slopes, and determine the optimal prediction model for the collapse of highway subgrade slopes;

[0009] S3: Collect real-time data of highway subgrade slopes, analyze and predict the real-time data of highway subgrade slopes based on the optimal prediction model for the collapse of highway subgrade slopes, judge the stability of highway subgrade slopes, and determine the prediction result of highway subgrade slope collapse;

[0010] S4: Analyze the reliability of the prediction results of highway subgrade slope collapse, judge the matching degree between the prediction results of highway subgrade slope collapse and the actual results of highway subgrade slope collapse, and determine the analysis results of the reliability of highway subgrade slope collapse monitoring.

[0011] Preferably, in the above S1, collect the historical data of highway subgrade slopes based on big data and perform the following operations:

[0012] Collect the historical design parameters of the highway subgrade slope to obtain the highway subgrade slope design data;

[0013] Collect the historical geological conditions of the highway subgrade slope to obtain the highway subgrade slope geological data;

[0014] Collect the historical hydrological conditions of the highway subgrade slope to obtain the highway subgrade slope hydrological data;

[0015] Collect the historical construction records of the highway subgrade slope to obtain the highway subgrade slope construction data;

[0016] Among them, based on the highway subgrade slope design data, highway subgrade slope geological data, highway subgrade slope hydrological data, and highway subgrade slope construction data, determine the historical data of highway subgrade slopes based on big data.

[0017] Preferably, in the above S1, process the historical data of highway subgrade slopes based on big data and perform the following operations:

[0018] Obtain the historical data of highway subgrade slopes based on big data;

[0019] Clean the historical data of highway subgrade slopes based on big data, including:

[0020] Check the consistency of the historical data of highway subgrade slopes based on big data;

[0021] According to the data consistency requirements, check whether the historical data of highway subgrade slopes based on big data contains inconsistent data that is useless for the analysis of the reliability of highway subgrade slope collapse monitoring, and remove the inconsistent data from the historical data of highway subgrade slopes based on big data;

[0022] Check for invalid values and missing values in the historical data of highway subgrade slopes based on big data;

[0023] According to the data validity and integrity requirements, check whether the historical data of highway subgrade slopes based on big data contains invalid values and missing values that are useless for the analysis of the reliability of highway subgrade slope collapse monitoring, and remove the invalid values and missing values from the historical data of highway subgrade slopes based on big data;

[0024] Determine the historical data of highway subgrade slopes that are useful for the reliability analysis of highway subgrade slope collapse monitoring.

[0025] Preferably, in the step S1, when processing the historical data of highway subgrade slopes based on big data, the following operations are also performed:

[0026] Obtain the historical data of highway subgrade slopes that are useful for the reliability analysis of highway subgrade slope collapse monitoring;

[0027] Convert the historical data of highway subgrade slopes that are useful for the reliability analysis of highway subgrade slope collapse monitoring to unify the formats among the historical data of highway subgrade slopes that are useful for the reliability analysis of highway subgrade slope collapse monitoring;

[0028] Eliminate the dimension differences among the historical data of highway subgrade slopes that are useful for the reliability analysis of highway subgrade slope collapse monitoring, and determine the standardized historical data of highway subgrade slopes;

[0029] Extract features from the standardized historical data of highway subgrade slopes;

[0030] Extract the features that can reflect the reliability analysis of highway subgrade slope collapse monitoring from the standardized historical data of highway subgrade slopes;

[0031] Determine the characteristic data of highway subgrade slopes based on big data.

[0032] Preferably, in the step S2, when constructing a highway subgrade slope collapse prediction model based on machine learning, the following operations are performed:

[0033] Obtain the characteristic data of highway subgrade slopes based on big data;

[0034] Divide the characteristic data of highway subgrade slopes based on big data to determine the training set and the test set;

[0035] According to the requirements of the reliability analysis of highway subgrade slope collapse monitoring, select a machine learning model framework suitable for highway subgrade slope collapse prediction;

[0036] Based on the training set, train the selected machine learning model framework suitable for highway subgrade slope collapse prediction to determine the highway subgrade slope collapse prediction model based on machine learning.

[0037] Preferably, in the step S2, when testing and optimizing the highway subgrade slope collapse prediction model, the following operations are performed:

[0038] Obtain the highway subgrade slope collapse prediction model based on machine learning;

[0039] Based on the test set, perform performance testing on the machine learning-based prediction model for highway subgrade slope collapse to determine whether the machine learning-based prediction model for highway subgrade slope collapse can achieve the expected results;

[0040] Determine the performance test results of the prediction model for highway subgrade slope collapse;

[0041] Based on the performance test results of the prediction model for highway subgrade slope collapse, adjust the parameters and optimize the structure of the machine learning-based prediction model for highway subgrade slope collapse, and perform iterative optimization on the machine learning-based prediction model for highway subgrade slope collapse;

[0042] Determine the optimal prediction model for highway subgrade slope collapse.

[0043] Preferably, in step S3, based on the optimal prediction model for highway subgrade slope collapse, analyze and predict the real-time data of the highway subgrade slope, and perform the following operations:

[0044] Obtain the optimal prediction model for highway subgrade slope collapse, and deploy the optimal prediction model for highway subgrade slope collapse in the actual monitoring work of highway subgrade slope collapse;

[0045] According to the requirements of the reliability analysis of highway subgrade slope collapse monitoring, collect the real-time data of the highway subgrade slope, and input the real-time data of the highway subgrade slope into the optimal prediction model for highway subgrade slope collapse;

[0046] Based on the optimal prediction model for highway subgrade slope collapse, analyze and predict the real-time data of the highway subgrade slope, judge the stability of the highway subgrade slope, and determine the prediction result of highway subgrade slope collapse;

[0047] Among them, the prediction result of highway subgrade slope collapse includes: the highway subgrade slope has good stability, that is, there is no collapse risk for the highway subgrade slope, or the highway subgrade slope has poor stability, that is, there is a collapse risk for the highway subgrade slope.

[0048] Preferably, in step S4, analyze the reliability of the prediction result of highway subgrade slope collapse, and perform the following operations:

[0049] Obtain the prediction result of highway subgrade slope collapse, and analyze the reliability of the prediction result of highway subgrade slope collapse;

[0050] Compare and analyze the prediction result of highway subgrade slope collapse with the actual result of highway subgrade slope collapse, judge the matching degree between the prediction result of highway subgrade slope collapse and the actual result of highway subgrade slope collapse, and determine the reliability analysis result of highway subgrade slope collapse monitoring;

[0051] If the prediction result of the highway subgrade slope collapse matches the actual result of the highway subgrade slope collapse, the analysis result of the reliability of the highway subgrade slope collapse monitoring is that the reliability of the highway subgrade slope collapse monitoring is high;

[0052] If the prediction result of the highway subgrade slope collapse does not match the actual result of the highway subgrade slope collapse, the analysis result of the reliability of the highway subgrade slope collapse monitoring is that the reliability of the highway subgrade slope collapse monitoring is low.

[0053] Preferably, comparing and analyzing the prediction result of the highway subgrade slope collapse with the actual result of the highway subgrade slope collapse includes:

[0054] Obtaining the first analysis data information and the second analysis data information by respectively performing result analysis data acquisition on the prediction result of the highway subgrade slope collapse and the actual result of the highway subgrade slope collapse;

[0055] Calculating the matching degree analysis according to the prediction result of the highway subgrade slope collapse and the actual result of the highway subgrade slope collapse to obtain the first matching value;

[0056] Respectively disassembling and performing data standardization processing on the first analysis data information and the second analysis data information to obtain a plurality of first analysis sub-data and a plurality of second analysis sub-data;

[0057] Corresponding the first analysis sub-data and the second analysis sub-data, and performing matching analysis calculation according to the corresponding result to obtain the second matching value;

[0058] Performing comprehensive analysis on the first matching value and the second matching value to obtain the matching degree between the prediction result of the highway subgrade slope collapse and the actual result of the highway subgrade slope collapse:

[0059] ;

[0060] In the above formula, is the matching degree between the prediction result of the highway subgrade slope collapse and the actual result of the highway subgrade slope collapse, is the first weight, is the parameter, is the th data in the prediction result of the highway subgrade slope collapse, is the th data in the actual result of the highway subgrade slope collapse, is the number of the first analysis sub-data, is the th position data in the th first analysis sub-data, is the position data in the th second analysis sub-data;

[0061] Furthermore, the reliability analysis result of the highway subgrade slope collapse monitoring is determined according to the matching degree combination result between the highway subgrade slope collapse prediction result and the actual result of the highway subgrade slope collapse.

[0062] Preferably, feature extraction is performed on the standardized historical data of the highway subgrade slope, including:

[0063] Change analysis is respectively performed on the standardized historical data of the highway subgrade slope according to the data types, and it is judged whether the standardized historical data of the highway subgrade slope has changed in combination with the time sequence to obtain an analysis and judgment result;

[0064] According to the analysis and judgment result, the standardized historical data of the highway subgrade slope is divided to obtain a first division result and a second division result, where the first division result is the standardized historical data of the highway subgrade slope that has changed, and the second division result is the standardized historical data of the highway subgrade slope that has not changed;

[0065] In the first division result, change feature analysis is performed on the sub-data of the first division result, and the corresponding standardized historical data of the highway subgrade slope is extracted according to the change features to obtain the sub-data features of the first division result;

[0066] In the second division result, statistical analysis is performed on the sub-data of the second division result to obtain a statistical analysis result, and the sub-data features of the second division result are determined according to the statistical analysis result.

[0067] Compared with the prior art, the beneficial effects of the present invention are:

[0068] By collecting the historical data of the highway subgrade slope, processing the historical data of the highway subgrade slope, determining the characteristic data of the highway subgrade slope, constructing a highway subgrade slope collapse prediction model based on machine learning according to the requirements of the reliability analysis of the highway subgrade slope collapse monitoring, testing and optimizing the highway subgrade slope collapse prediction model, determining the optimal highway subgrade slope collapse prediction model, collecting the real-time data of the highway subgrade slope, analyzing and predicting the real-time data of the highway subgrade slope based on the optimal highway subgrade slope collapse prediction model, judging the stability of the highway subgrade slope, determining the highway subgrade slope collapse prediction result, analyzing the reliability of the highway subgrade slope collapse prediction result, judging the matching degree between the highway subgrade slope collapse prediction result and the actual result of the highway subgrade slope collapse, and determining the reliability analysis result of the highway subgrade slope collapse monitoring, the reliability analysis of the highway subgrade slope collapse monitoring can be carried out, it can be clarified whether the highway subgrade slope collapse monitoring is accurate, it can provide an effective prediction for the highway subgrade slope collapse monitoring, and the monitoring effect of the highway subgrade slope collapse can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 It is a flowchart of the reliability analysis method for the collapse monitoring of highway subgrade slopes of the present invention;

[0070] Figure 2 It is an algorithm flowchart for analyzing the reliability of the present invention. Specific implementation manners

[0071] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0072] In order to solve the problems that the existing technology cannot perform reliability analysis on the collapse monitoring of highway subgrade slopes, cannot clarify whether the collapse monitoring of highway subgrade slopes is accurate, and cannot provide effective prediction for the collapse monitoring of highway subgrade slopes, resulting in poor collapse monitoring effect of highway subgrade slopes, please refer to Figure 1 - Figure 2 In this embodiment, the following technical solutions are provided:

[0073] The reliability analysis method for the collapse monitoring of highway subgrade slopes includes the following steps:

[0074] S1: Collect the historical data of highway subgrade slopes based on big data, process the historical data of highway subgrade slopes based on big data, and determine the characteristic data of highway subgrade slopes based on big data;

[0075] In this embodiment, collecting the historical data of highway subgrade slopes based on big data includes:

[0076] Collect the historical design parameters of the highway subgrade slope to obtain the highway subgrade slope design data;

[0077] Specifically, the historical design parameters of the highway subgrade slope include slope gradient, slope shape, and slope protection measures;

[0078] Collect the historical geological conditions of the highway subgrade slope to obtain the highway subgrade slope geological data;

[0079] Specifically, the historical geological conditions of the highway subgrade slope include the unit weight of soil, the internal friction angle, and the cohesion; among them, the unit weight of soil refers to the weight of soil per unit volume, including the weight of soil particles and the weight of water in pores. In the stability analysis of the subgrade slope, the unit weight of soil directly affects the bearing capacity and stability of the slope; the internal friction angle is the angle of frictional resistance between particles inside the soil mass, reflecting the internal friction characteristics of the soil mass. The larger the internal friction angle, the higher the shear strength of the soil mass and the better the stability of the slope; the cohesion is the bonding force between soil particles, reflecting the cohesive characteristics of the soil mass. The larger the cohesion, the better the overall stability of the soil mass;

[0080] Collect the historical hydrological conditions of the highway subgrade slope to obtain the hydrological data of the highway subgrade slope;

[0081] Specifically, the historical hydrological conditions of the highway subgrade slope include the hydraulic conductivity, the permeability coefficient, the water level, the water pressure, the water volume, the thickness of the aquifer, the influence radius of the dewatering level, the bursting coefficient, and the hydraulic gradient;

[0082] Among them, the hydraulic conductivity represents the seepage flow per unit area and is used to evaluate the water conduction ability of the highway subgrade slope; the permeability coefficient represents the seepage flow per unit area and reflects the seepage performance of the highway subgrade slope; the water level represents the elevation of the maximum static water level that the water body can reach under the condition of water balance; the water pressure represents the pressure of the water body and affects the stability of the highway subgrade slope; the water volume represents the total volume of water filling or the volume of water flowing out per unit time and is used to evaluate the change of water volume; the thickness of the aquifer represents the vertical distance from the top plate to the bottom plate of the aquifer and affects the water storage capacity of the highway subgrade slope; the influence radius of the dewatering level represents the radius of the water level lowering funnel centered on the dewatering and pressure reduction well hole and reflects the range of dewatering effect; the bursting coefficient represents the hydrostatic pressure borne by the water-resistant layer per unit area and thickness and is used to evaluate the stability of the water-resistant layer; the hydraulic gradient represents the head difference per unit seepage length in the water flow direction and reflects the driving force of the water flow;

[0083] Collect the historical construction records of the highway subgrade slope to obtain the construction data of the highway subgrade slope;

[0084] Specifically, the historical construction records of the highway subgrade slope include the construction progress records, the records of the use and inspection of construction materials, the records of equipment operation and maintenance, the records of quality inspection and control during the construction process, the records of the implementation of safety measures, and the records of project changes and acceptance;

[0085] Among them, based on the design data of the highway subgrade slope, the geological data of the highway subgrade slope, the hydrological data of the highway subgrade slope, and the construction data of the highway subgrade slope, the historical data of the highway subgrade slope based on big data is determined.

[0086] In this embodiment, processing the historical data of the highway subgrade slope based on big data includes:

[0087] Obtain the historical data of highway subgrade slopes based on big data;

[0088] Clean the historical data of highway subgrade slopes based on big data, including:

[0089] Check the consistency of the historical data of highway subgrade slopes based on big data;

[0090] According to the data consistency requirements, check whether the historical data of highway subgrade slopes based on big data contains inconsistent data that is useless for the reliability analysis of highway subgrade slope collapse monitoring, and remove the inconsistent data from the historical data of highway subgrade slopes based on big data;

[0091] Check for invalid values and missing values in the historical data of highway subgrade slopes based on big data;

[0092] According to the data validity and integrity requirements, check whether the historical data of highway subgrade slopes based on big data contains invalid values and missing values that are useless for the reliability analysis of highway subgrade slope collapse monitoring, and remove the invalid values and missing values from the historical data of highway subgrade slopes based on big data;

[0093] Determine the historical data of highway subgrade slopes that is useful for the reliability analysis of highway subgrade slope collapse monitoring;

[0094] It should be noted that data cleaning refers to the process of processing and organizing the historical data of highway subgrade slopes to improve the quality and usability of the historical data of highway subgrade slopes;

[0095] Among them, consistency checking is to check whether the data meets the requirements according to the reasonable value range and mutual relationship of each parameter, and find data that exceeds the normal range, is logically unreasonable or contradictory; for example, if a variable measured on a 1-7 scale shows a value of 0, or if the weight shows a negative number, it should be regarded as exceeding the normal value range; computer software such as SPSS, SAS, and Excel can automatically identify each variable value that exceeds the defined range; answers with logical inconsistencies may appear in various forms: for example, many respondents say they drive to work but report not having a car; or respondents report being heavy purchasers and users of a certain brand, but at the same time give a very low score on the familiarity scale; when inconsistencies are found, the questionnaire serial number, record serial number, variable name, error category, etc. should be listed for further verification and correction;

[0096] Therefore, by cleaning the historical data of highway subgrade slopes based on big data, inconsistent data, invalid values, and missing values that are useless for the reliability analysis of highway subgrade slope collapse monitoring in the historical data of highway subgrade slopes based on big data can be removed, and the historical data of highway subgrade slopes that are useful for the reliability analysis of highway subgrade slope collapse monitoring can be determined, which can improve the processing accuracy and efficiency of subsequent historical data of highway subgrade slopes and enhance the quality of the historical data of highway subgrade slopes.

[0097] Convert the historical data of highway subgrade slopes that are useful for the reliability analysis of highway subgrade slope collapse monitoring to unify the formats among the historical data of highway subgrade slopes that are useful for the reliability analysis of highway subgrade slope collapse monitoring;

[0098] Eliminate the dimensional differences among the historical data of highway subgrade slopes that are useful for the reliability analysis of highway subgrade slope collapse monitoring and determine the standardized historical data of highway subgrade slopes;

[0099] Extract features from the standardized historical data of highway subgrade slopes;

[0100] Extract the features from the standardized historical data of highway subgrade slopes that can reflect the reliability analysis of highway subgrade slope collapse monitoring;

[0101] Determine the feature data of highway subgrade slopes based on big data.

[0102] It should be noted that by converting and extracting features from the historical data of highway subgrade slopes that are useful for the reliability analysis of highway subgrade slope collapse monitoring, the feature data of highway subgrade slopes based on big data can be determined, which is convenient for subsequent construction of a highway subgrade slope collapse prediction model based on machine learning and provides data support for the construction of a highway subgrade slope collapse prediction model based on machine learning.

[0103] S2: According to the requirements of the reliability analysis of highway subgrade slope collapse monitoring, construct a highway subgrade slope collapse prediction model based on machine learning, test and optimize the highway subgrade slope collapse prediction model, and determine the optimal highway subgrade slope collapse prediction model;

[0104] In this embodiment, constructing a highway subgrade slope collapse prediction model based on machine learning includes:

[0105] Obtain the feature data of highway subgrade slopes based on big data;

[0106] Divide the feature data of highway subgrade slopes based on big data to determine the training set and the test set;

[0107] According to the requirements of the reliability analysis of highway subgrade slope collapse monitoring, select a machine learning model framework suitable for highway subgrade slope collapse prediction;

[0108] Based on the training set, train the selected machine learning model framework applicable to the prediction of highway subgrade slope collapse to determine a machine learning-based highway subgrade slope collapse prediction model.

[0109] In this embodiment, testing and optimizing the highway subgrade slope collapse prediction model includes:

[0110] Obtain a machine learning-based highway subgrade slope collapse prediction model;

[0111] Based on the test set, conduct a performance test on the machine learning-based highway subgrade slope collapse prediction model to determine whether the machine learning-based highway subgrade slope collapse prediction model can achieve the expected effect;

[0112] Determine the performance test results of the highway subgrade slope collapse prediction model;

[0113] Based on the performance test results of the highway subgrade slope collapse prediction model, adjust the parameters and optimize the structure of the machine learning-based highway subgrade slope collapse prediction model, and perform iterative optimization on the machine learning-based highway subgrade slope collapse prediction model;

[0114] Determine the optimal highway subgrade slope collapse prediction model.

[0115] S3: Collect real-time data of the highway subgrade slope, analyze and predict the real-time data of the highway subgrade slope based on the optimal highway subgrade slope collapse prediction model, judge the stability of the highway subgrade slope, and determine the highway subgrade slope collapse prediction result;

[0116] In this embodiment, analyzing and predicting the real-time data of the highway subgrade slope based on the optimal highway subgrade slope collapse prediction model includes:

[0117] Obtain the optimal highway subgrade slope collapse prediction model and deploy the optimal highway subgrade slope collapse prediction model in the actual highway subgrade slope collapse monitoring work;

[0118] According to the requirements of the reliability analysis of highway subgrade slope collapse monitoring, collect real-time data of the highway subgrade slope and input the real-time data of the highway subgrade slope into the optimal highway subgrade slope collapse prediction model;

[0119] Based on the optimal highway subgrade slope collapse prediction model, analyze and predict the real-time data of the highway subgrade slope, judge the stability of the highway subgrade slope, and determine the highway subgrade slope collapse prediction result;

[0120] Among them, the highway subgrade slope collapse prediction result includes: the highway subgrade slope has good stability, that is, there is no collapse risk for the highway subgrade slope, or the highway subgrade slope has poor stability, that is, there is a collapse risk for the highway subgrade slope.

[0121] S4: Analyze the reliability of the prediction result of the highway subgrade slope collapse, judge the matching degree between the prediction result of the highway subgrade slope collapse and the actual result of the highway subgrade slope collapse, and determine the analysis result of the reliability of the highway subgrade slope collapse monitoring.

[0122] In this embodiment, analyzing the reliability of the prediction result of the highway subgrade slope collapse includes:

[0123] Obtain the prediction result of the highway subgrade slope collapse and analyze the reliability of the prediction result of the highway subgrade slope collapse;

[0124] Compare and analyze the prediction result of the highway subgrade slope collapse with the actual result of the highway subgrade slope collapse, judge the matching degree between the prediction result of the highway subgrade slope collapse and the actual result of the highway subgrade slope collapse, and determine the analysis result of the reliability of the highway subgrade slope collapse monitoring;

[0125] Wherein, if the prediction result of the highway subgrade slope collapse matches the actual result of the highway subgrade slope collapse, the analysis result of the reliability of the highway subgrade slope collapse monitoring is that the reliability of the highway subgrade slope collapse monitoring is high;

[0126] Wherein, if the prediction result of the highway subgrade slope collapse does not match the actual result of the highway subgrade slope collapse, the analysis result of the reliability of the highway subgrade slope collapse monitoring is that the reliability of the highway subgrade slope collapse monitoring is low.

[0127] Specifically, analyze the reliability of the prediction result of the highway subgrade slope collapse. Among them, the analysis result of the reliability of the highway subgrade slope collapse monitoring is shown in Table 1:

[0128] Table 1: Analysis result of the reliability of the highway subgrade slope collapse monitoring

[0129] Based on the actual results of highway subgrade slope collapse, analyze the reliability of the prediction results of highway subgrade slope collapse Analysis results of the reliability of highway subgrade slope collapse monitoring The prediction results of highway subgrade slope collapse match the actual results of highway subgrade slope collapse The reliability of highway subgrade slope collapse monitoring is high The prediction results of highway subgrade slope collapse do not match the actual results of highway subgrade slope collapse The reliability of highway subgrade slope collapse monitoring is low ;

[0130] Therefore, based on the actual result of the highway subgrade slope collapse, analyze the reliability of the prediction result of the highway subgrade slope collapse, determine the analysis result of the reliability of the highway subgrade slope collapse monitoring, can conduct a reliability analysis on the highway subgrade slope collapse monitoring, can clarify whether the highway subgrade slope collapse monitoring is accurate. When the highway subgrade slope collapse monitoring is accurate, it can provide an effective prediction for the highway subgrade slope collapse monitoring, facilitate taking protective measures in time to avoid the occurrence of highway subgrade slope collapse accidents, and can improve the monitoring effect of the highway subgrade slope collapse.

[0131] In this embodiment, comparing and analyzing the prediction result of the highway subgrade slope collapse with the actual result of the highway subgrade slope collapse includes:

[0132] Data acquisition for result analysis is carried out separately for the prediction results of highway subgrade slope collapse and the actual results of highway subgrade slope collapse, obtaining the first analysis data information and the second analysis data information;

[0133] According to the prediction results of highway subgrade slope collapse and the actual results of highway subgrade slope collapse, a matching degree analysis and calculation is carried out to obtain the first matching value;

[0134] Separate disassembly and data standardization processing are carried out for the first analysis data information and the second analysis data information, obtaining multiple first analysis sub-data and multiple second analysis sub-data;

[0135] The first analysis sub-data and the second analysis sub-data are corresponded, and a matching analysis and calculation is carried out according to the corresponding results to obtain the second matching value;

[0136] A comprehensive analysis is carried out for the first matching value and the second matching value to obtain the matching degree between the prediction results of highway subgrade slope collapse and the actual results of highway subgrade slope collapse:

[0137] ;

[0138] In the above formula, is the matching degree between the prediction results of highway subgrade slope collapse and the actual results of highway subgrade slope collapse, is the first weight, is a parameter, is the th data in the prediction results of highway subgrade slope collapse, is the th data in the actual results of highway subgrade slope collapse, is the number of the first analysis sub-data, is the th position data in the th first analysis sub-data, is the position data in the

[0139] th second analysis sub-data;

[0140] The above technical solution obtains the result analysis data by separately analyzing the prediction results and the actual results of the highway subgrade slope collapse, so that when conducting the comparative analysis, it not only analyzes the matching degree based on the prediction results and the actual results of the highway subgrade slope collapse, but also combines the analysis data of the prediction results and the analysis data of the actual results of the highway subgrade slope collapse to analyze the matching degree, improving the comprehensiveness of the matching degree between the prediction results and the actual results of the highway subgrade slope collapse, making the matching degree between the prediction results and the actual results of the highway subgrade slope collapse more accurate, and further improving the accuracy of the analysis result of the monitoring reliability of the highway subgrade slope collapse.

[0141] In this embodiment, feature extraction is performed on the standardized historical data of the highway subgrade slope, including:

[0142] Perform change analysis on the standardized historical data of the highway subgrade slope according to the data types respectively, and combine the time sequence to judge whether the standardized historical data of the highway subgrade slope has changed to obtain the analysis and judgment result;

[0143] Divide the standardized historical data of the highway subgrade slope according to the analysis and judgment result to obtain the first division result and the second division result, where the first division result is the standardized historical data of the highway subgrade slope that has changed, and the second division result is the standardized historical data of the highway subgrade slope that has not changed;

[0144] In the first division result, perform change feature analysis on the sub-data of the first division result, and extract the corresponding standardized historical data of the highway subgrade slope according to the change feature to obtain the sub-data feature of the first division result;

[0145] In the second division result, perform statistical analysis on the sub-data of the second division result, obtain the statistical analysis result, and determine the sub-data feature of the second division result according to the statistical analysis result.

[0146] The above technical solution divides the standardized historical data of the highway subgrade slope, so that different feature extraction methods are used to determine the data features according to the features of the standardized historical data of the highway subgrade slope when performing feature extraction, which can not only improve the efficiency of feature extraction, but also ensure the accuracy of the data features, and further provide guarantee for analysis and prediction, avoid deviation effects on the prediction results of highway subgrade slope collapse, and ensure the analysis of the monitoring reliability of highway subgrade slope collapse.

[0147] It should be noted that in this text, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0148] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A reliability analysis method for monitoring highway embankment slope collapse, characterized in that: The steps include: S1: Collecting historical data of highway roadbed slopes based on big data, processing the historical data of highway roadbed slopes based on big data, and determining characteristic data of highway roadbed slopes based on big data; S2: According to the reliability analysis requirements of highway embankment slope collapse monitoring, a highway embankment slope collapse prediction model based on machine learning is constructed, and the highway embankment slope collapse prediction model is tested and optimized to determine the optimal highway embankment slope collapse prediction model; S3: Collect the real-time data of highway embankment slopes, analyze and predict the real-time data of highway embankment slopes based on the optimal highway embankment slope collapse prediction model, judge the stability of highway embankment slopes, and determine the prediction results of highway embankment slope collapse; S4: Analyze the reliability of the highway embankment slope collapse prediction results, determine the matching degree between the highway embankment slope collapse prediction results and the actual results of the highway embankment slope collapse, and determine the reliability analysis results of the highway embankment slope collapse monitoring; Compare and analyze the predicted results of highway embankment slope collapse with the actual results of highway embankment slope collapse, including: Obtaining result analysis data for the predicted result of highway embankment slope collapse and the actual result of highway embankment slope collapse respectively, to obtain first analysis data information and second analysis data information; A first matching value is obtained by performing matching degree analysis and calculation based on the predicted result of highway embankment slope collapse and the actual result of highway embankment slope collapse; Performing disassembly and data standardization processing on the first analysis data information and the second analysis data information respectively to obtain a plurality of first analysis sub-data and a plurality of second analysis sub-data; Corresponding the first analysis sub-data and the second analysis sub-data, and performing matching analysis calculation according to the corresponding result to obtain a second matching value; Comprehensively analyzing the first matching value and the second matching value, obtaining the matching degree between the predicted result of highway embankment slope collapse and the actual result of highway embankment slope collapse; In the above formula, W is the matching degree between the predicted result of highway embankment slope collapse and the actual result of highway embankment slope collapse, u is the first weight, m ​​is the parameter, A i is the i-th data in the prediction result of highway embankment slope collapse, B i is the ith data in the actual result of highway embankment slope collapse, n is the number of the first analysis sub-data, x k (p) is the p position data in the kth first analysis sub-data, y k (p) is the p position data in the k-th second analysis sub-data; Then, the reliability analysis results of highway roadbed slope collapse monitoring are determined according to the matching degree between the predicted results of highway roadbed slope collapse and the actual results of highway roadbed slope collapse and the result determination standard.

2. The highway embankment slope collapse monitoring reliability analysis method according to claim 1 is characterized in that: In S1, historical data of highway embankment slopes based on big data is collected, and the following operations are performed: Collect historical design parameters of highway embankment slopes to obtain highway embankment slope design data; Collect historical geological conditions of highway embankment slopes and obtain geological data of highway embankment slopes; Collect historical hydrological conditions of highway embankment slopes to obtain hydrological data of highway embankment slopes; Collect historical construction records of highway embankment slopes to obtain highway embankment slope construction data; Among them, based on the highway roadbed slope design data, highway roadbed slope geological data, highway roadbed slope hydrological data and highway roadbed slope construction data, the highway roadbed slope historical data based on big data is determined.

3. The highway embankment slope collapse monitoring reliability analysis method according to claim 2 is characterized in that: In S1, the historical data of highway embankment slope based on big data is processed, and the following operations are performed: Obtain historical data of highway embankment slopes based on big data; Clean the historical data of highway embankment slope based on big data, including: Conduct consistency check on historical data of highway embankment slopes based on big data; According to the data consistency requirements, check whether the historical data of highway embankment slopes based on big data contain inconsistent data that is useless for the reliability analysis of highway embankment slope collapse monitoring, and remove the inconsistent data in the historical data of highway embankment slopes based on big data; Check invalid and missing values ​​of historical data of highway embankment slopes based on big data; According to the requirements of data validity and integrity, check whether the historical data of highway embankment slopes based on big data contain invalid values ​​and missing values ​​that are useless for the reliability analysis of highway embankment slope collapse monitoring, and remove the invalid values ​​and missing values ​​in the historical data of highway embankment slopes based on big data; Determine the historical data of highway embankment slopes that are useful for reliability analysis of highway embankment slope collapse monitoring.

4. The highway embankment slope collapse monitoring reliability analysis method according to claim 3 is characterized in that: In S1, the historical data of highway embankment slope based on big data is processed, and the following operations are performed: Obtain historical data of highway embankment slopes that are useful for reliability analysis of highway embankment slope collapse monitoring; The historical data of highway embankment slopes useful for reliability analysis of highway embankment slope collapse monitoring are converted to unify the formats of the historical data of highway embankment slopes useful for reliability analysis of highway embankment slope collapse monitoring; Eliminate the dimensional differences between the historical data of highway embankment slopes that are useful for reliability analysis of highway embankment slope collapse monitoring, and determine standardized historical data of highway embankment slopes; Extract features from standardized historical data of highway embankment slopes; Extract features that can reflect the reliability analysis of highway embankment slope collapse monitoring from standardized highway embankment slope historical data; Determine highway embankment slope characteristic data based on big data.

5. The highway embankment slope collapse monitoring reliability analysis method according to claim 4 is characterized in that: In S2, a highway embankment slope collapse prediction model based on machine learning is constructed, and the following operations are performed: Obtain highway embankment slope characteristic data based on big data; Divide the highway embankment slope characteristic data based on big data and determine the training set and test set; According to the reliability analysis requirements of highway embankment slope collapse monitoring, a machine learning model framework suitable for highway embankment slope collapse prediction is selected; Based on the training set, the selected machine learning model framework suitable for highway roadbed slope collapse prediction is trained to determine the highway roadbed slope collapse prediction model based on machine learning.

6. The highway embankment slope collapse monitoring reliability analysis method according to claim 5 is characterized in that: In S2, the highway embankment slope collapse prediction model is tested and optimized, and the following operations are performed: Obtain a highway embankment slope collapse prediction model based on machine learning; Based on the test set, the performance of the highway roadbed slope collapse prediction model based on machine learning is tested to determine whether the highway roadbed slope collapse prediction model based on machine learning can achieve the expected effect; Determine the performance test results of the highway embankment slope collapse prediction model; Based on the performance test results of the highway embankment slope collapse prediction model, the parameters of the highway embankment slope collapse prediction model based on machine learning are adjusted and the structure is optimized, and the highway embankment slope collapse prediction model based on machine learning is repeatedly optimized; Determine the optimal highway embankment slope collapse prediction model.

7. The highway embankment slope collapse monitoring reliability analysis method according to claim 6 is characterized in that: In S3, the real-time data of the highway embankment slope is analyzed and predicted based on the optimal highway embankment slope collapse prediction model, and the following operations are performed: Obtain the optimal highway embankment slope collapse prediction model and deploy it in actual highway embankment slope collapse monitoring work; According to the reliability analysis requirements of highway embankment slope collapse monitoring, real-time data of highway embankment slopes are collected and input into the optimal highway embankment slope collapse prediction model; Based on the optimal highway embankment slope collapse prediction model, the real-time data of highway embankment slope is analyzed and predicted to determine the stability of highway embankment slope and determine the prediction result of highway embankment slope collapse; Among them, the prediction results of highway roadbed slope collapse include: the stability of the highway roadbed slope is good, that is, the highway roadbed slope has no risk of collapse or the stability of the highway roadbed slope is poor, that is, the highway roadbed slope has a risk of collapse.

8. The highway embankment slope collapse monitoring reliability analysis method according to claim 7 is characterized in that: In S4, the reliability of the prediction result of highway embankment slope collapse is analyzed, and the following operations are performed: Obtain the prediction results of highway embankment slope collapse and analyze the reliability of the prediction results of highway embankment slope collapse; Compare and analyze the predicted results of highway embankment slope collapse with the actual results of highway embankment slope collapse, judge the matching degree between the predicted results of highway embankment slope collapse and the actual results of highway embankment slope collapse, and determine the reliability analysis results of highway embankment slope collapse monitoring; Among them, the prediction result of highway embankment slope collapse matches the actual result of highway embankment slope collapse, and the reliability analysis result of highway embankment slope collapse monitoring is that the reliability of highway embankment slope collapse monitoring is high; Among them, if the prediction result of highway roadbed slope collapse does not match the actual result of highway roadbed slope collapse, the reliability analysis result of highway roadbed slope collapse monitoring is that the reliability of highway roadbed slope collapse monitoring is low.

9. The highway embankment slope collapse monitoring reliability analysis method according to claim 4 is characterized in that: Feature extraction of standardized highway embankment slope historical data, including: The standardized highway embankment slope historical data are analyzed for changes according to the data types, and the standardized highway embankment slope historical data are judged in time sequence whether they have changed, and the analysis and judgment results are obtained; According to the analysis and judgment results, the standardized historical data of the roadbed slope are divided to obtain a first division result and a second division result, wherein the first division result is the standardized historical data of the roadbed slope that has changed, and the second division result is the standardized historical data of the roadbed slope that has not changed; In the first division result, the change characteristics of the first division result sub-data are analyzed, and the corresponding standardized highway roadbed slope historical data are extracted according to the change characteristics to obtain the first division result sub-data characteristics; In the second division result, statistical analysis is performed on the second division result sub-data to obtain the statistical analysis result, and the second division result sub-data features are determined according to the statistical analysis result.

Images