Building deformation early warning method and device based on interference radar time sequence analysis

Abstract

The invention provides a building deformation early warning method and device based on interference radar time sequence analysis, and relates to the technical field of deformation monitoring, and the method comprises the steps: obtaining multi-moment interference echoes covering a building area, constructing an interference data cube, and obtaining a stable scatterer set through scattering stability screening; performing phase unwrapping, time domain smoothing and atmospheric compensation based on the stable scatterer complex echoes, extracting a deformation quantity time sequence of the building, performing projection conversion in combination with a building geometrical relationship and a radar incident angle, and constructing a deformation field covering a building area; whether the building is abnormal or not is judged according to comparison of the change rate and the change acceleration of the deformation field and a safety threshold value, and a deformation early warning result is generated; and the early warning result is comprehensively analyzed in combination with the building operation state and historical records, and building deformation early warning information is output. According to the invention, the technical defect of result distortion caused by unstable ground object coherence in building deformation early warning can be solved.

Description

Technical Field

[0001] This invention relates to the technical field of deformation monitoring, specifically to a method and device for early warning of building deformation based on interferometric radar time-series analysis. Background Technology

[0002] Interferometric radar time-series analysis processes interferometric complex image sequences of the same area under multi-temporal observation conditions, models the trajectory of the scattering phase of each pixel over time, and obtains the phase evolution caused only by the gradual deformation of the surface by separating atmospheric phase, orbital error, and ground object scattering noise. This allows for the reconstruction of the surface deformation process with millimeter or even sub-millimeter precision over long time scales. In this process, by constructing a high-coherence pixel set, performing temporal phase reconstruction, unifying the differential phase reference, and using spatial-temporal filtering to enhance the phase stability of stable scattering centers, the deformation signal is made prominent in random noise and environmental disturbances, providing high-precision time-series information for continuous deformation monitoring of buildings, landslides, foundation settlement areas, and infrastructure.

[0003] The application of interferometric radar time-series analysis in building deformation monitoring is reflected in the long-term continuous tracking of the scattering phase of buildings using multi-temporal interferometric images. By separating atmospheric disturbances, orbital errors, and random noise from the time-series phase sequence, the phase evolution caused only by the slow deformation of the building is extracted. This allows for the characterization of minute deformation trends of buildings under the influence of factors such as daily loads, groundwater changes, foundation settlement, structural fatigue, and external construction disturbances with millimeter-level precision. At the same time, by using high-coherence pixel screening and spatiotemporal filtering to enhance the phase stability of the building's scattering center, local deformation, tilt development, differential settlement, and potential structural anomalies on the building surface can be accurately identified from the time-series data. This provides a continuous, precise, and non-contact deformation monitoring method for building safety assessment, early warning decision-making, and operation and maintenance management.

[0004] Interferometric radar time-series analysis suffers from technical defects in building deformation early warning due to the instability of ground object coherence, leading to distorted results. When the monitoring area contains a large amount of vegetation, water surface, or loose ground, the scattering phase of these areas cannot remain stable in multi-temporal images, thus disturbing the reference benchmark of surrounding buildings. Especially in high-vegetation environments, leaves and tree crowns continuously generate subtle movements under the action of wind, causing local scattering centers to drift over time, resulting in random fluctuations in the phase sequence and masking the true slow deformation signal of the building. This introduces noise bias into the deformation time series during spatial filtering, temporal filtering, and phase reconstruction, ultimately reducing the sensitivity and reliability of building deformation early warning and increasing the risk of false alarms and missed alarms. Summary of the Invention

[0005] This invention provides a method and device for early warning of building deformation based on interferometric radar time series analysis, which can solve the technical defects in early warning of building deformation caused by the instability of ground object coherence.

[0006] In a first aspect, the present invention provides a method for early warning of building deformation based on interferometric radar time-series analysis, the method comprising: Multi-time interferometric echoes covering the building area are acquired to form an interferometric data cube, and stable scatterers are selected based on scattering stability to construct a set of stable scatterers; Phase unwrapping, time-domain smoothing, and atmospheric compensation are performed based on the complex echoes corresponding to the stable scatterer. High-frequency random disturbances and low-frequency deformation trends are separated from the complex echoes through time-series decomposition to obtain the time series of building deformation in the line-of-sight direction. The deformation time series is combined with the building geometry and radar incident angle for projection conversion, and a deformation field covering the building area is constructed through multiple deformation time series to identify overall settlement, local uneven settlement or facade torsional deformation. Based on the comparison between the rate of change and acceleration of the deformation field and the preset safety threshold, it is determined whether the building deformation has entered an abnormal state and a deformation early warning result is generated. Based on the deformation early warning results, a comprehensive analysis is performed on the building's operating status and historical records, and building deformation early warning information is output.

[0007] Based on the above technical solutions, preferably, the step of performing phase unwrapping, time-domain smoothing, and atmospheric compensation based on the complex echo corresponding to the stable scatterer, and separating high-frequency random disturbances and low-frequency deformation trends from the complex echo through time-series decomposition to obtain the time series of building deformation in the line-of-sight direction, specifically includes: The complex echoes of the stable scatterer in the interferometric data cube are extracted in order of observation time to form a complex echo time series; Phase unwrapping is performed on the complex echo time series to eliminate cross-cycle jumps and form a time-continuous phase unwrapped time series; The phase unwrapped time series is subjected to time-domain smoothing to suppress high-frequency noise, and an atmospheric phase field is constructed based on multiple phase unwrapped time series that have undergone time-domain smoothing to subtract the atmospheric phase delay component from the phase unwrapped time series to form an atmospheric compensated phase time series. A time-series decomposition model is constructed for the time-domain smoothing process to separate the high-frequency random disturbance phase component and the low-frequency deformation trend phase component, and the low-frequency deformation trend phase component is converted into the deformation time series.

[0008] Based on the above technical solutions, preferably, the step of combining the deformation time series with the building's geometric relationship and the radar incident angle for projection conversion, and constructing a deformation field covering the building area through multiple deformation time series to identify overall settlement, local uneven settlement, or facade torsional deformation, specifically includes: The pixel coordinates of the stable scatterer in the interferometric data cube are converted into three-dimensional coordinate positions in the building coordinate system through georegistration and three-dimensional geometric calibration. The projection relationship between the line of sight direction and the vertical direction, horizontal direction and normal direction of the building are established based on the building geometry and radar incident angle. The building coordinate system is established based on the interferometric data cube and the building structural data. The deformation time series for the line-of-sight direction of each stable scatterer is decomposed into a decomposed time series according to the projection relationship. The decomposed time series includes a vertical deformation time series, a building horizontal deformation time series, or a component normal deformation time series. At each observation time, the decomposed time series corresponding to the multiple stable scatterers are divided into multiple deformation analysis regions according to the geometric relationship of the building; In each deformation analysis region, a structural fitting model is constructed based on the spatial distribution and deformation values ​​of the stable scatterer to form a local continuous deformation distribution. A continuity constraint is applied between adjacent deformation analysis regions to ensure that the distribution of each local continuous deformation variable maintains spatial continuity within the entire building area, thereby forming a deformation field covering the building area at each observation time.

[0009] Based on the above technical solutions, preferably, the step of combining the deformation time series with the building's geometric relationship and the radar incident angle for projection conversion, and constructing a deformation field covering the building area through multiple deformation time series to identify overall settlement, local uneven settlement, or facade torsional deformation, specifically further includes: At each observation time, the vertical deformation located in the base region is extracted from the deformation field; Perform an overall plane fitting on the vertical deformation to form a planar deformation distribution that describes the overall vertical displacement trend of the foundation; Based on the aforementioned planar deformation distribution, the tilt vector used to characterize the overall settlement direction and overall tilt degree of the building is solved to identify the overall settlement; Based on the deformation field, the stable scattering body is divided into multiple vertical deformation analysis sequences according to the floor number and structural axis number. The vertical deformation distribution along each structural axis is calculated in each floor, and the vertical deformation distribution along the floor height is calculated on each structural axis. Local uneven settlement is identified by calculating the difference in vertical deformation between adjacent floors and adjacent structural axes and the spatial gradient of vertical deformation. Extract the component normal deformation variables located in the facade area from the deformation field, organize the component normal deformation variables into multiple component normal deformation variable profile sequences according to the building height direction, and identify the torsional deformation of the facade around the vertical axis of the building by comparing the difference distribution of component normal deformation variables along the height direction at different facade boundaries.

[0010] Based on the above technical solutions, preferably, the step of determining whether the building deformation has entered an abnormal state and generating a deformation early warning result by comparing the change rate and acceleration of the deformation field with a preset safety threshold specifically includes: At each observation time, the deformation time series of each stable scatterer in the deformation field covering the building area is fitted to obtain the corresponding deformation rate time series and deformation acceleration time series. The time series of deformation, the time series of deformation rate of change, and the time series of deformation acceleration are used together to construct a spatiotemporal deformation field containing dynamic deformation characteristics; Based on the building type, structural system, and material properties, set safety thresholds for deformation, rate of deformation change, and acceleration of deformation change for different areas of the building. The deformation, deformation rate, and deformation acceleration of each stable scatterer position in the spatiotemporal deformation field are sequentially compared with the corresponding preset safety thresholds to identify the deformation acceleration development region and the deformation exceeding limit region. Spatial clustering and spatial connectivity analysis are performed on the comparison results of adjacent stable scatterers to form continuous deformation risk regions. Based on the degree of exceeding of the limits of deformation, deformation rate of change and deformation acceleration in each risk region, the deformation risk regions are divided into different risk levels. The deformation warning result is generated by combining the risk level, the spatial range of the risk area, and the corresponding deformation type.

[0011] Based on the above technical solutions, preferably, the step of comprehensively analyzing and outputting building deformation early warning information based on the deformation early warning results combined with the building's operating status and historical records specifically includes: A spatiotemporal access channel is constructed to map the building's operating status and historical records to a spatiotemporal indexing system consistent with the building's coordinate system, and the spatial location and time window of each deformation risk area in the deformation early warning result are spatiotemporally aligned with the building's operating status and historical records. Based on the temporal evolution characteristics of the deformation, deformation rate of change, and deformation acceleration of the deformation risk area, and the relationship between the load condition changes in the building's operating state, external disturbance events, and construction activities, it is determined whether the deformation risk area belongs to the continuation of historical deformation trends, reactivation of existing defects, or new abnormal deformation. In conjunction with the previous warning results in the historical records and the corresponding disposal measures, it is assessed whether the deformation evolution of the deformation risk area has broken through the previous deformation envelope. The risk interpretation results of the deformation risk areas are differentiated and adjusted based on the functional zoning, component importance and structural safety level in the building's operation status to form a risk level ranking oriented towards operation and maintenance needs; The spatial location, risk level, deformation type, deformation amount, relationship with each preset safety threshold, deformation development trend, and correlation with the building's operating status and historical records of each deformation risk area are organized into building deformation early warning information and output.

[0012] Based on the above technical solutions, preferably, the step of acquiring multi-time interference echoes covering the building area to form an interference data cube, and screening stable scatterers based on scattering stability to construct a stable scatterer set, specifically includes: On the interferometric radar side, the observation geometry parameters and observation period are planned to obtain interferometric echoes covering the building facade, roof structure and surrounding hardened area at multiple observation times. Geometric registration and resampling are performed on the interferometric echoes at different observation times to construct an interferometric data cube to characterize the spatiotemporal response features of the building area. Calculate the scattering stability index for the complex reflection sequence corresponding to each pixel in the interferometric data cube; From the plurality of pixels, candidate pixels for stable scatterers are selected, wherein the candidate pixels for stable scatterers are pixels whose scattering stability value is higher than a preset first threshold and whose phase fluctuation value is less than a preset second threshold, based on the scattering stability index. A stable scatterer set is constructed by combining the candidate stable scatterer pixels.

[0013] In a second aspect of the invention, a building deformation early warning device based on interferometric radar time-series analysis is provided. The device is used to execute a building deformation early warning method based on interferometric radar time-series analysis as described above. The device includes an acquisition module, a processing module, and an output module, wherein: The acquisition module is used to acquire multi-time interference echoes covering the building area to form an interference data cube, and to screen stable scatterers based on scattering stability to construct a set of stable scatterers; The processing module is used to perform phase unwrapping, time-domain smoothing and atmospheric compensation based on the complex echo corresponding to the stable scatterer, and to separate high-frequency random disturbances and low-frequency deformation trends from the complex echo through time-series decomposition to obtain the time series of building deformation in the line-of-sight direction. The processing module is used to project and convert the deformation time series into building geometry and radar incident angle, and to construct a deformation field covering the building area through multiple deformation time series to identify overall settlement, local uneven settlement or facade torsional deformation. The processing module is used to determine whether the building deformation has entered an abnormal state based on the comparison between the change rate and change acceleration of the deformation field and a preset safety threshold, and to generate a deformation early warning result. The output module is used to perform comprehensive analysis based on the deformation early warning results, combined with the building's operating status and historical records, and output building deformation early warning information.

[0014] In a third aspect of the invention, an electronic device is provided, including a processor, a memory, a user interface, and a network interface, wherein the memory is used to store instructions, the user interface and the network interface are both used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any of the preceding embodiments.

[0015] In a fourth aspect, the present invention provides a computer-readable storage medium storing instructions that, when executed, perform the method as described in any of the preceding claims.

[0016] In summary, one or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages: 1. This invention introduces stable scatterer screening, temporal phase reconstruction, atmospheric compensation, spatiotemporal decomposition, and deformation field construction constrained by building geometry into the interferometric radar time-series analysis link. It strictly limits all calculation benchmarks to a set of stable scatterers with long-term stable scattering phases and continuously trackable time series. This effectively separates and enhances the true slow deformation signal of buildings from random noise, vegetation disturbance, water surface scattering instability, and low-coherence background caused by loose ground surfaces, thus avoiding phase fluctuations introduced by unstable ground object pixels from contaminating the building deformation reference benchmark. Simultaneously, through projection conversion, structural region division, continuity constraints, and dynamic feature threshold comparison, the deformation identification and early warning logic is entirely based on a deformation field driven by high-coherence pixels. This eliminates the dependence of building deformation early warning on overall scene coherence, instead relying on the time-series information of stable scatterers after screening and compensation, fundamentally eliminating deformation distortion, false alarms, and missed alarms caused by unstable ground object coherence.

[0017] 2. By performing phase unwrapping, time-domain smoothing, atmospheric compensation, and time-series decomposition on the complex echoes from stable scatterers to recover the time series of deformation variables along the line of sight, it is possible to construct a time-continuous, noise-suppressed phase series with atmospheric effects deducted, even when the interference phase is subject to cross-period jumps, atmospheric delays, and high-frequency noise at multiple times. This allows the true slow deformation trend of buildings to be reliably extracted from high-frequency random disturbances, thereby significantly improving the stability, reliability, and millimeter-level deformation recovery capability of the deformation time series.

[0018] 3. By combining the time series of deformation variables with the geometric relationship of the building and the radar incident angle for projection conversion and constructing a deformation field covering the building area, the deformation in the line of sight direction can be restored to the vertical deformation, horizontal deformation or component normal deformation with engineering significance. Through regional division, structural fitting and continuity constraints, a spatially continuous building deformation field is formed, which can accurately present the settlement, displacement and component deformation characteristics of the building at the overall scale and local scale, thereby realizing the spatial organization of deformation and the interpretation of overall deformation at the structural level.

[0019] 4. Based on the deformation field, further analysis of the vertical deformation of the foundation area, the sequence of vertical deformation of floors and structural axes, and the normal deformation profile of facade components can identify the overall settlement trend and tilt direction at the overall building scale, identify the differential settlement between different floors or structural axes at the local scale, and identify the torsional deformation around the vertical axis of the building at the facade scale. This allows for the layered analysis of different types of building deformation patterns, thereby improving the ability and scope of identifying complex building deformation mechanisms.

[0020] 5. Anomaly detection based on the comparison of deformation field change rate, change acceleration and safety threshold can identify whether building deformation is accelerating or exceeding limits from three dynamic dimensions: the absolute value of deformation, the trend of deformation change and the intensity of change. It can also extract deformation risk areas in the actual structural sense through spatial clustering and connectivity analysis, so that the early warning results are no longer limited to single-point exceedance, but are based on the deformation behavior of large areas to classify risks and determine risk levels, thereby significantly improving the sensitivity and accuracy of deformation anomaly detection. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating a building deformation early warning method based on interferometric radar time series analysis disclosed in an embodiment of the present invention; Figure 2 This is a schematic diagram of a building deformation early warning device based on interferometric radar time sequence analysis disclosed in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of the present invention.

[0022] Explanation of reference numerals in the attached drawings: 201, acquisition module; 202, processing module; 203, output module; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. Detailed Implementation

[0023] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0024] In the description of the embodiments of the present invention, words such as "for example" or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "for example" or "for instance" in the embodiments of the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Rather, the use of words such as "for example" or "for instance" is intended to present the relevant concepts in a specific manner.

[0025] In the description of the embodiments of the present invention, the term "multiple" means two or more. For example, multiple systems means two or more systems, and multiple screen terminals means two or more screen terminals. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0026] Interferometric radar time-series analysis processes interferometric complex images of the same area under multi-temporal conditions to construct a highly coherent pixel set and performs temporal phase reconstruction, differential phase reference unification, and spatial-temporal filtering to separate atmospheric phase, orbital errors, and random scattering noise. This allows for the acquisition of phase evolution caused solely by slow deformation of the ground surface or buildings, reconstructing deformation processes over long timescales with millimeter-level or even sub-millimeter-level accuracy. This enables continuous tracking of minute deformation trends of buildings under the influence of daily loads, groundwater changes, foundation settlement, and external construction disturbances. However, in monitoring areas containing abundant vegetation, water surfaces, or loose ground, the scattering phase in these areas is difficult to maintain temporal stability. Wind-induced swaying of vegetation and drift of the scattering center can cause random fluctuations in the phase sequence and contaminate the reference reference around the building. This causes the true deformation signal to be masked by noise during temporal filtering and phase reconstruction, thereby reducing the accuracy, sensitivity, and reliability of building deformation early warning and increasing the risk of false alarms and missed alarms.

[0027] This embodiment discloses a building deformation early warning method based on interferometric radar time series analysis, referring to... Figure 1 This includes the following steps S110-S150: S110: Obtain multi-time interference echoes covering the building area to form an interference data cube, and select stable scatterers based on scattering stability to construct a set of stable scatterers.

[0028] This invention discloses a building deformation early warning method based on interferometric radar time series analysis, which is applied to a server. The server includes, but is not limited to, electronic devices such as mobile phones, tablets, wearable devices, and PCs (Personal Computers), and can also be a backend server running a building deformation early warning method based on interferometric radar time series analysis. The server can be implemented using a standalone server or a server cluster composed of multiple servers.

[0029] In one possible implementation, interferometric echoes covering a building area at multiple times are acquired to form an interferometric data cube, and stable scatterers are selected based on scattering stability to construct a set of stable scatterers. Specifically, this includes: planning observation geometric parameters and observation periods on the interferometric radar side to acquire interferometric echoes covering the building facade, roof structure, and surrounding hardened area at multiple observation times; performing geometric registration and resampling processing on the interferometric echoes at different observation times to construct an interferometric data cube to characterize the spatiotemporal response features of the building area; calculating a scattering stability index for the complex reflection sequence corresponding to each pixel in the interferometric data cube; selecting candidate stable scatterers from multiple pixels, wherein the candidate stable scatterers are pixels whose scattering stability value is higher than a preset first threshold and whose phase fluctuation value is less than a preset second threshold, selected based on the scattering stability index; and constructing a set of stable scatterers by combining the candidate stable scatterers.

[0030] Specifically, when planning the observation geometry parameters and observation period on the interferometric radar side, the radar operating frequency band, polarization mode, incident angle, azimuth angle, spatial resolution, and time interval of repeated observations are determined to ensure that the radar maintains stable imaging of the same building area at multiple observation times. Interferometric echoes covering the building facade, roof structure, and surrounding hardened area are acquired at multiple observation times, and geometric registration is performed on the interferometric echoes at each observation time to align the range and azimuth coordinates to a unified ground coordinate system. Then, by resampling a unified pixel grid, each pixel corresponds to the same scattering position of the building area at all observation times. Thus, complex interferometric images with the same spatial dimension but from different observation times are stacked in time to form an interferometric data cube. Each voxel of the interferometric data cube contains the complex reflection sequence of the same pixel at multiple observation times, which is used to characterize the spatiotemporal response characteristics of the building area.

[0031] When calculating the scattering stability index for the complex reflection sequence corresponding to each pixel in the interferometric data cube, the stability of the pixel's scattering response in the time dimension is measured by calculating temporal coherence, phase standard deviation, and amplitude fluctuation coefficient. Temporal coherence characterizes the degree of coherence of the complex reflection sequence across multiple observation times; a higher value indicates a more stable position and physical properties of the scatterer. Phase standard deviation measures the amplitude of phase fluctuation over time; a smaller value indicates that the phase sequence is closer to the true deformation trajectory than random perturbation. Amplitude fluctuation coefficient describes the stability of scattering intensity at different observation times; a smaller value indicates a smoother change in scattering energy. The above indicators are combined to form the scattering stability index, which quantifies the scattering temporal stability of each pixel, thus providing a basis for subsequent selection of stable scatterers.

[0032] When selecting candidate pixels for stable scatterers from multiple pixels, pixels with scattering stability values ​​higher than a preset first threshold are selected as the coherent stable candidate set, and pixels with phase fluctuation values ​​lower than a preset second threshold are selected as the phase stable candidate set. The intersection of the two candidate sets is used to eliminate pixels with unstable scattering, severe phase fluctuations, or significant environmental disturbances. The preset first threshold ensures that pixels have sufficient coherence preservation capability in long-term observations, and the preset second threshold ensures that the phase changes of pixels are mainly driven by real structural deformation rather than noise fluctuations. The stable scatterer candidate pixels obtained by the joint threshold selection are often distributed in metal components, concrete components, and geometrically stable structural parts in buildings, providing a high-quality scattering foundation for subsequent deformation inversion.

[0033] When constructing a stable scatterer set by combining candidate pixels of stable scatterers, the spatial correlation of candidate pixels within the building area is analyzed to determine whether the candidate pixels are in a locally spatially consistent scattering cluster. Candidate pixels that only maintain coherence briefly at some observation times are eliminated based on their coherence persistence throughout the complete observation period. On this basis, candidate pixels located at key structural parts such as beam-column nodes, floor edges, roof panel connections, and rigid facade edges are further retained by combining the structural geometric information of the building. This makes the stable scatterer set have high coherence stability in the time dimension and form a dense scattering point layout covering the key structural areas of the building in the spatial dimension. This provides a stable, continuous, and physically meaningful data foundation for subsequent phase unwrapping, time-domain smoothing, atmospheric compensation, and deformation time series construction.

[0034] S120 performs phase unwrapping, time-domain smoothing, and atmospheric compensation based on the complex echoes corresponding to stable scatterers. It also separates high-frequency random disturbances and low-frequency deformation trends from the complex echoes through time-series decomposition to obtain the time series of building deformation in the line-of-sight direction.

[0035] In one possible implementation, phase unwrapping, time-domain smoothing, and atmospheric compensation are performed based on the complex echoes corresponding to the stable scatterer. High-frequency random disturbances and low-frequency deformation trends are separated from the complex echoes through time-series decomposition to obtain the deformation time series of the building in the line-of-sight direction. Specifically, this includes: extracting the complex echoes of the stable scatterer in the interferometric data cube according to the observation time sequence to form a complex echo time series; performing phase unwrapping on the complex echo time series to eliminate cross-period jumps and form a time-continuous phase-unwrapped time series; performing time-domain smoothing on the phase-unwrapped time series to suppress high-frequency noise, and constructing an atmospheric phase field based on multiple phase-unwrapped time series that have undergone time-domain smoothing to subtract the atmospheric phase delay component from the phase-unwrapped time series to form an atmospheric-compensated phase time series; constructing a time-series decomposition model after performing time-domain smoothing to separate the high-frequency random disturbance phase component and the low-frequency deformation trend phase component, and converting the low-frequency deformation trend phase component into a deformation time series.

[0036] Specifically, for the complex echoes of stable scatterers in the interferometric data cube, the complex pixel values ​​corresponding to each stable scatterer at all observation times are first read sequentially according to the time order of the observation times. The complex echoes of the same stable scatterer at different observation times are organized into a complex echo time series by time index. The complex echo consists of two parts: amplitude and phase. The amplitude is used to characterize the scattering intensity of the radar electromagnetic wave by the stable scatterer at that moment, reflecting the magnitude of its scattering energy. The phase is used to characterize the relative change in the round-trip propagation path of the radar signal, and is highly sensitive to displacement changes at the millimeter or even sub-millimeter level. By constructing a complex echo time series with the observation time as the horizontal axis and the complex echo as the vertical axis, the scattering intensity change and scattering phase change of each stable scatterer throughout the entire monitoring period are recorded on the same time trajectory. This provides a continuous, complete and physically consistent basic data sequence for subsequent phase unwrapping, time-domain smoothing, atmospheric compensation, and time series decomposition for the same stable scatterer.

[0037] Based on the phase component of the complex echo time series, the phase value corresponding to each observation moment is separated from the complex form to form a phase time series. The phase difference between adjacent observation moments is analyzed on the time axis to detect whether there are cross-cycle jumps caused by crossing an integer phase cycle. Since the radar phase itself is a circular quantity with a period of 2π, when the phase change caused by the actual deformation exceeds π, it will appear as an abrupt jump in the phase time series, that is, a sudden jump from a value close to π to close to -π or a reverse jump. This jump is not a sudden change in physical displacement, but a periodic back-turn phenomenon caused by phase entanglement. Therefore, it is necessary to unwrap the phase time series to map all phase samples from the circular interval to a continuous real number interval. Specifically, by comparing whether the phase difference between adjacent moments exceeds a preset phase difference threshold, if it exceeds, one or more integer multiples of 2π are automatically compensated, so that the processed phase unwrapped time series presents a continuous and smooth change trajectory in time, and no longer exhibits fake cross-cycle jumps, thus truly reflecting the cumulative phase evolution process of the stable scatterer throughout the entire monitoring period.

[0038] For the phase unwrapped time series, time-domain smoothing is performed along the time axis to suppress high-frequency noise components caused by measurement noise, system noise, and sporadic speckle fluctuations as much as possible, so that the slowly changing deformation information and large-scale atmospheric disturbance components are mainly retained. Time-domain smoothing can be performed using weighted average filtering, low-pass filtering, or smoothing methods based on local polynomial fitting under a sliding time window. Within each time window, the phase unwrapped values ​​of multiple adjacent observation times are used to jointly constrain the phase estimation at a single time moment, thereby reducing the influence of isolated anomalies on the overall phase evolution trajectory. After performing time-domain smoothing on multiple stable scatterers, the smoothed data of all stable scatterers at the same observation time are then processed. Phase values ​​are fitted or interpolated in space to construct a three-dimensional atmospheric phase field at the observation time. The atmospheric phase field is used to characterize the additional phase delay caused by the non-uniform spatial distribution of atmospheric refractive index. It has the characteristics of being spatially smooth and temporally gradual. After obtaining the atmospheric phase field at multiple times, for each stable scatterer, the atmospheric phase delay component at the corresponding position is subtracted from its phase unwrapping time series at each observation time. That is, the atmospheric phase field value is subtracted from the phase unwrapping value, thus forming the atmospheric compensated phase time series. This ensures that the main component retained in the compensated phase time series comes from the phase change caused by the real structural deformation, and no longer includes the pseudo-deformation component caused by the change of atmospheric refractive index.

[0039] For the compensated phase time series after atmospheric compensation, a time-series decomposition model is constructed to further decompose the compensated phase time series in the time domain into high-frequency random disturbance phase components and low-frequency deformation trend phase components. The high-frequency random disturbance phase components mainly originate from residual noise, local speckle evolution, and small unstructured disturbances, exhibiting an oscillating sequence with rapid alternation of positive and negative values ​​and a mean close to zero over short timescales. The low-frequency deformation trend phase components mainly correspond to the gradual deformation of buildings driven by factors such as foundation settlement, temperature expansion and contraction, long-term load, and structural fatigue, exhibiting a slowly changing monotonic or near-monotonic curve over long timescales. The time-series decomposition model can employ a combination of time-domain low-pass and high-pass filters, empirical mode decomposition, wavelet-based multi-scale decomposition, or Kalman filtering and state space decomposition. The recursive estimation method based on inter-frequency modeling, by limiting the time scales of the low-frequency subspace and the high-frequency subspace, makes the decomposed low-frequency deformation trend phase component smooth and stable, while the high-frequency random disturbance phase component is concentrated in the short-period band. After obtaining the low-frequency deformation trend phase component, according to the geometric relationship between the interferometric radar operating wavelength and the line-of-sight direction, the phase change in this phase component is converted according to the linear correspondence between deformation and phase change in the line-of-sight direction. The phase difference of each observation time relative to the reference time is converted into the displacement difference in the line-of-sight direction, thereby forming a time series of building deformation in the line-of-sight direction in chronological order. This deformation time series can be directly used to perform projection conversion and construct a deformation field covering the building area by combining the building geometry and the radar incident angle.

[0040] S130 combines the time series of deformation variables with the geometric relationship of the building and the radar incident angle for projection conversion, and constructs a deformation field covering the building area through multiple time series of deformation variables to identify overall settlement, local uneven settlement or facade torsional deformation.

[0041] In one possible implementation, the deformation time series is combined with the building's geometric relationships and radar incident angle for projection conversion. A deformation field covering the building area is constructed using multiple deformation time series to identify overall settlement, localized uneven settlement, or facade torsional deformation. Specifically, this includes: converting the pixel coordinates of stable scatterers in the interferometric data cube into three-dimensional coordinates in the building coordinate system through georegulation and 3D geometric calibration; and establishing projection relationships between the line-of-sight direction and the building's vertical, horizontal, and component normal directions based on the building's geometric relationships and radar incident angle. The building coordinate system is established based on the interferometric data cube and building structural data. The projection relationship for each stable scatterer is... The linear deformation time series is decomposed into a decomposed time series based on the projection relationship. The decomposed time series includes the vertical deformation time series, the horizontal deformation time series of the building, or the normal deformation time series of the components. At each observation time, the decomposed time series corresponding to multiple stable scatterers are divided into multiple deformation analysis regions according to the geometric relationship of the building. Within each deformation analysis region, a structural fitting model is constructed based on the spatial distribution of stable scatterers and the deformation values ​​to form a local continuous deformation distribution. Continuity constraints are applied between adjacent deformation analysis regions to ensure that the distribution of each local continuous deformation maintains spatial continuity within the entire building area, thereby forming a deformation field covering the building area at each observation time.

[0042] Specifically, firstly, the pixel coordinates of stable scatterers in the interferometric data cube are converted into three-dimensional coordinates in the building coordinate system through georegistration and 3D geometric calibration. The pixel coordinates in the interferometric data cube are typically projected in the range, azimuth, and height directions to form the radar's imaging coordinates to the ground. Georegistration converts the radar imaging coordinates into geographic coordinates by aligning with external geographic reference data. 3D geometric calibration registers the geographic coordinates with the building coordinate system in the building structural data. The building coordinate system defines the horizontal and vertical axes with the building's foundation plane as a reference, thus obtaining the three-dimensional coordinate position of each stable scatterer in the building coordinate system. Based on the building's geometric relationships and the radar's incident angle, a projection relationship is established using the angle between the interferometric radar's line-of-sight unit vector and the corresponding unit vectors in the building's vertical, horizontal, and component normal directions. The line-of-sight unit vector is denoted as:

[0043] The unit vectors for the vertical direction, horizontal direction, and normal direction of the building components are denoted as follows: , , ,in Determined by radar incident angle and azimuth angle, Aligned with the direction of the building's gravity, Along the main axis of the building's plan, Perpendicular to the plane where the component is located and pointing outward from the component, the projection relationship between the line of sight and each structural direction can be established by the direction cosine between the above vectors, so that the line of sight deformation for the same stable scattering body can be uniquely mapped to the deformation of the vertical direction of the building, the horizontal direction of the building, and the normal direction of the component.

[0044] For the time series of deformation along the line of sight of each stable scattering body, the aforementioned projection relationship is used to decompose the deformation along the line of sight into components in the vertical direction of the building, the horizontal direction of the building, and the normal direction of the components. This allows the stable scattering body to be visualized at time... The corresponding line-of-sight deformation is denoted as Then, the vertical deformation of the building, the horizontal deformation of the building, and the normal deformation of the components can be determined by the following formula:

[0045]

[0046]

[0047] in For the vertical deformation time series at time t The value of , The time series of horizontal deformation variables of the building at time t The value of , For the component normal deformation time series at time... The value of , , and Let be the dot product of the unit vector in the line of sight and the unit vector in the corresponding direction, representing the projection coefficient of the line of sight onto each structural direction, with values ​​ranging from -1 to 1. This is achieved by applying the projection coefficients to all moments within the entire observation period. By performing the above projection calculations, a set of decomposed time series can be formed for each stable scatterer, including the time series of vertical deformation, the time series of horizontal deformation of the building, and the time series of normal deformation of the components. This decomposes the original deformation time series, which was only in the line-of-sight direction, into multiple directional components with clear structural significance.

[0048] At each observation time, the decomposed time series of multiple stable scatterers at the corresponding time are divided into multiple deformation analysis regions according to the geometric relationship of the building. The geometric relationship of the building includes information such as floor division, structural axis division, facade division, and functional area division. By assigning stable scatterers belonging to the same floor, the same structural axis, the same facade area, or the same functional area to the same deformation analysis region, several regional-level deformation observation sets consistent with the structural logic of the building are formed at each observation time. In the process of division, the stable scatterers retain their three-dimensional coordinate positions in the building coordinate system and their corresponding vertical deformation, horizontal deformation, and component normal deformation. This ensures that each deformation analysis region has a sufficient number of observation points and can closely fit the actual structural layout of the building in space. This provides spatial support and structural semantic constraints for the subsequent establishment of a structural fitting model within the region and the acquisition of local continuous deformation distribution.

[0049] Within each deformation analysis region, a structural fitting model is constructed based on the spatial distribution of stable scatterers within the region and the corresponding deformation values, ensuring a continuous spatial distribution of deformation at discrete observation points within that region. For foundation or floor slab regions that approximate rigid plates, a planar fitting model can be used, representing the vertical deformation as a linear function of the horizontal coordinates in the building coordinate system. With coordinates At any moment The relationship can be represented as:

[0050] in For at any time Corresponding position The vertical deformation, , , For at any time The plane model parameters obtained by least squares fitting represent the overall vertical translation and the deformation gradients along the two horizontal directions, respectively. For facade areas or complex component areas, high-order polynomial surface fitting or spline surface fitting can be used to represent the component normal deformation or the building horizontal deformation as a nonlinear function of coordinates. The fitting accuracy and smoothness are balanced by adjusting the model order and regularization term. The establishment of the structural fitting model enables the deformation within each deformation analysis area to be spatially transformed from discrete point observations to a continuous field, laying the foundation for subsequent deformation field splicing and overall analysis within the entire building.

[0051] Continuity constraints are applied between adjacent deformation analysis regions to ensure that the distribution of local continuous deformation variables maintains spatial continuity across the entire building area, thus forming a deformation field covering the building region at each observation time. To achieve this, a set of boundary points is introduced at the boundaries between adjacent deformation analysis regions. The deformation variables given by the structural fitting models of different regions at these boundary points are differencing. A global constraint objective function is constructed by minimizing the sum of squared boundary differences. For example, at time... For any pair of adjacent regions i and j, the set of common boundary points Objective function can be defined

[0052] in For at any time The continuous constraint objective function, Let be the set of common boundary points between region i and region j. Let p be the deformation of the structural fitting model for region i at the boundary point p. For the deformation of the structural fitting model of region j at the boundary point p, by summing the objective functions of all adjacent region pairs and jointly optimizing the parameters of the structural fitting models of each region, the deformation of different regions at the common boundary is made as close as possible, thereby achieving a smooth splicing of the local continuous deformation distribution across the entire building area. Finally, at each observation time, a deformation field covering the building area that is spatially continuous and structurally consistent with the geometric relationship of the building is obtained, providing a complete spatial basis for identifying overall settlement, local uneven settlement, or facade torsional deformation.

[0053] In one possible implementation, the deformation time series is combined with the building's geometric relationships and radar incident angle for projection conversion. A deformation field covering the building area is constructed using multiple deformation time series to identify overall settlement, localized uneven settlement, or facade torsional deformation. Specifically, this includes: extracting vertical deformation variables located in the foundation area from the deformation field at each observation time; performing overall plane fitting on the vertical deformation variables to form a planar deformation variable distribution describing the overall vertical displacement trend of the foundation; solving for the tilt vector characterizing the overall settlement direction and overall tilt degree of the building based on the planar deformation variable distribution to identify overall settlement; and arranging the deformation field according to floor numbers and structural axes. The algorithm divides the stable scattering body into multiple vertical deformation analysis sequences, calculates the vertical deformation distribution along each structural axis within each floor, and calculates the vertical deformation distribution along the floor height on each structural axis. Local uneven settlement is identified by calculating the difference in vertical deformation between adjacent floors and adjacent structural axes, as well as the spatial gradient of vertical deformation. The normal deformation of components located in the facade area is extracted from the deformation field, and the normal deformation of components is organized into multiple normal deformation profile sequences according to the building height direction. The torsional deformation of the facade around the building's vertical axis is identified by comparing the difference in the distribution of normal deformation of components along the height direction at different facade boundaries.

[0054] Specifically, when extracting the vertical deformation of the foundation region from the deformation field at each observation time, the area representing the plane range of the foundation in the building coordinate system is taken as the foundation region. The planar projection range of the foundation slab, pile cap, and substructure in direct contact with the ground is determined using building structural data or design drawings. Within this range, the positions of all stable scatterers with vertical deformation are selected. The vertical deformation of these stable scatterers at the current observation time is taken as the set of vertical deformation sampling points for the foundation region. The vertical deformation describes the displacement of the foundation in the direction of gravity; positive values ​​usually indicate uplift, and negative values ​​indicate settlement. The planar coordinates of each sampling point in the foundation region in the building coordinate system are retained. With the corresponding vertical deformation At each observation time A set of discrete sampling data that can characterize the vertical deformation state of the basic region is formed, providing input data for subsequent overall plane fitting.

[0055] When performing global plane fitting on the vertical deformation of the foundation area, the discrete vertical deformation in the sampling point set is regarded as the values ​​of a plane function at the corresponding coordinate positions. A linear function defined in the building coordinate system plane is fitted using the least squares method to characterize the overall vertical displacement trend of the foundation; at the observation time... The corresponding planar deformation distribution can be expressed as:

[0056] in Indicates position At any moment The vertical deformation, This represents the average vertical displacement of the entire foundation. and These represent the vertical deformation gradients along the horizontal coordinate axes of the two buildings, respectively; the solution is obtained by minimizing the sum of squared fitting residuals from all sampling points in the basic region. , , The optimal value, i.e., minimizing:

[0057] in The set of sampling points within the base region, , Let p be the plane coordinates of the sampling point. The plane deformation distribution obtained by solving the above optimization problem not only describes the horizontal vertical displacement of the foundation as a whole, but also the trend of vertical deformation along different directions.

[0058] When solving for the tilt vector based on the distribution of planar deformation, the gradient of the planar function is regarded as a quantitative representation of the overall tilt of the foundation. The gradient vector is defined as:

[0059] in For at any time The planar gradient vector points in the direction of increasing vertical subsidence, and its magnitude represents the change in vertical displacement per unit horizontal distance. By constructing a three-dimensional tilt vector, the planar gradient is elevated to the building's three-dimensional coordinate system. The tilt vector can be represented as:

[0060] in The horizontal component is composed of , It was decided that the vertical component would be set to -1 to uniformly represent the downward tilt direction, through normalization. The unit vector of the overall settlement direction of the foundation can be obtained. The magnitude of the tilt vector is combined with the foundation plane dimensions to estimate the overall tilt angle. When the magnitude of the tilt vector continues to increase over time or the direction shifts significantly, it can be determined that the overall settlement trend of the building is aggravated or the settlement direction has changed, thus realizing the quantitative correlation between overall settlement identification and foundation plane deformation distribution.

[0061] When dividing stable scatterers into multiple vertical deformation analysis sequences based on floor numbers and structural axis numbers according to the deformation field, the floor height range and structural axis position defined in the building structural data are used as the division criteria. Each stable scatterer is mapped to the corresponding floor number and structural axis number according to its vertical and horizontal coordinates in the building coordinate system, and stable scatterers mapped to the same floor and the same structural axis constitute a vertical deformation analysis sequence. Within each floor, the vertical deformation distribution of the same floor located on different structural axes is statistically analyzed along each structural axis direction. By comparing the difference in vertical deformation at the same floor height for each structural axis, it is identified whether there are local areas within the floor that show significant downward deflection or upward lifting relative to the overall plane. Along the floor height direction on each structural axis, the vertical deformation distribution of different floors is statistically analyzed. By comparing the difference in vertical deformation at the same structural axis between adjacent floors, it is identified whether there are continuously increasing settlement zones along the height direction. The difference in vertical deformation can be expressed as:

[0062]

[0063] in Indicates at time The difference in vertical deformation between the i-th floor and the j-th floor on the specified axis. Indicates at time The difference in vertical deformation between the m-th and n-th axes on the same floor is further calculated in the plane using the spatial gradient field of the vertical deformation:

[0064] Regions with large gradient modulus are identified, and regions with significant vertical deformation differences and prominent spatial gradients are marked as local non-uniform settlement regions, thus achieving unification between the identification of local non-uniform settlement and the vertical deformation analysis sequence and spatial distribution of deformation field.

[0065] When extracting the component normal deformation variables from the deformation field in the facade region, the scattering points located on the facade plane in the stable scattering bodies are selected using the building facade geometry information. The facade region is identified by judging whether the three-dimensional coordinates of the stable scattering bodies satisfy the facade plane equation. The component normal deformation variables corresponding to these stable scattering bodies are regarded as the displacement of the facade region in the normal direction. The component normal deformation variables are used to describe the degree of convexity or concavity of the curtain wall, shear wall, or facade panel relative to the initial plane of the facade. When the component normal deformation variables are organized into multiple component normal deformation variable profile sequences according to the building height direction, the facade is divided into several height sections along the building height direction. Different facade boundary lines are divided on each height section according to the facade boundary position, such as the left boundary, right boundary, or the front and rear facade edges. The component normal deformation variables corresponding to each facade boundary line are collected on each height section to form a component normal deformation variable profile sequence that varies with height. This represents the normal deformation of the component at height z at the facade boundary b. The difference between the component normal deformation profile sequences at different facade boundaries is defined, for example, as a difference function for the boundary profiles of a pair of opposing facades.

[0066] when When the height z maintains the same sign and its amplitude gradually increases, it can be determined that the facade exhibits a torsional tendency around the building's vertical axis. A positive sign indicates that one facade is offset outward relative to the other, while a negative sign indicates offset in the opposite direction. This can be further analyzed by... The joint evolution analysis over time and height can identify the location, direction and degree of torsional deformation of the facade, thereby accurately mapping the identification of torsional deformation of the facade to the distribution of normal deformation of components in the deformation field.

[0067] S140, based on the comparison between the rate of change and acceleration of the deformation field and the preset safety threshold, determine whether the building deformation has entered an abnormal state and generate a deformation early warning result.

[0068] In one possible implementation, the building deformation is assessed based on a comparison of the deformation field's rate of change, acceleration, and a preset safety threshold to determine whether the building deformation has entered an abnormal state and to generate a deformation warning result. Specifically, this includes: performing a fitting process on the deformation time series of each stable scatterer in the deformation field covering the building area at each observation time to obtain the corresponding deformation rate of change time series and deformation acceleration of change time series; constructing a spatiotemporal deformation field containing dynamic deformation characteristics by combining the deformation time series, deformation rate of change time series, and deformation acceleration of change time series; and setting deformation parameters for different areas of the building based on the building type, structural system, and material properties. Safety thresholds, deformation rate safety thresholds, and deformation acceleration safety thresholds are established. The deformation, deformation rate, and deformation acceleration at each stable scatterer location in the spatiotemporal deformation field are sequentially compared with their corresponding preset safety thresholds to identify areas of accelerated deformation development and areas exceeding deformation limits. Spatial clustering and spatial connectivity analysis are performed on the comparison results of adjacent stable scatterers to form continuous deformation risk areas. Based on the degree to which the deformation, deformation rate, and deformation acceleration exceed the limits in each risk area, the deformation risk areas are divided into different risk levels. Deformation early warning results are generated by combining the risk level, the spatial extent of the risk area, and the corresponding deformation type.

[0069] Specifically, when performing fitting processing on the deformation time series of each stable scatterer in the deformation field covering the building area at each observation time, the deformation of the same stable scatterer at all observation times is organized into a complete deformation time series in chronological order. Within a preset time window, a linear or low-order polynomial curve is used to locally fit this deformation time series. The slope of the fitted curve is used as the deformation rate of change within the time window, and the curvature of the fitted curve or the difference in the rate of change between adjacent time windows is used as the deformation acceleration. The time window is updated by sliding throughout the monitoring period, so that at each stable scatterer location, a deformation rate of change time series describing the trend of deformation over time and a deformation acceleration time series describing whether the trend of deformation of change is accelerating or decelerating are obtained simultaneously. The deformation rate of change is used to characterize how fast the deformation changes per unit time, and the deformation acceleration is used to characterize the acceleration or deceleration of the deformation rate of change over time. This expands each stable scatterer from a single deformation time series into a time feature set containing three dynamic feature sequences.

[0070] When constructing a spatiotemporal deformation field containing dynamic deformation characteristics by combining the deformation time series, deformation rate time series, and deformation acceleration time series, the three-dimensional spatial position of the stable scatterer in the building coordinate system is used as a spatial index. The deformation time series, deformation rate time series, and deformation acceleration time series corresponding to this position are bound to the same spatial point as multidimensional time attributes. This ensures that each stable scatterer not only has fixed spatial coordinates but also possesses three time evolution curves describing the degree of deformation, deformation development rate, and degree of deformation intensification at that location throughout the entire monitoring period. By completing the above binding operation for all stable scatterers, the spatial dimension covers the building foundation area, superstructure area, and facade area, and the temporal dimension covers all observation times, forming a spatiotemporal deformation field that can simultaneously reflect deformation, deformation rate, and deformation acceleration. This spatiotemporal deformation field provides a unified data carrier for subsequent threshold comparison, risk area identification, and risk level classification.

[0071] When setting safety thresholds for deformation, rate of change of deformation, and acceleration of change of deformation for different areas of a building based on building type, structural system, and material properties, the safety performance requirements for the foundation area, main load-bearing component area, secondary load-bearing component area, and non-load-bearing enclosure structure area are first determined based on the building's structural design parameters, structural stress calculation results, and relevant standards and specifications. Then, considering the allowable deformation range and deformation development rate of different materials under long-term loads and environmental effects, the maximum allowable cumulative deformation within the monitoring period is set as the safety threshold for deformation, the maximum allowable change in deformation per unit time is set as the safety threshold for rate of change of deformation, and the maximum allowable increment of rate of change per unit time is set as the safety threshold for acceleration of change of deformation. At the same time, these safety thresholds are checked and adapted with reference to historical monitoring data and previous damage cases, so that the safety thresholds can cover deformation fluctuations under normal operating conditions and provide sufficiently sensitive over-limit judgment criteria when deformation enters an abnormal development stage.

[0072] To identify regions of accelerated deformation and regions of excessive deformation, the deformation, rate of change of deformation, and acceleration of change of deformation at each stable scatterer location in the spatiotemporal deformation field are compared sequentially with corresponding preset safety thresholds. Specifically, at each stable scatterer location, the deformation, rate of change of deformation, and acceleration of change of deformation are read at the current moment and compared item by item with the safety thresholds for deformation, rate of change of deformation, and acceleration of change of deformation corresponding to the structural region at that location. If the deformation has not yet exceeded the safety threshold but the rate of change of deformation is close to or exceeds the safety threshold for rate of change of deformation, the comparison is made accordingly. When the deformation change acceleration is positive and close to or exceeds the safe threshold for deformation change acceleration, the location of the stable scatterer is marked as a deformation acceleration development point, indicating that the deformation at this location has not yet exceeded the limit but there is a significant deterioration trend; when the deformation is close to or exceeds the safe threshold for deformation, regardless of whether the deformation change rate and deformation change acceleration exceed the limit, the location of the stable scatterer is marked as a deformation over-limit point, indicating that the location has reached or exceeded the allowable deformation range. By comparing the above thresholds throughout the entire building, a discrete set of deformation acceleration development points and a set of deformation over-limit points reflecting the deformation development state can be obtained.

[0073] When performing spatial clustering and spatial connectivity analysis on the comparison results of adjacent stable scatterers to form continuous deformation risk regions, all deformation acceleration points and deformation exceedance points are clustered according to their spatial location in the building coordinate system. Based on spatial distance thresholds or adjacency relationships based on the building's structural mesh, anomalous points that are close to each other or located in the same structural unit are aggregated into several candidate risk regions. Then, spatial connectivity analysis is performed on each candidate risk region, eliminating regions consisting of only a few isolated anomalous points and lacking spatial extension. Regions consisting of multiple adjacent anomalous points and exhibiting spatial continuity are identified as deformation risk regions. In each deformation... Within the risk zone, the degree of deformation exceeding the limit, the degree of deformation change rate exceeding the limit, and the degree of deformation change acceleration exceeding the limit of all anomalies within the zone are statistically analyzed. Taking into account the number of exceeding points, the intensity of exceeding the limit, and the duration, the deformation risk zone is divided into different risk levels. For example, a region with both large-scale deformation exceeding the limit and obvious accelerated development characteristics is classified as a high-risk level; a region where the deformation change rate and deformation change acceleration are close to the threshold but the deformation has not yet exceeded the limit is classified as a medium-risk level; and a region with only local slight exceeding the limit or short-term exceeding the limit is classified as a low-risk level. This ensures that the risk level classification strictly corresponds to the multidimensional dynamic characteristics in the spatiotemporal deformation field.

[0074] When generating deformation early warning results by combining risk level, spatial range of risk area, and corresponding deformation type, the spatial range of each deformation risk area in the building coordinate system is associated with its floor number, structural axis number, and component type. This clarifies the structural location and affected building functional area of ​​the risk area. Combined with deformation types identified in the previous deformation field analysis steps, such as overall settlement, local uneven settlement, or facade torsional deformation, each risk area is marked as a specific deformation type category. Based on this, the risk level is used as a severity indicator, the spatial range as an impact range indicator, and the deformation type as a pattern indicator. The evolution curves of deformation, deformation rate of change, and deformation acceleration over time at typical anomaly points are used to illustrate the development trend. These are combined to form a deformation early warning entry for each risk area. Multiple deformation early warning entries constitute a complete deformation early warning result, which is used in subsequent steps to comprehensively analyze the building's operating status and historical records, ultimately outputting building deformation early warning information.

[0075] S150 performs a comprehensive analysis based on the deformation early warning results, combined with the building's operating status and historical records, and outputs building deformation early warning information.

[0076] In one possible implementation, a comprehensive analysis is performed based on the deformation early warning results, combined with the building's operational status and historical records, to output building deformation early warning information. Specifically, this includes: constructing a spatiotemporal access channel to map the building's operational status and historical records to a spatiotemporal indexing system consistent with the building's coordinate system; and aligning the spatial location and time window of each deformation risk area in the deformation early warning results with the building's operational status and historical records in a spatiotemporal manner; and determining whether the deformation risk area belongs to a historical record based on the temporal evolution characteristics of the deformation, rate of change of deformation, and acceleration of deformation in the deformation risk area, and the sequential relationship between load changes, external disturbance events, and construction activities in the building's operational status. The system assesses whether the deformation evolution of a deformation risk area exceeds the previous deformation envelope, based on the continuation of historical deformation trends, reactivation of existing defects, or new abnormal deformations, combined with historical warning results and corresponding response measures. It also differentiates the risk interpretation results of deformation risk areas according to the functional zoning, component importance, and structural safety level in the building's operational status, forming a risk level ranking oriented towards operation and maintenance needs. Finally, it organizes and outputs building deformation warning information by describing the spatial location, risk level, deformation type, deformation amount, relationship with preset safety thresholds, deformation development trend, and correlation with the building's operational status and historical records for each deformation risk area.

[0077] Specifically, when constructing a spatiotemporal access channel to map building operational status and historical records to a spatiotemporal indexing system consistent with the building coordinate system, operational status data and historical record data from different data sources are standardized to the same spatial and temporal reference. Operational status data includes current and historical load condition records, equipment start-up and shutdown records, functional adjustment records, construction and renovation activity records, and external disturbance event records. Each of these records has a timestamp and associated location description, such as floor number, axis number, room number, or component number. By associating these location descriptions with pre-established floor divisions, structural axis grids, and component 3D models in the building coordinate system, the spatial location range of each operational status record in the building coordinate system is obtained, while retaining the time of occurrence or duration interval of the record. As a time index, historical records include past deformation monitoring results, past deformation early warning results, on-site inspection reports, structural appraisal reports, and reinforcement and maintenance records. These are also mapped to the same spatiotemporal index system through location mapping and timestamp association. Based on this, using the spatial location and time window of each deformation risk area in the deformation early warning results as an index, the spatiotemporal index system is used to retrieve operational status records and historical records that intersect with the spatial range and overlap with the time window. The relevant operational status events and historical handling processes are associated with the deformation risk areas one by one, thereby achieving spatiotemporal alignment between deformation early warning results and building operational status and historical records. The spatiotemporal access channel essentially encodes heterogeneous data through unified spatial coordinates and a unified time axis, enabling subsequent comprehensive analysis to compare deformation evolution and engineering background within the same spatiotemporal framework.

[0078] When judging the deformation risk area based on the temporal evolution characteristics of deformation, deformation rate of change, and deformation acceleration, and the chronological relationship between load condition changes in building operation, external disturbance events, and construction activities, the time series of representative stable scatterers or regional average deformation within the deformation risk area is used as the basis. Combined with the corresponding time series of deformation rate of change and deformation acceleration, the deformation evolution stage of the risk area within the monitoring period is analyzed, such as the slow monotonic evolution stage, the sudden increase stage, or the accelerated development stage. Subsequently, this temporal evolution process is compared with the temporal sequence of load condition changes, external disturbance events, and construction activities after spatiotemporal alignment. Load condition changes include concentrated equipment activation, increased storage load, and increased personnel usage intensity. External disturbance events include earthquakes, heavy rainfall, excavation of foundation pits around the foundation, and significant changes in groundwater level. Construction activities include drilling, demolition and alteration, rebar reinforcement, or foundation deepening. When a significant acceleration or abrupt change in deformation evolution occurs after an external action or construction event and there is a reasonable causal relationship in time, the deformation risk area can be judged to be a newly induced abnormal deformation caused by external factors. When the deformation evolution trend is consistent with the previous monitoring stage over a long time scale and there is no significant acceleration, but only a continuation of the original deformation trend, the deformation risk area can be judged to be a continuation of the historical deformation trend. When the historical record has already recorded defects or abnormal deformations at this location or adjacent locations and reinforcement or repair has been carried out, and the current deformation evolution characteristics are obviously reactivated or exceed the stable deformation level after reinforcement, the deformation risk area can be judged to be a reactivation of existing defects.

[0079] Based on this, and by combining historical warning results, on-site verification conclusions, and reinforcement and maintenance measures, we assess whether the deformation, rate of change, and acceleration of change of the current deformation risk area exceed the historical deformation envelope. If the current deformation state exceeds the extreme value range of deformation and dynamic characteristics recorded in previous monitoring cycles, it is determined that the deformation evolution has broken through the previous deformation envelope, indicating that the risk level has an upward trend.

[0080] When differentiating the risk interpretation results of deformation risk areas based on the functional zoning, component importance, and structural safety level of the building in its operational status, the spatial location of the deformation risk area in the building coordinate system is mapped onto the building's functional plan and structural layout drawing. This identifies whether the risk area is located in important functional zones such as densely populated areas, evacuation routes, machine rooms, and control rooms; critical load-bearing component areas such as main beams, main columns, core tubes, and foundation caps; secondary load-bearing component areas; or non-load-bearing enclosure structure areas. Furthermore, considering the structural safety level requirements for different component categories given in the design documents, different risk interpretations are assigned to areas with the same degree of deformation exceeding limits. For example, for deformation risk areas located in high-safety-level load-bearing components or important functional zones, even if their deformation, rate of change of deformation, or acceleration of change of deformation is only slightly higher than the safety limit, different risk interpretations are applied. Thresholds can also be used to increase the risk level in the interpretation, adjusting it from the original medium risk to high risk to meet higher requirements for structural reliability and personnel safety. For deformation risk areas located on non-load-bearing enclosure components or decorative components that can accept a certain degree of deformation, when the deformation slightly exceeds the limit and the rate and acceleration of deformation change are controlled within a reasonable range, the urgency in the interpretation can be appropriately reduced, defining it as a risk with limited impact on overall structural safety but potentially affecting function or durability. By incorporating the structural importance and functional use of deformation risk areas into the interpretation logic, the risk level ranking not only reflects the numerical degree of deformation itself but also reflects the comprehensive impact of deformation on building operation and safety, thereby forming a priority ranking result oriented towards operation and maintenance needs, guiding limited resources to be prioritized for high-risk and high-importance areas.

[0081] When organizing the spatial location, risk level, deformation type, relationship between deformation amount and preset safety thresholds, deformation development trend, and correlation with building operation status and historical records of each deformation risk area into building deformation early warning information, the multidimensional information obtained from the aforementioned analysis is archived according to region. A location description in the building coordinate system is given for each deformation risk area, including floor number, structural axis number, component type, and specific range identification in the building floor plan or elevation drawing. Simultaneously, the final determined risk level and corresponding deformation type of the area are marked, such as accelerated overall settlement, intensified local uneven settlement, or development of facade torsional deformation. A comparison is given between the deformation amount at a representative point or the regional average and the deformation safety threshold, and a comparison is given between the deformation change rate and the deformation change rate safety threshold. The system compares the deformation acceleration with the deformation acceleration safety threshold to visually demonstrate the degree of exceeding limits and the safety margin. Furthermore, it provides a textual description of the deformation development trend, indicating whether the risk area is stable, developing slowly, accelerating continuously, or undergoing a sudden change. It also clarifies the correspondence between this trend and recent load changes, construction activities, or external disturbances, as well as its relationship with past deformation envelopes, such as whether it indicates a resurgence of historical problems or a new risk point. Based on this, the above information is output in a structured form as building deformation early warning information, which can be presented in the form of reports, electronic documents, or alarm entries in an information system. This allows maintenance personnel to conduct on-site verification, further testing, structural assessment, and necessary reinforcement and maintenance decisions based on the early warning information, thus completing the entire closed-loop process from time-series interference data to executable early warning information.

[0082] This embodiment also discloses a building deformation early warning device based on interferometric radar time series analysis, referring to... Figure 2 The device includes an acquisition module 201, a processing module 202, and an output module 203. It is used to execute any of the above-described methods for early warning of building deformation based on interferometric radar time-series analysis, wherein: The acquisition module 201 is used to acquire multi-time interference echoes covering the building area to form an interference data cube, and to screen stable scatterers based on scattering stability to construct a set of stable scatterers; Processing module 202 is used to perform phase unwrapping, time-domain smoothing and atmospheric compensation based on the complex echo corresponding to the stable scatterer, and to separate high-frequency random disturbances and low-frequency deformation trends from the complex echo through time-series decomposition to obtain the time series of building deformation in the line-of-sight direction; Processing module 202 is used to combine the deformation time series with the building geometry and radar incident angle for projection conversion, and to construct a deformation field covering the building area through multiple deformation time series to identify overall settlement, local uneven settlement or facade torsional deformation. Processing module 202 is used to determine whether the building deformation has entered an abnormal state based on the comparison between the change rate and change acceleration of the deformation field and a preset safety threshold, and to generate a deformation early warning result. The output module 203 is used to perform comprehensive analysis based on the deformation early warning results, combined with the building's operating status and historical records, and output building deformation early warning information.

[0083] It should be noted that the above embodiments of the apparatus are only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0084] This embodiment also discloses an electronic device, as shown in the reference. Figure 3 The electronic device may include: at least one processor 301, at least one communication bus 302, user interface 303, network interface 304, and at least one memory 305.

[0085] The communication bus 302 is used to enable communication between these components.

[0086] The user interface 303 may include a display screen and a camera. Optionally, the user interface 303 may also include a standard wired interface and a wireless interface.

[0087] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0088] The processor 301 may include one or more processing cores. The processor 301 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 305, and by calling data stored in memory 305. Optionally, the processor 301 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 301 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 301 and may be implemented as a separate chip.

[0089] The memory 305 may include random access memory (RAM) or read-only memory. Optionally, the memory may include a non-transitory computer-readable storage medium. The memory 305 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 305 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch functionality, sound playback functionality, image playback functionality, etc.), instructions for implementing the various method embodiments described above, etc.; the data storage area may store data involved in the various method embodiments described above, etc. Optionally, the memory 305 may also be at least one storage device located remotely from the aforementioned processor 301. As a computer storage medium, the memory 305 may include an operating system, a network communication module, a user interface 303 module, and an application program for a building deformation early warning method based on interferometric radar time-series analysis.

[0090] exist Figure 3In the electronic device shown, the user interface 303 is mainly used to provide an input interface for the user and to obtain the user input data; while the processor 301 can be used to call an application program stored in the memory 305 that is a building deformation early warning method based on interferometric radar timing analysis. When executed by one or more processors 301, the electronic device executes one or more methods as described in the above embodiments.

[0091] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, as some steps can be performed in other orders or simultaneously according to the present invention. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0092] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0093] In the several embodiments provided by this invention, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some service interface; the indirect coupling or communication connection between apparatuses or units may be electrical or other forms.

[0094] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0095] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0096] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory 305 and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned memory 305 includes various media capable of storing program code, such as a USB flash drive, external hard drive, magnetic disk, or optical disk.

[0097] The present invention also discloses a computer-readable storage medium storing instructions. When executed by one or more processors 301, these instructions cause an electronic device to perform one or more methods as described in the above embodiments.

[0098] The above are merely exemplary embodiments of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Those skilled in the art will readily conceive of other embodiments of this disclosure upon considering the specification and the disclosure of practical truths. This invention is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A method for early warning of building deformation based on interferometric radar time series analysis, characterized in that, The method includes: Multi-time interferometric echoes covering the building area are acquired to form an interferometric data cube, and stable scatterers are selected based on scattering stability to construct a set of stable scatterers; Phase unwrapping, time-domain smoothing, and atmospheric compensation are performed based on the complex echoes corresponding to the stable scatterer. High-frequency random disturbances and low-frequency deformation trends are separated from the complex echoes through time-series decomposition to obtain the time series of building deformation in the line-of-sight direction. The deformation time series is combined with the building geometry and radar incident angle for projection conversion, and a deformation field covering the building area is constructed through multiple deformation time series to identify overall settlement, local uneven settlement or facade torsional deformation. Based on the comparison between the rate of change and acceleration of the deformation field and the preset safety threshold, it is determined whether the building deformation has entered an abnormal state and a deformation early warning result is generated. Based on the deformation early warning results, a comprehensive analysis is performed on the building's operating status and historical records, and building deformation early warning information is output.

2. The building deformation early warning method based on interferometric radar time series analysis according to claim 1, characterized in that, The process involves performing phase unwrapping, time-domain smoothing, and atmospheric compensation based on the complex echoes corresponding to the stable scatterer, and separating high-frequency random disturbances and low-frequency deformation trends from the complex echoes through time-series decomposition to obtain the time series of building deformation along the line-of-sight direction. Specifically, this includes: The complex echoes of the stable scatterer in the interferometric data cube are extracted in order of observation time to form a complex echo time series; Phase unwrapping is performed on the complex echo time series to eliminate cross-cycle jumps and form a time-continuous phase unwrapped time series; The phase unwrapped time series is subjected to time-domain smoothing to suppress high-frequency noise, and an atmospheric phase field is constructed based on multiple phase unwrapped time series that have undergone time-domain smoothing to subtract the atmospheric phase delay component from the phase unwrapped time series to form an atmospheric compensated phase time series. A time-series decomposition model is constructed for the time-domain smoothing process to separate the high-frequency random disturbance phase component and the low-frequency deformation trend phase component, and the low-frequency deformation trend phase component is converted into the deformation time series.

3. The building deformation early warning method based on interferometric radar time series analysis according to claim 1, characterized in that, The process of combining the deformation time series with the building's geometric relationship and radar incident angle for projection conversion, and constructing a deformation field covering the building area using multiple deformation time series to identify overall settlement, local uneven settlement, or facade torsional deformation, specifically includes: The pixel coordinates of the stable scatterer in the interferometric data cube are converted into three-dimensional coordinate positions in the building coordinate system through georegistration and three-dimensional geometric calibration. The projection relationship between the line of sight direction and the vertical direction, horizontal direction and normal direction of the building are established based on the building geometry and radar incident angle. The building coordinate system is established based on the interferometric data cube and the building structural data. The deformation time series for the line-of-sight direction of each stable scatterer is decomposed into a decomposed time series according to the projection relationship. The decomposed time series includes a vertical deformation time series, a building horizontal deformation time series, or a component normal deformation time series. At each observation time, the decomposed time series corresponding to the multiple stable scatterers are divided into multiple deformation analysis regions according to the geometric relationship of the building; In each deformation analysis region, a structural fitting model is constructed based on the spatial distribution and deformation values ​​of the stable scatterer to form a local continuous deformation distribution; A continuity constraint is applied between adjacent deformation analysis regions to ensure that the distribution of each local continuous deformation variable maintains spatial continuity within the entire building area, thereby forming a deformation field covering the building area at each observation time.

4. The building deformation early warning method based on interferometric radar time series analysis according to claim 3, characterized in that, The step of combining the deformation time series with the building's geometric relationship and radar incident angle for projection conversion, and constructing a deformation field covering the building area through multiple deformation time series to identify overall settlement, local uneven settlement, or facade torsional deformation, specifically also includes: At each observation time, the vertical deformation located in the base region is extracted from the deformation field; Perform an overall plane fitting on the vertical deformation to form a planar deformation distribution that describes the overall vertical displacement trend of the foundation; Based on the aforementioned planar deformation distribution, the tilt vector used to characterize the overall settlement direction and overall tilt degree of the building is solved to identify the overall settlement; Based on the deformation field, the stable scattering body is divided into multiple vertical deformation analysis sequences according to the floor number and structural axis number. The vertical deformation distribution along each structural axis is calculated in each floor, and the vertical deformation distribution along the floor height is calculated on each structural axis. Local uneven settlement is identified by calculating the difference in vertical deformation between adjacent floors and adjacent structural axes and the spatial gradient of vertical deformation. Extract the component normal deformation variables located in the facade area from the deformation field, organize the component normal deformation variables into multiple component normal deformation variable profile sequences according to the building height direction, and identify the torsional deformation of the facade around the vertical axis of the building by comparing the difference distribution of component normal deformation variables along the height direction at different facade boundaries.

5. The building deformation early warning method based on interferometric radar time series analysis according to claim 1, characterized in that, The step of determining whether the building deformation has entered an abnormal state and generating a deformation early warning result based on the comparison between the rate of change and acceleration of change of the deformation field and a preset safety threshold specifically includes: At each observation time, the deformation time series of each stable scatterer in the deformation field covering the building area is fitted to obtain the corresponding deformation rate time series and deformation acceleration time series. The time series of deformation, the time series of deformation rate of change, and the time series of deformation acceleration are used together to construct a spatiotemporal deformation field containing dynamic deformation characteristics; Based on the building type, structural system, and material properties, set safety thresholds for deformation, rate of deformation change, and acceleration of deformation change for different areas of the building. The deformation, deformation rate, and deformation acceleration of each stable scatterer position in the spatiotemporal deformation field are sequentially compared with the corresponding preset safety thresholds to identify the deformation acceleration development region and the deformation exceeding limit region. Spatial clustering and spatial connectivity analysis are performed on the comparison results of adjacent stable scatterers to form continuous deformation risk regions. Based on the degree of exceeding of the limits of deformation, deformation rate of change and deformation acceleration in each risk region, the deformation risk regions are divided into different risk levels. The deformation warning result is generated by combining the risk level, the spatial range of the risk area, and the corresponding deformation type.

6. The building deformation early warning method based on interferometric radar time series analysis according to claim 1, characterized in that, The process of comprehensively analyzing the deformation early warning results in conjunction with the building's operational status and historical records, and outputting building deformation early warning information, specifically includes: A spatiotemporal access channel is constructed to map the building's operating status and historical records to a spatiotemporal indexing system consistent with the building's coordinate system, and the spatial location and time window of each deformation risk area in the deformation early warning result are spatiotemporally aligned with the building's operating status and historical records. Based on the temporal evolution characteristics of the deformation, deformation rate of change, and deformation acceleration of the deformation risk area, and the relationship between the load condition changes in the building's operating state, external disturbance events, and construction activities, it is determined whether the deformation risk area belongs to the continuation of historical deformation trends, reactivation of existing defects, or new abnormal deformation. In conjunction with the previous warning results in the historical records and the corresponding disposal measures, it is assessed whether the deformation evolution of the deformation risk area has broken through the previous deformation envelope. The risk interpretation results of the deformation risk areas are differentiated and adjusted based on the functional zoning, component importance and structural safety level in the building's operation status to form a risk level ranking oriented towards operation and maintenance needs; The spatial location, risk level, deformation type, deformation amount, relationship with each preset safety threshold, deformation development trend, and correlation with the building's operating status and historical records of each deformation risk area are organized into building deformation early warning information and output.

7. The building deformation early warning method based on interferometric radar time series analysis according to claim 1, characterized in that, The process of acquiring multi-time interference echoes covering the building area to form an interference data cube, and then selecting stable scatterers based on scattering stability to construct a set of stable scatterers, specifically includes: On the interferometric radar side, the observation geometry parameters and observation period are planned to obtain interferometric echoes covering the building facade, roof structure and surrounding hardened area at multiple observation times. Geometric registration and resampling are performed on the interferometric echoes at different observation times to construct an interferometric data cube to characterize the spatiotemporal response features of the building area. Calculate the scattering stability index for the complex reflection sequence corresponding to each pixel in the interferometric data cube; From the plurality of pixels, candidate pixels for stable scatterers are selected, wherein the candidate pixels for stable scatterers are pixels whose scattering stability value is higher than a preset first threshold and whose phase fluctuation value is less than a preset second threshold, based on the scattering stability index. A stable scatterer set is constructed by combining the candidate stable scatterer pixels.

8. A building deformation early warning device based on interferometric radar time series analysis, characterized in that, The device is used to execute a building deformation early warning method based on interferometric radar time series analysis as described in any one of claims 1-7. The device includes an acquisition module, a processing module, and an output module, wherein: The acquisition module is used to acquire multi-time interference echoes covering the building area to form an interference data cube, and to screen stable scatterers based on scattering stability to construct a set of stable scatterers; The processing module is used to perform phase unwrapping, time-domain smoothing and atmospheric compensation based on the complex echo corresponding to the stable scatterer, and to separate high-frequency random disturbances and low-frequency deformation trends from the complex echo through time-series decomposition to obtain the time series of building deformation in the line-of-sight direction. The processing module is used to project and convert the deformation time series into building geometry and radar incident angle, and to construct a deformation field covering the building area through multiple deformation time series to identify overall settlement, local uneven settlement or facade torsional deformation. The processing module is used to determine whether the building deformation has entered an abnormal state based on the comparison between the change rate and change acceleration of the deformation field and a preset safety threshold, and to generate a deformation early warning result. The output module is used to perform comprehensive analysis based on the deformation early warning results, combined with the building's operating status and historical records, and output building deformation early warning information.

9. An electronic device, characterized in that, The device includes a processor, a communication bus, a user interface, a network interface, and a memory. The memory is used to store instructions. The user interface and the network interface are both used to communicate with other devices. The communication bus is used to enable communication between the components within the electronic device. The processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, perform the method as described in any one of claims 1-7.

Images