Multi-source data fusion and analysis method for monitoring frost heaving of high-speed railway roadbed in alpine region

Abstract

The invention relates to the technical field of high-speed railway subgrade frost heaving monitoring in alpine regions, in particular to a high-speed railway subgrade frost heaving monitoring multi-source data fusion and analysis method in alpine regions, which comprises the following steps: data source selection and data cleaning; standardizing and converting a data format; fusing the data; controlling and evaluating data quality; and comprehensively analyzing frost heaving deformation. According to the high-cold-region high-speed railway roadbed frost heaving monitoring multi-source data fusion and analysis method, a multi-source data fusion technology is applied, fusion and analysis of a large amount of frost heaving monitoring data are achieved, more accurate, complete and reliable roadbed frost heaving information is obtained, and scientific basis and technical support are provided for design, construction and operation of high-cold-region high-speed railways.

Description

Technical Field

[0001] This invention relates to the field of frost heave monitoring technology for high-speed railway subgrade in cold regions, specifically to a method for multi-source data fusion and analysis for frost heave monitoring of high-speed railway subgrade in cold regions. Background Technology

[0002] In the construction of high-speed railways, all countries in the world face the challenge of ensuring high-speed, safe, and smooth train operation while reducing track maintenance. my country's high-speed railway construction standards are strict, and the construction quality is directly related to the safety of train operation in the later stages. With the gradual coverage of the "four vertical and four horizontal" high-speed railway network, the problem of frost heave of high-speed railway subgrades in large areas of seasonally frozen soil in Northeast, Northwest, and North my country has become increasingly prominent, becoming a global problem restricting the safe operation of high-speed railways.

[0003] In high-altitude permafrost regions, roadbed frost heave deformation is one of the main factors affecting line smoothness and traffic safety. As a special soil-water system containing ice crystals, high-altitude permafrost has a seasonal active layer that freezes in winter and thaws in summer. Under the combined action of high-speed impact loads and seasonal freeze-thaw processes, roadbed frost heave deformation becomes a key issue affecting the stability and traffic safety of high-speed railway lines. Therefore, real-time monitoring and dynamic analysis of roadbed frost heave deformation has become an important measure to ensure the safe operation of high-speed railways.

[0004] Currently, the main approach to monitoring frost heave in high-speed railway subgrades in cold regions is to conduct real-time monitoring and dynamic analysis of representative subgrade sections. The monitoring results serve as an objective basis for early warning and forecasting of high-speed railway safety and in-depth research on frost heave issues. However, frost heave monitoring of high-speed railways in cold regions of northern my country faces numerous challenges: a wide range of monitoring items, diverse data sources, a massive amount of frost heave monitoring data, inconsistent data formats, significant influence from external environmental factors, long and frequent observation periods, and the potential generation of abnormal data during the observation process. These issues mean that if the frost heave monitoring data cannot be integrated and analyzed in a timely and accurate manner, the monitoring work will lose its engineering significance.

[0005] Furthermore, due to the diverse sources and formats of monitoring data, and the significant influence of external environmental conditions, the frost heave deformation of roadbeds is generally large and its variation patterns are complex, making data analysis exceptionally difficult. During the data analysis process, it is not only necessary to compare roadbed frost heave monitoring data from different sources, formats, and periods to analyze the changing trends, absolute frost heave amounts, differential frost heave amounts, and frost heave deformation rates, but also to comprehensively consider various influencing factors, such as roadbed fill material, moisture content, temperature, train speed, and foundation treatment methods. This further increases the complexity and challenge of data analysis.

[0006] In conclusion, developing a multi-source data fusion and analysis method for monitoring frost heave of high-speed railway subgrade in cold regions is of great significance for accurately obtaining subgrade frost heave information and ensuring the safe operation of high-speed railways. Summary of the Invention

[0007] To address the aforementioned issues, this invention provides a method for multi-source data fusion and analysis of frost heave monitoring data for high-speed railway subgrades in cold regions. This method utilizes multi-source data fusion technology to integrate and analyze a large amount of frost heave monitoring data, thereby obtaining more accurate, complete, and reliable subgrade frost heave information. This provides a scientific basis and technical support for the design, construction, and operation of high-speed railways in cold regions.

[0008] The technical solution of the present invention is as follows:

[0009] A method for multi-source data fusion and analysis of frost heave monitoring for high-speed railway subgrade in high-altitude and cold regions includes the following steps:

[0010] S1: Data source selection and data cleaning. Select frost heave monitoring data from multiple sources and clean the data to remove errors, duplicates and invalid data.

[0011] S2: Data format standardization and conversion, standardizing and normalizing the cleaned data to unify the data format and units;

[0012] S3: Data fusion, which uses data fusion tools and technologies to merge standardized multi-source data to generate a fused dataset;

[0013] S4: Data quality control and evaluation, which involves quality control and evaluation of the merged dataset;

[0014] S5: Comprehensive analysis of frost heave deformation. Based on the fusion dataset after quality control, it comprehensively analyzes multiple influencing factors to achieve the analysis, prediction and determination of frost heave deformation of roadbed.

[0015] Furthermore, in step S1, the frost heave monitoring data from multiple sources include field monitoring data and non-contact monitoring data; field monitoring data includes manual observation data, automated monitoring data, ground-penetrating radar data, and data from the railway bureau's dynamic inspection vehicle; non-contact monitoring data includes laboratory experimental data, remote sensing data, and geographic information system data.

[0016] Furthermore, the data cleaning described in step S1 includes the following sub-steps: defining the cleaning objective, handling missing data, handling abnormal data, handling duplicate data, and verifying data quality.

[0017] Furthermore, the standardization and normalization process in step S2 includes at least one of data type conversion, data format conversion, data encoding conversion, data value normalization, data range normalization, and data distribution normalization.

[0018] Furthermore, the data fusion tools and techniques in step S3 include at least one of ETL tools, data warehouses, feature layer fusion techniques, data mining, and machine learning.

[0019] Furthermore, the quality control and assessment in step S4 includes an assessment of the data integrity, accuracy, and consistency.

[0020] Furthermore, in step S5, various influencing factors include roadbed fill material, moisture content, temperature, train speed, and foundation treatment method.

[0021] Optionally, the frost heave deformation analysis in step S5 includes establishing a mathematical model to predict and simulate the frost heave deformation of the roadbed.

[0022] Optionally, the frost heave deformation analysis in step S5 includes obtaining surface information using remote sensing technology and processing and analyzing it in conjunction with geographic information system technology.

[0023] Optionally, the frost heave deformation analysis in step S5 includes analyzing the trend of frost heave in the roadbed, the absolute frost heave amount, the differential frost heave amount, and the frost heave deformation rate.

[0024] The beneficial effects of this invention are as follows:

[0025] 1. This invention discloses a method for fusing and analyzing multi-source data on frost heave monitoring of high-speed railway subgrade in high-altitude and cold regions. This method selects and cleans data sources, eliminating erroneous, duplicate, and invalid data to ensure the accuracy and reliability of data entering subsequent processing, laying a solid foundation for subsequent analysis. The cleaned data undergoes standardization and normalization, unifying data formats and units to ensure consistency and comparability across different sources, further improving data quality. Various data fusion tools and technologies are employed, such as ETL tools, data warehouses, feature layer fusion technology, data mining, and machine learning, to effectively fuse the standardized multi-source data, generating a fused dataset. This dataset integrates data from various sources, including on-site monitoring (manual observation, automated monitoring, ground-penetrating radar, and railway bureau dynamic inspection vehicle data) and non-contact monitoring (laboratory experiments, remote sensing, and geographic information system data), yielding more comprehensive, accurate, and valuable information.

[0026] 2. This invention discloses a method for multi-source data fusion and analysis of high-speed railway subgrade frost heave monitoring in high-altitude and cold regions. This method performs quality control and evaluation on the fused dataset, ensuring data integrity, accuracy, and consistency. It utilizes data quality control technology to reduce errors and deviations during the data fusion process, ensuring the reliability of the final data used for analysis and providing a guarantee for accurate analysis of subgrade frost heave deformation.

[0027] 3. This invention discloses a multi-source data fusion and analysis method for monitoring frost heave of high-speed railway subgrade in high-altitude and cold regions. This method comprehensively analyzes various influencing factors such as subgrade fill material, moisture content, temperature, train speed, and foundation treatment methods. It can more comprehensively and objectively consider the effects of various factors on subgrade frost heave deformation, making the judgment of frost heave phenomena more reasonable and accurate. On the one hand, it predicts and simulates subgrade frost heave deformation by establishing a mathematical model, providing a scientific prediction of the development trend of frost heave deformation. On the other hand, it uses remote sensing technology to obtain surface information and combines it with geographic information system technology for processing and analysis, providing more comprehensive information support for analysis. At the same time, it can also analyze the changing trend of subgrade frost heave, absolute frost heave amount, differential frost heave amount, and frost heave deformation rate, etc., to comprehensively and deeply understand the law of subgrade frost heave deformation.

[0028] 4. This invention discloses a multi-source data fusion and analysis method for monitoring frost heave of high-speed railway subgrade in cold regions. This method is simple and convenient to use, and has strong applicability. It can effectively solve the problems existing in the monitoring of frost heave of high-speed railway subgrade in cold regions, such as many monitoring items, wide range of data sources, large data volume, inconsistent formats, susceptibility to external environmental influences, and the existence of abnormal data. It provides accurate, complete and reliable subgrade frost heave information reference and suggestions for the design, construction and operation of high-speed railways in cold regions, ensuring the stability of high-speed railway lines and traffic safety. Attached Figure Description

[0029] Figure 1 This is a flowchart of a method for multi-source data fusion and analysis for monitoring frost heave of high-speed railway subgrade in cold regions, according to an embodiment of the present invention. Detailed Implementation

[0030] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of this application. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, certain operations related to this application are not shown or described in the specification. This is to avoid obscuring the core parts of this application with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0031] Furthermore, the features, operations, or characteristics described in the specification can be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can be rearranged or adjusted in a manner obvious to those skilled in the art. Therefore, the various orders in the specification and drawings are only for the clear description of a particular embodiment and do not imply a necessary order, unless otherwise stated that a particular order must be followed.

[0032] like Figure 1 As shown, the method for multi-source data fusion and analysis of high-speed railway subgrade frost heave monitoring in high-altitude and cold regions includes the following steps:

[0033] S1. Data Source Selection and Data Cleaning

[0034] Choose appropriate data sources for frost heave monitoring, such as field monitoring data (manual observation, automated monitoring, and data from railway bureau's dynamic inspection vehicles), laboratory test data, remote sensing data, geographic information system data, and ground-penetrating radar data. Consider factors such as data quality, coverage, and monitoring frequency to ensure the scientific validity and rationality of data fusion.

[0035] The first step is to clean the frost heave monitoring data from different data sources. This involves correcting or deleting erroneous, duplicate, and invalid information from the multi-source data to ensure the accuracy and consistency of the data and improve the reliability and usability of data fusion. This includes removing, filling in, and filtering duplicate data.

[0036] The basic steps of data cleaning are: define the cleaning goal, handle missing data, handle abnormal data, handle duplicate data, verify data quality, record and report data processing results, and save the cleaned data.

[0037] In one specific embodiment, when selecting a data source, the comprehensiveness, accuracy, real-time nature, and availability of the data must be comprehensively considered. For frost heave monitoring of high-speed railway subgrade in high-altitude and cold regions, the data sources mainly include two categories: on-site monitoring data and non-contact monitoring data.

[0038] The on-site monitoring data includes manual observation data, automated monitoring data, ground-penetrating radar data, and data from the railway bureau's dynamic inspection vehicle.

[0039] Manual observation data is obtained through regular manual inspections and measurements of the roadbed frost heave, including intuitive indicators such as frost heave height and crack width. Although this data is highly subjective, it can provide intuitive information about the roadbed condition.

[0040] Automated monitoring data are parameters related to roadbed frost heave that are collected in real time using automated sensors (such as displacement sensors, temperature sensors, humidity sensors, etc.). These data have high precision and real-time performance, and can continuously reflect the dynamic changes of roadbed frost heave.

[0041] Ground-penetrating radar (GPR) data is obtained by using GPR to detect the internal structure of the roadbed and identify information such as the distribution, thickness, and water content of the permafrost layer, providing in-depth geological evidence for analyzing roadbed frost heave.

[0042] The data from the railway bureau's dynamic inspection vehicle is obtained by using a dedicated dynamic inspection vehicle to dynamically inspect the railway line, acquiring data such as track geometry and wheel-rail interaction forces, which indirectly reflects the impact of roadbed frost heave on track smoothness.

[0043] Non-contact monitoring data includes laboratory experimental data, remote sensing data, and geographic information system data.

[0044] Laboratory experimental data are obtained by conducting frost heave tests on roadbed materials in a laboratory environment that simulates high-altitude and cold conditions, in order to obtain the frost heave characteristics and patterns of the materials under different conditions.

[0045] Remote sensing data is obtained by using satellite remote sensing or UAV remote sensing technology to acquire information such as surface temperature, vegetation cover, and soil moisture of the roadbed and surrounding areas, providing macroscopic background data for the analysis of roadbed frost heave.

[0046] Geographic Information System (GIS) data combines GIS technology to perform spatial analysis and visualization of roadbed frost heave monitoring data, helping to understand the spatial distribution and evolution of frost heave phenomena.

[0047] Data cleaning is a crucial step in ensuring data quality, and it mainly includes the following sub-steps:

[0048] Define the cleaning objectives: Based on the analysis requirements, determine the types and indicators that need to be cleaned, such as focusing only on data directly related to roadbed frost heave;

[0049] Handling missing data: For missing data points, interpolation, average imputation, or model-based prediction can be used to impute them, thereby reducing the impact of missing data on the analysis results.

[0050] Handling outlier data: Identify and remove or correct outliers in the data, such as identifying outliers through statistical methods (e.g., standard deviation method, quartile method), and decide whether to remove or correct these outliers based on the actual situation;

[0051] Handling duplicate data: For data points with duplicate records, deduplication is performed based on timestamps, location information, or other unique identifiers to ensure that each data point is analyzed only once;

[0052] Verify data quality: Verify the cleaned data using data quality checking tools or manual checks to ensure the accuracy, consistency and integrity of the data. Data quality thresholds can be set, and data that does not meet the requirements can be further processed or reported.

[0053] Record and report data processing results: Record the data cleaning process and results in detail, including data comparison before and after cleaning, problems found during cleaning and their handling methods, etc., to provide a basis for subsequent data analysis and interpretation;

[0054] Save the cleaned data: Store the cleaned data in a secure and reliable data storage medium for subsequent data analysis and application. At the same time, establish a data backup mechanism to prevent data loss or damage.

[0055] S2. Data format standardization and conversion

[0056] Standardize and normalize the data formats and units from different data sources to convert raw data from different data sources into a usable, unified data format, ensuring data consistency and comparability.

[0057] By using data transformation technology, frost heave-related indicators from different data sources can be unified to facilitate subsequent analysis and comparison.

[0058] Establish standardized rules and processes for data to ensure the standardization and repeatability of data fusion.

[0059] This includes data type conversion, data format conversion, data encoding conversion, data value normalization, data range normalization, and data distribution normalization.

[0060] (1) Data format conversion must ensure that the data type is consistent with the expectation, such as converting a string to a date format or a numeric format. Unify data formats, such as unifying date formats and ensuring consistency in text encoding.

[0061] (2) Data standardization and normalization: Data needs to be standardized so that variables with different dimensions can be compared and analyzed. Data normalization scales the data to a specific range to improve the performance of the data analysis model.

[0062] The main purpose of the data format standardization and conversion stage is to transform raw data from different data sources with different formats and units into a standardized format suitable for subsequent analysis and processing. This process is crucial for ensuring data consistency and comparability. In a specific embodiment, the specific steps include, but are not limited to, the following aspects:

[0063] 1. Data type conversion:

[0064] Perform unified conversion of data types from different data sources. For example, convert string-type timestamps to date-time types to facilitate time series analysis; convert text-formatted numeric values ​​to numeric types for mathematical operations and statistical analysis.

[0065] 2. Data format conversion:

[0066] Standardize date and time formats. Different data sources may use different date and time representations, such as "YYYY-MM-DD" and "MM / DD / YYYY". These need to be standardized to a single format for time comparison and analysis. Standardize text encoding formats to ensure that all text data uses the same character encoding (such as UTF-8) to avoid garbled characters caused by inconsistent encoding.

[0067] 3. Data encoding conversion:

[0068] For categorized or coded data, such as roadbed filler type and foundation treatment method, it is necessary to ensure that all data sources use the same coding system. If different data sources use different codes, mapping and conversion are required to unify the coding standard.

[0069] 4. Data value normalization:

[0070] Normalization is performed on numerical data to eliminate the influence of different units on the data analysis results. Commonly used normalization methods include min-max normalization (scaling the data to the [0,1] interval) and Z-score standardization (converting based on mean and standard deviation).

[0071] 5. Data range normalization:

[0072] For data with specific value ranges, such as temperature ranges or moisture content ranges, they can be normalized to a uniform range, such as [0,1] or [-1,1], for comparison and analysis.

[0073] 6. Data distribution normalization:

[0074] When there are significant differences in the data distribution from different data sources, data distribution normalization methods, such as logarithmic transformation and Box-Cox transformation, can be used to make the data distribution closer to a normal distribution, thereby improving the stability and accuracy of data analysis.

[0075] 7. Establish data standardization rules and processes:

[0076] Develop detailed data standardization rules and process documents, clarify the transformation methods and standards for each step, ensure the standardization and repeatability of the data fusion process, and train all personnel involved in data processing to ensure that they understand and follow the data standardization rules and processes.

[0077] 8. Implementation and Verification:

[0078] Use programming tools or data processing software (such as Python, R, Excel, etc.) to standardize and transform the data format, verify the transformed data, check whether the data meets the expected standard format, and ensure the accuracy and completeness of the transformation process.

[0079] By following the steps above, we can ensure the consistency and comparability of data from different data sources in terms of format and units, laying a solid foundation for subsequent data fusion and analysis.

[0080] S3, Data Fusion Tools and Technologies

[0081] By selecting appropriate data fusion tools, such as ETL (Extract, Transform, Load) tools and data warehouses, and employing feature-layer fusion technology, frost heave monitoring data from different sources and of different types can be fused, analyzed, and processed to generate more comprehensive, accurate, and useful information, and obtain more complete and valuable data monitoring information.

[0082] We utilize cutting-edge technologies such as data mining and machine learning to conduct in-depth analysis and mining of the fused data.

[0083] Pay attention to the updates and development of data fusion technologies, and continuously improve the ability and level of data fusion.

[0084] In monitoring frost heave of high-speed railway subgrades in high-altitude and cold regions, data fusion tools and technologies are crucial for ensuring the effective integration and utilization of multi-source data. Specifically, data fusion tools and technologies include the following aspects:

[0085] 1. Application of ETL tools

[0086] ETL (Extract, Transform, Load) tools are fundamental tools in the data fusion process. They are responsible for extracting data from different data sources, performing necessary transformations, and finally loading the data into the target database or data warehouse.

[0087] In monitoring roadbed frost heave in high-altitude and cold regions, ETL tools can automatically process data from various sources, including field monitoring equipment (such as displacement sensors and temperature sensors), non-contact monitoring methods (such as remote sensing data and geographic information system data), and laboratory experimental data.

[0088] ETL processes can ensure data consistency and accuracy, providing a reliable data foundation for subsequent analysis.

[0089] Data extraction: Automatically extracts data from various monitoring devices, databases, and files, supporting scheduled or real-time extraction modes.

[0090] Data transformation: Preprocessing operations such as cleaning, format standardization, and encoding conversion are performed on the data to ensure that the data meets the analysis requirements. For example, timestamps collected from different devices are standardized to a standard time format, and text encoding is standardized to UTF-8.

[0091] Data loading: Load the processed data into the data warehouse or target database to facilitate subsequent querying and analysis.

[0092] 2. Building a Data Warehouse

[0093] A data warehouse is a subject-oriented, integrated, stable, and time-varying collection of data used to support management decisions. In the monitoring of roadbed frost heave in high-altitude and cold regions, building a data warehouse can achieve centralized storage and management of multi-source data, and provide efficient data query and analysis capabilities.

[0094] Data integration: Integrating data from different data sources to eliminate data redundancy and inconsistency.

[0095] Data storage: Employ efficient data storage structures, such as columnar storage or distributed file systems, to improve data access speed.

[0096] Data Analysis: Supports complex data analysis operations, such as OLAP (Online Analytical Processing) and data mining, helping users discover potential patterns in data.

[0097] 3. Feature layer fusion technology

[0098] Feature layer fusion technology is an information integration method performed at the data feature level. By extracting key features from different data sources and fusing them, a more comprehensive and accurate information representation can be obtained. In the monitoring of roadbed frost heave in high-altitude and cold regions, feature layer fusion technology can be used to fuse roadbed deformation features, temperature features, etc. from different monitoring methods.

[0099] Feature extraction: Extract key features related to roadbed frost heave deformation from the raw data, such as displacement and temperature gradient.

[0100] Feature selection: Select the most representative features for fusion based on the analysis requirements to reduce data redundancy.

[0101] Feature fusion: Different features are fused using methods such as weighted average and principal component analysis (PCA) to form a comprehensive feature vector.

[0102] 4. Data Mining and Machine Learning

[0103] Data mining and machine learning technologies can automatically discover patterns, regularities, and anomalies from massive amounts of data, providing strong support for the analysis of roadbed frost heave deformation. In the monitoring of roadbed frost heave in high-altitude and cold regions, these technologies can be used for data classification, clustering, prediction, and other operations.

[0104] Data classification: The data is classified according to the different types and degrees of roadbed frost heave deformation to facilitate targeted analysis.

[0105] Data clustering: Grouping similar roadbed frost heave deformation data into clusters to discover potential structures in the data.

[0106] Predictive analysis: Establish mathematical models to predict and simulate frost heave deformation of roadbeds, and identify potential safety hazards in advance.

[0107] Commonly used machine learning algorithms include linear regression, support vector machines (SVM), and neural networks.

[0108] 5. Real-time data analysis and processing technology

[0109] In monitoring roadbed frost heave in high-altitude and cold regions, real-time data analysis and processing technology can ensure immediate response and processing of monitoring data. Through stream processing technology and real-time database systems, newly arriving data can be cleaned, transformed, and analyzed in real time, providing timely support for operation and maintenance.

[0110] Stream processing technology: Employs stream processing frameworks such as Apache Kafka and Apache Flink to perform real-time data processing and analysis.

[0111] Real-time database: Use a real-time database system (such as TimescaleDB) that supports high-concurrency read and write operations to store and manage real-time monitoring data.

[0112] 6. Visualization technology

[0113] Visualization technology can present complex data analysis results to users in an intuitive and easy-to-understand form of graphics and charts, improving decision-making efficiency. In the monitoring of roadbed frost heave in high-altitude and cold regions, information such as the spatial distribution and temporal changes of roadbed frost heave deformation can be displayed through geographic information systems (GIS), dashboards, reports, and other forms.

[0114] GIS visualization: Combining GIS technology to display the spatial distribution and evolution of roadbed frost heave deformation.

[0115] Dashboard and Reports: Key indicators and statistical results are displayed through dashboards and reports, making it easy for users to quickly understand the frost heave deformation of the roadbed.

[0116] In summary, the data fusion tools and technologies used in monitoring frost heave of high-speed railway subgrades in cold regions encompass multiple aspects, including ETL tools, data warehouses, feature layer fusion technology, data mining and machine learning, real-time data analysis and processing technology, and visualization technology. The comprehensive application of these tools and technologies can ensure the effective integration and utilization of multi-source data, providing strong support for the accurate analysis of subgrade frost heave deformation.

[0117] S4. Data Quality Control and Assessment

[0118] Quality control is performed on the merged data, including assessments of data integrity, accuracy, and consistency.

[0119] Data quality control technology can be used to reduce errors and biases in the data fusion process.

[0120] In the multi-source data fusion and analysis method for monitoring frost heave of high-speed railway subgrade in high-altitude and cold regions, data quality control and evaluation are key steps to ensure the accuracy and reliability of the analysis results. Specific steps include the following aspects:

[0121] 1. Data integrity assessment:

[0122] Missing data check: Perform a comprehensive check on the merged dataset to identify and record any missing data points.

[0123] For missing data, appropriate interpolation methods (such as linear interpolation, spline interpolation) or model-based prediction methods are used to fill in the missing data, depending on the data type and analysis requirements, in order to reduce the impact of missing data on the analysis results.

[0124] Record integrity verification: Ensure that each monitoring record contains complete information fields, such as timestamp, location information, monitoring indicators, etc., to avoid analysis bias caused by incomplete records.

[0125] 2. Data accuracy assessment:

[0126] Outlier detection and handling: Using statistical methods (such as standard deviation method, quartile method) or machine learning algorithms to identify outliers in the dataset.

[0127] For identified outliers, they are removed, corrected, or retained (e.g., recorded as special events) depending on the actual situation to ensure the accuracy of the dataset.

[0128] Data comparison and verification: The merged data is compared with the original data source or other independent data sources to verify the consistency and accuracy of the data.

[0129] Cross-validation allows for the timely detection and correction of errors that may be introduced during the data fusion process.

[0130] 3. Data consistency assessment:

[0131] Time consistency check: Ensure that all monitoring data are consistent in time to avoid data misalignment or analysis bias caused by time asynchrony.

[0132] For data with inconsistent timestamps, time synchronization is performed.

[0133] Spatial consistency verification: For data with spatial location information, verify the rationality of its spatial distribution.

[0134] Geographic Information System (GIS) technology is used to check whether data points fall within the expected monitoring area, thus avoiding analytical errors caused by spatial location mistakes.

[0135] 4. Data Quality Assessment Report:

[0136] Evaluation index setting: Based on the analysis needs and data characteristics, set appropriate data quality evaluation indexes, such as data completeness rate, accuracy rate, consistency rate, etc.

[0137] Assessment Results Record: Record the process and results of the data quality assessment in detail, including the calculation results of the assessment indicators, the problems found and the handling methods, etc.

[0138] These records not only provide a basis for subsequent data analysis and interpretation, but also help to continuously improve data quality.

[0139] Data quality feedback mechanism: Establish a data quality feedback mechanism to promptly provide feedback on evaluation results to all stages of data collection, processing, and analysis.

[0140] Through feedback mechanisms, we continuously optimize data collection schemes, data processing procedures, and analysis methods to improve data quality.

[0141] 5. Utilize data quality control techniques:

[0142] Data cleaning and preprocessing: Before data fusion, rigorous data cleaning and preprocessing are performed to remove erroneous, duplicate, and invalid data, thereby improving data quality.

[0143] At the same time, the data is standardized and normalized to unify the data format and units, thereby reducing analysis errors caused by inconsistent data formats.

[0144] Data fusion algorithm optimization: Select appropriate data fusion algorithms and technologies, such as ETL tools, data warehouses, feature layer fusion technology, etc., to ensure the effective integration of multi-source data.

[0145] By optimizing algorithm parameters and fusion strategies, errors and biases in the data fusion process can be reduced.

[0146] Continuous monitoring and improvement: Establish a continuous monitoring mechanism for data quality and conduct regular quality assessments of the merged data.

[0147] Based on the evaluation results, we will adjust the data collection, processing, and analysis strategies in a timely manner to continuously improve data quality and the accuracy of analysis results.

[0148] Through the detailed data quality control and evaluation steps described above, it can be ensured that the data used in the multi-source data fusion and analysis method for monitoring frost heave of high-speed railway subgrade in cold regions has a high degree of integrity, accuracy and consistency, providing reliable data support for subsequent analysis, prediction and judgment of subgrade frost heave deformation.

[0149] S5. Comprehensive analysis of multiple influencing factors

[0150] In addition to factors such as roadbed fill material, moisture content, and temperature, the impact of factors such as train operation and foundation treatment on the frost heave deformation of high-speed railway subgrade should also be considered. A comprehensive comparative analysis should be conducted to objectively and reasonably determine the frost heave phenomenon.

[0151] By using the above-mentioned multi-source data fusion method, we can gain a more comprehensive understanding of frost heave and provide richer data support for the analysis of frost heave deformation.

[0152] Methods to address frost heave deformation of high-speed railway subgrade in high-altitude and cold regions:

[0153] 1. The frost heave deformation of the roadbed is monitored in real time through manual observation, automated monitoring, and railway bureau dynamic inspection vehicles. Simultaneously, laboratory experiments are conducted to study the physical and thermodynamic properties of the soil, providing fundamental data for analyzing roadbed frost heave deformation. Based on field monitoring and laboratory results, and considering various influencing factors such as roadbed fill material, moisture content, and temperature changes, a mathematical model is established to predict and simulate the frost heave deformation of high-speed railway roadbeds.

[0154] 2. Utilize remote sensing technology to obtain surface information of permafrost areas, and combine it with geographic information system technology to process and analyze frost heave monitoring data, providing more comprehensive information for the analysis of roadbed frost heave deformation.

[0155] 3. Conduct a comprehensive analysis of the impact of factors such as roadbed fill material, moisture content, thermometer, train speed, and foundation treatment method on the frost heave deformation of the roadbed in high-speed railways in cold regions.

[0156] In the analysis of frost heave deformation of high-speed railway subgrades in high-altitude and cold regions, a comprehensive analysis of multiple influencing factors is crucial. These factors not only directly relate to the degree of frost heave deformation but also profoundly affect its trend and rate. Specifically, the main influencing factors include, but are not limited to, the following:

[0157] 1. Subgrade filler: Different subgrade fillers have different physical and chemical properties, which can vary significantly under low temperature conditions. For example, some fillers are prone to volume changes during freeze-thaw cycles, which can lead to frost heave of the subgrade. Therefore, when analyzing frost heave deformation of the subgrade, factors such as the type, gradation, and compaction degree of the filler must be considered in detail.

[0158] 2. Moisture content: Water is one of the main driving forces of frost heave in roadbeds. Changes in the moisture content of roadbeds directly affect their frost heave characteristics. In cold regions, low winter temperatures cause the water in the roadbed to freeze and expand in volume, thus triggering frost heave. Therefore, accurate monitoring and analysis of changes in the moisture content of roadbeds are of great significance for predicting and assessing frost heave deformation.

[0159] 3. Temperature: Temperature is a key factor affecting the frost heave of roadbeds. Winters in high-altitude and cold regions are long and cold, and the roadbed will experience low temperatures for a long time. Temperature changes not only affect the freezing and melting process of water in the roadbed, but also indirectly affect the deformation of the roadbed by affecting the thermal expansion and contraction properties of materials. Therefore, when analyzing the frost heave deformation of roadbeds, the historical changes and future trends of temperature must be fully considered.

[0160] 4. Train speed: The high-speed impact load generated by train operation will have a dynamic impact on the roadbed, especially under freeze-thaw cycle conditions, where this impact is more significant. Changes in train speed will change the stress state of the roadbed, thereby affecting its frost heave deformation characteristics. Therefore, when analyzing the frost heave deformation of the roadbed, it is necessary to consider the impact of train speed and its changes on the roadbed.

[0161] 5. Foundation treatment methods: Foundation treatment methods have a significant impact on the frost heave deformation of the roadbed. Different foundation treatment methods (such as replacement, reinforcement, drainage, etc.) will change the physical and mechanical properties of the foundation, thereby affecting its ability to resist frost heave deformation. Therefore, when analyzing the frost heave deformation of the roadbed, it is necessary to understand the foundation treatment methods and their effects in detail.

[0162] To comprehensively and accurately analyze the impact of the above-mentioned factors on the frost heave deformation of the roadbed, the following comprehensive analysis method can be adopted:

[0163] 1. Establish a mathematical model: Based on field monitoring data and laboratory experimental results, establish a mathematical model for roadbed frost heave deformation that considers multiple factors. This model should be able to simulate the roadbed frost heave deformation process under different conditions and predict its future trend.

[0164] 2. Multi-factor sensitivity analysis: By changing various parameters in the mathematical model (such as filler properties, moisture content, temperature, etc.), the influence of different factors on the frost heave deformation of the roadbed is analyzed. This helps to identify key factors and provides a basis for optimizing design and construction measures.

[0165] 3. Data Mining and Machine Learning: By utilizing data mining and machine learning techniques, patterns, regularities, and anomalies can be automatically discovered from massive amounts of monitoring data. This helps to gain a deeper understanding of the mechanism of roadbed frost heave deformation and improve the accuracy of predictions.

[0166] 4. Visualization and Analysis: By combining Geographic Information System (GIS) technology, the data analysis results are presented to users in an intuitive and easy-to-understand graphical and chart format, which helps improve decision-making efficiency and promotes communication and collaboration between different departments.

[0167] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the embodiments of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0168] The above examples illustrate the present invention only to aid in understanding it and are not intended to limit the scope of the invention. Those skilled in the art can make various simple deductions, modifications, or substitutions based on the principles of this invention.

Claims

1. A method for multi-source data fusion and analysis of frost heave monitoring for high-speed railway subgrade in high-altitude and cold regions, characterized in that, Includes the following steps: S1: Data source selection and data cleaning. Select frost heave monitoring data from multiple sources and clean the data to remove errors, duplicates and invalid data. S2: Data format standardization and conversion, standardizing and normalizing the cleaned data to unify the data format and units; S3: Data fusion, which uses data fusion tools and technologies to merge standardized multi-source data to generate a fused dataset; S4: Data quality control and evaluation, which involves quality control and evaluation of the merged dataset; S5: Comprehensive analysis of frost heave deformation. Based on the fusion dataset after quality control, it comprehensively analyzes multiple influencing factors to achieve the analysis, prediction and determination of frost heave deformation of roadbed.

2. The method for multi-source data fusion and analysis of high-speed railway subgrade frost heave monitoring in high-altitude and cold regions as described in claim 1, characterized in that, In step S1, the frost heave monitoring data from multiple sources include field monitoring data and non-contact monitoring data; field monitoring data includes manual observation data, automated monitoring data, ground-penetrating radar data, and data from the railway bureau's dynamic inspection vehicle; non-contact monitoring data includes laboratory experimental data, remote sensing data, and geographic information system data.

3. The method for multi-source data fusion and analysis of high-speed railway subgrade frost heave monitoring in high-altitude and cold regions as described in claim 1, characterized in that, The data cleaning described in step S1 includes the following sub-steps: defining the cleaning objective, handling missing data, handling abnormal data, handling duplicate data, and verifying data quality.

4. The method for multi-source data fusion and analysis of high-speed railway subgrade frost heave monitoring in high-altitude and cold regions as described in claim 1, characterized in that, The standardization and normalization process in step S2 includes at least one of the following: data type conversion, data format conversion, data encoding conversion, data value normalization, data range normalization, and data distribution normalization.

5. The method for multi-source data fusion and analysis of high-speed railway subgrade frost heave monitoring in high-altitude and cold regions as described in claim 1, characterized in that, The data fusion tools and techniques in step S3 include at least one of ETL tools, data warehouses, feature layer fusion techniques, data mining, and machine learning.

6. The method for multi-source data fusion and analysis of high-speed railway subgrade frost heave monitoring in high-altitude and cold regions as described in claim 1, characterized in that, The quality control and assessment in step S4 includes an assessment of the data integrity, accuracy, and consistency.

7. The method for multi-source data fusion and analysis of high-speed railway subgrade frost heave monitoring in high-altitude and cold regions as described in claim 1, characterized in that, The various influencing factors in step S5 include roadbed fill material, moisture content, temperature, train speed, and foundation treatment method.

8. The method for multi-source data fusion and analysis of high-speed railway subgrade frost heave monitoring in high-altitude and cold regions as described in claim 1, characterized in that, Step S5 involves analyzing frost heave deformation, including establishing a mathematical model to predict and simulate frost heave deformation of the roadbed.

9. The method for multi-source data fusion and analysis of high-speed railway subgrade frost heave monitoring in high-altitude and cold regions as described in claim 1, characterized in that, Step S5, the analysis of frost heave deformation, includes using remote sensing technology to obtain surface information and combining it with geographic information system technology for processing and analysis.

10. The method for multi-source data fusion and analysis of high-speed railway subgrade frost heave monitoring in high-altitude and cold regions as described in claim 1, characterized in that, Step S5 involves analyzing the frost heave deformation, including the trend of frost heave in the roadbed, the absolute frost heave amount, the differential frost heave amount, and the frost heave deformation rate.

Images