Conventional railway self-sustaining Beidou ground-based augmentation system

Abstract

The present invention discloses a self-sustaining Beidou ground-based augmentation system for conventional railways, comprising a reference station network subsystem, a data center subsystem, a data communication subsystem, and a user application subsystem. The reference station network subsystem constructs an improved Delaunay triangulation grid based on the railway's strip-shaped characteristics; the data center subsystem performs automated satellite observation data quality analysis, station selection, and dynamic networking; the data center subsystem uses an adaptive weighted filtering method to perform baseline solution and constraint adjustment, and then utilizes a precise single-point positioning algorithm of an adaptive robust Kalman filter, adopting a combined analysis scheme to achieve automatic maintenance and updating of the reference network with a one-day cycle; the data center subsystem utilizes the original data and real-time solution data stream to provide users with static data and real-time differential services, thereby providing high-precision control benchmarks and position services for railway surveying, operation, and maintenance.

Description

Technical Field

[0001] The present invention relates to the field of railway satellite positioning technology, and in particular to a self-sustaining Beidou ground-based augmentation system for conventional railways. Background Art

[0002] Conventional railways often conduct surveys using a network of conductors. Due to errors in alignment and leveling, as well as the structure of the control network, track deviation and measurement errors can increase over time, significantly impacting the safety and comfort of train operations. Due to the lack of a benchmark, these deviations are difficult to detect. Consequently, conventional railway surveying and operational maintenance are currently difficult, labor-intensive, and inaccurate.

[0003] The BeiDou-3 satellite system was completed in June 2020, marking a new phase in the development of my country's satellite navigation system. The BeiDou satellite system features three constellations of satellites: MEO, IGSO, and GEO. They provide multi-frequency signals and offer crucial support for the country's integrated positioning, navigation, and timing system. However, in the existing railway network, due to the strong technical foundation of high-speed railways, research is currently underway on the deployment and networking of ground-based augmentation systems (GABSs) based on the existing high-speed railway network. For example, application number 201920792216.4, titled "Installation Platform for BeiDou Ground-Based Augmentation Stations," discloses an installation platform for BeiDou GABS stations suitable for high-speed railways. In a paper titled "Research on the Construction of a BeiDou Ground-Based Augmentation System for Railways and Site Selection of Reference Stations" published in the journal Railway Computer Applications, a general description of base station site selection and observation data is provided. However, due to the older construction period of conventional railways, the GABS systems and networking methods suitable for high-speed railways are not applicable to conventional railway networks.

[0004] Although with the rise of high-speed railways, conventional railways have been gradually compressed in the passenger network, they still occupy a major position in the freight network. Therefore, it is extremely necessary to build a self-sustaining Beidou ground-based augmentation system for conventional railways. Summary of the Invention

[0005] Therefore, the purpose of the present invention is to provide a self-sustaining Beidou ground-based augmentation system for conventional railways, which adopts a Beidou reference station network to provide a high-precision reference framework for the survey, operation and maintenance of conventional railways; and adopts an improved Delaunay triangulation network that takes into account the strip characteristics of the railway to construct a foundation augmentation network topology structure, thereby improving the network stability, baseline solution accuracy and system rationality.

[0006] In order to achieve the above-mentioned object, the present invention provides a conventional railway self-sustaining Beidou ground-based augmentation system, which includes a reference station network subsystem, a data center subsystem, a data communication subsystem and a user application subsystem;

[0007] The reference station network subsystem uses the Delaunay triangulation method that takes into account the railway strip characteristics to construct the Beidou reference station network;

[0008] The data center subsystem acquires BeiDou data from the reference station network subsystem, performs automated data quality analysis and dynamic networking on the acquired BeiDou data, and utilizes a combination of baseline solution based on adaptive factor weighted filtering and adaptive robust Kalman filtering (PPP) to dynamically maintain and update the reference network.

[0009] The data communication subsystem establishes communication between the base station network, data center and user application subsystem through 4G network and cloud virtual LAN technology;

[0010] The user application subsystem uses the original data and real-time solution data stream of the data center subsystem to provide users with static data and real-time differential services.

[0011] Further preferably, the Delaunay triangulation method taking into account the strip-shaped characteristics of the railway is used to construct the Beidou reference station network according to the following steps:

[0012] S101, for the point set V composed of Beidou reference stations, the vertical coordinate Y axis is the main point position. Arrange in ascending order and recursively call the split-merge algorithm until a triangulation is generated, dividing the point set V into two subsets with roughly equal number of points. and ;

[0013] S102, creating Delaunay triangulations in each of the two subsets, optimizing the generated triangulations using a local optimization algorithm, calculating the convex hull of each triangulation, finding the bottom line and top line of the convex hull connecting the two sets, and merging the two triangulations of the two sets upward from the bottom line to the top line of the convex hull. During the merging process, if a vertex in a triangulation is within the circumcircle of another triangle, deleting the Delaunay edge and reconstructing the correct edge using the Delaunay criterion, so that the circumcircle of any triangle does not contain any other point in the set V;

[0014] S103. Use the Beidou reference stations along the conventional railway to construct a Delaunay triangulation. Based on the railway's strip characteristics and the direction of the railway line, starting from the first reference point, the two nearest reference points are searched for each reference point in sequence to construct a new strip triangulation, called a "nearby triangulation."

[0015] S104. The repeated baseline edges of the Delaunay triangulation and the "adjacent triangulation" are used as the definite baseline edges. The baseline edges where the two triangulations intersect are compared with the baseline solution residuals, and the baseline edge with the smaller average residual is selected as the definite baseline edge. After deleting the baselines that exceed the set threshold distance from all the definite baseline edges, the remaining baseline edges form a mesh shape, thereby forming a final mesh shape.

[0016] Further preferably, the data center subsystem performs automated data quality analysis and dynamic networking on the acquired Beidou data, including the following processes:

[0017] S201. Determine whether a satellite participates in data solution based on the satellite status in the Beidou data acquired at each reference station, wherein the Beidou data includes a multipath effect value, an ionospheric effect value, a cycle slip ratio, a data integrity rate, and a satellite signal-to-noise ratio;

[0018] S202: Perform a seven-day data check on each reference station in the Beidou data acquired by each reference station; the seven-day data check includes calculating an average of a data quality index of each reference station within seven days, and if a multipath effect value, an ionospheric effect value, a cycle slip ratio, and a data integrity rate of the current reference station simultaneously meet a first preset condition, then determining that the current reference station participates in the networking;

[0019] S203. If the Beidou data of the current reference station cannot simultaneously meet the first preset condition during data detection within seven days, a 24-hour data detection is performed. The 24-hour data detection includes real-time calculation of the average value of data quality indicators from 24 hours ago to the current moment. If the multipath effect value, ionospheric effect value, cycle slip ratio, and data integrity rate of the current reference station simultaneously meet the second preset condition, the current reference station is judged to participate in the networking; otherwise, the current reference station is prohibited from participating in the networking and an alarm is issued.

[0020] Further preferably, when judging whether a satellite participates in data solution according to the satellite status, the following method is adopted:

[0021] For satellites with an elevation angle below 35 degrees, when the L1 and L2 SNRs are less than 40, the satellite is prohibited from participating in the solution; for satellites with an elevation angle above 35 degrees, when the L1 and L2 SNRs are less than 35, the satellite is prohibited from participating in the solution;

[0022] Further preferably, the first preset conditions include: multipath effect value mp1<0.5m, mp2<0.65m, mp3<0.65m, ionospheric effect value IOD<200cm / min, cycle slip ratio o / slps>500, and data integrity rate>98%.

[0023] Further preferably, the second preset conditions include: multipath effect value mp1<0.55m, mp2<0.70m, mp3<0.70m, ionospheric effect value IOD<400cm / min, cycle slip ratio o / slps>200, and data integrity rate>95%.

[0024] Further preferably, the data center subsystem utilizes a combined analysis of a baseline solution based on adaptive factor weighted filtering and an adaptive robust Kalman filter PPP method to achieve dynamic maintenance and update of the baseline network, including the following methods:

[0025] S301. Based on the adaptive factor weighted filtering method, the double-difference tropospheric delay and ionospheric delay are estimated as parameters. A baseline solution is performed with a one-day cycle. The baseline solution results are subjected to network adjustment. The post-test variance and unit weighted mean error of the observation values ​​are calculated using the network adjustment results of the day. The coordinates of the known points are introduced to obtain the constrained adjustment coordinates (X, Y, Z). line ;

[0026] S302. Use the BeiDou satellite precise point positioning (PPP) algorithm and the robust adaptive Kalman filter method to solve the PPP single-day solution. Use the daily PPP single-day solution coordinates as the prior coordinates and substitute them into the next day's single-day solution as the initial filter value. Automatically perform PPP coordinate statistics using a 7-day window, and take the average of the daily PPP coordinates for the 7 days before 0:00 on the current day as the PPP weekly solution (X, Y, Z) for the current day. PPP .

[0027] S303, (X, Y, Z) PPP AND (X, Y, Z) line Processing: When the difference between the two coordinate components is less than 5mm, the average value of the coordinates is taken as the final coordinates of the day (X, Y, Z) day When the coordinate component difference is greater than or equal to 5mm, the coordinate with the smaller 3D residual value is used as the final coordinate (X, Y, Z) of the day. day , and ultimately achieve automatic maintenance and updating of the Beidou reference network with a cycle of one day.

[0028] Further preferably, the robust adaptive Kalman filtering method for solving the PPP single-day solution includes the following process:

[0029] Use IGG3 equivalent weight function to obtain robust equivalent weight and adaptive equivalent weight matrix ,in, is the state prediction value, is the weight matrix of the state prediction value, the adaptive factor The calculation is done using the following formula:

[0030] ;

[0031] Where c=2.5, , is the epoch solution;

[0032] Using the robust Kalman filter gain matrix ,in is the coefficient matrix;

[0033] The result of robust adaptive filtering is:

[0034] ;

[0035] Where, is the parameter value to be determined after filtering, is the predicted value of the parameter to be determined, is the gain matrix, is the observation noise vector, is the filter error variance matrix.

[0036] Further preferably, the reference station network subsystem has adjacent reference stations with a distance of 15-25 km, and is deployed on the roofs or courtyards of work areas; it also includes encryption points, which are evenly distributed between the reference points at intervals of 5 km, for regularly manually collecting Beidou observation data, maintaining the benchmark, and providing a control benchmark for railway survey and operation and maintenance;

[0037] The manual collection of Beidou observation data includes manual measurement every quarter, conducting third-level control measurements on benchmark points and encrypted points along the benchmark network, and further inspecting the self-maintenance of the Beidou benchmark network. Ultimately, the Beidou benchmark network maintenance accuracy is 5mm.

[0038] Furthermore, the benchmark includes a Beidou data observation device and a Beidou data acquisition device. The Beidou data observation device includes a high-precision 3D choke antenna, a steel observation column, and a concrete reinforced ground anchor base; the Beidou data acquisition device includes a cabinet, a Beidou low-power anti-interference receiver, a UPS power supply, a lithium battery, and a radiator.

[0039] Further preferably, the data communication subsystem is composed of three parts: a distributed cloud server network in the data center, a base station communication network formed from the base station to the data center, and a user communication network broadcast from the data center to the mobile station users;

[0040] The data center distributed cloud server network is deployed in a distributed manner;

[0041] The base station communication network establishes communication with the data center through 4G network and cloud virtual local area network technology to form a virtual local area network to achieve remote access and monitoring of the base station;

[0042] The user communication network uses the Ntrip protocol to send RTCMv3 differential data to the user in real time through the 4G network.

[0043] Further preferably, the user application subsystem is divided into a post-service system and a real-time service system.

[0044] The post-event service system authorizes users to log in to the post-event service website through HTTP, independently select observation data to download, and perform post-event static and dynamic differential analysis.

[0045] The real-time service system establishes a virtual reference station and a short baseline in real time based on the location sent by the user, and uses the Ntrip network protocol to send RTCMv3 pseudorange / carrier differential information to the user in real time based on the user's decimeter / centimeter level accuracy requirements.

[0046] The self-sustaining Beidou ground-based augmentation system for conventional railways disclosed in this application has at least the following advantages over existing technologies:

[0047] 1. An improved Delaunay triangulation network that takes into account the railway's strip-like characteristics is used to construct the foundation-enhanced network topology, improving network stability, baseline solution accuracy, and system rationality. This provides a high-precision benchmark framework for the survey, operation, and maintenance of conventional railways.

[0048] 2. Use high-precision full-satellite full-frequency 3D choke antenna and concrete reinforced ground anchor base to ensure the stability of the base station.

[0049] 3. Use additional encryption points for manual measurement to maintain and verify the Beidou reference network and provide a control benchmark for railway survey and operation and maintenance.

[0050] 4. The data center subsystem uses automated data quality analysis to evaluate site data quality in terms of MP1, MP2, O / SLPs, signal-to-noise ratio, ionospheric impact, and data integrity, and flexibly selects sites based on this to achieve dynamic networking.

[0051] 5. Adopting the adaptive weighted filtering method, baseline solution and network adjustment are performed on a one-day basis. The Beidou satellite precise single-point positioning technology based on the adaptive robust Kalman filter is used for combined analysis to obtain the optimal solution, thus realizing high-precision automatic maintenance and updating of the Beidou reference network.

[0052] 6. Baseline calculation, network adjustment and precise single-point positioning are all automated, and the benchmark maintenance accuracy is 5mm.

[0053] 7. The user application subsystem is divided into post-service system and real-time service system. The real-time service system can provide pseudo-range or carrier phase differential information according to the user's accuracy requirements to achieve diversified user use.

[0054] 8. Carrier phase differential RTK accuracy is ±(10+1×10-6×D)mm in the horizontal plane and ±(20+1×10-6×D)mm in the vertical plane, where D is the baseline length in kilometers. This accuracy fully meets the requirements of surveying, operation, and maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 This is a structural schematic diagram of a conventional railway self-sustaining Beidou ground-based augmentation system provided by the present invention.

[0056] Figure 2 This is a schematic diagram of the benchmark self-sustaining process in the conventional railway self-sustaining Beidou ground-based augmentation system provided by the present invention.

[0057] 1. Base station network subsystem, 2. Data center subsystem, 3. User application subsystem, 4. Data communication subsystem, 5. Cloud virtual LAN, 6. Real-time service system, 7. Post-service system, 8. Distributed cloud server network, 9. Baseline self-maintenance system, 10. VRS solution system, 11. User communication network. DETAILED DESCRIPTION

[0058] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0059] like Figure 1 As shown, the conventional railway self-sustaining BeiDou ground-based augmentation system provided by an embodiment of the present invention includes a reference station network subsystem 1, a data center subsystem 2, a data communication subsystem 4 and a user application subsystem 3;

[0060] The base station network subsystem 1 uses the Delaunay triangulation method that takes into account the railway strip characteristics to construct the Beidou base station network;

[0061] Furthermore, the Delaunay triangulation method that takes into account the strip-shaped characteristics of the railway is used to construct the Beidou base station network according to the following steps:

[0062] S101, for the point set V composed of Beidou reference stations, the vertical coordinate Y axis is the main point position. Arrange in ascending order and recursively call the split-merge algorithm until a triangulation is generated, dividing the point set V into two subsets with roughly equal number of points. and ;in Contains the points in set V with the smaller half of the X value, The point that contains the other half;

[0063] S102, creating Delaunay triangulations in each of the two subsets, optimizing the generated triangulations using a local optimization algorithm, calculating the convex hull of each triangulation, finding the bottom line and top line of the convex hull connecting the two sets, and merging the two triangulations of the two sets upward from the bottom line to the top line of the convex hull. During the merging process, if a vertex in a triangulation is within the circumcircle of another triangle, deleting the Delaunay edge and reconstructing the correct edge using the Delaunay criterion, so that the circumcircle of any triangle does not contain any other point in the set V;

[0064] S103. Use the Beidou reference stations along the conventional railway to construct a Delaunay triangulation. Based on the railway's strip characteristics and the direction of the railway line, starting from the first reference point, the two nearest reference points are searched for each reference point in sequence to construct a new strip triangulation, called a "nearby triangulation."

[0065] S104. The repeated baseline edges of the Delaunay triangulation and the "adjacent triangulation" are used as the definite baseline edges. The baseline edges where the two triangulations intersect are compared with the baseline solution residuals, and the baseline edge with the smaller average residual is selected as the definite baseline edge. After deleting the baselines that exceed the set threshold distance from all the definite baseline edges, the remaining baseline edges form a mesh shape, thereby forming a final mesh shape.

[0066] Data Center Subsystem 2 acquires BeiDou data from the Reference Station Network Subsystem, performs automated data quality analysis and dynamic networking on the acquired BeiDou data, and utilizes a combination of baseline solution based on adaptive factor weighted filtering and adaptive robust Kalman filtering (PPP) to dynamically maintain and update the reference network.

[0067] Further preferably, the data center subsystem performs automated data quality analysis and dynamic networking on the acquired Beidou data, including the following processes:

[0068] S201. Determine, based on the satellite status, whether a satellite participates in data solution in the BeiDou data acquired at each reference station; the BeiDou data includes a multipath effect value, an ionospheric effect value, a cycle slip ratio, a data integrity rate, and a satellite signal-to-noise ratio; and further, determine, based on the satellite status, whether a satellite participates in data solution by using the following method:

[0069] For satellites with an elevation angle below 35 degrees, when the L1 and L2 SNRs are less than 40, the satellite is prohibited from participating in the solution; for satellites with an elevation angle above 35 degrees, when the L1 and L2 SNRs are less than 35, the satellite is prohibited from participating in the solution;

[0070] S202: Perform a seven-day data check on each reference station in the Beidou data acquired by each reference station; the seven-day data check includes calculating an average of a data quality index of each reference station within seven days, and if a multipath effect value, an ionospheric effect value, a cycle slip ratio, and a data integrity rate of the current reference station simultaneously meet a first preset condition, then determining that the current reference station participates in the networking;

[0071] S203. If, during data testing within the past seven days, the Beidou data at the current reference station does not simultaneously meet the first preset conditions, which include: multipath effect values ​​mp1 < 0.5 m, mp2 < 0.65 m, mp3 < 0.65 m, ionospheric effect value IOD < 200 cm / min, cycle slip ratio o / slps > 500, and data integrity rate > 98%, then a 24-hour data test is performed. The 24-hour data test includes real-time calculation of the average data quality indicators from 24 hours ago to the current time.

[0072] If the multipath effect value, ionospheric effect value, cycle slip ratio, and data integrity rate of the current reference station simultaneously meet the second preset conditions (the second preset conditions include: multipath effect value mp1 < 0.55m, mp2 < 0.70m, mp3 < 0.70m, ionospheric effect value IOD < 400cm / min, cycle slip ratio o / slps > 200, and data integrity rate > 95%), the current reference station is determined to participate in the network. Otherwise, the current reference station is prohibited from participating in the network and an alarm is issued.

[0073] The data center subsystem 2 uses a baseline self-sustaining system 9. The baseline self-sustaining system 9 uses a combination of adaptive factor weighted filtering-based baseline solution and adaptive robust Kalman filtering (PPP) method analysis to achieve dynamic maintenance and update of the baseline network, including the following methods:

[0074] S301. Based on the adaptive factor weighted filtering method, the double-difference tropospheric delay and ionospheric delay are estimated as parameters. A baseline solution is performed with a one-day cycle. The baseline solution results are subjected to network adjustment. The post-test variance and unit weighted mean error of the observation values ​​are calculated using the network adjustment results of the day. The coordinates of the known points are introduced to obtain the constrained adjustment coordinates (X, Y, Z). line ;

[0075] S302. Use the BeiDou satellite precise point positioning (PPP) algorithm and the robust adaptive Kalman filter method to solve the PPP single-day solution. Use the daily PPP single-day solution coordinates as the prior coordinates and substitute them into the next day's single-day solution as the initial filter value. Automatically perform PPP coordinate statistics using a 7-day window, and take the average of the daily PPP coordinates for the 7 days before 0:00 on the current day as the PPP weekly solution (X, Y, Z) for the current day. PPP .

[0076] S303, (X, Y, Z)PPP AND (X, Y, Z) line Processing: When the difference between the two coordinate components is less than 5mm, the average value of the coordinates is taken as the final coordinates of the day (X, Y, Z) day When the coordinate component difference is greater than or equal to 5mm, the coordinate with the smaller 3D residual value is used as the final coordinate (X, Y, Z) of the day. day , and ultimately achieve automatic maintenance and updating of the Beidou reference network with a cycle of one day.

[0077] The robust adaptive Kalman filter method for solving the PPP single-day solution includes the following steps:

[0078] Use IGG3 equivalent weight function to obtain robust equivalent weight and adaptive equivalent weight matrix ,in, is the state prediction value, is the weight matrix of the state prediction value, the adaptive factor The calculation is done using the following formula:

[0079] ;

[0080] Where c=2.5, , is the epoch solution;

[0081] Using the robust Kalman filter gain matrix ,in is the coefficient matrix;

[0082] The result of robust adaptive filtering is:

[0083] ;

[0084] Where, is the parameter value to be determined after filtering, is the predicted value of the parameter to be determined, is the gain matrix, is the observation noise vector, is the filter error variance matrix.

[0085] The daily PPP single-day solution coordinates are used as the prior coordinates and substituted into the next day's single-day solution as the initial filter value. The data center subsystem automatically performs PPP coordinate statistics with a 7-day window, and takes the average of the daily PPP coordinates of the 7 days before 0:00 on the current day as the PPP weekly solution (X, Y, Z) for the current day. PPP .

[0086] Change (X, Y, Z) PPP AND (X, Y, Z) lineProcessing: When the difference between the two coordinate components is less than 5mm, the average value of the coordinates is taken as the final coordinates of the day (X, Y, Z) day When the coordinate component difference is greater than 5mm, the coordinate with the smaller 3D residual value is used as the final coordinate (X, Y, Z) of the day. day , and ultimately achieve automatic maintenance and updating of the Beidou reference network with a cycle of one day.

[0087] The data communication subsystem 4 establishes communication between the base station network, data center and user application subsystem through 4G network and cloud virtual LAN technology;

[0088] The user application subsystem 3 uses the original data and real-time solution data stream of the data center subsystem to provide users with static data and real-time differential services.

[0089] The base station network subsystem 1 has base stations with a distance of 15-25 km between adjacent points, all located on rooftops or in yards within work areas. It also includes encryption points, which are evenly distributed between base stations at 5 km intervals. These points are used to manually collect Beidou observation data on a regular basis, maintain the base, and provide a control benchmark for railway surveys and operations.

[0090] Manual collection of Beidou observation data, including manual measurement every quarter, third-level control measurement of benchmark points and encryption points along the benchmark network, and further inspection of the self-maintenance of the Beidou benchmark network. Finally, the Beidou benchmark network maintenance accuracy is 5mm.

[0091] The benchmark includes a BeiDou data observation device and a BeiDou data acquisition device. The BeiDou data observation device includes a high-precision 3D choke antenna, a steel observation column, and a concrete reinforced ground anchor base. The BeiDou data acquisition device includes a cabinet, a BeiDou low-power anti-interference receiver, a UPS power supply, a lithium battery, and a radiator.

[0092] The data communication subsystem 4 consists of three parts: the distributed cloud server network 8 of the data center, the base station communication network established from the base station to the data center, and the user communication network 11 broadcast from the data center to the rover users;

[0093] The distributed cloud server network in the data center is deployed in a distributed manner;

[0094] The data center uses the VRS solver 10 to broadcast RTCMv3 differential data to rover users via the user communication network, using the Ntrip protocol over a 4G network. The VRS solver 10 utilizes high-precision, self-sustaining coordinates and a real-time baseline network to construct an atmospheric correction model, thereby establishing a virtual reference station based on the user's location in real time.

[0095] The base station communication network establishes communication with the data center through 4G network and cloud virtual LAN technology, forming a cloud virtual LAN 5 to achieve remote access and monitoring of the base station;

[0096] The user application subsystem is divided into the post-service system 7 and the real-time service system 6.

[0097] The post-service system 7 authorizes users to log in to the post-service website through HTTP, independently select observation data to download, and perform post-static and dynamic differential analysis.

[0098] Real-time service system 6, based on the position sent by the user, establishes a virtual reference station in real time, builds a short baseline, and uses the Ntrip network protocol to send RTCMv3 pseudo-range / carrier differential information to the user in real time according to the user's decimeter / centimeter level accuracy requirements. Its carrier phase differential RTK accuracy is ±(10+1×10 -6 ×D) mm, elevation ±(20+1×10 -6 ×D)mm, D is the baseline length in km.

[0099] The specific implementation process of this system is as follows Figure 2 As shown:

[0100] S1. Obtain Beidou reference station;

[0101] S2, construct Delauany triangulation and improve the reference mesh shape based on the railway strip features;

[0102] S3, perform automated data quality analysis and dynamic networking;

[0103] S4, performing precise point positioning using adaptive factor weighting filtering and adaptive robust Kalman filtering;

[0104] S5. Perform post-calculation on the encrypted points, wherein regular measurement data needs to be obtained for the encrypted points;

[0105] S6. After settlement, the Beidou benchmark 5mm accuracy is maintained;

[0106] S7. Complete high-precision carrier phase differential positioning measurement based on Beidou.

[0107] Obviously, the above embodiments are merely examples for clarity of explanation and are not intended to limit the implementation methods. Those skilled in the art will readily appreciate that other variations or modifications based on the above descriptions are possible. It is not necessary and impossible to enumerate all implementation methods here. Obvious variations or modifications arising therefrom remain within the scope of protection of the present invention.

Claims

1. A self-sustaining Beidou ground-based augmentation system for conventional railways, characterized in that: Including base station network subsystem, data center subsystem, data communication subsystem and user application subsystem; The reference station network subsystem uses the Delaunay triangulation method that takes into account the railway strip characteristics to construct the Beidou reference station network; The data center subsystem obtains BeiDou data from the reference station network subsystem, performs automated data quality analysis and dynamic networking on the acquired BeiDou data, and utilizes a combination of baseline solution based on adaptive factor weighted filtering and adaptive robust Kalman filtering (PPP) to dynamically maintain and update the reference network. The following steps are included: S301. Based on the adaptive factor weighted filtering method, the double-difference tropospheric delay and ionospheric delay are estimated as parameters. A baseline solution is performed with a one-day cycle. The baseline solution results are subjected to network adjustment. The post-test variance and unit weighted mean error of the observation values ​​are calculated using the network adjustment results of the day. The coordinates of the known points are introduced to obtain the constrained adjustment coordinates (X, Y, Z). line ; S302. Use the BeiDou satellite precise point positioning (PPP) algorithm and the robust adaptive Kalman filter method to solve the PPP single-day solution. Use the daily PPP single-day solution coordinates as the prior coordinates and substitute them into the next day's single-day solution as the initial filter value. Automatically perform PPP coordinate statistics using a 7-day window, and take the average of the daily PPP coordinates for the 7 days before 0:00 on the current day as the PPP weekly solution (X, Y, Z) for the current day. PPP ; S303, (X, Y, Z) PPP AND (X, Y, Z) line Processing: When the difference between the two coordinate components is less than 5mm, the average value of the coordinates is taken as the final coordinates of the day (X, Y, Z) day When the coordinate component difference is greater than or equal to 5mm, the coordinate with the smaller 3D residual value is used as the final coordinate (X, Y, Z) of the day. day , ultimately achieving automatic maintenance and updating of the BeiDou reference network with a one-day cycle; The data communication subsystem establishes communication between the base station network, data center and user application subsystem through 4G network and cloud virtual LAN technology; The user application subsystem uses the original data and real-time solution data stream of the data center subsystem to provide users with static data and real-time differential services.

2. The conventional railway self-sustaining Beidou ground-based augmentation system according to claim 1 is characterized by: The Delaunay triangulation method taking into account the railway strip characteristics is used to construct the Beidou reference station network according to the following steps: S101, for the point set V composed of Beidou reference stations, the vertical coordinate Y axis is the main point position. Arrange in ascending order and recursively call the split-merge algorithm until a triangulation is generated, dividing the point set V into two subsets with roughly equal number of points. and ; S102, creating Delaunay triangulations in each of the two subsets, optimizing the generated triangulations using a local optimization algorithm, calculating the convex hull of each triangulation, finding the bottom line and top line of the convex hull connecting the two sets, and merging the two triangulations of the two sets upward from the bottom line to the top line of the convex hull. During the merging process, if a vertex in a triangulation is within the circumcircle of another triangle, deleting the Delaunay edge and reconstructing the correct edge using the Delaunay criterion, so that the circumcircle of any triangle does not contain any other point in the set V; S103. Use the Beidou reference stations along conventional railways to construct a Delaunay triangulation. Based on the railway's strip-shaped characteristics and the direction of the railway, start from the first reference point and search for the two nearest reference points for each reference point in sequence to construct a new strip-shaped triangulation, called a "proximity triangulation." S104. The repeated baseline edges of the Delaunay triangulation and the "adjacent triangulation" are used as the final baseline edges. The baseline edges where the two triangulations intersect are compared with the baseline solution residuals, and the baseline edge with the smaller average residual is selected as the final baseline edge. After deleting the baselines that exceed the set threshold distance from all the final baseline edges, the remaining baseline edges form a mesh shape, thereby forming a final mesh shape.

3. The conventional railway self-sustaining Beidou ground-based augmentation system according to claim 1 is characterized by: The data center subsystem performs automated data quality analysis and dynamic networking on the acquired Beidou data. The following processes are included: S201. Determine whether a satellite participates in data solution based on the satellite status in the Beidou data acquired at each reference station, wherein the Beidou data includes a multipath effect value, an ionospheric effect value, a cycle slip ratio, a data integrity rate, and a satellite signal-to-noise ratio; S202. Detect the data within seven days for each base station in the BeiDou data obtained from each base station. The seven-day data detection includes: calculating the average value of the data quality index of each reference station within seven days, and if the multipath effect value, ionospheric effect value, cycle slip ratio, and data integrity rate of the current reference station simultaneously meet the first preset condition, then determining that the current reference station participates in the network; S203. If the Beidou data of the current reference station cannot simultaneously meet the first preset condition during data detection within seven days, a 24-hour data detection is performed. The 24-hour data detection includes real-time calculation of the average value of data quality indicators from 24 hours ago to the current moment. If the multipath effect value, ionospheric effect value, cycle slip ratio, and data integrity rate of the current reference station simultaneously meet the second preset condition, the current reference station is judged to participate in the networking; otherwise, the current reference station is prohibited from participating in the networking and an alarm is issued.

4. The conventional railway self-sustaining Beidou ground-based augmentation system according to claim 3 is characterized by: When judging whether a satellite participates in data solution according to the satellite status, the following method is used: For satellites with an elevation angle below 35 degrees, when the L1 and L2 SNRs are less than 40, the satellite is prohibited from participating in the solution; for satellites with an elevation angle above 35 degrees, when the L1 and L2 SNRs are less than 35, the satellite is prohibited from participating in the solution.

5. The conventional railway self-sustaining Beidou ground-based augmentation system according to claim 3 is characterized by: The first preset conditions include: multipath effect value mp1<0.5m, mp2<0.65m, mp3<0.65m, ionospheric effect value IOD<200cm / min, cycle slip ratio o / slps>500, and data integrity rate>98%.

6. The conventional railway self-sustaining Beidou ground-based augmentation system according to claim 3 is characterized by: The second preset conditions include: multipath effect value mp1<0.55m, mp2<0.70m, mp3<0.70m, ionospheric effect value IOD<400cm / min, cycle slip ratio o / slps>200, and data integrity rate>95%.

7. The conventional railway self-sustaining Beidou ground-based augmentation system according to claim 1 is characterized by: The robust adaptive Kalman filter method for solving the PPP single-day solution includes the following steps: Use IGG3 equivalent weight function to obtain robust equivalent weight and adaptive equivalent weight matrix ,in, is the state prediction value, is the weight matrix of the state prediction value, the adaptive factor The calculation is done using the following formula: Where c=2.5, , is the epoch solution; Using the robust Kalman filter gain matrix ,in is the coefficient matrix; The result of robust adaptive filtering is: Where, is the parameter value to be determined after filtering, is the predicted value of the parameter to be determined, is the gain matrix, is the observation noise vector, is the filter error variance matrix.

8. The conventional railway self-sustaining Beidou ground-based augmentation system according to any one of claims 1 to 7, characterized in that: The base station network subsystem has base stations with a distance of 15-25 km between adjacent points, all of which are located on rooftops or in yards of work areas. It also includes encryption points, which are evenly distributed between base stations at intervals of 5 km. These points are used to regularly collect Beidou observation data manually, maintain the baseline, and provide a control baseline for railway surveys and operations. The manual collection of Beidou observation data includes manual measurement every quarter, conducting third-level control measurements on benchmark points and encrypted points along the benchmark network, and further inspecting the self-maintenance of the Beidou benchmark network. Ultimately, the Beidou benchmark network maintenance accuracy is 5mm.

9. The conventional railway self-sustaining Beidou ground-based augmentation system according to any one of claims 1 to 7, characterized in that: The data communication subsystem consists of three parts: a distributed cloud server network in the data center, a base station communication network from the base station to the data center, and a user communication network broadcast from the data center to the rover users. The data center distributed cloud server network is deployed in a distributed manner; The base station communication network establishes communication with the data center through 4G network and cloud virtual local area network technology to form a virtual local area network to achieve remote access and monitoring of the base station; The user communication network uses the Ntrip protocol to send RTCMv3 differential data to the user in real time through the 4G network.

10. The conventional railway self-sustaining Beidou ground-based augmentation system according to any one of claims 1 to 7, characterized in that: The user application subsystem is divided into a post-service system and a real-time service system; The post-event service system authorizes users to log in to the post-event service website through HTTP, independently select observation data to download, and perform post-event static and dynamic differential analysis; The real-time service system establishes a virtual reference station and a short baseline in real time based on the location sent by the user, and uses the Ntrip network protocol to send RTCMv3 pseudorange / carrier differential information to the user in real time based on the user's decimeter / centimeter level accuracy requirements.

Images