Landslide prediction method and system based on GNSS receiver

Abstract

The invention relates to the technical field of satellite ground equipment and the like, and provides a landslide prediction method and system based on a GNSS receiver, and the receiver synchronously obtains a three-frequency carrier signal, constructs a non-combined double-difference model, and carries out the hierarchical processing of a troposphere dry component and a zenith wet component; and two groups of ultra-wide lane ambiguity are generated based on original observation, cascade rounding is performed according to the length of the wavelength after a noise threshold is met, and a whole cycle result is output. And writing a whole-cycle result into integer linear constraints, and reducing the dimension of the narrow lane ambiguity into one dimension and thoroughly removing an ionosphere item in cooperation with geometric correlation and ionosphere-eliminating equivalent transformation. Baseline displacement, zenith moisture content and residual narrow lane ambiguity are used as states, extended Kalman filtering is implemented by combining an elevation angle weighted noise matrix, a narrow lane floating point solution is obtained, and displacement is updated after a fixed integer is checked through an LAMBDA algorithm and a ratio. And finally, the displacement rate and the hourly rainfall are synthesized into a risk index according to historical weights, a landslide early warning is output when the double indexes exceed the threshold, and high-reliability risk judgment of the landslide is realized.

Description

Technical Field

[0001] The present invention relates to the technical fields of satellite ground equipment, electronic digital data processing, and the like, and in particular to a landslide prediction method and system based on a GNSS receiver. Background Art

[0002] Early prediction of landslides relies on centimeter-level or even millimeter-level monitoring of tiny surface deformations. Traditional single-frequency or dual-frequency GNSS real-time dynamic positioning is easily limited by factors such as ionospheric delay, tropospheric wet component and insufficient ambiguity fixation rate, and it is difficult to work stably for a long time under medium-long baseline conditions. As systems such as Beidou and GPS begin to broadcast three-frequency signals, the combination of multi-frequency ultra-wide lanes provides an opportunity to improve the reliability of rapid ambiguity fixation; however, existing methods often separate the ultra-wide lane and narrow lane solutions, and do not fully utilize the integer results to reduce the dimension of the narrow lane model, resulting in limited solution efficiency and success rate. In addition, when the ionospheric activity is strong, the traditional geometric correlation model needs to introduce ionospheric parameters, and the model error is difficult to match accurately; although the geometrically independent model can offset the geometric terms, it loses the ability to directly estimate the baseline vector. To this end, the existing technology proposes to merge the geometric correlation and the ionospheric elimination target into an equivalent transformation, but there is still a lack of solutions that integrate the ultra-wide lane integer constraints, Kalman filtering and rainfall triggering mechanisms.

[0003] In view of the above shortcomings, there is an urgent need for a multi-frequency GNSS landslide prediction method that can estimate the baseline displacement in real time, suppress the ionospheric and tropospheric errors, and give a landslide risk judgment in combination with rainfall information while ensuring high-reliability fixation of the ambiguity. Summary of the invention

[0004] In view of the deficiencies in the above-mentioned prior art, the present invention provides a landslide prediction method and system based on a GNSS receiver, so as to achieve high-reliability fixation of ambiguity, estimate baseline displacement in real time, suppress ionosphere and troposphere errors, and give a landslide risk judgment in combination with rainfall information.

[0005] In a first aspect, the present invention provides a landslide prediction method based on a GNSS receiver, comprising: A three-frequency non-combined double-difference observation model is established according to the three-segment carrier frequency signals received by the receiver, and an original observation data set including baseline displacement, tropospheric delay, ionospheric delay and three-dimensional narrow lane ambiguity is obtained; the tropospheric delay includes a dry component model value and a zenith wet component to be estimated; two sets of ultra-wide lane ambiguities are combined using the original observation data set, and floating-point estimation values ​​and variances of the two sets of ultra-wide lane ambiguities are calculated; if the equivalent carrier noise corresponding to the variance of the floating-point estimation value is lower than a preset threshold, the ultra-wide lane ambiguity is cascaded and rounded according to the wavelength from long to short, and the ultra-wide lane ambiguity integer results that have passed the inspection are output to ensure that the overall fixation success rate meets the system requirements; The integer result of the ultra-wide lane ambiguity is written into the integer linear constraint and back-substituted into the three-frequency non-combined double-difference observation model, and the three-dimensional narrow lane ambiguity to be estimated is reduced to one dimension; geometric correlation and ionospheric elimination equivalent transformation are applied to the three-frequency non-combined double-difference observation model, and the ionospheric delay is eliminated by using the integer linear constraint to obtain a simplified observation model containing only baseline displacement, tropospheric zenith wet component and a single narrow lane ambiguity; the baseline displacement, zenith wet component and residual narrow lane ambiguity are used as state quantities, combined with the height angle weighted observation noise covariance matrix, and the extended Kalman filter model is used to recursively obtain the narrow lane ambiguity floating point solution; The narrow lane ambiguity floating-point solution is input into the LAMBDA algorithm for narrow lane ambiguity integer fixation, and is tested according to the fixing strategy based on ratio test. If it passes, the integer solution is accepted and the baseline displacement is updated. The displacement rate is calculated based on the updated baseline displacement, and the hazard index is synthesized with the hourly rainfall provided by the rainfall monitor according to the historical calibration weight. When the displacement rate and the hazard index exceed their respective thresholds at the same time, a landslide warning sign is generated.

[0006] In a second aspect, the present invention provides a landslide prediction system based on a GNSS receiver, comprising: A rainfall monitor, arranged at one end of the top crossbar of the mounting bracket, is used to monitor the hourly rainfall at the landslide monitoring location; A receiver, arranged in the middle of the top crossbar, connected to and communicating with the rainfall monitor, and running the landslide prediction method based on the GNSS receiver according to any one of the above claims; A lightning rod, arranged at the other end of the top crossbar and higher than the rainfall monitor and the receiver; A GNSS displacement monitoring station is arranged in the middle of the mounting bracket and is connected to communicate with the receiver; A solar power supply system is arranged in the middle of the mounting bracket and above the GNSS displacement monitoring station to form a rain shield, and is used to supply power to the GNSS displacement monitoring station, the receiver and the rainfall monitor.

[0007] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a landslide prediction method and system based on a GNSS receiver. A three-frequency non-combined double-difference observation model is established according to three carrier frequency signals received by the receiver to obtain an original observation data set including baseline displacement, tropospheric delay, ionospheric delay and three-dimensional narrow lane ambiguity, wherein the tropospheric delay includes a dry component model value and a zenith wet component to be estimated. Two groups of ultra-wide lane ambiguities are combined using the original observation data set, and floating-point estimation values ​​and variances of the two groups of ultra-wide lane ambiguities are calculated. If the equivalent carrier noise corresponding to the variance of the floating-point estimation value is lower than a preset threshold, the ultra-wide lane ambiguity is cascaded and rounded according to the wavelength from long to short, and an ultra-wide lane ambiguity integer result that has passed the inspection is output to ensure that the overall fixation success rate meets the system requirements. The ultra-wide lane ambiguity integer result is written into an integer linear constraint and back-substituted into the three-frequency non-combined double-difference observation model, and the three-dimensional narrow lane ambiguity to be estimated is reduced to one dimension. A geometric phase is applied to the three-frequency non-combined double-difference observation model. The invention relates to an observation model of the present invention, which is a simplified observation model of the present invention. The present invention relates to an observation model of the present invention, which is a method for determining the displacement of the narrow lane ambiguity and the ionosphere ... BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their description are used to explain the present invention and do not constitute an improper limitation of the present invention. Some specific embodiments of the present invention will be described in detail in an exemplary and non-restrictive manner with reference to the drawings. The same reference numerals in the drawings indicate the same or similar components or parts. It should be understood by those skilled in the art that these drawings are not necessarily drawn to scale. In the drawings: Figure 1 It is a flow chart of a landslide prediction method based on a GNSS receiver according to an embodiment of the present invention; Figure 2 It is a schematic diagram of the architecture of a landslide prediction system based on a GNSS receiver according to an embodiment of the present invention; Figure 3It is a structural schematic diagram of a receiver according to an embodiment of the present invention.

[0009] Description of reference numerals: 1. Rainfall monitor; 2. Mounting bracket; 20. Top crossbar; 3. Receiver; 4. Lightning rod; 5. GNSS displacement monitoring station; 6. Solar power supply system. DETAILED DESCRIPTION

[0010] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only an embodiment of a part of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of the present invention.

[0011] Embodiment 1 See also Figure 1-Figure 3 This embodiment provides a landslide prediction method based on a GNSS receiver, comprising the following steps: S101. A three-frequency non-combined double-difference observation model is established according to three carrier frequency signals received by the receiver to obtain an original observation data set including baseline displacement, tropospheric delay, ionospheric delay and three-dimensional narrow lane ambiguity; the tropospheric delay includes a dry component model value and a zenith wet component to be estimated; two sets of ultra-wide lane ambiguities are combined using the original observation data set and floating-point estimation values ​​and variances of the two sets of ultra-wide lane ambiguities are calculated; if the equivalent carrier noise corresponding to the variance of the floating-point estimation value is lower than a preset threshold, the ultra-wide lane ambiguity is cascaded and rounded according to the wavelength from long to short and the ultra-wide lane ambiguity integer result that has passed the inspection is output to ensure that the overall fixation success rate meets the system requirements; S102, writing the integer result of the ultra-wide lane ambiguity into the integer linear constraint and back-substituting it into the three-frequency non-combined double-difference observation model, reducing the dimension of the three-dimensional narrow lane ambiguity to be estimated to one dimension; applying geometric correlation and ionospheric elimination equivalent transformation to the three-frequency non-combined double-difference observation model, and using the integer linear constraint to eliminate the ionospheric delay, to obtain a simplified observation model containing only baseline displacement, tropospheric zenith wet component and a single narrow lane ambiguity; using the baseline displacement, zenith wet component and the remaining narrow lane ambiguity as state quantities, combined with the height angle weighted observation noise covariance matrix, using the extended Kalman filter model to recursively obtain the narrow lane ambiguity floating point solution; S103, inputting the narrow lane ambiguity floating point solution into the LAMBDA algorithm to fix the narrow lane ambiguity integer, and performing the test according to the fixing strategy based on the ratio test. If it passes, the integer solution is accepted and the baseline displacement is updated. The displacement rate is calculated based on the updated baseline displacement, and the hazard index is synthesized with the hourly rainfall provided by the rainfall monitor according to the historical calibration weight. When the displacement rate and the hazard index exceed their respective thresholds at the same time, a landslide warning mark is generated.

[0012] It should be noted that in the prior art, if single-frequency or dual-frequency GNSS positioning is used for medium-long baseline landslide monitoring, insufficient correction of ionospheric delay and tropospheric wet component will amplify the positioning error; the success rate of ambiguity fixation is reduced due to multipath and noise, resulting in the inability to continuously obtain centimeter-level displacement. In addition, in the prior art, multi-frequency methods often separate the ultra-wide lane fixation, narrow lane fixation, and rainfall triggering, resulting in long links and poor real-time performance. In this embodiment, a three-frequency non-combined double-difference observation model is used to retain the geometric information of the satellite and the receiver at the data level, and the tropospheric dry component and the zenith wet component are layered and modeled to lay the foundation for controllable atmospheric errors; then two groups of ultra-wide lanes are combined and the variance threshold is used as the noise criterion to complete the cascade rounding, and the most easily fixed long wavelength information is first converted into a high-confidence integer prior; then the whole cycle results are written into the integer linear constraints and injected into the original model together with the geometric correlation and ionospheric elimination equivalent transformation, so as to achieve the simultaneous completion of the three goals of ambiguity reduction, ionospheric elimination, and geometric preservation, and eliminate the ambiguity. In addition to the rank deficiency risk of traditional step-by-step calculations, the extended Kalman filter performs joint recursion on the baseline displacement, zenith wet component and single narrow lane ambiguity, so that the wet delay can be updated in real time with meteorological changes; the LAMBDA algorithm is then used in conjunction with the ratio test to complete the narrow lane integer fixation and update the baseline; finally, the displacement rate and hourly rainfall are combined into a hazard index according to the historical calibration weights, and the dual indicators simultaneously exceed the threshold to trigger an early warning, thereby ensuring that the ambiguity is highly reliable fixed, while being able to estimate the baseline displacement in real time, suppress ionosphere and troposphere errors, and give a landslide risk judgment based on rainfall information.

[0013] In some preferred embodiments, the establishment of the three-frequency non-combined double-difference observation model includes: obtaining the original carrier phase observation and pseudo-range observation at each frequency according to the three carrier frequency signals received by the receiver; performing intra-station satellite-to-satellite single-difference calculation on the original carrier phase observation and pseudo-range observation to eliminate the receiver clock error and obtain a single-difference observation; performing inter-station single-difference calculation on the single-difference observation to further eliminate the satellite clock error and obtain a double-difference observation, which retains the geometric distance information between the satellite and the receiver and serves as the original observation data set. It should be noted that in this embodiment, the original carrier phase observation and pseudo-range observation are extracted according to the frequency points to ensure that the subsequent differential operation has a complete information basis; the intra-station single-difference operation performs a difference operation on different satellite observations within the same receiver to instantly eliminate the receiver clock error and avoid the oscillation noise of the receiver itself from polluting the subsequent ambiguity estimation; the single-difference observation that has been de-clocked is then subjected to inter-station single-difference, which can synchronously offset the satellite clock error and achieve strict time synchronization between two stations. The double-difference observations after two-level difference still retain the geometric distance information between the satellite and the receiver, which can provide a complete geometric column vector for the subsequent geometric correlation and ionospheric elimination equivalent transformation, so that the baseline displacement can directly participate in the extended Kalman filter solution. Through this embodiment, the original observation data set is completely stripped of clock error noise at the input end, thereby providing high signal-to-noise ratio and geometrically complete basic data for subsequent links (ultra-wide lane rounding, ambiguity dimension reduction, ionospheric elimination), improving the rapid and reliable fixation of ambiguities and baseline accuracy from the source.

[0014] In some preferred embodiments, the tropospheric delay is modeled in a hierarchical manner: the dry component model value is calculated in advance through a standard atmospheric model and field-measured meteorological data, and remains fixed within the epoch, while the zenith wet component is estimated in real time as a random variable and dynamically updated and then added to the extended Kalman filter model to recursively obtain the narrow lane ambiguity floating-point solution. It should be noted that in this embodiment, the dry component is calculated once using a standard atmospheric model combined with field meteorological data, and remains constant throughout the epoch to ensure that the stable and predictable part of the troposphere is fully compensated; the rapidly changing zenith wet component is written into the extended Kalman filter model as a random variable, and is estimated and dynamically updated in real time through random walk noise, so that the model can sensitively respond to sudden increases and decreases in wet delay, thereby avoiding the systematic deviation caused by the traditional method of treating the wet component as a constant, and avoiding parameter redundancy caused by repeated estimation of the dry component, thereby improving model observability and filtering convergence speed. The estimated value of the wet component output by the hierarchical modeling is recursively derived together with the baseline displacement and narrow lane ambiguity inside the filter, which significantly reduces the impact of the wet delay error on ambiguity fixation and displacement estimation, and fundamentally improves the accuracy and reliability of early landslide prediction.

[0015] In some preferred embodiments, the two sets of ultra-wide lane ambiguities select a combination with an effective wavelength greater than the original carrier wavelength to improve the noise resistance of the observed quantity and reduce the ambiguity estimation error during rounding; the calculation of the floating-point estimation values ​​of the two sets of ultra-wide lane ambiguities is subjected to noise suppression by multi-epoch smoothing and averaging. It should be noted that in this embodiment, by combining two sets of ultra-wide lanes with an effective wavelength greater than the original carrier wavelength, the carrier wavelength can be magnified to the meter level, significantly improving the noise resistance of the observed quantity, and greatly reducing the impact of the same noise level on the rounding success rate; at the same time, the two sets of combinations are linearly independent of each other, and can provide redundancy for cascade fixation. In addition, when calculating the ultra-wide lane floating-point estimation value, multi-epoch smoothing and averaging are introduced to expand the instantaneous observation errors in the time domain and offset each other, thereby reducing the variance and improving the ratio test pass rate. The long wavelength design can solve the noise amplification problem, and multi-epoch smoothing can solve the instantaneous fluctuation problem. The superposition of the two makes the ultra-wide lane integer prior reach high credibility, laying a reliable foundation for the subsequent narrow lane dimensionality reduction and integer fixation, and effectively solving the problem of limited fixation success rate caused by the separation of ultra-wide lanes and narrow lanes.

[0016] In some preferred embodiments, when performing ultra-wide lane ambiguity cascade rounding, the preset threshold of the equivalent carrier noise is a threshold calibrated in advance based on historical landslide monitoring experience data; when the variance of the floating-point estimate is higher than the preset threshold of the equivalent carrier noise, the delayed smoothing strategy of the observed quantity is automatically enabled, the number of historical observation epochs is increased, and the floating-point estimate is recalculated. It should be noted that the threshold is calibrated offline by long-term landslide monitoring historical data, which can reflect the upper limit of the noise of this site under normal working conditions; once the real-time floating-point variance exceeds the limit, the delayed smoothing is automatically started, the number of epochs is increased to re-estimate the floating-point value, and the rounding is performed after the noise is suppressed, so that the whole-week fixation is always carried out in a controllable noise environment, which not only maintains a high fixation success rate, but also avoids false fixation caused by temporary environmental deterioration.

[0017] In some preferred embodiments, when writing the integer results of ultra-wide lane ambiguity into integer linear constraints, the integer linear constraints use a transformation matrix that can reduce the ambiguity dimension from three dimensions to one dimension, and the transformation matrix is ​​directly constructed from the integer results of ultra-wide lane ambiguity. It should be noted that in this embodiment, the ultra-wide lane integer results are directly written into integer linear constraints to construct a transformation matrix that can reduce the three-dimensional narrow lane ambiguity to one dimension, so as to achieve dimensionality reduction and simultaneous injection of integer priors, reduce the state quantity of the extended Kalman filter, improve real-time performance, significantly reduce the variance of the remaining narrow lane ambiguity, improve the pass rate of ratio test, and effectively solve the problem of separation of ultra-wide lane and narrow lane solution and limited efficiency.

[0018] In some preferred embodiments, the geometric correlation and ionospheric elimination equivalent transformation are realized by a one-time overall linear transformation, and two conditions are satisfied simultaneously during the transformation process: the ionospheric delay in the observation model is completely eliminated, while the geometric relationship between the satellite and the receiver is retained. It should be noted that when the ionospheric activity is strong, if the ionospheric parameters are explicitly estimated in the observation model, new strongly correlated unknowns will be introduced; if a geometrically independent combination is used, the baseline estimation capability will be lost. In this embodiment, by clearly satisfying the two conditions of geometric correlation and ionospheric elimination through a one-time overall linear transformation, the ionospheric delay is completely eliminated, and the geometric relationship between the satellite and the receiver is completely retained, avoiding the rank deficiency and numerical noise amplification caused by multiple combinations, so that the baseline displacement can still be directly solved, and the accuracy is not degraded due to ionospheric activity, which effectively solves the problem that ionospheric processing and geometric information are difficult to obtain at the same time.

[0019] In some preferred embodiments, the state variables of the extended Kalman filter model include baseline displacement, tropospheric zenith wet component and single narrow lane ambiguity after dimensionality reduction, and the observation noise covariance matrix is ​​determined by an adaptive weighting method related to the satellite elevation angle to adjust the weights of different satellite observations in real time. It should be noted that the landslide monitoring area is often blocked, and the multipath and low elevation signal noise is large. If a fixed weight is given to all satellite observations, the noise of low-quality observations will be introduced into the state solution. In this embodiment, an adaptive weighting strategy related to the satellite elevation angle is adopted in the extended Kalman filter, and the weight of high-elevation satellites is high and the weight of low-elevation satellites is low, which reflects the signal quality in real time; at the same time, the state quantity only retains the baseline displacement, the zenith wet component and the single narrow lane ambiguity after dimensionality reduction, ensuring observability and computational efficiency, thereby enhancing the robustness of the filter to environmental changes and ensuring that the centimeter-level displacement solution can still be stably output under complex mountain conditions.

[0020] In some preferred embodiments, the LAMBDA algorithm performs integer fixation of narrow lane ambiguity. When the fixation strategy based on the ratio test is used for the test, it includes: comparing the residual ratio of the best integer candidate solution and the suboptimal integer candidate solution with a preset threshold to determine whether the fixation is successful; when the ratio is higher than the preset threshold, the integer solution is confirmed to be a reliable solution, otherwise the narrow lane ambiguity floating point solution of the next epoch is continued to be recursively deduced. It should be noted that the wrong fixation of narrow lane ambiguity will directly pollute the displacement result and trigger a false alarm. In this embodiment, the LAMBDA algorithm is used in combination with the ratio test: the reliability of the fixation is determined by comparing the residual ratio of the best and suboptimal integer candidates with the threshold, and the threshold is reversed by the preset failure probability, which not only gives the upper bound of the quantitative risk, but also automatically maintains the floating point state recursion when the ratio is insufficient, so as to strictly control the probability of wrong fixation within an acceptable range. At the same time, it can quickly pass the test in a small variance environment after the ultra-wide lane dimensionality reduction, so as to achieve both high fixation rate and low false alarm rate, and effectively solve the problem of insufficient ambiguity fixation rate.

[0021] In some preferred embodiments, the displacement rate calculation after the baseline displacement is updated is performed by dividing the displacement change by the time interval of adjacent epochs, and further processed by multi-epoch sliding window averaging to eliminate random fluctuations in the calculation results of a single epoch. It should be noted that the displacement rate is an important dynamic indicator for landslide triggering, but the single epoch difference is susceptible to occasional noise. In this embodiment, after calculating the displacement rate, a multi-epoch sliding window average is introduced to smooth out random peaks and retain the true deformation trend, thereby reducing the false alarm rate and avoiding misjudgment of the hazard index due to abnormal instantaneous observations; at the same time, the window length can be adaptively set according to the sampling rate, which not only maintains timeliness but also improves stability, and meets the requirements of early warning for continuous and reliable speed thresholds.

[0022] In some preferred embodiments, the hourly rainfall data of the rainfall monitor is sent to the receiver in real time via a wired or wireless communication link. The receiver has a built-in abnormal data detection module, which automatically removes or interpolates the obviously abnormal rainfall data. It should be noted that rainfall data is an important component of the hazard index, but rainfall calculation is susceptible to failures or communication interference. In this embodiment, an abnormal data detection module is configured at the receiver end to automatically remove or interpolate the obviously abnormal or missing rainfall data to ensure that the rainfall sequence is continuous and reasonable, thereby avoiding extremely erroneous data from amplifying the hazard index and causing false alarms, and also ensuring that risk assessment is not interrupted during short-term communication interruptions, effectively solving the problem of lack of quality control in the rainfall trigger link.

[0023] In some preferred embodiments, in the synthesis of the hazard index, the historical calibration weight is an empirical weight obtained based on the statistical regression analysis of long-term monitoring data. When the long-term historical data of the monitoring point changes, the empirical weight is automatically adjusted to adapt to the change in site conditions and reflect the actual landslide risk. It should be noted that the landslide triggering mechanism may change over time due to changes in vegetation, drainage projects, etc., and the fixed weights will gradually become inaccurate. In this embodiment, the historical calibration weights are automatically updated through the statistical regression of long-term monitoring data, so that the hazard index dynamically fits the actual site. When the long-term characteristics of the monitoring point change, the weight can be adaptively adjusted to avoid risk assessment deviations caused by model aging and ensure the accuracy of early warning.

[0024] In some preferred embodiments, when the displacement rate and the hazard index exceed their respective thresholds at the same time, the generated landslide warning mark is uploaded to the remote cloud platform through the GNSS displacement monitoring station connected to the receiver for communication. The remote cloud platform has a multi-level warning push function, which pushes the warning message to management terminals of different levels according to the degree of danger. It should be noted that the warning information must be delivered to the appropriate level in the shortest time. In this embodiment, when the displacement rate and the hazard index both exceed the limit, the warning mark is uploaded to the cloud platform through the GNSS displacement monitoring station, and the cloud platform pushes it to different management terminals according to the hazard level, thereby avoiding information flooding and ensuring that key decision makers receive high-risk alerts as soon as possible. At the same time, the general management terminal only receives necessary information to improve response efficiency.

[0025] Embodiment 2 See also Figure 1-Figure 3 This embodiment provides a landslide prediction system based on a GNSS receiver, comprising: A rainfall monitor 1, which is arranged at one end of a top crossbar 20 of a mounting bracket 2, is used to monitor the hourly rainfall at a landslide monitoring location; A receiver 3 is arranged in the middle of the top crossbar, connected to communicate with the rainfall monitor, and runs the landslide prediction method based on the GNSS receiver; A lightning rod 4, arranged at the other end of the top crossbar and higher than the rainfall monitor and the receiver; A GNSS displacement monitoring station 5 is arranged in the middle of the mounting bracket and is connected to communicate with the receiver; The solar power supply system 6 is arranged in the middle of the mounting bracket and above the GNSS displacement monitoring station to form a rain shield, and is used to supply power to the GNSS displacement monitoring station, the receiver and the rainfall monitor.

[0026] It should be noted that the rainfall monitor is installed at one end of the top crossbar, at the highest point of the bracket and away from the metal antenna, which can avoid shielding and splashing errors; hourly rainfall is collected in real time as an external factor triggering landslides, providing reliable input for the hazard index, which can solve the problem of underreporting of traditional deformation single criterion. The lightning rod is placed at the other end of the crossbar and higher than the rainfall monitor and the receiver, which can provide a lightning drainage channel for the whole machine, reduce the risk of damage to the receiver circuit by direct lightning and induced lightning, and ensure that the monitoring link is uninterrupted during heavy rain. The GNSS displacement monitoring station is arranged at the low center of gravity in the middle of the bracket to reduce the interference of wind load and vibration on the centimeter-level displacement solution; at the same time, the GNSS displacement monitoring station can communicate with the receiver in two directions, and upload the warning signs generated by the receiver to the cloud through the cellular or NB-IoT network to realize remote multi-station collaboration. The solar power supply system is installed above the monitoring station to form a rain shield, which not only ensures the endurance on rainy days, but also enables the rainfall sensor, receiver, and displacement monitoring station to share an independent power supply circuit.

[0027] It should also be noted that when the receiver runs the above-mentioned landslide prediction method based on GNSS receiver, the geometric information of satellite and receiver is retained at the data level by using the three-frequency non-combined double-difference observation model, and the tropospheric dry component and the zenith wet component are layered and modeled to lay the foundation for the controllable atmospheric error; then, two sets of ultra-wide lane combinations are used and the variance threshold is used as the noise criterion to complete the cascade rounding, and the most easily fixed long-wavelength information is first converted into a high-reliability integer prior; then the whole cycle results are written into the integer linear constraint and injected into the original model together with the geometric correlation and the ionospheric elimination equivalent transformation to achieve ambiguity reduction and ionospheric elimination. , geometry maintenance, and the three objectives are completed synchronously to eliminate the rank deficiency risk of traditional step-by-step operations; the extended Kalman filter is used to jointly recursively perform the baseline displacement, zenith wet component and single narrow lane ambiguity, so that the wet delay can be updated in real time with meteorological changes; then the LAMBDA algorithm is used in conjunction with the ratio test to complete the narrow lane integer fixation and update the baseline; finally, the displacement rate and hourly rainfall are combined with the historical calibration weights to synthesize the hazard index, and the double indicators exceed the threshold at the same time to trigger an early warning, thereby ensuring the high reliability of the ambiguity fixation while estimating the baseline displacement in real time, suppressing the ionosphere and troposphere errors, and giving a landslide risk judgment based on the rainfall information.

[0028] It should be pointed out that the above embodiments are only preferred specific implementation modes of the present invention, and the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention. The protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A landslide prediction method based on GNSS receiver, characterized in that: include: A three-frequency non-combined double-difference observation model is established according to the three-segment carrier frequency signals received by the receiver, and an original observation data set including baseline displacement, tropospheric delay, ionospheric delay and three-dimensional narrow lane ambiguity is obtained; the tropospheric delay includes a dry component model value and a zenith wet component to be estimated; two sets of ultra-wide lane ambiguities are combined using the original observation data set, and floating-point estimation values ​​and variances of the two sets of ultra-wide lane ambiguities are calculated; if the equivalent carrier noise corresponding to the variance of the floating-point estimation value is lower than a preset threshold, the ultra-wide lane ambiguity is cascaded and rounded according to the wavelength from long to short, and the ultra-wide lane ambiguity integer results that have passed the inspection are output to ensure that the overall fixation success rate meets the system requirements; The ultra-wide lane ambiguity integer result is written into the integer linear constraint and back-substituted into the three-frequency non-combined double-difference observation model to reduce the dimension of the three-dimensional narrow lane ambiguity to be estimated to one dimension; Applying geometric correlation and ionospheric equivalent transformation to the triple-frequency non-combined double-difference observation model, and using the integer linear constraint to eliminate ionospheric delay, a simplified observation model containing only baseline displacement, tropospheric zenith wet component and single narrow lane ambiguity is obtained; using baseline displacement, zenith wet component and residual narrow lane ambiguity as state quantities, combined with the height angle weighted observation noise covariance matrix, an extended Kalman filter model is used to recursively obtain a narrow lane ambiguity floating point solution; The narrow lane ambiguity floating-point solution is input into the LAMBDA algorithm for narrow lane ambiguity integer fixation, and is tested according to the fixing strategy based on ratio test. If it passes, the integer solution is accepted and the baseline displacement is updated. The displacement rate is calculated based on the updated baseline displacement, and the hazard index is synthesized with the hourly rainfall provided by the rainfall monitor according to the historical calibration weight. When the displacement rate and the hazard index exceed their respective thresholds at the same time, a landslide warning sign is generated.

2. The landslide prediction method according to claim 1, characterized in that: The establishment of the three-frequency non-combined double-difference observation model includes: obtaining original carrier phase observations and pseudo-range observations on each frequency according to three carrier frequency signals received by the receiver; performing intra-station satellite-to-satellite single-difference calculations on the original carrier phase observations and pseudo-range observations to eliminate receiver clock error and obtain single-difference observations; performing inter-station single-difference calculations on the single-difference observations to further eliminate satellite clock error and obtain double-difference observations, which retain the geometric distance information between the satellite and the receiver and serve as the original observation data set.

3. The landslide prediction method according to claim 1, characterized in that: The tropospheric delay is modeled in a layered manner: the dry component model value is calculated in advance using a standard atmospheric model and field measured meteorological data and remains fixed within the epoch, while the zenith wet component is estimated in real time as a random variable and dynamically updated and then added to the extended Kalman filter model to recursively obtain the narrow lane ambiguity floating-point solution.

4. The landslide prediction method according to claim 1, characterized in that: The two sets of ultra-wide lane ambiguities select a combination whose effective wavelength is greater than the original carrier wavelength to improve the noise resistance of the observation and reduce the ambiguity estimation error during rounding; the calculation of the floating-point estimation values ​​of the two sets of ultra-wide lane ambiguities is noise suppressed through multi-epoch smoothing and averaging processing.

5. The landslide prediction method according to claim 1, characterized in that: When performing ultra-wide lane ambiguity cascade rounding, the preset threshold of the equivalent carrier noise is a threshold calibrated in advance based on historical landslide monitoring experience data; when the variance of the floating-point estimate is higher than the preset threshold of the equivalent carrier noise, the delayed smoothing strategy of the observation quantity is automatically enabled, the number of historical observation epochs is increased, and the floating-point estimate is recalculated.

6. The landslide prediction method according to claim 1, characterized in that: When writing the ultra-wide lane ambiguity integer results into integer linear constraints, the integer linear constraints use a transformation matrix that can reduce the ambiguity dimension from three dimensions to one dimension, and the transformation matrix is ​​directly constructed from the ultra-wide lane ambiguity integer results.

7. The landslide prediction method according to claim 1, characterized in that: The geometric correlation and ionospheric elimination equivalent transformation is achieved through a one-time overall linear transformation, and two conditions are met simultaneously during the transformation process: the ionospheric delay in the observation model is completely eliminated, and the geometric relationship between the satellite and the receiver is retained.

8. The landslide prediction method according to claim 1, characterized in that: The state variables of the extended Kalman filter model include baseline displacement, tropospheric zenith wet component and single narrow lane ambiguity after dimensionality reduction, and the observation noise covariance matrix is ​​determined by an adaptive weighting method related to the satellite altitude angle to adjust the weights of different satellite observations in real time.

9. The landslide prediction method according to claim 1, characterized in that: The LAMBDA algorithm performs narrow lane ambiguity integer fixation. When the fixation strategy based on ratio test is used for inspection, it includes: judging whether the fixation is successful by comparing the residual ratio of the best integer candidate solution and the second best integer candidate solution with the preset threshold; when the ratio is higher than the preset threshold, the integer solution is confirmed to be a reliable solution, otherwise the narrow lane ambiguity floating point solution of the next epoch is recursively deduced.

10. A landslide prediction system based on GNSS receiver, characterized in that: include: A rainfall monitor, arranged at one end of the top crossbar of the mounting bracket, is used to monitor the hourly rainfall at the landslide monitoring location; A receiver, arranged in the middle of the top crossbar, connected to and communicated with the rainfall monitor, and running the landslide prediction method based on the GNSS receiver according to any one of claims 1 to 9; A lightning rod, arranged at the other end of the top crossbar and higher than the rainfall monitor and the receiver; A GNSS displacement monitoring station is arranged in the middle of the mounting bracket and is connected to communicate with the receiver; A solar power supply system is arranged in the middle of the mounting bracket and above the GNSS displacement monitoring station to form a rain shield, and is used to supply power to the GNSS displacement monitoring station, the receiver and the rainfall monitor.

Images