Ground surface deformation monitoring method and system based on distributed millimeter wave radar

Abstract

The application discloses a ground surface deformation monitoring method and system based on distributed millimeter wave radar, relates to the technical field of geological disaster monitoring, and comprises the following steps: cross-node closed-loop observation calibration is carried out based on distributed monitoring initialization configuration, phase bias, clock drift and installation angle error are corrected, and a closed-loop self-calibration monitoring frame is obtained; phase timing arrangement is carried out on the closed-loop self-calibration monitoring frame, and propagation environment disturbance separation is carried out in combination with micro-meteorological observation records and stable reference scattering point residuals, so that a disturbance decoupling displacement candidate set is obtained; same-name monitoring point matching and cross-node space-time alignment are carried out according to the disturbance decoupling displacement candidate set, deformation trajectory solving is carried out based on cross-node closed-loop phase self-consistent constraints, and a cross-node fusion deformation time sequence set is obtained; and the application realizes the collaborative correction of phase bias, clock drift and installation angle error by carrying out closed-loop consistency verification on cross-node observation relationship data and carrying out cross-node observation consistency influence evaluation and iterative compensation.

Description

Technical Field

[0001] This invention relates to the field of geological disaster monitoring technology, and in particular to a method and system for monitoring surface deformation based on distributed millimeter-wave radar. Background Technology

[0002] Surface deformation monitoring falls under the intersection of geological disaster monitoring and radar measurement. Existing conventional solutions mostly employ ground-based millimeter-wave radar or multi-point radar to conduct periodic observations of target areas. These observations are combined with unified time synchronization, external parameter calibration, phase timing processing, reference scattering point verification, displacement calculation, and trend interpretation to generate monitoring results. Continuous monitoring and reporting are then conducted in scenarios such as engineering slopes, mining areas, and infrastructure perimeters.

[0003] In the scenario of long-term collaborative operation of distributed nodes, on the one hand, conventional methods have a weak characterization of the coupling effect of phase offset, clock drift and installation angle error, and the consistency of cross-node observations is prone to fluctuation with the running time; on the other hand, conventional methods have limited separation depth between micro-meteorological propagation disturbances and surface deformation phases, which can easily affect the stability of continuous reconstruction of cross-node deformation trajectories and subsequent evolution trend identification. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a method for monitoring surface deformation based on distributed millimeter-wave radar, which solves the problems of fluctuating cross-node observation consistency and insufficient phase separation between propagation environment disturbances and surface deformation, leading to decreased monitoring stability.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for monitoring surface deformation based on distributed millimeter-wave radar, which includes collecting topographic mapping data and risk zoning data and performing initial timing calibration and spatial extrinsic parameter calibration, and obtaining distributed monitoring initialization configuration; Based on the distributed monitoring initialization configuration, cross-node closed-loop observation calibration is performed, and phase offset, clock drift and installation angle error are corrected to obtain closed-loop self-calibration monitoring frames. Phase timing of the closed-loop self-calibration monitoring frames is processed, and propagation environment disturbances are separated by combining micrometeorological observation records and stable reference scattering point residuals to obtain a candidate set of disturbance decoupling displacements; Based on the candidate set of disturbance decoupling displacement, the same-name monitoring points are matched and cross-node spatiotemporal alignment is performed. Then, the deformation trajectory is solved based on the cross-node closed-loop phase self-consistency constraint to obtain the cross-node fused deformation time series set. The evolution trend of surface deformation is identified by cross-node fused deformation time series, and a surface deformation monitoring report is generated.

[0007] In a preferred embodiment of the surface deformation monitoring method based on distributed millimeter-wave radar described in this invention, the steps for obtaining the distributed monitoring initialization configuration are as follows: Based on topographic mapping data, historical geological disaster records, and environmental factors, potential risks are classified, and risk zoning data is generated. A unified clock is established for the nodes of the distributed millimeter-wave radar, and time deviation calibration, spatial position parameter calibration, and attitude parameter calibration are performed between nodes. At the same time, combined with terrain mapping data and risk zoning data, a distributed monitoring initialization configuration is generated.

[0008] As a preferred embodiment of the surface deformation monitoring method based on distributed millimeter-wave radar described in this invention, the terrain mapping data includes elevation information, surface undulation, terrain slope, obstacle distribution, and infrastructure location information of the target area. The risk zoning data includes the risk level and emergency response strategy for each region.

[0009] As a preferred embodiment of the surface deformation monitoring method based on distributed millimeter-wave radar described in this invention, the steps of performing cross-node closed-loop observation calibration based on distributed monitoring initialization configuration, and correcting phase offset, clock drift, and installation angle errors to obtain a closed-loop self-calibration monitoring frame are as follows. Based on the distributed monitoring initialization configuration, cross-node synchronous calibration and observation relationship establishment are performed between the nodes of the distributed millimeter-wave radar to obtain cross-node observation relationship data. For each node in the cross-node observation relationship data, a closed-loop consistency check is performed on the observation phase of the same stable reference scattering point, and the phase offset is iteratively corrected according to the closed-loop residual to generate calibrated cross-node observation relationship data. Based on the calibrated cross-node observation relationship data, the impact of cross-node observation consistency on node clock drift and installation angle error is evaluated and iteratively compensated to obtain a closed-loop self-calibration monitoring frame.

[0010] As a preferred embodiment of the surface deformation monitoring method based on distributed millimeter-wave radar described in this invention, the step of performing phase timing organization on the closed-loop self-calibration monitoring frames refers to sorting and organizing the closed-loop self-calibration monitoring frames according to time identifiers and node identifiers, removing distorted position segments, and generating a phase timing sequence sorted by time identifiers and node identifiers.

[0011] As a preferred embodiment of the surface deformation monitoring method based on distributed millimeter-wave radar described in this invention, the step of separating propagating environmental disturbances by combining micro-meteorological observation records and stable reference scattering point residuals to obtain a disturbance decoupling displacement candidate set is as follows: By aligning the micrometeorological observation records and stable reference scattering point residuals within the same time window according to the phase time sequence, and extracting the synchronization characteristics between temperature and humidity changes and reference point residual changes, information on propagation environment disturbance discrimination is obtained. Based on the propagation environment disturbance discrimination information, the propagation environment disturbance components in the phase time series are identified and stripped, while retaining the phase changes related to surface deformation, to obtain the propagation environment disturbance separation information. The displacement candidate quantity is calculated and the temporal validity is screened for the propagation environment disturbance separation information, and the corresponding node identifier and time identifier are written into it to obtain the disturbance decoupling displacement candidate set.

[0012] As a preferred embodiment of the surface deformation monitoring method based on distributed millimeter-wave radar described in this invention, the steps of matching corresponding monitoring points and cross-node spatiotemporal alignment based on the disturbance decoupling displacement candidate set, and solving for deformation trajectories based on cross-node closed-loop phase self-consistency constraints to obtain a cross-node fused deformation time series set are as follows. The candidate set of disturbance decoupling displacements is aggregated and the correspondence is determined according to the node identifier, time identifier and spatial location association features to generate matching data of monitoring points with the same name; By combining the initial configuration of distributed monitoring with the matching data of monitoring points with the same name, the cross-node observation records corresponding to the monitoring points with the same name are spatiotemporally aligned and the phase self-consistency is verified to generate cross-node closed-loop phase self-consistency constraint data. Deformation trajectory solving and time continuity reconstruction are performed on the cross-node closed-loop phase self-consistent constraint data, and deformation time series records are written according to the monitoring points with the same name to obtain the cross-node fused deformation time series set.

[0013] As a preferred embodiment of the surface deformation monitoring method based on distributed millimeter-wave radar described in this invention, the steps for identifying the surface deformation evolution trend of the cross-node fused deformation time series set are as follows: For the cross-node fused deformation time series set, the time series is sorted and missing segments are marked according to the monitoring points with the same name and the time sequence. The displacement change rate, change acceleration and stage duration information are extracted to generate surface deformation evolution characteristic data. Based on the surface deformation evolution characteristic data, the surface deformation evolution trend is identified for each monitoring point with the same name, and the trends of continuous increase, continuous decrease and phased acceleration are determined to generate surface deformation evolution trend identification information.

[0014] As a preferred embodiment of the surface deformation monitoring method based on distributed millimeter-wave radar described in this invention, the surface deformation monitoring report includes trend determination of monitoring points and description of regional trend changes.

[0015] Secondly, the present invention provides a surface deformation monitoring system based on distributed millimeter-wave radar, including an initialization configuration module, which collects topographic mapping data and risk zoning data and performs initial timing calibration and spatial extrinsic parameter calibration to obtain distributed monitoring initialization configuration; The closed-loop calibration module performs cross-node closed-loop observation calibration based on the distributed monitoring initialization configuration, and corrects phase offset, clock drift and installation angle error to obtain closed-loop self-calibration monitoring frames. The disturbance separation module performs phase timing processing on the closed-loop self-calibration monitoring frames and separates the propagation environment disturbances by combining micro-meteorological observation records and stable reference scattering point residuals, thereby obtaining a candidate set of disturbance decoupling displacements. The trajectory reconstruction module performs matching of corresponding monitoring points and cross-node spatiotemporal alignment based on the candidate set of disturbance decoupling displacements, and performs deformation trajectory solving based on cross-node closed-loop phase self-consistency constraints to obtain a cross-node fused deformation time series set. The trend monitoring module identifies the evolution trend of surface deformation from the cross-node fused deformation time series and generates a surface deformation monitoring report.

[0016] The beneficial effects of this invention are as follows: By performing closed-loop consistency verification on cross-node observation relationship data and conducting cross-node observation consistency impact assessment and iterative compensation, the phase offset, clock drift and installation angle error are collaboratively corrected, enabling the closed-loop self-calibration monitoring frame to form a stable observation basis; by driving the propagation environment disturbance discrimination information to separate propagation environment disturbances and calculate displacement candidate quantities, the extraction of surface deformation-related phase changes and the construction of disturbance decoupling displacement candidate sets are realized, which are used to support deformation trajectory solving and surface deformation evolution trend identification, ultimately improving the continuity, reliability and engineering application value of surface deformation monitoring reports. Attached Figure Description

[0017] 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 only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a surface deformation monitoring method based on distributed millimeter-wave radar.

[0019] Figure 2 This is a schematic diagram of a surface deformation monitoring system based on distributed millimeter-wave radar.

[0020] Figure 3 A flowchart for cross-node closed-loop observation calibration. Detailed Implementation

[0021] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] Many 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 different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0023] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0024] Reference Figures 1-3 This is one embodiment of the present invention, which provides a method for monitoring surface deformation based on distributed millimeter-wave radar, including the following steps: S1. Collect topographic mapping data and risk zoning data, perform initial time synchronization calibration and spatial extrinsic parameter calibration, and obtain the distributed monitoring initialization configuration.

[0025] Topographic mapping data includes elevation information, surface undulation, terrain slope, obstacle distribution, and infrastructure location information for the target area.

[0026] Based on topographic mapping data, historical geological disaster records, and environmental factors, potential risks are identified, and risk zoning data is generated.

[0027] Furthermore, coordinate unification, time information verification, and monitoring range trimming are performed on topographic mapping data, historical geological disaster records, and environmental factors to generate spatial alignment information. Based on the spatial alignment information, elevation, slope, aspect, and surface undulation features are extracted from the topographic mapping data to generate a set of topographic features. Based on the spatial alignment information, disaster type, location, frequency, and impact range are extracted from historical geological disaster records to generate a set of historical geological disaster features. The set of topographic features, the set of historical geological disaster features, and environmental factors are overlaid and analyzed according to a unified spatial grid, and the risk assessment value corresponding to each spatial grid is calculated. Based on the risk assessment value corresponding to each spatial grid, potential risks are classified and potential risk zoning identifiers are generated. Based on the potential risk zoning identifiers, the zoning boundaries are sorted and adjacent areas are merged to generate risk zoning data. For example, the risk assessment value corresponding to spatial grid A1 is 18 and a low potential risk partition identifier is generated; the risk assessment value corresponding to spatial grid A2 is 22 and a medium potential risk partition identifier is generated; the risk assessment values ​​corresponding to spatial grid A3 and spatial grid A4 are 68 and 72 respectively and the same high potential risk partition identifier is generated. Based on the high potential risk partition identifier, the partition boundaries of spatial grid A3 and spatial grid A4 are sorted and adjacent areas are merged to generate risk partition data.

[0028] Risk zoning data includes the risk level and emergency response strategy for each region.

[0029] It should be noted that risk level and emergency response strategy refer to the risk classification and corresponding handling rules determined based on the risk assessment value corresponding to each spatial grid, the identification information of the surface deformation evolution trend, and the scope of impact. For example, low risk level corresponds to routine monitoring, medium risk level corresponds to intensified monitoring and on-site verification, high risk level corresponds to early warning issuance and key area duty, and extremely high risk level corresponds to emergency evacuation and joint response.

[0030] A unified clock is established for the nodes of the distributed millimeter-wave radar, and time deviation calibration, spatial position parameter calibration, and attitude parameter calibration are performed between nodes. At the same time, combined with terrain mapping data and risk zoning data, a distributed monitoring initialization configuration is generated.

[0031] Furthermore, a unified clock configuration is first completed for the nodes of the distributed millimeter-wave radar, enabling each node to output observation time information under the same time reference. Then, the observation time information under the same time reference is used to calculate and correct the time deviation calibration between nodes. After the time deviation calibration between nodes is completed, spatial position parameter calibration is carried out in combination with the observation information of the node installation location, and the spatial position parameter calibration is spatially correlated with the topographic mapping data. Based on the spatial correlation, attitude parameter calibration is completed in combination with the terrain direction and slope characteristics in the topographic mapping data. Subsequently, the unified clock, the time deviation calibration between nodes, the spatial position parameter calibration, the attitude parameter calibration, the topographic mapping data, and the risk zoning data are jointly configured to determine the node coverage, the correspondence between monitoring areas, and the monitoring priority, and generate the distributed monitoring initialization configuration.

[0032] S2. Based on the distributed monitoring initialization configuration, perform cross-node closed-loop observation calibration and correct phase offset, clock drift and installation angle error to obtain closed-loop self-calibration monitoring frame.

[0033] Based on the distributed monitoring initialization configuration, cross-node synchronous calibration and observation relationship establishment are performed between the nodes of the distributed millimeter-wave radar to obtain cross-node observation relationship data.

[0034] Furthermore, based on the distributed monitoring initialization configuration, node identifiers, unified clocks, spatial location parameter calibrations, attitude parameter calibrations, topographic mapping data, and risk zoning data are read. The same observation time between nodes of the distributed millimeter-wave radar is organized according to the unified clock, and time alignment is completed, forming a set of synchronous observation times required for cross-node synchronous calibration. The set of synchronous observation times is overlaid with spatial location parameter calibrations and attitude parameter calibrations to unify the observation directions of each node to the spatial coordinate range corresponding to the topographic mapping data, obtaining comparable spatial observation positions between nodes. Combined with risk zoning data, common observation areas and common observation targets are selected within the overlapping areas of node coverage. Observation time matching, spatial location matching, and observation record correspondence between nodes are completed in the common observation areas and common observation targets, forming observation correlation terms between node pairs. The set of synchronous observation times, comparable spatial observation positions between nodes, and observation correlation terms between node pairs are summarized to complete cross-node synchronous calibration and establish observation relationships, obtaining cross-node observation relationship data.

[0035] By combining risk zoning data, common observation areas and common observation targets are selected within the overlapping area of ​​node coverage. Specifically, the overlapping area of ​​node coverage is spatially overlaid with the risk zoning data. Spatial grids with high monitoring priority and located near the boundaries of the same or adjacent risk zoning are retained. Spatial grids with terrain obstruction, observation distances exceeding the effective range, or observation directions not meeting the conditions are eliminated by combining spatial position parameter calibration and attitude parameter calibration, thus forming a common observation area. Within the common observation area, stable reference scattering points that can be observed simultaneously by multiple nodes and whose observation records are continuous and stable are selected according to the same observation time corresponding to a unified clock. These are used as common observation targets (such as stable reference scattering points located near the boundary of high potential risk zoning and within the overlapping area of ​​node coverage).

[0036] For each node in the cross-node observation relationship data, a closed-loop consistency check is performed on the observation phase for the same stable reference scattering point, and the phase offset is iteratively corrected according to the closed-loop residual to generate calibrated cross-node observation relationship data.

[0037] Furthermore, the nodes in the cross-node observation relationship data are grouped according to the same stable reference scattering point. The observation phases at the same observation time or after time alignment are extracted, and a closed-loop path is established according to the node observation relationship in the cross-node observation relationship data. The observation phase difference between adjacent nodes is calculated on the closed-loop path, and the closed-loop residual is obtained by summing the results. The closed-loop residual is used to perform closed-loop consistency verification with a consistency threshold. If the closed-loop consistency verification fails, the closed-loop residual is allocated as a phase offset correction amount according to the observation stability of the observation phase of each node in the closed-loop path. The phase offset is iteratively corrected, and the phase offset correction amount is applied to the observation phase. The closed-loop residual is then recalculated, and the closed-loop consistency verification continues. After the closed-loop residual meets the consistency threshold, the observation phase that has passed the closed-loop consistency verification, the phase offset correction result, and the corresponding node observation relationship are written into the cross-node observation relationship data to generate calibrated cross-node observation relationship data.

[0038] It should be noted that a stable reference scattering point refers to a reference target point whose spatial location is relatively fixed during the monitoring period, whose backscattering characteristics are stable, which can be continuously and repeatedly observed by multiple nodes, and whose observed phase changes are mainly affected by phase offset between nodes and disturbances in the propagation environment, and whose influence from surface deformation is negligible or known and correctable.

[0039] Based on the calibrated cross-node observation relationship data, the impact of cross-node observation consistency on node clock drift and installation angle error is evaluated and iteratively compensated to obtain a closed-loop self-calibration monitoring frame.

[0040] Furthermore, based on the calibrated cross-node observation relationship data, continuous observation records are organized according to node identifier, observation time, and the same stable reference scattering point. Observation time information, observation direction information, observation phase, and phase closed-loop residual are extracted. The node clock drift estimate, installation angle error estimate, and the mean absolute value of the phase closed-loop residual are calculated. The three items are normalized to the baseline, squared, summed, averaged, and squared respectively to obtain the cross-node observation consistency impact score. Cross-node observation consistency impact assessment and iterative compensation are performed according to the cross-node observation consistency impact score. The observation time information, observation direction information, and observation phase are written back and the phase closed-loop residual is updated. After the compensation convergence condition is met, the data is encapsulated according to the node identifier and observation time to obtain the closed-loop self-calibration monitoring frame.

[0041] The expression for assessing the impact of cross-node observation consistency is: ; in, It is the first The impact of cross-node observation consistency on the score; It is the node number in a distributed millimeter-wave radar network; It is the first The clock drift estimate of a node represents the node's time base relative to a unified clock; It is the clock drift reference value, used for... Perform normalization processing; It is the first The estimated installation angle error of each node represents the angular deviation of the actual installation posture of the node relative to the calibrated posture. It is the reference value for installation angle error; It is the first The phase closed-loop residual of each node in the closed-loop phase consistency verification; It is the phase closed-loop residual reference value.

[0042] It should be noted that the clock drift reference value is a time scale parameter used to normalize the estimated node clock drift value, and it is preset according to the uniform clock accuracy requirements, sampling period and cross-node synchronization allowable error range.

[0043] The installation angle error estimate is the angular deviation of the actual installation attitude of the node relative to the calibrated attitude. It is calculated by inverting the cross-node observation consistency deviation on the same stable reference scattering point using calibrated cross-node observation relationship data.

[0044] The installation angle error reference value is an angular scale parameter used to normalize the estimated installation angle error value. It is preset according to the accuracy requirements of attitude parameter calibration, on-site installation tolerance, and allowable deviation range of observation direction.

[0045] The phase closed-loop residual reference value is a phase scale parameter used to normalize the phase closed-loop residual. It is set comprehensively according to the closed-loop consistency verification threshold, the phase measurement noise level, and the observation stability requirements of the stable reference scattering point.

[0046] S3. The phase timing of the closed-loop self-calibration monitoring frames is sorted out, and the propagation environment disturbance is separated by combining the micro-meteorological observation records and the residual of the stable reference scattering point to obtain the candidate set of disturbance decoupling displacement.

[0047] The closed-loop self-calibration monitoring frames are sorted and their phase sequences are expanded and organized according to time and node identifiers. Distorted position segments are removed to generate a phase time sequence sorted by time and node identifiers.

[0048] Furthermore, time markers, node markers, and observation phases are extracted from the closed-loop self-calibration monitoring frames. These frames are then sorted primarily by time markers and secondarily by node markers to form a continuous frame arrangement. The observation phases of the same node within the continuous frame arrangement are connected sequentially by time markers. For observation phases with periodic discontinuities, phase sequence unfolding is performed, and the observation positions of different nodes under the same time marker are simultaneously aligned during the unfolding process. By combining the continuity of time intervals after the continuous frame arrangement, the continuity of observation phase changes, and the phase jump amplitude of adjacent frames, distorted position segments are identified and removed from the phase sequence unfolding results. The observation phases after removing distorted position segments are then reorganized according to time markers and node markers to generate a phase time series sequence sorted by time markers and node markers.

[0049] It should be noted that a distortion phase segment refers to a segment in the observed phase corresponding to a closed-loop self-calibration monitoring frame where the continuity of phase changes is abnormal, the jump amplitude is abnormal, or it is inconsistent with adjacent time markers due to signal obstruction, low signal-to-noise ratio, sudden interference, or unwrapping error. For example, an observed phase segment where the same node experiences a short-term large phase jump under continuous time markers and then falls back in the next frame.

[0050] By aligning the micrometeorological observation records and stable reference scattering point residuals within the same time window with the phase time sequence, and extracting the synchronization characteristics between temperature and humidity changes and reference point residual changes, information on propagation environment disturbance discrimination is obtained.

[0051] Furthermore, based on the phase time series, time windows are divided according to time markers. Phase change segments within the same time window are extracted, and temperature and humidity observations are read from micrometeorological observation records and reference point residual change sequences are read from stable reference scattering point residuals according to the start and end times of the same time window. The micrometeorological observation records, stable reference scattering point residuals, and phase time series are resampled at a unified time interval and timestamped to form a one-to-one correspondence between temperature and humidity change sequences and reference point residual change sequences within the same time window. The amplitude, rate, amplitude, and rate of temperature and humidity change, as well as the reference point residual change amplitude and rate of change, are calculated within the same time window. Synchronicity analysis is performed on temperature and humidity changes and reference point residual changes to extract synchronous change direction, synchronous change intensity, time delay relationship, and duration characteristics. The synchronous change direction, synchronous change intensity, time delay relationship, and duration characteristics are associated and labeled with phase change segments to obtain propagation environment disturbance discrimination information.

[0052] It should be noted that the synchronous characteristic of temperature and humidity changes and reference point residual changes refers to the corresponding characteristics of temperature and humidity changes and stable reference scattering point residual changes in terms of direction, magnitude, rate of change, temporal sequence, and duration within the same time window.

[0053] Based on the propagation environment disturbance discrimination information, the propagation environment disturbance components in the phase time series are identified and stripped, while retaining the phase changes related to surface deformation, thus obtaining the propagation environment disturbance separation information.

[0054] Furthermore, based on the phase time series, time windows are divided according to time markers. Phase change segments within the same time window are extracted, and temperature and humidity observations are read from micrometeorological observation records and reference point residual change sequences are read from stable reference scattering point residuals according to the start and end times of the same time window. The micrometeorological observation records, stable reference scattering point residuals, and phase time series are resampled at a unified time interval and timestamped to form a one-to-one correspondence between temperature and humidity change sequences and reference point residual change sequences within the same time window. The amplitude, rate, amplitude, and rate of temperature and humidity change, as well as the reference point residual change amplitude and rate of change, are calculated within the same time window. Synchronicity analysis is performed on temperature and humidity changes and reference point residual changes to extract synchronous change direction, synchronous change intensity, time delay relationship, and duration characteristics. The synchronous change direction, synchronous change intensity, time delay relationship, and duration characteristics are associated and labeled with phase change segments to obtain propagation environment disturbance discrimination information.

[0055] It should be noted that the propagation environment disturbance discrimination information is used to identify the propagation environment disturbance components in the phase change and to provide a basis for subsequent propagation environment disturbance separation. This avoids the propagation environment disturbance caused by temperature and humidity changes being misjudged as surface deformation-related phase changes, and improves the reliability of the disturbance decoupling displacement candidate set as well as the accuracy and stability of the surface deformation monitoring results.

[0056] The displacement candidate quantity is calculated and the temporal validity is screened for the propagation environment disturbance separation information, and the corresponding node identifier and time identifier are written into it to obtain the disturbance decoupling displacement candidate set.

[0057] Furthermore, the phase change sequence after propagation environment disturbance separation is extracted from the node observation records, and displacement candidate quantity calculation is performed according to the phase change sequence to form a displacement candidate quantity sequence arranged in the sampling order. The temporal validity of the displacement candidate quantity sequence is screened, and the time continuity, displacement change continuity, displacement amplitude rationality and consistency between adjacent times are checked. Abnormal displacement candidate quantities are eliminated and valid displacement candidate quantities are retained. The valid displacement candidate quantities and their corresponding node identifiers and time identifiers are written one by one to obtain the disturbance decoupling displacement candidate set.

[0058] S4. Based on the candidate set of disturbance decoupling displacement, perform same-name monitoring point matching and cross-node spatiotemporal alignment, and perform deformation trajectory solution based on cross-node closed-loop phase self-consistency constraint to obtain cross-node fused deformation time series set.

[0059] The candidate set of disturbance decoupling displacements is aggregated and the correspondence is determined according to the node identifier, time identifier and spatial location association features, and the matching data of the same monitoring point is generated.

[0060] Furthermore, the candidate set of disturbance decoupling displacements is organized by time marker and a candidate point list is established by node marker. The spatial location, candidate displacement amount, and change trajectory under adjacent time markers of each candidate point are extracted to form spatial location association features for judgment. Within the same time marker or adjacent time markers, candidate points of different nodes are aggregated based on spatial location association features to obtain candidate point aggregation results. The correspondence of each candidate point aggregation result is judged to determine whether candidate points of different nodes correspond to the same monitoring location and maintain continuous temporal correspondence (for example, under the same time marker, if the spatial distance between the candidate point corresponding to node 1 and the candidate point corresponding to node 2 is within the allowable matching range, the direction of change of the candidate displacement amount is consistent, and the difference of the candidate displacement amount is within the allowable difference range, and the spatial location association features continue to meet the matching conditions under subsequent adjacent time markers, then the candidate point corresponding to node 1 and the candidate point corresponding to node 2 are determined to correspond to the same monitoring location and maintain continuous temporal correspondence), forming a correspondence relationship of the same-name monitoring points. The correspondence relationship of the same-name monitoring points is associated with the corresponding node marker, time marker, spatial location association features, and candidate displacement amount and written to generate the same-name monitoring point matching data.

[0061] It should be noted that spatial location association features refer to the combined features used to characterize the spatial proximity relationship, relative position change relationship, and temporal consistency relationship of different node candidate points in the disturbance decoupling displacement candidate set under the same time mark or adjacent time marks.

[0062] By combining the initial configuration of distributed monitoring with the matching data of monitoring points with the same name, the cross-node observation records corresponding to the monitoring points with the same name are spatiotemporally aligned and the phase self-consistency is verified, generating cross-node closed-loop phase self-consistency constraint data.

[0063] Furthermore, combining the unified clock, spatial location parameter calibration, attitude parameter calibration, and node coverage relationship in the distributed monitoring initialization configuration, the corresponding relationship of the same-named monitoring points, node identifiers, and time identifiers in the same-named monitoring point matching data are read. The cross-node observation records corresponding to the same-named monitoring points are time-aligned according to the unified clock, and the observation positions and observation directions of each node are mapped to the same spatial coordinates using spatial location parameter calibration and attitude parameter calibration to complete spatiotemporal alignment. In the spatiotemporally aligned cross-node observation records, cross-node closed-loop observation combinations are established with the same-named monitoring points as units. The phase difference, phase closed-loop residual, and phase change continuity relationship of the observation phase of each node under the same or adjacent time identifiers are calculated to verify phase self-consistency. The corresponding relationship of the same-named monitoring points, spatiotemporal alignment relationship, phase closed-loop residual, and phase continuity relationship that have passed the phase self-consistency verification are jointly written into the constraint terms to generate cross-node closed-loop phase self-consistent constraint data.

[0064] It should be noted that the cross-node observation record corresponding to the same monitoring point refers to the associated record of the observation time information, observation location, observation direction and observation phase formed by the same monitoring points that are determined to be the same monitoring location and maintain continuous time in the matching data of the same monitoring points, under the same time mark or adjacent time mark at different nodes.

[0065] Cross-node closed-loop phase self-consistent constraint data refers to constraint data formed by cross-node observation records corresponding to the same monitoring points after spatiotemporal alignment and phase self-consistency verification. It includes the correspondence between the same monitoring points, spatiotemporal alignment relationship, phase closed-loop residual and phase continuity relationship. It is used to provide cross-node consistent constraints for subsequent deformation trajectory solution and time continuity reconstruction, thereby reducing the propagation of cross-node observation bias and improving the continuity, accuracy and stability of cross-node fused deformation time series set.

[0066] Deformation trajectory solving and time continuity reconstruction are performed on the cross-node closed-loop phase self-consistent constraint data, and deformation time series records are written according to the monitoring points with the same name to obtain the cross-node fused deformation time series set.

[0067] Furthermore, the cross-node closed-loop phase self-consistent constraint data are grouped by monitoring points with the same name. The correspondence between monitoring points with the same name, the spatiotemporal alignment relationship, the phase closed-loop residual, the phase continuity relationship, the node identifier, and the time identifier are extracted, and sorted by time identifier to form a solution sequence of monitoring points with the same name. In the solution sequence of monitoring points with the same name, the spatiotemporal alignment relationship is used to unify the spatial orientation of the observed phase of different nodes. The deformation trajectory is solved by combining the phase closed-loop residual and the phase continuity relationship to form the deformation trajectory points corresponding to each time identifier. The time discontinuity positions in the deformation trajectory point sequence are combined with the phase continuity relationship of adjacent time identifiers to reconstruct the temporal continuity and form a continuous deformation trajectory. The continuous deformation trajectory is written into the deformation time series record by monitoring points with the same name, and the corresponding time identifier, node identifier, and constraint relationship are associated in the deformation time series record. The deformation time series records of multiple monitoring points with the same name are summarized and organized to obtain a cross-node fused deformation time series set.

[0068] It should be noted that deformation time series records refer to continuous time series data records that are organized and recorded according to monitoring points with the same name, including deformation, corresponding node identifiers, and associated constraint relationships under each time marker.

[0069] Cross-node fusion deformation time series refers to a data set that is formed by summarizing deformation time series records from multiple monitoring points with the same name. It contains continuous deformation time series information after cross-node alignment and is used to characterize the spatiotemporal evolution process of surface deformation in the monitoring area. This reduces the impact of single-node observation interruption or local deviation on deformation analysis and improves the continuity, accuracy and stability of surface deformation monitoring.

[0070] S5. Identify the evolution trend of surface deformation from the cross-node fused deformation time series set and generate a surface deformation monitoring report.

[0071] For the cross-node fused deformation time series set, the time series is sorted and missing segments are marked according to the monitoring points with the same name and time sequence. The displacement change rate, change acceleration and stage duration information are extracted to generate surface deformation evolution characteristic data.

[0072] Furthermore, the cross-node fused deformation time series set is grouped by monitoring points with the same name, and the deformation time series records are arranged in chronological order within each monitoring point with the same name. Based on the continuity of time markers, missing segments are marked for the time gap position, gap length, and adjacent deformation before and after the gap, forming a time series sequence of monitoring points with the same name containing time series information and missing segment annotation information. Based on the time series sequence of monitoring points with the same name containing time series information and missing segment annotation information, the deformation difference is calculated according to adjacent time markers, and the displacement change rate is extracted by combining the time interval. The change acceleration is extracted according to the change rate of displacement under continuous time markers, and the stage duration information is extracted for continuous time segments where the displacement change rate and change acceleration are in the same direction or have the same change state. The monitoring point markers, time order, missing segment annotations, displacement change rate, change acceleration, and stage duration information are associated and written to generate surface deformation evolution characteristic data.

[0073] Based on the surface deformation evolution characteristic data, the surface deformation evolution trend is identified for each monitoring point with the same name, and the trends of continuous increase, continuous decrease and phased acceleration are determined to generate surface deformation evolution trend identification information.

[0074] Furthermore, based on the surface deformation evolution characteristic data, the monitoring points are grouped according to their names, and the displacement change rate, acceleration, stage duration information, and missing segment annotations are read in chronological order to form a trend identification sequence of the monitoring points. Based on the trend identification sequence of the monitoring points, the consistency of change direction and the continuity of change intensity are analyzed for continuous time segments. Continuous time segments with a continuous unidirectional displacement change rate and a stage duration that meets the continuous determination criteria are identified as having a continuous increasing trend or a continuous decreasing trend. Continuous time segments with an acceleration characteristic that remains within the continuous time segment are identified as having a staged acceleration trend. The information of the corresponding monitoring points, time segments, displacement change rate, acceleration, and stage duration for continuous increasing trends, continuous decreasing trends, and staged acceleration trends is written to generate surface deformation evolution trend identification information.

[0075] The spatial regional aggregation and temporal stage summarization of the surface deformation evolution trend identification information are combined, and the corresponding deformation time series records in the cross-node fusion deformation time series set are used as the basis for monitoring to generate a surface deformation monitoring report.

[0076] Furthermore, the surface deformation evolution trend identification information is organized according to the same monitoring point, trend type, and time segment. Based on the spatial location relationship of the same monitoring points, spatially adjacent records with consistent trend types and intersecting or continuous time segments are merged into spatial regions to form a spatial region merging result. Within the spatial region merging result, the continuously increasing trend, continuously decreasing trend, and phased acceleration trend are summarized in chronological order to form a summary record containing spatial range, trend type, and phase start and end times. Based on the summary record, the corresponding deformation time series record is retrieved from the cross-node fused deformation time series set, and the corresponding deformation time series record is associated with the spatial region merging result and the time segment summary result as the monitoring basis. The spatial range, trend type, phase start and end time, and corresponding deformation time series record in the monitoring basis are summarized and written to generate a surface deformation monitoring report.

[0077] The surface deformation monitoring report includes trend determination of monitoring points and explanation of regional trend changes.

[0078] It should be noted that the monitoring point trend determination content refers to the determination record of the continuous increasing trend, continuous decreasing trend, or phased acceleration trend formed by each monitoring point with the same name within the corresponding time period, as well as the corresponding time period, displacement change rate, change acceleration, and phase duration information; the regional trend change description refers to the textual summary of the surface deformation evolution trend identification information after spatial regional aggregation and time phase summarization, which is used to describe the trend type, change stage, and change process characteristics within the corresponding spatial range.

[0079] This embodiment also provides a surface deformation monitoring system based on distributed millimeter-wave radar, including: an initialization configuration module, which collects topographic mapping data and risk zoning data and performs initial timing calibration and spatial extrinsic parameter calibration to obtain distributed monitoring initialization configuration; The closed-loop calibration module performs cross-node closed-loop observation calibration based on the distributed monitoring initialization configuration, and corrects phase offset, clock drift and installation angle error to obtain closed-loop self-calibration monitoring frames. The disturbance separation module performs phase timing processing on the closed-loop self-calibration monitoring frames and separates the propagation environment disturbances by combining micro-meteorological observation records and stable reference scattering point residuals, thereby obtaining a candidate set of disturbance decoupling displacements. The trajectory reconstruction module performs matching of corresponding monitoring points and cross-node spatiotemporal alignment based on the candidate set of disturbance decoupling displacements, and performs deformation trajectory solving based on cross-node closed-loop phase self-consistency constraints to obtain a cross-node fused deformation time series set. The trend monitoring module identifies the evolution trend of surface deformation from the cross-node fused deformation time series and generates a surface deformation monitoring report.

[0080] In summary, this invention achieves coordinated correction of phase offset, clock drift, and installation angle errors by performing closed-loop consistency verification on cross-node observation relationship data and conducting cross-node observation consistency impact assessment and iterative compensation, thus forming a stable observation basis for closed-loop self-calibration monitoring frames. Furthermore, by driving propagation environment disturbance separation and displacement candidate quantity calculation through propagation environment disturbance discrimination information, it realizes the extraction of surface deformation-related phase changes and the construction of disturbance decoupling displacement candidate sets, which are used to support deformation trajectory solving and surface deformation evolution trend identification, ultimately improving the continuity, reliability, and engineering application value of surface deformation monitoring reports.

[0081] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for monitoring surface deformation based on distributed millimeter-wave radar, characterized in that: include, Collect topographic mapping data and risk zoning data, perform initial time synchronization calibration and spatial extrinsic parameter calibration, and obtain the initial configuration for distributed monitoring; Based on the distributed monitoring initialization configuration, cross-node closed-loop observation calibration is performed, and phase offset, clock drift and installation angle error are corrected to obtain closed-loop self-calibration monitoring frames. Phase timing of the closed-loop self-calibration monitoring frames is processed, and propagation environment disturbances are separated by combining micrometeorological observation records and stable reference scattering point residuals to obtain a candidate set of disturbance decoupling displacements; Based on the candidate set of disturbance decoupling displacement, the same-name monitoring points are matched and cross-node spatiotemporal alignment is performed. Then, the deformation trajectory is solved based on the cross-node closed-loop phase self-consistency constraint to obtain the cross-node fused deformation time series set. The evolution trend of surface deformation is identified by analyzing cross-node fused deformation time series sets, and a surface deformation monitoring report is generated. The process of performing cross-node closed-loop observation calibration based on distributed monitoring initialization configuration, and correcting phase offset, clock drift, and installation angle errors to obtain a closed-loop self-calibration monitoring frame, is as follows. Based on the distributed monitoring initialization configuration, cross-node synchronous calibration and observation relationship establishment are performed between the nodes of the distributed millimeter-wave radar to obtain cross-node observation relationship data. For each node in the cross-node observation relationship data, a closed-loop consistency check is performed on the observation phase of the same stable reference scattering point, and the phase offset is iteratively corrected according to the closed-loop residual to generate calibrated cross-node observation relationship data. Based on the calibrated cross-node observation relationship data, the impact of cross-node observation consistency on node clock drift and installation angle error is evaluated and iteratively compensated to obtain a closed-loop self-calibration monitoring frame.

2. The surface deformation monitoring method based on distributed millimeter-wave radar as described in claim 1, characterized in that: The steps to obtain the distributed monitoring initialization configuration are as follows: Based on topographic mapping data, historical geological disaster records, and environmental factors, potential risks are classified, and risk zoning data is generated. A unified clock is established for the nodes of the distributed millimeter-wave radar, and time deviation calibration, spatial position parameter calibration, and attitude parameter calibration are performed between nodes. At the same time, combined with terrain mapping data and risk zoning data, a distributed monitoring initialization configuration is generated.

3. The surface deformation monitoring method based on distributed millimeter-wave radar as described in claim 2, characterized in that: The topographic mapping data includes elevation information, surface relief, terrain slope, obstacle distribution, and infrastructure location information of the target area; The risk zoning data includes the risk level and emergency response strategy for each region.

4. The surface deformation monitoring method based on distributed millimeter-wave radar as described in claim 1, characterized in that: The phase timing organization of the closed-loop self-calibration monitoring frame refers to sorting and organizing the closed-loop self-calibration monitoring frame according to the time identifier and node identifier, removing distorted position segments, and generating a phase timing sequence sorted by the time identifier and node identifier.

5. The surface deformation monitoring method based on distributed millimeter-wave radar as described in claim 4, characterized in that: The process of separating propagating environmental disturbances by combining micrometeorological observation records and the residuals of stable reference scattering points to obtain a candidate set of disturbance decoupling displacements is as follows: By aligning the micrometeorological observation records and stable reference scattering point residuals within the same time window according to the phase time sequence, and extracting the synchronization characteristics between temperature and humidity changes and reference point residual changes, information on propagation environment disturbance discrimination is obtained. Based on the propagation environment disturbance discrimination information, the propagation environment disturbance components in the phase time series are identified and stripped, while retaining the phase changes related to surface deformation, to obtain the propagation environment disturbance separation information. The displacement candidate quantity is calculated and the temporal validity is screened for the propagation environment disturbance separation information, and the corresponding node identifier and time identifier are written into it to obtain the disturbance decoupling displacement candidate set.

6. The surface deformation monitoring method based on distributed millimeter-wave radar as described in claim 1, characterized in that: The steps are as follows: matching corresponding monitoring points and cross-node spatiotemporal alignment based on the candidate set of disturbance decoupling displacements, and solving for deformation trajectories based on cross-node closed-loop phase self-consistency constraints to obtain a cross-node fused deformation time series set. The candidate set of disturbance decoupling displacements is aggregated and the correspondence is determined according to the node identifier, time identifier and spatial location association features to generate matching data of monitoring points with the same name; By combining the initial configuration of distributed monitoring with the matching data of monitoring points with the same name, the cross-node observation records corresponding to the monitoring points with the same name are spatiotemporally aligned and the phase self-consistency is verified to generate cross-node closed-loop phase self-consistency constraint data. Deformation trajectory solving and time continuity reconstruction are performed on the cross-node closed-loop phase self-consistent constraint data, and deformation time series records are written according to the monitoring points with the same name to obtain the cross-node fused deformation time series set.

7. The surface deformation monitoring method based on distributed millimeter-wave radar as described in claim 1, characterized in that: The steps for identifying the surface deformation evolution trend of the cross-node fused deformation time series are as follows: For the cross-node fused deformation time series set, the time series is sorted and missing segments are marked according to the monitoring points with the same name and the time sequence. The displacement change rate, change acceleration and stage duration information are extracted to generate surface deformation evolution characteristic data. Based on the surface deformation evolution characteristic data, the surface deformation evolution trend is identified for each monitoring point with the same name, and the trends of continuous increase, continuous decrease and phased acceleration are determined to generate surface deformation evolution trend identification information.

8. The surface deformation monitoring method based on distributed millimeter-wave radar as described in claim 7, characterized in that: The surface deformation monitoring report includes trend determination of monitoring points and description of regional trend changes.

9. A surface deformation monitoring system based on distributed millimeter-wave radar, based on the surface deformation monitoring method based on distributed millimeter-wave radar according to any one of claims 1 to 8, characterized in that: include, The initialization configuration module collects topographic mapping data and risk zoning data, performs initial time synchronization calibration and spatial extrinsic parameter calibration, and obtains the initial configuration for distributed monitoring. The closed-loop calibration module performs cross-node closed-loop observation calibration based on the distributed monitoring initialization configuration, and corrects phase offset, clock drift and installation angle error to obtain closed-loop self-calibration monitoring frames. The disturbance separation module performs phase timing processing on the closed-loop self-calibration monitoring frames and separates the propagation environment disturbances by combining micro-meteorological observation records and stable reference scattering point residuals, thereby obtaining a candidate set of disturbance decoupling displacements. The trajectory reconstruction module performs matching of corresponding monitoring points and cross-node spatiotemporal alignment based on the candidate set of disturbance decoupling displacements, and performs deformation trajectory solving based on cross-node closed-loop phase self-consistency constraints to obtain a cross-node fused deformation time series set. The trend monitoring module identifies the evolution trend of surface deformation from the cross-node fused deformation time series and generates a surface deformation monitoring report.

Images