Slope landslide monitoring method and system

Abstract

The application discloses a kind of side slope landslide monitoring method and system, the monitoring method includes to the echo signal after pulse compression is handled, detection and calculation, obtain rainfall data;According to the accumulated rainfall, determine landslide confidence data;The echo signal after pulse compression is carried out digital beam forming imaging, obtain two-dimensional complex scattering image;According to rainfall intensity, two-dimensional complex scattering image is compensated;Extract PS point and carry out deformation inversion, obtain deformation data;According to deformation data and accumulated rainfall, real-time evaluation comprehensive landslide risk grade, and then determine acquisition frequency, according to deformation data, determine initial landslide risk grade;According to landslide confidence data, the initial landslide risk grade is corrected, according to the corrected landslide risk grade, landslide risk early warning is carried out.The application can realize deformation monitoring and rainfall monitoring simultaneously by using a set of radar equipment, effectively reduces the procurement, installation and operation and maintenance cost of a set of equipment, reduces the cost, reduces the monitoring failure rate.

Description

Technical Field

[0001] This invention belongs to the field of radar monitoring technology, and in particular relates to a method and system for monitoring slope landslides. Background Technology

[0002] In practical slope monitoring applications, the vast majority of landslide geological disasters are caused by rainfall. Therefore, in the design of slope landslide monitoring systems, the measurement of key landslide triggers (such as rainfall) and the monitoring of key specific manifestations of the landslide process (such as surface deformation) are indispensable. In the construction of slope landslide monitoring systems, deformation monitoring and rainfall monitoring both require separate equipment procurement, installation, and maintenance to ensure the fidelity of monitoring data and the accuracy of early warning information, which places a certain burden on system construction and application costs.

[0003] Existing sources of rainfall information for slope areas include:

[0004] (1) Automatic weather station: mainly refers to small area rain gauges such as tipping bucket rain gauges, piezoelectric rain gauges, and photoelectric rain gauges. The sensing area is very limited, only a few square meters, and it is impossible to obtain real rainfall data over a large area. The monitoring method of using points to represent areas and multiple points to represent areas is only a data fitting algorithm. The accuracy of the fitted data is highly limited by the density of monitoring points. High-density deployment of monitoring points brings high cost pressure to the monitoring system. At the same time, real rainfall often exhibits non-uniform and non-linear characteristics in a wide area, and the rainfall data in local areas is very likely to be seriously distorted.

[0005] (2) Rain measuring radar: It can obtain the location, intensity and speed of rainfall in real time and is one of the important sources of rainfall information for meteorological departments. However, the construction cost is very high, and it is only for important urban areas with very low coverage.

[0006] (3) Meteorological satellites: They estimate precipitation by observing cloud layers and water vapor content. They are one of the main sources of rainfall information for meteorological departments. However, they have problems such as inaccurate precipitation data, inaccurate precipitation time data, and inaccurate precipitation area data. They are not suitable for the high precision and real-time requirements of rainfall information in slope monitoring.

[0007] In practice, due to insufficient monitoring budget, some construction parties tend to use rainfall information released by meteorological departments to obtain rainfall information. As a result, the inaccuracy of some rainfall information leads to great unreliability of early warning information in actual slope monitoring.

[0008] Slope radar is a radar device used to monitor slope stability. It monitors the deformation and displacement of the slope surface by collecting phase changes on the slope surface. Environmental changes (rainfall) as well as natural dynamics and human activities on the slope surface have a strong impact on the imaging of radar echoes. Therefore, there is a certain interference in the measurement of phase interference. The method of self-correction and compensation based on its own data has certain limitations. At the same time, existing slope monitoring radars cannot directly sense changes in the external environment and need to use external environmental sensing data to filter out interference signals.

[0009] In actual slope monitoring scenarios, most operations are conducted in the field. Power supply and grid conditions for monitoring equipment are extremely limited, necessitating solar power solutions for most. However, limitations in current solar panel conversion efficiency, effective sunshine duration, and the size and cost of solar panel installation necessitate low-power monitoring equipment. Furthermore, radar equipment consumes tens of watts of power, making it impossible to effectively and universally apply solar power solutions under conventional conditions.

[0010] Currently, radar equipment employs an intermittent sleep mode to reduce average power consumption throughout the day. However, the frequency of signal acquisition needs to match the actual landslide risk at the monitoring site. Otherwise, too high a frequency leads to excessive power consumption, while too low a frequency fails to meet the real-time requirements of monitoring data during periods of high landslide risk. Currently, the radar's acquisition frequency is mainly controlled by assessing the risk level based on its own deformation data. However, deformation data is susceptible to interference, resulting in limited accuracy. Schemes that fuse external information with radar deformation data to improve the accuracy of risk level assessment increase the system's external dependence and also amplify the problem of reduced data accuracy due to external interference.

[0011] The patent document CN115993586A discloses a slope radar monitoring method with micro-deformation monitoring and moving target detection. This method includes pulse compression of the echo signal from the target area acquired by the radar to obtain a pulse-compressed output signal. If monitoring micro-deformation in the target area, the two-dimensional image formed by the pulse-compressed output signal is used to form a complex scattering image pair. The distance after phase compensation following fitting of the stable PS points is calculated to obtain the directional displacement accuracy information of the micro-deformation in the target area. If detecting moving targets in the target area, when at least one moving target is detected, the pulse-compressed output signal is processed through two-pulse cancellation and Fourier transform to form a range-velocity spectrum. The motion information of the moving targets in the range-velocity spectrum is calculated using the unit-average constant false alarm rate method, and the continuous motion trajectory corresponding to each moving target is output. While this method improves the accuracy of deformation monitoring and moving target detection, it does not incorporate rainfall data during deformation monitoring. Rainfall is a major cause of slope landslides, and relying solely on deformation data may result in missed or false alarms. Furthermore, the radar operates in continuous mode, leading to high power consumption. Summary of the Invention

[0012] The purpose of this invention is to provide a slope landslide monitoring method and system to solve at least one of the following problems in the prior art: the high construction and operation and maintenance costs caused by slope monitoring and rainfall monitoring being two separate types of equipment; the difficulty in matching high accuracy rainfall information with high construction costs; and inaccurate deformation monitoring data.

[0013] This invention solves the above-mentioned technical problems through the following technical solution: a slope landslide monitoring method, comprising:

[0014] Acquire the echo signal from the MIMO radar device and perform pulse compression on the echo signal;

[0015] The pulse-compressed echo signal is processed, detected, and calculated to obtain rainfall data; wherein, the rainfall data includes rainfall intensity and cumulative rainfall;

[0016] The landslide confidence data is determined based on the cumulative rainfall.

[0017] Digital beamforming imaging is performed on the pulse-compressed echo signal to obtain a two-dimensional complex scattering image;

[0018] The two-dimensional complex scattering image is compensated based on the rainfall intensity;

[0019] PS points are extracted from the compensated two-dimensional complex scattering image, and deformation inversion is performed based on the PS points to obtain deformation data;

[0020] The comprehensive landslide risk level is assessed in real time based on the deformation data and cumulative rainfall, and the data collection frequency is determined accordingly. The operation of the MIMO radar equipment is then controlled based on the data collection frequency.

[0021] The initial landslide risk level is determined based on the deformation data.

[0022] The initial landslide risk level is corrected based on the landslide confidence data, and a landslide risk warning is issued based on the corrected landslide risk level.

[0023] Furthermore, the processing, detection, and calculation of the pulse-compressed echo signal includes:

[0024] Clutter suppression is applied to the echo signal after pulse compression.

[0025] Perform FFT on the echo signal after clutter suppression to obtain the frequency domain signal and power spectrum;

[0026] By performing constant false alarm rate (CFAR) detection on the frequency domain signal, the precipitation target can be obtained.

[0027] Calculate precipitation-related parameters based on the stated precipitation target;

[0028] Smooth, check, and filter precipitation-related parameters;

[0029] Rainfall intensity and cumulative rainfall are calculated based on the power spectrum and the selected precipitation-related parameters.

[0030] Furthermore, the formula for calculating the rainfall intensity is:

[0031] ;

[0032] Where R represents rainfall intensity; a and b are both empirical coefficients; Z represents reflectivity factor; C represents radar constant; and r represents the distance from the precipitation target to the radar. This represents the power spectrum.

[0033] Further, determining landslide confidence data based on the cumulative rainfall includes:

[0034] If the cumulative rainfall in time T1 is less than AR1 and the cumulative rainfall in time T2 is less than AR2, then the landslide confidence data is 0.

[0035] If the cumulative rainfall during AR1 ≤ T1 < AR3, or the cumulative rainfall during AR2 ≤ T2, then the landslide confidence data is 0.

[0036] If the cumulative rainfall during AR3 ≤ T1 is less than AR4, or if the cumulative rainfall during AR3 ≤ T2, then the landslide confidence data is 1.

[0037] If the cumulative rainfall during AR4 ≤ T1 is less than AR5, or if the cumulative rainfall during AR4 ≤ T3, then the landslide confidence data is 1.

[0038] If the cumulative rainfall within AR5 ≤ T1, or the cumulative rainfall within AR6 ≤ T3, then the landslide confidence data is 2.

[0039] Where T1, T2 and T3 represent time periods, AR1, AR2, AR3, AR4, AR5 and AR6 represent rainfall thresholds, T3 < T2 < T1, AR1 < AR2 < AR3, AR4 < AR6 < AR5.

[0040] Further, the two-dimensional complex scattering image is compensated according to the rainfall intensity, including:

[0041] The attenuation coefficient is calculated based on the rainfall intensity, and the specific calculation formula is as follows:

[0042] ;

[0043] in, R represents the attenuation coefficient; R represents the rainfall intensity. and All represent empirical coefficients;

[0044] The radar echo path integral attenuation value is calculated based on the attenuation coefficient, and the specific calculation formula is as follows:

[0045] ;

[0046] in, This represents the radar echo path integral attenuation value; Represents the distance variable; The value represents the unit distance; r represents the distance from the precipitation target to the radar.

[0047] The radar echo path integral attenuation value is superimposed with the echo intensity of the two-dimensional complex scattering image to obtain the compensated two-dimensional complex scattering image.

[0048] Furthermore, based on the deformation data and cumulative rainfall, the comprehensive landslide risk level is assessed in real time, thereby determining the data collection frequency, including:

[0049] Deformation level and rainfall level are determined based on the deformation data and the cumulative rainfall, respectively.

[0050] The comprehensive landslide risk level is assessed in real time based on the deformation level and rainfall level, thereby determining the data collection frequency, specifically including:

[0051] If the deformation level is Level 1 and / or the rainfall level is Level 1, then the comprehensive landslide risk level is Level 0, and the data collection frequency is f1.

[0052] If the deformation level is level 2 and the rainfall level is not greater than level 2, or if the deformation level is not greater than level 2 and the rainfall level is level 2, then the comprehensive landslide risk level is level 1, and the data collection frequency is f2.

[0053] If both the deformation level and the rainfall level are level 2, or the deformation level is level 3 and the rainfall level is no greater than level 3, or the deformation level is no greater than level 3 and the rainfall level is level 3, then the comprehensive landslide risk level is level 2, and the data collection frequency is f3.

[0054] If both the deformation level and the rainfall level are level three, or the deformation level is level four and the rainfall level is no greater than level four, or the deformation level is no greater than level four and the rainfall level is level four, then the comprehensive landslide risk level is level three, and the data collection frequency is f4.

[0055] If both the deformation level and the rainfall level are level four, then the comprehensive landslide risk level is level four, and the data collection frequency is f5.

[0056] Where f1 < f2 < f3 < f4 < f5.

[0057] Further, the deformation level and rainfall level are determined based on the deformation data and the cumulative rainfall, respectively, including:

[0058] If the cumulative deformation within time T4 is less than AL1, then the deformation level is level one.

[0059] If the cumulative deformation within the time interval AL1≤T4 <AL2, then the deformation level is level two.

[0060] If the cumulative deformation within the time interval AL2≤T4 is <AL3, then the deformation level is level three.

[0061] If the cumulative deformation within the time interval AL3≤T4, then the deformation level is level four;

[0062] If the cumulative rainfall within time T2 is less than AR2, then the rainfall level is Level 1.

[0063] If the cumulative rainfall within the time interval AR2≤T2 is <AR3, then the rainfall level is Level II;

[0064] If the cumulative rainfall within the time interval AR3≤T2 is <AR4, then the rainfall level is level three;

[0065] If the cumulative rainfall within the time interval AR4≤T2, then the rainfall level is level four;

[0066] Where T4 and T2 represent time periods; AL1, AL2, and AL3 represent deformation thresholds; and AR2, AR3, and AR4 represent rainfall thresholds.

[0067] Furthermore, the corrected landslide risk level is equal to the sum of the landslide confidence data and the initial landslide risk level.

[0068] Based on the same concept, the present invention also provides a slope landslide monitoring system, including a MIMO radar device and a host computer, wherein the MIMO radar device communicates with the host computer;

[0069] The MIMO radar device is used to acquire the echo signal of the MIMO radar device and to perform pulse compression on the echo signal.

[0070] The pulse-compressed echo signal is processed, detected, and calculated to obtain rainfall data; wherein, the rainfall data includes rainfall intensity and cumulative rainfall;

[0071] Digital beamforming imaging is performed on the pulse-compressed echo signal to obtain a two-dimensional complex scattering image;

[0072] The two-dimensional complex scattering image is compensated based on the rainfall intensity;

[0073] PS points are extracted from the compensated two-dimensional complex scattering image, and deformation inversion is performed based on the PS points to obtain deformation data;

[0074] The comprehensive landslide risk level is assessed in real time based on the deformation data and cumulative rainfall, and the data collection frequency is determined accordingly. The operation of the MIMO radar equipment is then controlled based on the data collection frequency.

[0075] The host computer is used to determine the initial landslide risk level based on the deformation data; determine landslide confidence data based on the cumulative rainfall; correct the initial landslide risk level based on the landslide confidence data; and issue a landslide risk warning based on the corrected landslide risk level.

[0076] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0077] This invention uses a single radar device to simultaneously monitor deformation and rainfall, effectively reducing the procurement, installation, and maintenance costs of a single device, thus lowering costs and reducing the monitoring failure rate.

[0078] This invention, based on the original functions of radar, achieves large-scale, high-frequency, high-resolution, and high-precision monitoring coverage of the monitored area without adding hardware equipment. It can acquire more accurate and real-time rainfall data, and the rainfall monitoring range is perfectly matched with the slope monitoring range, enabling more accurate correction of landslide risk levels and helping to improve the accuracy of slope landslide monitoring and early warning. By using rainfall data that is perfectly matched with the slope monitoring range to effectively compensate for the two-dimensional complex scattering image, the accuracy of deformation monitoring is improved.

[0079] This invention performs a comprehensive self-assessment of landslide risk levels and adaptive data acquisition frequency based on deformation and rainfall data acquired by radar itself. It is unaffected by communication signal interruptions or poor signal strength, effectively reducing the communication traffic required for monitoring data during low-risk landslide periods in long-term monitoring, as well as the computing and storage resources of the platform (e.g., host computer). This significantly reduces monitoring and maintenance costs and decreases the probability of monitoring interruptions due to insufficient computing and storage resources on the platform. During high-risk landslide periods, higher-frequency data acquisition and analysis are performed, matching the real-time nature of the monitoring data with the urgency of the high-risk period, ensuring the effectiveness and timeliness of the overall monitoring data, and thus achieving more accurate landslide risk early warning.

[0080] Compared to existing technologies (i.e., patent document with publication number CN115993586A), this invention combines deformation data with rainfall data, which can eliminate interference from non-rainfall-induced factors, triggering early warnings when rainfall and deformation occur simultaneously, thus reducing false alarms; it also provides early warnings when deformation does not reach the threshold but rainfall exceeds the standard, thus reducing missed alarms; in addition, this invention determines the radar equipment's acquisition frequency based on deformation data and rainfall data, so that the radar equipment is not always in continuous operation mode, reducing power consumption. Attached Figure Description

[0081] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only one embodiment of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0082] Figure 1 This is a flowchart of the slope landslide monitoring method in an embodiment of the present invention. Detailed Implementation

[0083] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0084] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0085] Example 1

[0086] like Figure 1As shown, the slope landslide monitoring method provided in this embodiment of the invention includes the following steps:

[0087] S1: Acquire the echo signal from the MIMO radar device and perform pulse compression on the echo signal.

[0088] MIMO (Multiple-Input Multiple-Output) radar equipment employs multiple transmitting antennas and multiple receiving antennas. At the transmitting end, different transmitting antennas transmit mutually orthogonal signals, which propagate in space and illuminate the target. At the receiving end, multiple receiving antennas simultaneously receive signals reflected back from the target (i.e., echo signals). By processing and analyzing these echo signals, information about the target is obtained by utilizing the correlation and differences between the signals.

[0089] The MIMO radar equipment transmits signals based on the acquisition frequency and receives echo signals. The echo signals are then pulse-compressed and processed simultaneously for both deformation monitoring and rainfall monitoring. Pulse compression allows for high range resolution while preserving the high energy of long pulses, ensuring long-range detection capabilities. This helps in distinguishing minute displacements during deformation monitoring and differentiating precipitation layers at different altitudes during rainfall monitoring, thereby improving the accuracy of both measurements.

[0090] S2: Process, detect, and calculate the echo signal after pulse compression to obtain rainfall data.

[0091] In a specific embodiment of the present invention, the processing, detection, and calculation of the pulse-compressed echo signal include:

[0092] S2.1: Clutter suppression is applied to the echo signal after pulse compression.

[0093] By using filtering and gain adjustment, clutter is suppressed in the pulse-compressed echo signal, improving the signal-to-noise ratio and signal-to-clutter ratio of the echo signal while preserving the effective signal. This avoids nonlinear distortion of the amplifier caused by strong clutter, balances the signal strength at different distances, and ensures the accuracy of rainfall data monitoring.

[0094] S2.2: Perform FFT (Fast Fourier Transform) operation on the echo signal after clutter suppression to obtain the frequency domain signal and power spectrum.

[0095] By performing an FFT operation on the clutter-suppressed echo signal, the time-domain echo signal is converted to the frequency domain, thus obtaining the frequency-domain signal. The power spectrum of the echoes is then calculated by performing FFT operations on multiple clutter-suppressed echo signals.

[0096] S2.3: Perform constant false alarm rate (CFAR) detection on the frequency domain signal to obtain the precipitation target.

[0097] Based on the statistical characteristics of noise and clutter in a local area, a detection threshold is adaptively set, and a constant false alarm rate (CFAR) detection method is used to detect targets in the frequency domain, enabling the detection of real precipitation targets in noisy and cluttered environments. Specifically, the average intensity of the neighboring cells around the target is first estimated, and then this average value is multiplied by a fixed threshold to obtain the detection threshold.

[0098] S2.4: Calculate precipitation-related parameters based on precipitation targets.

[0099] In this embodiment, the parameters related to precipitation include echo intensity, velocity, spectral width, reflectivity factor, and distance from the target to the radar.

[0100] S2.5: Smooth, check, and filter parameters related to precipitation.

[0101] To reduce data fluctuations and noise, methods such as mean filtering or median filtering are used to smooth precipitation-related parameters before inspection and screening. For example, it is checked whether the echo intensity is within a reasonable range, whether the velocity conforms to physical laws, and whether the spectral width is abnormal. Echo intensities that exceed reasonable ranges, velocities that do not conform to physical laws, and abnormal spectral widths are marked or removed, and repaired or supplemented as needed.

[0102] S2.6: Calculate rainfall intensity and cumulative rainfall based on the power spectrum and the screened precipitation-related parameters.

[0103] In this embodiment, the formula for calculating rainfall intensity is:

[0104] (1)

[0105] (2)

[0106] Where R represents rainfall intensity; a and b are both empirical coefficients that vary with precipitation type (rain, snow, hail) and regional climate differences; Z represents reflectivity factor; C represents radar constant, which is related to radar parameters; and r represents the distance from the precipitation target to the radar, in km. This represents the power spectrum. In this embodiment, if the layer is rain, then a is set to 100 and b is set to 1.6.

[0107] Rainfall is obtained by taking parameters related to precipitation and then accumulating them over time to obtain cumulative rainfall, such as cumulative rainfall in 24 hours, cumulative rainfall in 6 hours, and cumulative rainfall in 3 hours.

[0108] S3: Determine landslide confidence data based on cumulative rainfall.

[0109] To correct the initial landslide risk level based on rainfall data and improve the accuracy of landslide risk level assessment, landslide confidence data were first determined based on cumulative rainfall, as shown in Table 1:

[0110]

[0111] In Table 1, T1, T2, and T3 represent time periods, and AR1, AR2, AR3, AR4, AR5, and AR6 represent rainfall thresholds, where T3 < T2 < T1, AR1 < AR2 < AR3, and AR4 < AR6 < AR5. In this embodiment, T1 is set to 24 hours, T2 to 6 hours, and T3 to 3 hours; AR1 is set to 20 mm, AR2 to 30 mm, AR3 to 50 mm, AR4 to 100 mm, AR5 to 200 mm, and AR6 to 150 mm.

[0112] Landslide confidence data is determined by rainfall data. The initial landslide risk level of each local area is corrected by the corresponding thresholds of indicators such as deformation displacement, velocity, acceleration, reciprocal velocity, and displacement curve tangent angle. Finally, landslide risk warnings are pushed out according to the corrected landslide risk level to improve the accuracy of the warning.

[0113] S4: Perform digital beamforming (DBF) imaging on the pulse-compressed echo signal to obtain a two-dimensional complex scattering image.

[0114] Two-dimensional high-resolution imaging (range + azimuth) using DBF imaging can simultaneously achieve high precision, high resolution, and strong anti-interference capability, which is beneficial for accurately improving the phase change in the deformed region.

[0115] S5: Compensate the two-dimensional complex scattering image based on the rainfall intensity.

[0116] To improve the accuracy of deformation monitoring, the two-dimensional complex scattering image is compensated based on rainfall intensity. The specific compensation steps are as follows:

[0117] S5.1: Calculate the attenuation coefficient based on rainfall intensity. The specific calculation formula is as follows:

[0118] (3)

[0119] in, R represents the attenuation coefficient; R represents the rainfall intensity. and These are all empirical coefficients, related to the radar's operating frequency. Taking the X-band as an example, .

[0120] S5.2: Calculate the radar echo path integral attenuation value based on the attenuation coefficient. The specific calculation formula is as follows:

[0121] (4)

[0122] in, This represents the radar echo path integral attenuation value; Represents the distance variable; 'r' represents the distance per unit distance; 'r' represents the distance from the precipitation target to the radar.

[0123] S5.3: Integral attenuation value over radar echo path The compensated two-dimensional complex scattering image is obtained by superimposing the echo intensity of the two-dimensional complex scattering image.

[0124] The compensated echo intensity is equal to the sum of the radar echo path integral attenuation value and the echo intensity before compensation.

[0125] The echo signals of slope monitoring radar are affected by rainfall, and existing slope monitoring radars lack effective solutions to eliminate and compensate for this impact. Therefore, this invention employs rainfall data that highly matches the resolution and location information of the radar image data used for deformation monitoring to effectively compensate for the two-dimensional complex scattering image, thereby improving the intensity and phase stability of the monitoring points and ultimately enhancing the accuracy of deformation monitoring.

[0126] S6: Extract PS points (Permanent Scatterer) from the compensated two-dimensional complex scattering image, and perform deformation inversion based on the PS points to obtain deformation data.

[0127] High-quality points with coherence coefficients greater than the coefficient threshold and that meet the amplitude deviation requirements are extracted from the compensated two-dimensional complex scattering image and used as PS points.

[0128] In a two-dimensional complex scattering image, a point refers to a single resolvable cell. The size of the resolvable cell is determined by the radar's range resolution and angular resolution. The two-dimensional complex scattering image contains information such as the range, angle, echo intensity, and phase of each imaged target. In the compensated two-dimensional complex scattering image, points whose echo intensity reaches an intensity threshold (e.g., a signal-to-noise ratio higher than 100 dB) and whose echo intensity and phase are stable within a certain time period are considered high-quality points. Phase stability means that the phase fluctuation of the same target in multiple scans within a certain time period is within a very small range (i.e., within the phase threshold range). The coefficient threshold is usually adjusted according to environmental differences; in this embodiment, the coefficient threshold is set to 0.7~0.9. The more stable the phase, the higher the coherence coefficient.

[0129] Amplitude deviation refers to the difference between the echo intensity of the current frame and the average (or other reference value) of the echo intensity of multiple frames over a certain period of time. It is mainly used to evaluate the stability of the target echo intensity.

[0130] Differential interferometry is performed on the phase of the PS point to obtain the phase difference; atmospheric phase compensation is performed on the compensated two-dimensional complex scattering image using the phase difference of PS points with high echo intensity (i.e., echo intensity reaches the intensity threshold) and stable phase at different distance segments.

[0131] Atmospheric phase compensation technology and deformation inversion are both existing technologies. The phase difference of points with high echo intensity and stable phase in each range segment is taken as the phase fluctuation caused by atmospheric interference, and this phase difference is used as the phase compensation value for all points in the current range segment. For range segments where the phase compensation value cannot be selected, the phase difference of points determined before and after is combined with the distance difference to perform a linear function correspondence to determine the atmospheric phase compensation value for this range segment.

[0132] Based on the two-dimensional complex scattering image after atmospheric phase compensation, the deformation displacement is calculated using the inversion formula:

[0133] (5)

[0134] in, Indicates deformation displacement; Indicates the radar wavelength; This indicates the phase difference.

[0135] S7: Based on deformation data and cumulative rainfall, assess the comprehensive landslide risk level in real time, determine the data collection frequency, and control the operation of the MIMO radar equipment based on the data collection frequency.

[0136] In a specific embodiment of the present invention, the comprehensive landslide risk level is assessed in real time based on deformation data and cumulative rainfall, thereby determining the data collection frequency, including:

[0137] S7.1: Determine the deformation level and rainfall level based on the deformation data and cumulative rainfall, respectively, as shown in Table 2:

[0138]

[0139] In Table 2, T4 and T2 represent time periods; AL1, AL2, and AL3 represent deformation thresholds; and AR2, AR3, and AR4 represent rainfall thresholds. In this embodiment, T4 is set to 12 hours, T2 to 6 hours, AL1 to 5 mm, AL2 to 10 mm, AL3 to 50 mm, AR2 to 30 mm, AR3 to 50 mm, and AR4 to 100 mm.

[0140] S7.2: Real-time assessment of the comprehensive landslide risk level based on deformation level and rainfall level, thereby determining the data collection frequency, specifically including:

[0141] If the deformation level is Level 1 and / or the rainfall level is Level 1, then the comprehensive landslide risk level is Level 0, and the data collection frequency is f1.

[0142] If the deformation level is level 2 and the rainfall level is not greater than level 2, or if the deformation level is not greater than level 2 and the rainfall level is level 2, then the comprehensive landslide risk level is level 1, and the data collection frequency is f2.

[0143] If both the deformation level and the rainfall level are level 2, or the deformation level is level 3 and the rainfall level is no greater than level 3, or the deformation level is no greater than level 3 and the rainfall level is level 3, then the comprehensive landslide risk level is level 2, and the data collection frequency is f3.

[0144] If both the deformation level and the rainfall level are level three, or the deformation level is level four and the rainfall level is no greater than level four, or the deformation level is no greater than level four and the rainfall level is level four, then the comprehensive landslide risk level is level three, and the data collection frequency is f4.

[0145] If both the deformation level and the rainfall level are level four, then the overall landslide risk level is level four, and the data collection frequency is f5.

[0146] Wherein, f1 < f2 < f3 < f4 < f5. In this embodiment, f1 is set to 1 hour / time, f2 to 15 minutes / time, f3 to 5 minutes / time, f4 to 1 minute / time, and f5 to 10 seconds / time.

[0147] Long-term deformation data identifies high-risk slopes, while short-term rainfall data filters triggering events, reducing the waste of monitoring resources. A comprehensive landslide risk assessment based on deformation and rainfall levels comprehensively reflects the intrinsic mechanisms and external causes of landslides, shifting landslide risk assessment from passive response to proactive prediction. This provides a more reliable basis for disaster prevention and mitigation decisions, and significantly improves accuracy, timeliness, and reliability through multi-source data fusion. By simultaneously acquiring deformation and rainfall data within the radar equipment, the radar signal acquisition frequency is determined. Based on this acquisition frequency, the signal generation, transmission, reception, and processing of the MIMO radar equipment are controlled, ensuring that the signal acquisition frequency matches the actual landslide risk status at the monitoring site. This reduces power consumption while meeting the real-time requirements of monitoring data during high landslide risk periods.

[0148] S8: Determine the initial landslide risk level based on deformation data.

[0149] Deformation data includes deformation displacement, velocity, acceleration, reciprocal of velocity, and tangent angle of displacement curve. The host computer performs a preliminary assessment of the landslide risk level of each local area based on the long-term deformation data according to the threshold. The specific assessment process is based on existing technology.

[0150] S9: Correct the initial landslide risk level based on landslide confidence data, and issue a landslide risk warning based on the corrected landslide risk level.

[0151] The corrected landslide risk level is equal to the sum of the landslide confidence data and the initial landslide risk level. For example, if the initial landslide risk level is level three (corresponding to a value of 3) and the landslide confidence data is 1, then the corrected landslide risk level is equal to 4, and a red alert is issued according to Table 3.

[0152] Landslide risk warnings are issued based on the corrected landslide risk levels, as shown in Table 3:

[0153]

[0154] Because existing slope monitoring radar and rainfall monitoring equipment are two separate types of devices, and slope surface deformation monitoring and rainfall information monitoring are two indispensable monitoring components in a slope landslide monitoring system, it is necessary to purchase these two types of equipment separately in actual slope landslide monitoring systems. This results in disadvantages such as a large number of devices used in the system, high procurement costs, high installation costs, high operation and maintenance costs, high probability of monitoring failures, and low accuracy of rainfall information acquisition. This invention uses the echo signal from a single MIMO radar device to simultaneously process both deformation monitoring and rainfall monitoring signals after pulse compression, thereby achieving simultaneous slope deformation monitoring and rainfall monitoring. This effectively reduces the costs of purchasing, installing, and maintaining a single set of equipment and lowers the probability of monitoring interruptions due to hardware failures.

[0155] Because real-world rainfall often exhibits non-uniform and non-linear characteristics across a wide area, existing point-based local rainfall data collection technologies suffer from insufficient data accuracy. Meanwhile, high-accuracy rainfall radar is prohibitively expensive, limiting its widespread coverage. Meteorological satellite data also suffers from inaccuracies in precipitation magnitude, timing, and geographical distribution. To address these technical problems, this invention, building upon the existing functions of slope monitoring radar, achieves large-scale, high-frequency, high-resolution, and high-precision monitoring coverage of the monitored area without adding hardware. It acquires more accurate and real-time rainfall data, and the rainfall monitoring range perfectly matches the slope monitoring range, enabling more accurate correction of landslide risk levels and improving the accuracy of slope landslide monitoring and early warning.

[0156] Because existing landslide risk level assessments are based on composite judgments of long-term, multi-source data, and single sensor devices do not have the capability to collect multi-source data, current landslide risk level assessments are basically conducted by the platform (e.g., a host computer). That is, the platform sends data collection and upload frequencies to the monitoring terminals according to the determined landslide risk level. However, since many slope monitoring scenarios are located in remote field environments, the problems of intermittent data interruption and poor signal have not been completely resolved. As a result, the platform cannot send the corresponding data collection strategies to the monitoring terminals in a timely manner, and cannot achieve the adaptation between landslide risk level and collection frequency. Consequently, it cannot accurately perceive the key deformation data of high landslide risk levels, ultimately leading to inaccurate monitoring and early warning. To address the aforementioned issues, this invention utilizes deformation and rainfall data from a single MIMO radar device to perform comprehensive landslide risk level self-assessment and adaptive data acquisition frequency. This is unaffected by communication signal interruptions or malfunctions, effectively reducing the communication traffic required for monitoring data during low-risk landslide periods in long-term monitoring, as well as the computing and storage resources required on the platform. This significantly reduces monitoring and maintenance costs and decreases the probability of monitoring interruptions due to insufficient computing and storage resources on the platform. During high-risk landslide periods, higher-frequency data acquisition and analysis are performed, matching the real-time nature of the monitoring data with the urgency of the high-risk period, ensuring the effectiveness and timeliness of the overall monitoring data, thereby achieving more accurate landslide risk early warning.

[0157] Example 2

[0158] The slope landslide monitoring system provided in this embodiment of the invention includes a MIMO radar device and a host computer, and the MIMO radar device communicates with the host computer.

[0159] MIMO radar equipment is used for:

[0160] Acquire the echo signal from the MIMO radar equipment and perform pulse compression on the echo signal;

[0161] The pulse-compressed echo signal is processed, detected, and calculated to obtain rainfall data, which includes rainfall intensity and cumulative rainfall.

[0162] Digital beamforming imaging is performed on the pulse-compressed echo signal to obtain a two-dimensional complex scattering image;

[0163] Compensation is applied to the two-dimensional complex scattering image based on rainfall intensity;

[0164] PS points are extracted from the compensated two-dimensional complex scattering image, and deformation inversion is performed based on the PS points to obtain deformation data;

[0165] The comprehensive landslide risk level is assessed in real time based on deformation data and cumulative rainfall, and the data collection frequency is then determined. The operation of the MIMO radar equipment is then controlled based on the data collection frequency.

[0166] The host computer is used to: determine the initial landslide risk level based on deformation data; determine landslide confidence data based on cumulative rainfall; correct the initial landslide risk level based on the landslide confidence data; and issue a landslide risk warning based on the corrected landslide risk level.

[0167] In some specific embodiments of the present invention, the landslide monitoring system may incorporate the features of the landslide monitoring method in Embodiment 1 of the present invention, and vice versa.

[0168] The above description only discloses specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or modifications that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for monitoring slope landslides, characterized in that, The monitoring method includes: Acquire the echo signal from the MIMO radar device and perform pulse compression on the echo signal; The pulse-compressed echo signal is processed, detected, and calculated to obtain rainfall data; wherein, the rainfall data includes rainfall intensity and cumulative rainfall; The landslide confidence data is determined based on the cumulative rainfall. Digital beamforming imaging is performed on the pulse-compressed echo signal to obtain a two-dimensional complex scattering image; The two-dimensional complex scattering image is compensated based on the rainfall intensity; PS points are extracted from the compensated two-dimensional complex scattering image, and deformation inversion is performed based on the PS points to obtain deformation data; The comprehensive landslide risk level is assessed in real time based on the deformation data and cumulative rainfall, and the data collection frequency is determined accordingly. The operation of the MIMO radar equipment is then controlled based on the data collection frequency. The initial landslide risk level is determined based on the deformation data. The initial landslide risk level is corrected based on the landslide confidence data, and a landslide risk warning is issued based on the corrected landslide risk level.

2. The slope landslide monitoring method according to claim 1, characterized in that, The processing, detection, and calculation of the pulse-compressed echo signal include: Clutter suppression is applied to the echo signal after pulse compression. Perform FFT on the echo signal after clutter suppression to obtain the frequency domain signal and power spectrum; By performing constant false alarm rate (CFAR) detection on the frequency domain signal, the precipitation target can be obtained. Calculate precipitation-related parameters based on the stated precipitation target; Smooth, check, and filter precipitation-related parameters; Rainfall intensity and cumulative rainfall are calculated based on the power spectrum and the selected precipitation-related parameters.

3. The slope landslide monitoring method according to claim 2, characterized in that, The formula for calculating the rainfall intensity is: ； Where R represents rainfall intensity; a and b are both empirical coefficients; Z represents reflectivity factor; C represents radar constant; and r represents the distance from the precipitation target to the radar. This represents the power spectrum.

4. The slope landslide monitoring method according to claim 1, characterized in that, The landslide confidence data is determined based on the cumulative rainfall, including: If the cumulative rainfall in time T1 is less than AR1 and the cumulative rainfall in time T2 is less than AR2, then the landslide confidence data is 0. If the cumulative rainfall during AR1 ≤ T1 < AR3, or the cumulative rainfall during AR2 ≤ T2, then the landslide confidence data is 0. If the cumulative rainfall during AR3 ≤ T1 is less than AR4, or if the cumulative rainfall during AR3 ≤ T2, then the landslide confidence data is 1. If the cumulative rainfall during AR4 ≤ T1 is less than AR5, or if the cumulative rainfall during AR4 ≤ T3, then the landslide confidence data is 1. If the cumulative rainfall within AR5 ≤ T1, or the cumulative rainfall within AR6 ≤ T3, then the landslide confidence data is 2. Where T1, T2 and T3 represent time periods, AR1, AR2, AR3, AR4, AR5 and AR6 represent rainfall thresholds, T3 < T2 < T1, AR1 < AR2 < AR3, AR4 < AR6 < AR5.

5. The slope landslide monitoring method according to claim 1, characterized in that, Compensating the two-dimensional complex scattering image based on the rainfall intensity includes: The attenuation coefficient is calculated based on the rainfall intensity, and the specific calculation formula is as follows: ； in, R represents the attenuation coefficient; R represents the rainfall intensity. and All represent empirical coefficients; The radar echo path integral attenuation value is calculated based on the attenuation coefficient, and the specific calculation formula is as follows: ； in, This represents the radar echo path integral attenuation value; Represents the distance variable; The value represents the unit distance; r represents the distance from the precipitation target to the radar. The radar echo path integral attenuation value is superimposed with the echo intensity of the two-dimensional complex scattering image to obtain the compensated two-dimensional complex scattering image.

6. The slope landslide monitoring method according to any one of claims 1 to 5, characterized in that, Based on the deformation data and cumulative rainfall, a comprehensive landslide risk level is assessed in real time, thereby determining the data collection frequency, including: Deformation level and rainfall level are determined based on the deformation data and the cumulative rainfall, respectively. The comprehensive landslide risk level is assessed in real time based on the deformation level and rainfall level, and then the data collection frequency is determined.

7. The slope landslide monitoring method according to claim 6, characterized in that, Determining the deformation level and rainfall level based on the deformation data and the cumulative rainfall, respectively, includes: If the cumulative deformation within time T4 is less than AL1, then the deformation level is level one. If the cumulative deformation within the time interval AL1≤T4 <AL2, then the deformation level is level two. If the cumulative deformation within the time interval AL2≤T4 is <AL3, then the deformation level is level three. If the cumulative deformation within the time interval AL3≤T4, then the deformation level is level four; If the cumulative rainfall within time T2 is less than AR2, then the rainfall level is Level 1. If the cumulative rainfall within the time interval AR2≤T2 is <AR3, then the rainfall level is Level II; If the cumulative rainfall within the time interval AR3≤T2 is <AR4, then the rainfall level is level three; If the cumulative rainfall within the time interval AR4≤T2, then the rainfall level is level four; Where T4 and T2 represent time periods; AL1, AL2, and AL3 represent deformation thresholds; and AR2, AR3, and AR4 represent rainfall thresholds.

8. The slope landslide monitoring method according to claim 1, characterized in that, The corrected landslide risk level is equal to the sum of the landslide confidence data and the initial landslide risk level.

9. A slope landslide monitoring system, characterized in that, The monitoring system includes a MIMO radar device and a host computer, and the MIMO radar device communicates with the host computer. The MIMO radar device is used to acquire the echo signal of the MIMO radar device and to perform pulse compression on the echo signal. The pulse-compressed echo signal is processed, detected, and calculated to obtain rainfall data; wherein, the rainfall data includes rainfall intensity and cumulative rainfall; Digital beamforming imaging is performed on the pulse-compressed echo signal to obtain a two-dimensional complex scattering image; The two-dimensional complex scattering image is compensated based on the rainfall intensity; PS points are extracted from the compensated two-dimensional complex scattering image, and deformation inversion is performed based on the PS points to obtain deformation data; The comprehensive landslide risk level is assessed in real time based on the deformation data and cumulative rainfall, and the data collection frequency is determined accordingly. The operation of the MIMO radar equipment is then controlled based on the data collection frequency. The host computer is used to determine the initial landslide risk level based on the deformation data; determine landslide confidence data based on the cumulative rainfall; correct the initial landslide risk level based on the landslide confidence data; and issue a landslide risk warning based on the corrected landslide risk level.

Images