Railway CORS station selection method

Abstract

The invention discloses a railway CORS station selection method. At present, an experiential method is mostly adopted during site selection of the CORS. The method comprises the following steps: collecting DEM data of a target area, and calculating gradient, slope direction and surface roughness information of a research area; generating a DOM according to the remote sensing image type, and calculating a vegetation index; processing the SAR data through InSAR to obtain a deformation graph of the target area; communication network coverage conditions are collected and marked on a map, and a network signal coverage range map of the area is obtained; preprocessing the site selection factor of the CORS station; intelligent site selection is carried out according to the determined multiple factors; and according to the calculated site selection suitability plane graph, selecting a most suitable position as a CORS station deployment position every 50km along the railway, and guiding the construction of the CORS station. According to the method, the defects of a traditional on-site reconnaissance site selection method are overcome, and the arrangement position of the CORS station can be optimized.

Description

Technical Field

[0001] The invention belongs to the technical field of GNSS measurement, and in particular relates to a method for selecting a railway CORS station. Background Art

[0002] There are few high-level national control points along the railways in western my country, and the existing high-level control points are far away from the designed lines, which cannot meet the measurement requirements of the railway precision control network. The construction of CORS (Continuous Operational Reference System) stations can solve the reliability problem of the control benchmark results during the railway construction and operation periods by continuously observing and dynamically maintaining the benchmark control network. The CORS station is mainly composed of three parts: a data processing center, a continuously operating GNSS base station and a mobile station. The system is based on advanced network technology and communication technology to connect each GNSS base station and the data processing center. The data processing center calculates, processes and broadcasts the observation data on each base station, and users obtain differential information through GNSS receivers, thereby achieving precise positioning of the mobile station.

[0003] At present, CORS stations are mostly sited using empirical methods. For example, railway CORS stations are usually sited every 50km. The site is stable, the nearby field of vision is wide, there is no obvious obstruction, and the communication is good. The purpose of CORS station construction is to serve as a stable benchmark to provide benchmark information. If the site is likely to deform, the benchmark will be unstable, which not only wastes hardware equipment resources, but also fails to provide reliable differential information. Obviously, building CORS stations in areas where deformation has not occurred in history is a simple and effective method. Synthetic aperture radar interferometry (InSAR) is a satellite remote sensing technology developed in the past three decades. It can obtain surface information of the target area. By processing the archived satellite images of the target area, deformation information of the area can be obtained, and the site can be avoided in advance. Secondly, when the surface slope is too large or the surface is uneven, the difficulty of building the benchmark station will increase. The land needs to be leveled first, which will consume more manpower and material resources. Then, there is usually a certain density of vegetation near the line, which will cause multipath effects, which may affect the accuracy of GNSS measurement. Therefore, when selecting a CORS station, it is necessary to avoid densely vegetated areas to reduce the impact of multipath effects. Finally, the CORS station calculates, processes and broadcasts the observation data at each base station, and communication capabilities are also indispensable. Therefore, when choosing the location for building a CORS station, the deformation impact, deployment difficulty, observation conditions and continuous operation requirements should be comprehensively considered. Summary of the invention

[0004] In order to make up for the deficiencies of the prior art, the present invention provides a railway CORS station site selection method to solve the deficiencies of the traditional on-site survey site selection method and optimize the CORS station layout location.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] A railway CORS station selection method, characterized in that it comprises the following steps:

[0007] Step 1: Collect DEM data of the target area and calculate the slope, aspect, and surface roughness information of the study area;

[0008] Step 2: Collect remote sensing images of the target area, select corresponding professional software to generate DOM according to the image type, and calculate the vegetation index;

[0009] Step 3: Collect archived SAR images of the target area, and process the acquired SAR data using InSAR technology to obtain a deformation map of the target area;

[0010] Step 4: Collect the communication network coverage of the target area and mark it on the map to obtain the network signal coverage map of the area;

[0011] Step 5: Preprocessing of CORS station site selection factors;

[0012] Step 6: Multi-factor intelligent site selection determined according to the above steps;

[0013] Step 7: Based on the calculated site suitability plan, select the most suitable location every 50 km along the railway as the CORS station deployment location to guide the construction of CORS stations.

[0014] Furthermore, in step 1, the DEM data acquisition methods include the following: 1) downloading publicly available global free DEM; 2) collecting satellite images of the survey area to obtain the DEM of the target area; 3) using unmanned aerial vehicles / manned aircraft to fly over the survey area to obtain Lidar point cloud data or image data of the survey area, and obtaining the DEM of the target area through post-processing.

[0015] Further, the production process of step 2 is as follows:

[0016] The acquired original aerial photographs are subjected to data preprocessing to ensure that all image colors remain consistent and the image quality is clear. After aerial triangulation, a stereo model is created on a fully digital photogrammetry workstation. The generated DEM is then used to perform fully digital differential correction on the original images to generate orthophotos. The orthophotos are then mosaicked, segmented, and trimmed. Finally, after passing inspection, the DOM result data is output.

[0017] Furthermore, in step 2, after obtaining the DOM data, the vegetation index is calculated by calculating the information of the red, green and blue bands in the DOM. The calculation formula is: NDVI = (IR-R) / (IR+R), where IR represents the near-infrared band and R represents the infrared band.

[0018] Further, in step 3, the D-InSAR data processing flow is as follows:

[0019] 1) Data acquisition: Collecting synthetic aperture radar data;

[0020] 2) Image preprocessing: preprocess the collected SAR images;

[0021] 3) Generate interference pattern: Use the pre-processed SAR image to perform interference processing and generate interference pattern;

[0022] 4) Phase unwrapping: There may be phase discontinuity in the interference pattern, and phase unwrapping is required to obtain accurate phase information;

[0023] 5) Deformation calculation: Use the unwrapped phase information to calculate the deformation of the surface.

[0024] Further, in step 3, the annual deformation rate is calculated according to the following formula:

[0025]

[0026] Where N is the number of interference patterns; ψ i , △T i are the unwrapping phase and time interval of the i-th interference pattern respectively; V ψ Represents the average phase deformation rate; after the estimation is completed, the phase deformation rate is converted into the surface deformation rate and geocoded into the geographic coordinate system to obtain the final surface historical deformation rate map.

[0027] Furthermore, in step four, there are three types of communication networks, namely China Mobile, China Unicom and China Telecom. You can check the network status in your area through the official website or customer service.

[0028] Furthermore, in step five, the spatial resolution of each factor needs to be unified to 10 meters;

[0029] In order to convert the slope, vegetation index, and deformation factors into their corresponding site suitability, the site factors are preprocessed based on the normalization idea:

[0030]

[0031] In the formula, slope, vi, and defor represent the slope, vegetation index, and historical deformation site selection factor, respectively; max(Y) and min(Y) represent the maximum and minimum values, respectively; Y norm Represents the result after preprocessing of the corresponding location factor.

[0032] Furthermore, in step five, the reclassification method is used to pre-process the slope direction, that is, the sunny slope, semi-sunny slope, semi-shaded slope and shady slope are assigned values ​​of 0.8, 0.6, 0.4 and 0.2, respectively, indicating that the longer a certain position of the landslide is exposed to sunlight, the more suitable it is for the construction of a CORS station.

[0033] Furthermore, in step six, intelligent site selection is performed through the analytic hierarchy process, which specifically includes four steps:

[0034] Establish a hierarchical model, construct a judgment matrix, calculate relative weights, and calculate the target layer results;

[0035] The judgment matrix is:

[0036]

[0037] The first to fourth rows represent the four site selection factors of historical deformation, aspect, slope and vegetation index respectively. The calculated relative weight values ​​of the four site selection factors are 0.558, 0.263, 0.122 and 0.057 respectively. After passing the consistency test, they are used as the relative weight value of each site selection factor.

[0038] Beneficial effects of the present invention:

[0039] Aiming at the site selection and construction of railway CORS stations in complex and dangerous mountainous areas, the traditional manual on-site site selection method is difficult, time-consuming and costly. The present invention proposes a multi-factor intelligent decision-making site selection method based on the assistance of InSAR deformation and other information. The method uses five site selection factors, namely historical deformation, slope aspect, slope, vegetation index and communication conditions, to characterize the four requirements of CORS station site selection, namely stability and reliability, continuous operation, deployment difficulty and observation conditions. The method is combined with the hierarchical analysis method to intelligently evaluate the suitability distribution of CORS stations in the survey area and recommend specific site locations, which has high engineering application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is the operation flow chart of the present invention;

[0041] Figure 2 This is the D-InSAR data processing flow chart;

[0042] Figure 3 It is a DEM production process based on aerial survey method;

[0043] Figure 4This is a flowchart of DOM production based on aerial survey. DETAILED DESCRIPTION

[0044] The present invention is described in detail below in conjunction with specific implementation modes.

[0045] like Figure 1 As shown, the railway CORS station selection method of the present invention includes the following steps:

[0046] Step 1: Collect DEM data of the target area and calculate the slope, aspect, and surface roughness information of the study area using ARCGIS and other software;

[0047] There are many ways to obtain DEM, and the accuracy of DEM obtained by different methods is different. The methods for obtaining DEM data include the following: 1) Download the public global free DEM, such as SRTM (Shuttle Radar Topography Mission), with a resolution of 30 meters; Japan's ASTER Global Digital Elevation Model (GDEM), with a resolution of 12.5 meters. 2) Collect satellite images of the survey area, such as Worldview, SPOT and other satellite image data, and obtain the DEM of the target area through field image control point measurement and indoor aerial triangulation encryption. The resolution and accuracy are related to the image data; 3) UAV / manned aircraft fly over the survey area to obtain Lidar point cloud data or image data of the survey area, and obtain the DEM of the target area through post-processing. After obtaining the DEM data, the slope, aspect, and surface roughness information of the study area are calculated through ARCGIS and other software.

[0048] Step 2: Collect remote sensing images (or aerial images) of the target area, select corresponding professional software according to the image type to generate DOM (digital orthophoto), and calculate the vegetation index;

[0049] Step 2 production process is as follows:

[0050] The acquired original aerial images are subjected to data preprocessing, including image even lighting and color processing to eliminate problems such as over-darkness and over-exposure, so that all images maintain consistent color and clear image quality. After aerial triangulation, a stereo model is created on a fully digital photogrammetry workstation, and then the generated DEM is used to perform full digital differential correction on the original image to generate an orthophoto. The orthophoto is then mosaicked, segmented, and trimmed, and finally, after passing the inspection, the DOM results data are output. After obtaining the DOM data, the vegetation index is calculated by calculating the information of the red, green, and blue bands in the DOM. The calculation formula is: NDVI = (IR-R) / (IR+R), where IR represents the near-infrared band and R represents the infrared band.

[0051] Step 3: Collect archived SAR (Synthetic Aperture Radar) images of the target area, and process the acquired SAR data using InSAR technology to obtain a deformation map of the target area; commonly used technologies include Stacking, PS-InSAR, SBAS-InSAR, and SqueeSAR, which can effectively eliminate various errors in the data processing process and obtain more accurate research results on the surface;

[0052] There are many types of archived SAR data, such as the common Sentinel satellite, ALOS satellite, TerraSAR satellite, Radarsat and other satellites. The deformation map of the target area is obtained by processing the SAR data through InSAR technology. Figure 2 As shown in the figure, the D-InSAR data processing flow is as follows:

[0053] 1) Data acquisition: Collecting synthetic aperture radar (SAR) data, which is usually acquired by satellites or aircraft and can be radar images acquired at different times or from different perspectives;

[0054] 2) Image preprocessing: preprocess the collected SAR images, including radiation correction, atmospheric correction, noise removal, etc., to reduce noise and errors in the data;

[0055] 3) Generate interference pattern: Use the pre-processed SAR image to perform interference processing and generate interference pattern;

[0056] 4) Phase unwrapping: There may be phase discontinuity in the interference pattern, and phase unwrapping is required to obtain accurate phase information;

[0057] 5) Deformation calculation: Use the unwrapped phase information to calculate the deformation of the surface. Differential InSAR is the most basic data processing method. With the advancement of technology, Stacking, PS-InSAR, SBAS-InSAR, and SqueeSAR have emerged. They can overcome the errors of traditional technology such as time and space decoherence and atmospheric delay to the greatest extent, thereby obtaining high-precision time series and annual rate results, and have been widely studied in the field of geoscience. For example, Stacking technology can effectively obtain the linear deformation of the surface and effectively weaken the influence of atmospheric errors.

[0058] In step 3, before stacking, the SAR image is preprocessed, including cropping, multi-viewing, registration, unwrapping, removing terrain error, atmospheric error, and orbit error. Finally, the deformation map after unwrapping and error removal is obtained. Finally, the annual deformation rate is calculated according to the following formula:

[0059]

[0060] Where N is the number of interference patterns; ψ i , △T i are the unwrapping phase and time interval of the i-th interference pattern respectively; V ψ Represents the average phase deformation rate; after the estimation is completed, the phase deformation rate is converted into the surface deformation rate (unit: mm / y), and it is geocoded to the geographic coordinate system to obtain the final surface historical deformation rate map.

[0061] Step 4: Collect the communication network coverage of the target area and mark it on the map to obtain the network signal coverage map of the area; there are three types of communication networks, namely China Mobile, China Unicom and China Telecom. You can check the network conditions in your area through the official website or customer service.

[0062] Step 5: Preprocessing of CORS station site selection factors;

[0063] The spatial resolutions of the five site selection factors are inconsistent. For subsequent calculations, the spatial resolutions of the factors need to be unified to 10 meters (if the spatial resolutions of the acquired data are all high, a higher resolution can be used as the unified resolution). The factors with high spatial resolutions are downsampled using the weighted average method, while the low-resolution historical site selection factors are upsampled using the nearest neighbor interpolation method.

[0064] The meaning and value range of each site selection factor are also different. For the construction of CORS stations, communication conditions are indispensable, and communication conditions have the highest priority. The site selection must be within the range of signals. The slope indicates the steepness of the area, and the value range is 0°-90°. The greater the slope, the more difficult it is to build a CORS station; the vegetation index reflects the density of vegetation at the location. The larger the value, the denser the vegetation coverage and the worse the GNSS observation conditions; the deformation information represents the stability of the area. For the site selection of CORS stations, stable and non-deformed areas should be selected; the slope direction indicates the orientation of the location, and the value range is 0°-360°. Since CORS stations in remote areas need to be powered by solar energy, the south slope is preferred when selecting a location. In order to convert the slope, vegetation index, and deformation site selection factors into their corresponding site suitability, the site selection factors are preprocessed based on the normalization idea:

[0065]

[0066] In the formula, slope, vi, and defor represent the slope, vegetation index, and historical deformation site selection factor, respectively; max(Y) and min(Y) represent the maximum and minimum values, respectively; Y norm Represents the result after preprocessing of the corresponding location factor.

[0067] For the deformation factor, 0 represents regional stability, and non-zero values ​​represent regional uplift or subsidence deformation. The deformation value is taken as the absolute value during normalization. The larger the absolute value, the weaker the stability. For the aspect factor, the aspect value ranges from 0 to 360°. The illumination duration of different slopes is different, and it can be divided into four categories: sunny slope (135° to 225°), semi-sunny slope (225° to 315°), semi-shaded slope (45° to 135°) and shady slope (0° to 45° or 315° to 360°). The aspect value has a non-linear relationship with the selected area, so it is not suitable to use the normalization method for preprocessing. The reclassification method is used to preprocess the aspect, that is, the sunny slope, semi-sunny slope, semi-shaded slope and shady slope are assigned values ​​of 0.8, 0.6, 0.4 and 0.2 respectively, indicating that the longer the illumination time at a certain position of the landslide is, the more suitable it is to build a CORS station.

[0068] Step 6: Multi-factor intelligent site selection determined according to the above steps;

[0069] In order to better measure the importance of each factor, the analytic hierarchy process (AHP for short) is used. It refers to a decision-making method that decomposes the elements related to decision-making into goals, criteria, plans and other levels, and conducts qualitative and quantitative analysis on this basis. This method is suitable for solving complex multi-factor decision-making problems in the CORS station site selection task. Through AHP, multiple site selection factors can be systematically evaluated and compared, which can improve the accuracy and scientificity of decision-making and make the decision-making process more intelligent. In the AHP method, it is specifically reflected in the importance of different criteria through the judgment matrix, and then the relative weight of each criterion can be calculated.

[0070] Intelligent site selection through hierarchical analysis method includes 4 steps:

[0071] Establish a hierarchical model, construct a judgment matrix, calculate relative weights, and calculate the target layer results;

[0072] Establishing a hierarchical model is to determine the contents of three levels. Specifically for CORS station site selection, the plan level represents different construction locations; the criterion level corresponds to the criteria used in the selection process, namely the five site selection factors; and the target level is the suitability of the corresponding site selection location. Hierarchical analysis is to analyze whether a certain location is suitable for building a CORS station according to the criteria.

[0073] The judgment matrix represents the relative importance of the factors in site selection, usually expressed on a scale of 1 to 9. Assuming there are n factors in site selection, the size of the judgment matrix P is n×n, and the matrix element P ijIndicates the importance of site selection factor i compared to site selection factor j. In order to ensure the consistency of the judgment matrix, the matrix elements need to satisfy Pij·Pji=1. For the construction of CORS stations, communication is the first priority. Without communication, the station cannot be built, so its importance is different from other factors. Secondly, for CORS stations, the stability of the foundation is primary, so the deformation factor is extremely important. In addition, for CORS stations, providing navigation or differential services may occur at different times, and continuous operation is also important, so the slope factor is also very important. The slope determines the illumination time of the solar panel. The slope and vegetation coverage will affect the observation conditions of the CORS station. Among them, a large slope will increase the difficulty of building a CORS station, so the slope is more important than the vegetation coverage factor. Based on the above information, the judgment matrix is:

[0074]

[0075] The first to fourth rows represent the four site selection factors of historical deformation, aspect, slope and vegetation index respectively. The calculated relative weight values ​​of the four site selection factors are 0.558, 0.263, 0.122 and 0.057 respectively. After passing the consistency test, they are used as the relative weight value of each site selection factor.

[0076] In the CORS station site selection work, the target layer result is the site suitability, and its value range is [0, 1]. The specific calculation formula is as follows:

[0077]

[0078] In the formula, the subscript p represents a location in the site selection area; S represents the suitability of the site selection location; x i ,ω i They represent the value and relative weight corresponding to the i-th site selection factor respectively; the γ value is 0 or 1, which is used to control whether to calculate the site suitability of the location. When there is no communication signal at a certain place, the value is 0, and when there is a communication signal, the value is 1. The site selection suitability map thus generated reflects the spatial location distribution suitable for the construction of CORS stations. Combined with this distribution map, a suitable CORS station site can be selected.

[0079] Step 7: Based on the calculated site suitability plan, select the most suitable location every 50 km along the railway as the CORS station deployment location to guide the construction of CORS stations. Specific embodiment:

[0081] In a mountainous railway in the west, the terrain is steep and the vegetation is dense in some areas, making it very difficult to conduct manual site surveys. Combining the method described in this paper, a site suitability map for station construction was generated near the line by comprehensively considering deformation, slope aspect, slope, vegetation and communication, saving a lot of manpower for CORS station site selection. After selecting an optimal location every 50 km on the map, after on-site surveys and verification, all the selected locations can be used to build CORS stations.

[0082] The content of the present invention is not limited to the embodiments listed, and any equivalent changes made to the technical solution of the present invention by ordinary technicians in this field after reading the description of the present invention are covered by the claims of the present invention.

Claims

1. A method for selecting a railway CORS station, characterized in that: The steps include: Step 1: Collect DEM data of the target area and calculate the slope, aspect, and surface roughness information of the study area; Step 2: Collect remote sensing images of the target area, select corresponding professional software according to the image type to generate DOM, and calculate the vegetation index; Step 3: Collect archived SAR images of the target area, and process the acquired SAR data using InSAR technology to obtain a deformation map of the target area; Step 4: Collect the communication network coverage of the target area and mark it on the map to obtain the network signal coverage map of the area; Step 5: Preprocessing of CORS station site selection factors; Step 6: Multi-factor intelligent site selection determined according to the above steps; Step 7: Based on the calculated site suitability plan, select the most suitable location every 50 km along the railway as the CORS station deployment location to guide the construction of CORS stations.

2. A railway CORS station selection method according to claim 1, characterized in that: In step 1, the DEM data acquisition methods include the following: 1) Download the publicly available global DEM for free; 2) Collect satellite images of the survey area to obtain the DEM of the target area; 3) Unmanned aerial vehicles / manned aircraft fly over the survey area to obtain Lidar point cloud data or image data of the survey area, and obtain the DEM of the target area through post-processing.

3. A railway CORS station selection method according to claim 2, characterized in that: Step 2 production process is as follows: The acquired original aerial photographs are subjected to data preprocessing to ensure that all image colors remain consistent and the image quality is clear. After aerial triangulation, a stereo model is created on a fully digital photogrammetry workstation. The generated DEM is then used to perform fully digital differential correction on the original images to generate orthophotos. The orthophotos are then mosaicked, segmented, and trimmed. Finally, after passing inspection, the DOM result data is output.

4. A railway CORS station selection method according to claim 3, characterized in that: In step 2, after obtaining the DOM data, the vegetation index is calculated by calculating the information of the red, green and blue bands in the DOM. The calculation formula is: NDVI = (IR-R) / (IR+R), where IR represents the near-infrared band and R represents the infrared band.

5. A railway CORS station selection method according to claim 4, characterized in that: In step 3, the D-InSAR data processing flow is as follows: 1) Data acquisition: Collecting synthetic aperture radar data; 2) Image preprocessing: preprocess the collected SAR images; 3) Generate interference pattern: Use the pre-processed SAR image to perform interference processing and generate interference pattern; 4) Phase unwrapping: There may be phase discontinuity in the interference pattern, and phase unwrapping is required to obtain accurate phase information; 5) Deformation calculation: Use the unwrapped phase information to calculate the deformation of the surface.

6. A railway CORS station selection method according to claim 5, characterized in that: In step 3, the annual deformation rate is calculated according to the following formula: Where N is the number of interference patterns; ψ i , △T i are the unwrapping phase and time interval of the i-th interference pattern respectively; V ψ Represents the average phase deformation rate; after the estimation is completed, the phase deformation rate is converted into the surface deformation rate and geocoded into the geographic coordinate system to obtain the final surface historical deformation rate map.

7. A railway CORS station selection method according to claim 6, characterized in that: In step 4, there are three types of communication networks, namely China Mobile, China Unicom and China Telecom. You can check the network conditions in your area through the official website or customer service.

8. A railway CORS station selection method according to claim 7, characterized in that: In step 5, the spatial resolution of each factor needs to be unified to 10 meters; In order to convert the slope, vegetation index, and deformation factors into their corresponding site suitability, the site factors are preprocessed based on the normalization idea: In the formula, slope, vi, and defor represent the slope, vegetation index, and historical deformation site selection factor, respectively; max(Y) and min(Y) represent the maximum and minimum values, respectively; Y norm Represents the result after preprocessing of the corresponding location factor.

9. A railway CORS station selection method according to claim 8, characterized in that: In step five, the reclassification method is used to pre-process the slope direction, that is, the sunny slope, semi-sunny slope, semi-shaded slope and shady slope are assigned values ​​of 0.8, 0.6, 0.4 and 0.2 respectively, indicating that the longer a certain position of the landslide is exposed to sunlight, the more suitable it is for the construction of a CORS station.

10. A railway CORS station selection method according to claim 9, characterized in that: In step six, intelligent site selection is performed through the analytic hierarchy process, which includes four steps: Establish a hierarchical model, construct a judgment matrix, calculate relative weights, and calculate the target layer results; The judgment matrix is: The first to fourth rows represent the four site selection factors of historical deformation, aspect, slope and vegetation index respectively. The calculated relative weight values ​​of the four site selection factors are 0.558, 0.263, 0.122 and 0.057 respectively. After passing the consistency test, they are used as the relative weight value of each site selection factor.

Images