Real-time landslide mass millimeter-level displacement monitoring method based on Beidou No.3 double-frequency observation

Abstract

The invention belongs to the technical field of geological disaster monitoring, and particularly relates to a landslide mass millimeter-level displacement real-time monitoring method based on Beidou No.3 double-frequency observation. According to the method, a GNSS monitoring station is arranged in a landslide mass deformation sensitive area, a differential base station is arranged in a stable bedrock area to construct a short baseline network, and Beidou B1C / B2a double-frequency observation values are synchronously collected. Multi-path errors are suppressed through a hardware path suppression and algorithm correction dual mechanism, a carrier phase observation value is corrected, localized RTK calculation is executed on an embedded platform, common errors are eliminated in combination with a wide and narrow lane combination technology and a double-difference model, and high-precision displacement data are output through self-adaptive dynamic filtering based on a PDOP value. According to the invention, millimeter-level monitoring precision is realized, the anti-interference capability and environmental adaptability are high, early creep deformation of the landslide can be accurately captured, and reliable support is provided for disaster early warning.

Description

Technical Field

[0001] This invention belongs to the field of geological disaster monitoring technology, and in particular relates to a method for real-time monitoring of millimeter-level displacement of landslide bodies based on BeiDou-3 dual-frequency observation. Background Technology

[0002] Landslides pose a serious threat to people's lives and property, and existing monitoring technologies have multiple limitations. Traditional manual monitoring methods, such as total stations and rangefinders, require on-site personnel, are constrained by inclement weather and complex terrain, and are typically limited to once a day. They are also unable to capture transient deformation signals and pose significant risks when operating in high-risk environments. While sensor network solutions, such as inclinometers and crack gauges, can achieve automated monitoring, they can only obtain local point information and cannot comprehensively reflect the overall displacement trend of the landslide body. They also face problems such as complex installation and high maintenance costs.

[0003] Satellite navigation technology offers a new approach to solving these problems; however, the currently widely used single-frequency GNSS systems still face severe challenges in landslide monitoring. Ionospheric delay significantly amplifies positioning errors under specific space weather conditions, and multipath effects are particularly pronounced in complex mountainous environments, causing signal disturbances that make it difficult to meet millimeter-level monitoring requirements for overall positioning accuracy and effectively capture the critical evolutionary stages of early landslide creep. Although dual-frequency GNSS technology partially overcomes ionospheric errors through multi-frequency combination, existing solutions still have significant drawbacks. Reliance on GPS or GLONASS systems leads to unstable signal coverage in remote mountainous areas; data processing relies on cloud servers, resulting in degraded real-time performance; the effectiveness of multipath suppression mechanisms is limited in complex terrain; and fixed filter parameter designs are difficult to adapt to dynamic environmental changes such as signal obstruction and meteorological interference. These technical bottlenecks severely restrict the early warning capabilities for landslide disasters, urgently requiring the development of new monitoring methods with independent controllability, millimeter-level accuracy, and strong anti-interference characteristics. Summary of the Invention

[0004] To address the technical problems mentioned above, this invention proposes a real-time monitoring method for millimeter-level displacement of landslides based on dual-frequency observation of the BeiDou-3 system.

[0005] To achieve the above objectives, the technical solution adopted by the present invention includes the following steps:

[0006] S1. Deploy GNSS monitoring stations in deformation-sensitive areas on the landslide surface and differential reference stations in geologically stable bedrock areas to construct a short baseline monitoring network;

[0007] S2. The raw observation values ​​of BeiDou frequency points are collected synchronously by the differential reference station and the GNSS monitoring station. The PTP protocol is used to achieve sub-second time synchronization. The dual-frequency pseudorange and carrier phase observation values ​​are obtained synchronously at a sampling rate of 5Hz, and the observation data is transmitted back to the embedded processing platform of the monitoring station in real time.

[0008] S3. Multipath error is suppressed through a dual mechanism of hardware path suppression and algorithm correction, and the carrier phase observation value is corrected.

[0009] The hardware layer uses a 40cm diameter metal damping plate with a 15° edge tilt angle installed at the bottom of the monitoring station antenna to suppress ground reflection signals;

[0010] The algorithm layer is based on satellite elevation angle. The multipath correction amount is calculated dynamically, and the correction amount formula is as follows: The coefficient k is obtained through 24-hour static calibration on-site, with a value range of [0.05, 0.15], and the correction amount is used to correct the carrier phase observation value. ,in For carrier phase observations, For the corrected carrier phase observations;

[0011] S4. Perform localized RTK calculation on the embedded processing platform of the monitoring station, dynamically output the three-dimensional relative displacement, and eliminate common errors based on the double-difference observation model;

[0012] S5. Calculate the instantaneous horizontal displacement from the three-dimensional relative displacement obtained by the localized RTK solution, and perform adaptive dynamic filtering on the displacement sequence formed by the horizontal displacement to obtain stable and reliable high-precision displacement data.

[0013] Preferably, in step S1, the distance between the GNSS monitoring station and the differential reference station is ≤5km. The GNSS monitoring station is equipped with a Beidou-3 B1C and B2a dual-frequency receiver and an anti-multipath choke antenna. The location of the differential reference station meets the requirements of satellite signal obstruction angle <10° and avoids strong electromagnetic interference sources.

[0014] Preferably, step S4 involves performing localized RTK calculations on the embedded processing platform of the monitoring station to dynamically output the three-dimensional relative displacement. The specific steps for eliminating common errors based on the double-difference observation model include:

[0015] S41. Perform localized RTK calculation on the embedded platform of the monitoring station;

[0016] S42. Using the wide-narrow lane combination technique, the observed value for the wide lane combination is: The observed values ​​for the narrow alley combination are: ,in, The frequency of the B1C point, is the frequency of the B2a frequency point; is the corrected carrier phase observation value of the B1C frequency point, is the corrected carrier phase observation value of the B2a frequency point;

[0017] S43. Based on the double-difference observation value, calculate the three-dimensional relative displacement component of the monitoring station relative to the reference station.

[0018] Preferably, the implementation of step S5 for calculating the instantaneous horizontal displacement of the three-dimensional relative displacement obtained by the localized RTK solution and performing adaptive dynamic filtering on the displacement sequence composed of the horizontal displacement to obtain stable, reliable and high-precision displacement data is as follows:

[0019] S51. First, process the horizontal displacement amount based on the dynamic filtering algorithm for positioning quality assessment and calculate the instantaneous horizontal displacement: , where are the relative displacement values of the x-axis and y-axis respectively;

[0020] S52. Obtain the positioning quality factor PDOP value and dynamically adjust the filtering parameters; when PDOP ≤ 2, use moving average filtering; when 2 < PDOP ≤ 5, use Kalman filtering; when PDOP > 5, enable wavelet threshold denoising;

[0021] S53. Finally, perform variable parameter filtering to obtain the displacement , where, is the variable filtering coefficient, is the horizontal displacement sequence obtained after variable parameter filtering and weighted summation processing at the previous moment.

[0022] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0023] 1. High monitoring accuracy: Using the Beidou-3 dual-frequency signal and short-baseline RTK solution, the horizontal displacement monitoring accuracy is improved to the millimeter level, and the early weak creep of landslides can be captured.

[0024] 2. Strong anti-interference ability: By combining the hardware anti-jamming of the choke ring and the multipath correction model, the multipath error in the complex mountain environment is significantly reduced, and the data availability in bad weather is improved.

[0025] 3. Strong environmental adaptability: The adaptive filtering algorithm dynamically adjusts the filtering intensity according to the positioning quality, strengthens noise reduction when the signal is blocked, and retains the true displacement details in the open environment, ensuring the monitoring reliability in different environments. Description of the Drawings

[0026] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a schematic diagram of the structural process of a real-time monitoring method for millimeter-level displacement of landslides based on dual-frequency observation of BeiDou-3. Detailed Implementation

[0028] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described below in conjunction with the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0029] Numerous specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways than those described herein, and therefore the invention is not limited to the specific embodiments disclosed in the following specification.

[0030] In this embodiment, to overcome the problems of existing landslide monitoring technologies, such as inability to achieve millimeter-level accuracy, insufficient real-time performance, weak anti-interference capability, and poor adaptability to complex mountainous environments, this invention provides a real-time monitoring method for millimeter-level displacement of landslide bodies based on BeiDou-3 dual-frequency observation. This method achieves real-time high-precision displacement monitoring by deploying a short-baseline monitoring network, introducing a dual-frequency synchronous observation mechanism, constructing a multi-path suppression model combining hardware and algorithms, implementing localized RTK calculation, and designing an adaptive dynamic filtering algorithm. Specific implementation details are as follows... Figure 1 As shown.

[0031] Firstly, in this embodiment, multiple GNSS monitoring stations are deployed in the deformation-sensitive area of ​​the landslide surface, and differential reference stations are deployed in the bedrock area with stable geological structure and high safety level. The distance between the monitoring stations and the reference stations is limited to ≤5km to ensure high ambiguity fixation rate and stable baseline calculation of BeiDou dual-frequency carrier phase solution. The monitoring stations are equipped with BeiDou-3 B1C / B2a dual-frequency receivers and anti-multipath choke coil antennas, and the location of the reference stations must meet the requirement that the satellite signal obstruction angle is <10°.

[0032] Sub-second time synchronization is achieved by synchronously acquiring raw observation values ​​of BeiDou B1C and B2a frequencies from the base station and monitoring station. Simultaneous acquisition of dual-frequency pseudorange and carrier phase observation values ​​at a 5Hz sampling rate, with PTP protocol used to ensure time synchronization and an error of <1ms. Specifically, the base station and monitoring station are equipped with high-precision receivers supporting both BeiDou-3 B1C and B2a dual-frequency signals. Utilizing their simultaneous acquisition capability of dual-frequency carrier phase and pseudorange observation values, the consistency of observation frequencies is ensured at the physical level. Simultaneously, by deploying the PTP (Precision Time Protocol) mechanism in the inter-station communication network, nanosecond-level timestamp propagation, hardware clock calibration, and periodic comparison of time synchronization signals are achieved, strictly controlling the overall system time synchronization error to within 1ms. With continuous and effective time synchronization, the system acquires raw pseudorange and carrier phase observations at frequencies B1C and B2a in real time using a high sampling frequency of 5Hz. This enables the monitoring system to capture subtle dynamic displacement changes and transient disturbances in the landslide body, and transmits the synchronized observation data back to the embedded processing platform at the monitoring station with millisecond-level delays, achieving real-time continuous calculation of deformation information. This dual-frequency synchronization, high-speed sampling, and precise time synchronization observation mode ensures data timing consistency and significantly improves the stability and reliability of subsequent differential calculations.

[0033] To effectively suppress multipath errors in complex mountainous environments and address the severe impact of ground reflection and multi-source interference on GNSS monitoring accuracy, this embodiment employs a dual-layer suppression mechanism combining hardware path suppression and algorithm correction. The dual mechanism of hardware path suppression and algorithm correction suppresses multipath errors and corrects carrier phase observations. The hardware layer uses a 40cm diameter metal path suppression plate with a 15° edge tilt angle mounted at the bottom of the monitoring station antenna to suppress ground reflection signals. The algorithm layer is based on the satellite elevation angle... The multipath correction amount is calculated dynamically, and the correction amount formula is as follows: The coefficient k is obtained through 24-hour static calibration on-site, with a value range of [0.05, 0.15], and the correction amount is used to correct the carrier phase observation value. ,in For carrier phase observations, To correct the carrier phase observations, hardware suppression reduces interference from strong reflection paths, while dynamic algorithm correction further compensates for the limitations of hardware suppression, achieving carrier phase stability even in environments with low satellite elevation angles and weak signals. This dual mechanism significantly reduces carrier phase observation noise, providing reliable input for millimeter-level displacement calculations.

[0034] Then, in order to achieve fast, stable, and high-precision displacement calculation at the landslide monitoring site and solve the problems of poor real-time performance, dependence on external networks, and susceptibility to communication interference in traditional cloud RTK calculations, the present invention adopts an embedded local RTK calculation architecture to dynamically output three-dimensional relative displacement amounts and eliminate common errors based on the double-difference observation model. Specifically, after the monitoring station receives the synchronized and corrected Beidou B1C and B2a dual-frequency carrier phase observation values, it first constructs a double-difference observation model between the reference station and the monitoring station in the local processing platform using the wide-lane and narrow-lane combination technology. The wide-lane combined observation value is: , and the narrow-lane combined observation value is: , where is the frequency of the B1C frequency point, is the frequency of the B2a frequency point; is the corrected carrier phase observation value of the B1C frequency point, is the corrected carrier phase observation value of the B2a frequency point; based on the double-difference observation value, the three-dimensional relative displacement components of the monitoring station relative to the reference station are calculated, and common errors such as satellite clock error, receiver clock error, and ionospheric delay are eliminated, thereby significantly improving the purity of the carrier phase observation value.

[0035] Finally, the instantaneous horizontal displacement is calculated from the three-dimensional relative displacement amount obtained by the local RTK calculation, and an adaptive dynamic filtering process is performed on the displacement sequence formed by the horizontal displacement to obtain stable and reliable high-precision displacement data. Its implementation is as follows: First, the horizontal displacement amount is processed based on a dynamic filtering algorithm for positioning quality evaluation to calculate the instantaneous horizontal displacement: , where are the relative displacement values of the x-axis and y-axis respectively; the positioning quality factor PDOP value is obtained, and the filtering parameters are dynamically adjusted; when PDOP ≤ 2, moving average filtering is used; when 2 < PDOP ≤ 5, Kalman filtering is used; when PDOP > 5, wavelet threshold denoising is enabled; finally, variable parameter filtering is performed to obtain the displacement , where is the variable filtering coefficient, is the horizontal displacement sequence obtained after variable parameter filtering and weighted summation processing at the previous moment. Specifically, the system first extracts the plane components from the displacement components in the x, y, and z directions obtained by RTK solution, and constructs the instantaneous horizontal displacement for the x and y displacements. This instantaneous horizontal displacement reflects the dynamic changes of the landslide body in the plane direction and is an important basic quantity for judging the weak creep and sudden displacement of the landslide. On this basis, the system introduces the positioning quality factor PDOP as the core adaptive evaluation index of the filtering strategy. Since PDOP can reflect the impact of the current geometric distribution on the positioning accuracy, it is suitable as the basis for dynamically adjusting the filtering strength. When PDOP ≤ 2, it indicates good distribution and high observation accuracy, and the system uses moving average filtering to slightly smooth the instantaneous horizontal displacement to retain the true displacement details to the greatest extent. When 2 < PDOP ≤ 5, the stability of the observed data decreases, and the system enables Kalman filtering, comprehensively considering the historical displacement state and the current observation value, while suppressing random fluctuations and maintaining the response speed. When PDOP > 5, the structure of the observed data deteriorates significantly, the carrier observation noise and the risk of jumps increase, and the system switches to wavelet threshold denoising. By decomposing and threshold suppressing the displacement sequence, it effectively filters out the sudden noise peaks and prevents incorrect displacement information from entering the subsequent monitoring model. After completing the above multi-strategy filtering, the system performs variable parameter filtering on the smoothed displacement values to ensure the continuity and stability of the sequence. The variable parameter filtering is carried out recursively and adjusted in real time according to PDOP and the displacement change rate to optimize the filtering balance point under different environmental conditions. is the output of variable parameter filtering and weighted summation at the previous moment, which is used to strengthen the sequence coherence and reduce the jumps caused by measurement fluctuations. Through this chain-type adaptive dynamic filtering mechanism, the ability to stably output high-precision displacement results under the background of complex terrain, weak signals, and meteorological interference is achieved, providing a reliable data basis for early warning of landslides.

[0036] The above are only the preferred embodiments of the present invention, and are not limitations of the present invention in other forms. Any person skilled in the art may use the disclosed technical content to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change, and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A method for real-time monitoring of millimeter-level displacement of landslides based on dual-frequency observation of BeiDou-3, characterized in that, The steps include: S1. Deploy GNSS monitoring stations in the deformation-sensitive areas on the surface of the landslide body, and deploy differential reference stations in the bedrock area with stable geological structures to construct a short-baseline monitoring network; S2. Synchronously collect the original observation values of the Beidou frequency points through the differential reference stations and GNSS monitoring stations, implement sub-second-level time synchronization using the PTP protocol, synchronously obtain dual-frequency pseudorange and carrier phase observation values at a sampling rate of 5 Hz, and transmit the observation data back to the embedded processing platform of the monitoring station in real time; S3. Suppress the multipath error through a dual mechanism of hardware multipath suppression and algorithm correction, and correct the carrier phase observation values; At the hardware layer, a metal multipath suppression plate with a diameter of 40 cm and an edge inclination angle of 15° is installed at the bottom of the antenna of the monitoring station to suppress the ground reflection signal; The algorithm layer is based on satellite elevation angle. The multipath correction amount is calculated dynamically, and the correction amount formula is as follows: The coefficient k is obtained through 24-hour static calibration on-site, with a value range of [0.05, 0.15], and the correction amount is used to correct the carrier phase observation value. ,in For carrier phase observations, For the corrected carrier phase observations; S4. Perform local RTK solution at the embedded processing platform of the monitoring station, dynamically output the three-dimensional relative displacement, and eliminate the common error based on the double-difference observation model; S5. Calculate the instantaneous horizontal displacement of the three-dimensional relative displacement obtained by the local RTK solution, and perform adaptive dynamic filtering on the displacement sequence composed of the horizontal displacement to obtain stable and reliable high-precision displacement data.

2. The method for real-time monitoring of millimeter-level displacement of landslides based on BeiDou-3 dual-frequency observation as described in claim 1, characterized in that, In step S1, the distance between the GNSS monitoring station and the differential reference station is ≤ 5 km. The GNSS monitoring station is equipped with a Beidou-3 B1C and B2a dual-frequency receiver and an anti-multipath choke antenna. The location selection of the differential reference station satisfies that the satellite signal occlusion angle < 10° and avoids strong electromagnetic interference sources.

3. The method for real-time monitoring of millimeter-level displacement of landslides based on BeiDou-3 dual-frequency observation as described in claim 1, characterized in that, The specific steps of step S4 for performing local RTK solution at the embedded processing platform of the monitoring station, dynamically outputting the three-dimensional relative displacement, and eliminating the common error based on the double-difference observation model include: S41. Perform local RTK solution at the embedded platform of the monitoring station; S42. Using the wide-narrow lane combination technique, the observed value for the wide lane combination is: The observed values ​​for the narrow alley combination are: ,in, The frequency of the B1C point, This refers to the frequency of point B2a. The corrected carrier phase observation value for the B1C frequency point. The corrected carrier phase observation value for frequency B2a; S43. Based on the double-difference observations, solve the three-dimensional relative displacement components of the monitoring station relative to the reference station.

4. The method for real-time monitoring of millimeter-level displacement of landslides based on BeiDou-3 dual-frequency observation as described in claim 1, characterized in that, The implementation of step S5 for calculating the instantaneous horizontal displacement of the three-dimensional relative displacement obtained by the local RTK solution and performing adaptive dynamic filtering on the displacement sequence composed of the horizontal displacement to obtain stable and reliable high-precision displacement data is as follows: S51. First, the horizontal displacement is processed based on the dynamic filtering algorithm of the positioning quality assessment to calculate the instantaneous horizontal displacement: ,in These are the relative displacement values ​​along the x-axis and y-axis, respectively. S52. Obtain the positioning quality factor PDOP value and dynamically adjust the filtering parameters; when PDOP ≤ 2, use moving average filtering; when 2 < PDOP ≤ 5, use Kalman filtering; when PDOP > 5, enable wavelet threshold denoising; S53. Finally, variable parameter filtering is performed to obtain the displacement. ,in, For variable filter coefficients, This is the horizontal displacement sequence obtained from the previous time step after variable parameter filtering and weighted summation.

Images