Geotechnical engineering slope stability real-time monitoring method and system

Abstract

The invention discloses a geotechnical engineering slope stability real-time monitoring method and a geotechnical engineering slope stability real-time monitoring system, which are characterized in that a blind source separation technology combining independent component analysis and physical constraint is introduced, an original displacement time sequence and an environment temperature time sequence of a plurality of GNSS (Global Navigation Satellite System) measuring points are regarded as multi-channel mixed signals, and statistical independence among signal sources is utilized to monitor the stability of a slope in real time. Periodic environment noise, instrument random noise and drift and real slope deformation signals are effectively separated from the mixed observation signals; and then, through correlation verification with physical quantities such as the field environment temperature and the like, automatic calibration is performed on the separated source signals, and temperature effect source signals and long-term creep source signals with physical labels are accurately identified and extracted, so that accurate elimination of noise components and high-fidelity reconstruction of pure deformation signals are realized. Therefore, the accuracy of real-time monitoring of the slope stability of geotechnical engineering can be effectively improved, and a solid data basis and a decision basis are provided for early warning and prevention of slope disasters.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of intelligent monitoring, and more particularly, to a real-time monitoring method and system for slope stability in geotechnical engineering. BACKGROUND

[0002] The stability of the slope in geotechnical engineering is directly related to the safety of people's lives and property and the smooth operation of the project, so it is crucial to monitor and warn it in real time and accurately. However, the traditional slope monitoring method faces many challenges in practical application, especially in complex and variable field environments, the reliability of the monitoring data is often disturbed.

[0003] The existing high-precision monitoring system based on GNSS is seriously affected by various noises when obtaining millimeter-level deformation of the slope. On the one hand, the change of environmental temperature will cause the thermal expansion and contraction of the monitoring pier, observation instrument and even the surface rock-soil body of the slope, which presents a periodic day-night fluctuation in the displacement data. The amplitude of this periodic noise is much larger than the weak real creep deformation in the early stage of slope instability, making it difficult for analysts to separate the weak but continuously growing "weak signal" from the "strong background". On the other hand, the instrument's inherent random noise and long-term zero drift problem also exist universally. Even in the case of constant environmental conditions, the GNSS solution still contains high-frequency random "spikes" and small, one-way zero drift over time (months or years), making the instantaneous velocity and acceleration calculation results extremely unreliable. This low-frequency drift is very similar in form to the long-term slow creep of the slope, and it is easy to be confused, which may misjudge the real deformation as an instrument problem or vice versa. In addition, the real slope deformation signal has a very low signal-to-noise ratio in the early creep stage, and the signal strength only increases significantly when it enters the acceleration stage, but by then the best warning opportunity has been missed. At the same time, the real deformation is not simply linear growth, often coupled with external factors such as rainfall and reservoir water level changes, and presents a complex nonlinear step growth. These complex signals are easily smoothed out by traditional filtering methods while smoothing noise, resulting in delayed or missed warnings.

[0004] Therefore, an optimized real-time monitoring method for slope stability in geotechnical engineering is expected. SUMMARY

[0005] To solve the above technical problems, the present application is proposed. Embodiments of the present application provide a geotechnical engineering slope stability real-time monitoring method and system, which introduces a blind source separation technology combining independent component analysis and physical constraints, takes the original displacement time series and environmental temperature time series of multiple GNSS measuring points as multi-channel mixed signals, uses the statistical independence between signal sources to effectively separate the periodic environmental noise, instrument random noise and drift from the mixed observation signals and the real slope deformation signals; then, through correlation verification with physical quantities such as on-site environmental temperature, the separated source signals are automatically calibrated, the temperature effect source signals and long-term creep source signals with physical labels are accurately identified and extracted, so as to realize accurate removal of noise components and high-fidelity reconstruction of pure deformation signals. In this way, the accuracy and reliability of geotechnical engineering slope stability real-time monitoring can be effectively improved, and a solid data foundation and decision basis are provided for early warning and prevention of slope disasters.

[0006] According to one aspect of the present application, a geotechnical engineering slope stability real-time monitoring method is provided, which comprises: obtaining original displacement time series and on-site environmental temperature time series of multiple GNSS measuring points; performing centering processing and matrix construction on the original displacement time series of the multiple GNSS measuring points to obtain an observation signal matrix; performing source signal separation based on independent component analysis on the observation signal matrix to obtain an estimated independent source signal matrix and an estimated mixing matrix; performing source signal automatic calibration on the estimated independent source signal matrix and the on-site environmental temperature time series to obtain source signals with physical labels; based on the source signals with physical labels, performing high-fidelity deformation signal reconstruction on the estimated mixing matrix and the observation signal matrix to obtain a high-fidelity deformation signal matrix; performing trend analysis on the high-fidelity deformation signal matrix to obtain a deformation rate and an acceleration rate.

[0007] According to another aspect of the present application, a geotechnical engineering slope stability real-time monitoring system is provided, which comprises: a data acquisition module for obtaining original displacement time series and on-site environmental temperature time series of multiple GNSS measuring points; a centering processing and matrix construction module for performing centering processing and matrix construction on the original displacement time series of the multiple GNSS measuring points to obtain an observation signal matrix; a source signal separation module for performing source signal separation based on independent component analysis on the observation signal matrix to obtain an estimated independent source signal matrix and an estimated mixing matrix; a source signal automatic calibration module, configured to perform source signal automatic calibration on the estimated independent source signal matrix and the field environment temperature time series to obtain a source signal with a physical label; a high-fidelity deformation signal reconstruction module, configured to perform high-fidelity deformation signal reconstruction on the estimated mixing matrix and the observation signal matrix based on the source signal with the physical label to obtain a high-fidelity deformation signal matrix; a trend analysis module, configured to perform trend analysis on the high-fidelity deformation signal matrix to obtain a deformation rate and an acceleration rate.

[0008] Compared with the prior art, the rock-soil engineering slope stability real-time monitoring method and system provided by the present application can effectively improve the accuracy and reliability of rock-soil engineering slope stability real-time monitoring, and provide a solid data foundation and decision basis for early warning and prevention of slope disasters. BRIEF DESCRIPTION OF DRAWINGS

[0009] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description of embodiments of the present application taken in conjunction with the accompanying drawings. The accompanying drawings are provided to aid in understanding the present application and are incorporated in and constitute a part of this specification, illustrate embodiments of the present application and serve to explain the principles of the present application. The drawings are not intended to limit the present application to specific embodiments. In the drawings, like reference numerals refer to like elements or steps throughout.

[0010] Figure 1 a flowchart of the rock-soil engineering slope stability real-time monitoring method according to the embodiments of the present application; Figure 2 a data flow schematic diagram of the rock-soil engineering slope stability real-time monitoring method according to the embodiments of the present application; Figure 3 a block diagram of the rock-soil engineering slope stability real-time monitoring system according to the embodiments of the present application. DETAILED DESCRIPTION

[0011] Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part but not all of the embodiments of the present application, and the present application can be implemented in many different ways. Therefore, the attached drawings should not be used to limit and define the present application and the present application should cover all changes falling within the scope of the present application.

[0012] As shown in the present application and claims, unless the context clearly indicates otherwise, the words "one", "an", "a", and / or "the" do not mean "only one", but can include a plurality or "one or more" unless the context clearly indicates otherwise. Generally, the terms "comprise" and "include" only indicate the inclusion of the steps and elements explicitly identified, and these steps and elements do not constitute an exclusive list, and the method or device can also include other steps or elements.

[0013] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0014] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in sequence. On the contrary, various steps can be processed in reverse order or simultaneously as needed. Meanwhile, other operations can be added to these processes, or one or more steps of operations can be removed from these processes.

[0015] Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part but not all of the embodiments of the present application, and the present application can be implemented in many different ways. Therefore, the attached drawings should not be used to limit and define the present application and the present application should cover all changes falling within the scope of the present application.

[0016] In the technical solution of the present application, a real-time monitoring method for slope stability in geotechnical engineering is proposed. Figure 1 A flowchart of the real-time monitoring method for slope stability in geotechnical engineering according to the embodiments of the present application. Figure 2 A system architecture diagram of the real-time monitoring method for slope stability in geotechnical engineering according to the embodiments of the present application. As Figure 1 and Figure 2As shown, the geotechnical engineering slope stability real-time monitoring method according to the embodiments of the present application comprises the following steps: S1, obtaining original displacement time series and on-site environmental temperature time series of a plurality of GNSS measuring points; S2, performing centering processing and matrix construction on the original displacement time series of the plurality of GNSS measuring points to obtain an observation signal matrix; S3, performing source signal separation based on independent component analysis on the observation signal matrix to obtain an estimated independent source signal matrix and an estimated mixing matrix; S4, performing source signal automatic labeling on the estimated independent source signal matrix and the on-site environmental temperature time series to obtain a source signal with a physical label; S5, based on the source signal with the physical label, performing high-fidelity deformation signal reconstruction on the estimated mixing matrix and the observation signal matrix to obtain a high-fidelity deformation signal matrix; and S6, performing trend analysis on the high-fidelity deformation signal matrix to obtain a deformation rate and an acceleration rate.

[0017] In particular, the S1, obtaining original displacement time series and on-site environmental temperature time series of a plurality of GNSS measuring points. It should be understood that obtaining original displacement time series of a plurality of GNSS measuring points is to directly capture the actual deformation of the slope surface, which is the core indicator for evaluating the stability of the slope. However, the real deformation signal of the slope is often submerged in various noises, such as the periodic "day and night" fluctuation pseudo-displacement caused by the thermal expansion and contraction of the monitoring pier, the observation instrument and even the surface rock-soil body of the slope, and the instrument inherent random noise and long-term zero drift. The amplitude of these noises is sometimes much larger than the weak real creep deformation in the early stage of slope instability, making it difficult for analysts to accurately identify the true deformation trend. Therefore, in order to accurately separate the pure deformation signal from these mixed signals and effectively eliminate the temperature effect and other interference, it is indispensable to synchronously obtain high-precision GNSS displacement data and on-site environmental temperature data. The acquisition of the temperature time series is crucial for the subsequent source signal automatic labeling link. It serves as an external physical quantity, which can help to distinguish and compensate for the thermal conduction lag effect, ensure accurate identification of the temperature effect component, and thus ensure the fidelity of the final deformation signal.

[0018] Among them, the GNSS measuring point refers to a global navigation satellite system receiving point installed on a geotechnical engineering slope for monitoring the displacement change thereof. They accurately determine their own positions by receiving satellite signals and record the change of positions over time.

[0019] In specific implementation, on the geotechnical engineering slope to be monitored, multiple GNSS (Global Navigation Satellite System) receivers are strategically deployed as monitoring points. These GNSS points continuously and in real-time collect their position information in three-dimensional space, and through high-precision calculation, obtain the displacement data of each point over time, forming their respective original displacement time series; at the same time, environmental temperature sensors are synchronously deployed on the slope site or near the monitoring points to record the environmental temperature time series in real-time and continuously. It is worth mentioning that this process is a continuous process, ensuring that the GNSS displacement data and the environmental temperature data are collected with high precision and synchronization on the time axis, so that subsequent analysis can accurately correlate displacement changes with temperature fluctuations. This synchronous collection mode is the key to realizing subsequent automatic source signal calibration, which allows the system to accurately identify and eliminate displacement interference caused by temperature effects based on the optimal correlation between temperature signals and independent source signals in the data processing stage.

[0020] In particular, the S2, the original displacement time series of multiple GNSS measuring points are centralized and matrix constructed to obtain the observation signal matrix. In the technical solution of the present application, in order to enable the subsequent independent component analysis to be effectively carried out and the source signal with physical meaning to be extracted, the original GNSS displacement time series needs to be preprocessed. Among them, the centralization processing is a common data preprocessing technology, and the purpose is to eliminate the mean (or direct current component) in the time series. By subtracting the average value of the sequence from each data point, the average value of the processed time series is zero. This helps to highlight the volatility characteristics of the data and meets the requirement of some signal processing algorithms (such as independent component analysis) that the data mean is zero, thereby improving the accuracy and efficiency of the analysis; the matrix construction refers to organizing multiple independent, preprocessed (such as centralized) time series data into a two-dimensional or multi-dimensional mathematical matrix form according to a specific arrangement rule. In this step, usually the time series of each GNSS measuring point is taken as a row or a column of the matrix, and the observation values of all measuring points at all time points are collected in a matrix to form a unified data structure, which is convenient for subsequent batch calculation and multivariate analysis. Specifically, the centralization processing is a common precondition of the independent component analysis algorithm, which can eliminate the mean component of the data and ensure that the signal fluctuates around zero mean, which helps to simplify the model construction and enables the algorithm to focus more on the second-order and high-order statistical characteristics of the data, thereby better separating the independent source signal. The matrix construction is to integrate the dispersed time series data from different GNSS measuring points into a unified observation signal matrix, which provides a standardized input format for multivariate signal processing (such as ICA), so that the displacement changes of multiple measuring points at different time points can be analyzed simultaneously, and the overall deformation mode of the slope can be captured. Through these preprocessing, the direct current bias or fixed offset in the data can be effectively removed, the accuracy and robustness of subsequent analysis can be improved, and the foundation for accurately identifying the driving factors of slope deformation is laid.

[0021] In particular, the S3 performs source signal separation based on independent component analysis on the observation signal matrix to obtain an estimated independent source signal matrix and an estimated mixing matrix. The original displacement time series observed by the GNSS measurement points is often the result of the joint action of multiple physical processes. These physical processes include long-term creep of the slope itself, thermal expansion and contraction caused by environmental temperature changes, soil softening caused by rainfall infiltration, geological structure movement, and potential local instability deformation. These independent physical source signals are linearly superimposed in an unknown way when they reach each GNSS measurement point, forming mixed observation signals. Traditional signal processing methods cannot effectively distinguish and separate these mixed source signals, making it difficult to accurately identify the key deformation components that truly reflect the stability state of the slope. Independent component analysis (ICA) provides a powerful blind source separation capability, which assumes that the observed signals are linearly mixed by an unknown mixing matrix from statistically independent non-Gaussian source signals. Through ICA, the independent source signals and their mixing methods can be deduced in reverse, thereby disentangling the complex coupling effects in the slope deformation data and revealing the single and pure physical driving factors behind them. This separation process is crucial for subsequent identification of different deformation components (such as temperature effects and long-term creep) and accurate reconstruction of deformation signals, which helps improve the monitoring system's ability to capture weak pre-stability signals.

[0022] In implementation, first, the observation signal matrix is centralized and whitened to obtain a whitened signal matrix, a whitening matrix, and a de-whitening matrix. This step is a standard preprocessing step of the ICA algorithm, aiming to convert the observation signals into data with unit variance and independence, thereby simplifying the subsequent iterative optimization process and improving the convergence speed and accuracy of the algorithm. Here, whitening is a linear transformation that aims to eliminate the second-order statistical correlation of the observation signals, i.e., to make the whitened signal components mutually independent and each component with a variance of 1. This can be achieved by eigenvalue decomposition or singular value decomposition of the observation signal matrix. Specifically, first, the covariance matrix of the observation signal matrix is calculated; then, the covariance matrix is subjected to eigenvalue decomposition: this process is represented by the formula: , wherein, is the eigenvector matrix, is a diagonal matrix composed of eigenvalues; The whitening matrix can be constructed by the following formula: , wherein, is the whitening matrix; Then, the observation signal matrix is multiplied by the whitening matrix to obtain the whitened signal matrix, which is represented by the formula: , wherein, is the whitened signal matrix, is the observed signal matrix; each row (or column, depending on the organization of the matrix) of the whitened signal matrix is mutually uncorrelated and has unit variance. Meanwhile, in order to reconstruct later, the de-whitening matrix is also needed, which is usually the inverse or pseudo-inverse of the whitening matrix: , wherein, is the de-whitening matrix; Then, the whitened signal matrix is iteratively optimized to obtain the orthogonal rotation matrix. It should be understood that, after the whitening process, although the signal components are no longer correlated, they can still not be statistically independent. The core task of ICA is to find an orthogonal rotation matrix in the whitened space, so that after the orthogonal rotation matrix acts on the whitened signal, the non-Gaussianity (i.e. statistical independence) of the signal can be maximized. This process is usually implemented through an iterative optimization algorithm, such as FastICA, Infomax, etc. Specifically, the ICA algorithm will update the orthogonal rotation matrix through iteration to maximize a non-Gaussianity measure (such as negentropy or kurtosis), until convergence. Assuming that the update rule in the iteration process is: , wherein, is a specific ICA optimization algorithm, which will adjust the elements of the orthogonal rotation matrix in each iteration and perform orthogonalization processing. The iteration will continue until the orthogonal rotation matrix reaches the convergence condition, i.e. the change of the orthogonal rotation matrix between consecutive iterations is less than a pre-set threshold, or the maximum number of iterations is reached. The final converged matrix is the orthogonal rotation matrix; Further, based on the orthogonal rotation matrix, the whitening matrix and the de-whitening matrix, the observed signal matrix is reconstructed to obtain the estimated independent source signal matrix and the estimated mixing matrix. The independent source signal matrix is obtained by multiplying the whitened signal matrix by the final obtained orthogonal rotation matrix, which is represented by the formula as: , wherein, is the independent source signal matrix, and the matrix each row (or column) of the matrix represents an estimated independent source signal time series, and these source signals are statistically independent of each other and have specific physical meanings (such as temperature effect, long-term creep, etc.); The estimated mixing matrix describes how the independent source signals are superimposed to form the observed signal. It can be obtained by the inverse of the de-whitening matrix and the orthogonal rotation matrix, which is represented by the formula as: , wherein, is the transpose of the orthogonal rotation matrix, is the estimated mixing matrix, each column of which represents the contribution weight of an independent source signal to the individual observation, i.e. describes how source signals mix to form the observation.

[0023] In particular, the S4, the estimated independent source signal matrix and the field environment temperature time series are subjected to source signal automatic labeling to obtain source signals with physical labels. Although independent component analysis can decompose complex mixed observation signals into statistically independent source signals, these separated source signals do not have direct physical meaning labels. For example, a certain source signal can be caused by environmental temperature changes, another can be caused by long-term creep of the slope soil body, and others can represent real structural instability precursors, while others can be just system noise. Without physical label identification and attribution of these source signals, subsequent noise removal, deformation reconstruction and stability evaluation cannot be targeted.

[0024] In specific implementation, first, based on the field environment temperature time series, the estimated independent source signal matrix is subjected to external physical quantity strong correlation component labeling to obtain the index of the temperature effect source signal and the index set of the remaining source signals that have not been labeled. It should be understood that in the field of geotechnical engineering monitoring, the conventional method of directly comparing the separated source signals with the external temperature signal using the Pearson correlation coefficient to label the temperature effect component has a fundamental physical limitation. The underlying logic assumes that the response of the slope structure to environmental temperature changes is instantaneous and synchronous. However, in actual physical scenarios, the conduction of heat from the slope surface to the internal monitoring pier (or sensor) is a process that takes time, which is affected by the heat capacity and thermal conductivity of the slope body material, and inevitably leads to a time delay in the thermal deformation response of the structure relative to the environmental air temperature, i.e. a thermal conduction lag effect. Therefore, the source signal waveform representing the temperature effect and the measured temperature signal waveform will exhibit a phase difference on the time axis. The conventional instantaneous correlation calculation method fails to consider this time lag, and when the phase difference exists, it will significantly underestimate the true correlation strength between the two, and even possibly draw the wrong conclusion that they are not correlated, leading to misjudgment or failure to identify the temperature effect component. To overcome the above defects, a time-lag optimal correlation labeling method based on cross-correlation spectral analysis is proposed. This method no longer relies on instantaneous linear correlation, but rather delves into the frequency domain structure of the signals, accurately captures and compensates for the time delay by quantifying the phase relationship between the signals, thereby achieving robust labeling of physical correlation.

[0025] In this process, first, the time series of the field environment temperature and the estimated independent source signal matrix are subjected to time domain-frequency domain signal transformation to obtain the Fourier spectrum set of the independent source signal and the Fourier spectrum of the temperature signal. It should be understood that direct comparison in the time domain cannot effectively deal with the delay problem between signals. Converting the signals to the frequency domain can decompose them into a series of sine waves of different frequencies superimposed, and the time delay between signals is intuitively manifested as a phase difference in different frequency components in the frequency domain. Therefore, entering the frequency domain is a prerequisite for analyzing and quantifying this time lag relationship.

[0026] Specifically, the time series of each independent source signal separated by the ICA algorithm and the temperature time series collected synchronously are subjected to Fourier transform, mapping these signals from the time domain to the frequency domain to obtain their respective Fourier spectra. This process can be represented by the formula: , Wherein, represents time; represents frequency; is the time series function of the th independent source signal; is a natural constant; is an imaginary unit; is the representation of the source signal in the frequency domain, i.e., the Fourier spectrum, and the same operation is performed on the temperature signal to obtain the Fourier spectrum of the temperature signal. In this way, the analysis perspective is shifted from the comparison of time domain waveforms to the comparison of frequency domain spectrum structures, laying the foundation for subsequent calculation of time lag through phase information, which is the starting link of the entire time lag compensation correlation algorithm. It should be understood that the original time series data is transformed into a frequency domain complex spectrum containing amplitude and phase information, which makes the periodic characteristics and potential phase relationship within the signal visible, facilitating subsequent deeper mining of the relationship between signals; Next, the cross power spectral density between the Fourier spectrum of the temperature signal and the Fourier spectrum of each independent source signal in the Fourier spectrum set of the independent source signal is calculated to obtain a cross power spectral density matrix. It should be understood that after obtaining the Fourier spectrum of each signal, a mathematical tool is needed that can measure the energy correlation strength and phase difference of two signals at each frequency. Cross power spectral density is designed for this purpose, which fuses the common information and relative phase information in the two signal spectra through complex multiplication.

[0027] Specifically, the Fourier spectrum of each source signal is multiplied by the complex conjugate of the Fourier spectrum of the temperature signal , thereby obtaining the cross power spectral density between the two; this process can be represented by the formula: , where, is the cross power spectral density of the jth source signal and the temperature signal; is the Fourier spectrum of the jth source signal; is the complex conjugate of the Fourier spectrum of the temperature signal. In this way, an intermediate spectral function is generated, which completely encodes all the correlation information of the two original time series under the linear system assumption, and is the core bridge connecting the frequency domain phase difference and the time domain time delay; Through the execution of this step, a new complex spectral matrix is generated. The magnitude of the modulus of at a certain frequency reflects the correlation strength of the source signal and the temperature signal at that frequency component; while its phase angle at that frequency directly corresponds to the phase difference of the two at that frequency component; It should be understood that, considering that the cross power spectrum itself is complete in information, but is still in the frequency domain and is not intuitive enough. In order to obtain a correlation measure that can be directly used for decision-making and has clear physical meaning, the phase information in the frequency domain needs to be converted back to the time domain, represented as a specific time delay, and the "optimal" delay that maximizes the correlation needs to be found.

[0028] Specifically, first, the calculated cross power spectral density is subjected to inverse Fourier transform and normalized, thereby obtaining the cross-correlation function with time delay as the variable; this process is represented by the formula: , Here, represents the correlation coefficient of the jth source signal and the temperature signal when the time lag is ; represents the inverse Fourier transform operator; and are the standard deviations of the corresponding signals, and N is the number of points of the signal; Then, based on the cross power spectral density matrix, the time-lag optimal correlation of the on-site environmental temperature time series and the estimated independent source signal matrix is quantified and calibrated to obtain the index of the temperature effect source signal and the index set of the remaining source signals that have not been calibrated. That is, first, the peak value of this cross-correlation function is searched, which is the time-lag optimal correlation coefficient , and the corresponding abscissa is the optimal time lag ; this process is represented by the formula: , Finally, all source signals are traversed to find the one with the largest time-lag optimal correlation coefficient, and its index determined as the temperature effect component and compared with a preset strong correlation threshold to confirm the effectiveness of the calibration; the process is expressed by the formula: , In the above formula, is the maximum correlation coefficient of the jth source signal and the temperature signal under the optimal time lag; is a time delay variable; is the optimal time lag for achieving maximum correlation; indicates finding the maximum value at all possible time lags . is finally determined as the index of the temperature effect source signal.

[0029] In this way, the problem of low correlation caused by time delay is solved. The purpose is not limited to zero delay, but to systematically find and determine a delay time point that can maximize the physical correlation between the two in a wide delay range, and use the maximum correlation value as the final basis for calibration.

[0030] In summary, the preferred mechanism can not only accurately identify the temperature effect component disguised by time delay, output its index and the index set U_1 of the remaining uncalibrated signals, but also provides an important parameter—optimal time lag with important physical significance, which can be used to verify and analyze the thermal response characteristics of the slope monitoring system.

[0031] The technical effect achieved by the improved mechanism is to significantly improve the accuracy and robustness of temperature effect source signal identification in complex physical scenarios with thermal conduction lag effect. By actively searching and compensating for time delay, this method can reveal and quantify the real physical correlation ignored by traditional methods, avoiding false calibration due to mismatched signal phases. At the same time, by more accurately separating the temperature noise component, it ensures that the subsequent signal reconstruction process can output pure deformation signals with higher fidelity. This directly relates to the core performance of the real-time monitoring system for rock and soil engineering slope stability, can effectively reduce the false positive rate and false negative rate of the early warning model, improve the ability to capture weak pre-stability deformation signals, and thus provide a solid data foundation for earlier and more reliable disaster warning.

[0032] Further, internal time sequence feature analysis and trend component calibration are performed on the index set of the remaining uncalibrated source signals and the estimated independent source signal matrix to obtain the index of the long-term creep source signal and the final remaining source signal index set.

[0033] In particular, the S5 reconstructs the estimated mixing matrix and the observation signal matrix based on the source signals with physical labels to obtain a high-fidelity deformation signal matrix. It should be understood that, although independent component analysis (ICA) can decompose the mixed observation signal into statistically independent source signals, and the subsequent source signal automatic labeling step has given physical labels to these source signals, distinguishing the true deformation signal from the noise (such as displacement caused by temperature effect). However, the observation signal matrix is still the result of the linear mixing of all source signals (including the identified noise and the true deformation component). In order to accurately evaluate the stability of the slope, it is necessary to remove these identified noise components from the total observation signal to obtain pure deformation signals that are not affected by these interference factors. If this step is not performed, the noise components will continue to exist in the data, which may mask the weak precursory deformation signal of instability, leading to misjudgment or omission, seriously affecting the reliability of the monitoring system and the timeliness of the early warning. Therefore, based on the source signals with physical labels, the fidelity of the deformation signal can be effectively improved, so that it more accurately reflects the physical deformation process of the slope itself.

[0034] In specific implementation, first, the source signals with physical labels and the estimated mixing matrix are selectively re-projected with noise components to obtain a projection noise matrix. That is, the contributions of these noises in the original observation signal are accurately calculated using these identified noise source signals and their corresponding mixing methods. In this process, first, all source signals marked as noise are extracted from the independent source signal matrix to form a noise source signal sub-matrix; at the same time, column vectors corresponding to these noise source signals are extracted from the estimated mixing matrix to form a noise mixing matrix sub-matrix; then, by multiplying the noise mixing matrix sub-matrix with the noise source signal sub-matrix, a projection noise matrix can be obtained, which represents the displacement caused by all source signals identified as noise at the observation points; this process is represented by the formula: , wherein, is the column of the estimated mixing matrix A corresponding to the noise source signal, is the source signal row in the source signal with physical labels identified as noise, is the projection noise matrix, which accurately quantifies the displacement caused by the noise factor at each GNSS observation point, providing a basis for subsequent removal of noise from the total observation; Further, the projection noise matrix and the observation signal matrix are differentially reconstructed and the pure deformation signal is extracted to obtain a high-fidelity deformation signal matrix. That is, after the projection noise matrix is calculated, the influence of noise is subtracted from the original observation signal matrix using this information, thereby obtaining a pure and high-fidelity deformation signal. In this process, first, the projection noise matrix is subtracted from the original observation signal matrix point by point; the original observation signal matrix contains all mixed displacement components, including the true slope deformation and various noises, and by subtracting the quantized noise contribution, the remaining part is the pure deformation signal without noise interference, that is, the high-fidelity deformation signal matrix; this process is expressed by the formula: , wherein, is a high-fidelity deformation signal matrix, which contains the actual displacement of the slope at each measuring point, which is not disturbed by external environmental noise (such as temperature effect). It can more accurately reflect the internal stability state of the slope, and provide a pure data basis for subsequent deformation rate and acceleration rate analysis, thereby significantly improving the accuracy and reliability of slope instability warning.

[0035] In particular, S6, the high-fidelity deformation signal matrix is analyzed to obtain the deformation rate and the acceleration rate. It should be understood that only obtaining the displacement data of the slope is not enough to fully evaluate its stability state. The displacement of the slope itself may be within an acceptable range, but if the displacement rate is too fast or there is an acceleration trend, it may indicate a potential instability risk. The deformation rate (i.e., the speed of displacement change over time) can intuitively reflect the current activity level of the slope deformation, and the deformation acceleration rate (i.e., the speed of deformation rate change over time) is a key indicator for measuring the change trend of the slope stability, especially when the acceleration rate shows a continuous increasing trend, which is often regarded as an important precursor of the impending instability of the slope. By analyzing the trend of the high-fidelity deformation signal matrix, these dynamic parameters with physical meaning can be extracted from the pure deformation data which has been carefully processed and removed from noise interference. This enables the monitoring system to obtain dynamic rate and acceleration rate information from static displacement data, and more comprehensively describes the motion state of the slope.

[0036] In specific implementation, the high-fidelity deformation signal matrix is first-order and second-order differentiated to obtain the deformation rate and the acceleration rate. In the technical solution of the present application, the first-order differentiation aims to calculate the instantaneous change rate of the displacement time series of each measuring point in the high-fidelity deformation signal matrix, that is, the deformation rate. For discrete time series data, numerical differentiation method is usually used to approximate the calculation of the derivative. Assuming that the displacement time series of a measuring point j in the high-fidelity deformation signal matrix is wherein, represents the high-fidelity displacement value of the measuring point j at time t, and the time interval is The central difference method (or other suitable numerical differentiation method, such as forward difference or backward difference) is used to calculate the deformation rate The process is expressed by the formula: , For the start and end points of the time series, forward difference or backward difference can be used for processing: , , By performing the above first-order derivation operation on the time series of each measuring point in the high-fidelity deformation signal matrix, a deformation rate matrix similar in size to the original matrix can be obtained. The second-order derivation aims to calculate the second-order change rate of the displacement time series of each measuring point in the high-fidelity deformation signal matrix, i.e., the deformation acceleration. This can be obtained by performing first-order derivation again on the calculated deformation rate time series, or directly by performing second-order numerical differentiation on the original displacement time series. The central difference method is used to calculate the deformation acceleration The process is expressed by the formula: , Similarly, for the boundary points of the time series, adaptive difference formulas need to be used. By performing the above second-order derivation operation on the time series of each measuring point in the high-fidelity deformation signal matrix, a deformation acceleration matrix can finally be obtained.

[0037] In summary, the real-time monitoring method for geotechnical engineering slope stability according to the embodiments of the present application is illustrated, which introduces a blind source separation technology combining independent component analysis and physical constraints, treats the original displacement time series of multiple GNSS measuring points and the environmental temperature time series as multi-channel mixed signals, uses the statistical independence between signal sources to effectively separate periodic environmental noise, instrument random noise and drift from the mixed observation signals and the real slope deformation signal; then, through correlation verification with physical quantities such as on-site environmental temperature, the separated source signals are automatically calibrated, and the temperature effect source signal and the long-term creep source signal with physical labels are accurately identified and extracted, thereby realizing precise removal of noise components and high-fidelity reconstruction of pure deformation signals. In this way, the accuracy and reliability of real-time monitoring of geotechnical engineering slope stability can be effectively improved, providing a solid data foundation and decision basis for early warning and prevention of slope disasters.

[0038] Further, a real-time monitoring system for geotechnical engineering slope stability is provided.

[0039] Figure 3 The block diagram of the real-time monitoring system for geotechnical engineering slope stability according to the embodiments of the present application is shown. As Figure 3As shown, the geotechnical engineering slope stability real-time monitoring system 300 according to the embodiments of the present application comprises: a data acquisition module 310, configured to acquire original displacement time series and field environment temperature time series of a plurality of GNSS measuring points; a centralization processing and matrix construction module 320, configured to perform centralization processing and matrix construction on the original displacement time series of the plurality of GNSS measuring points to obtain an observation signal matrix; a source signal separation module 330, configured to perform source signal separation on the observation signal matrix based on independent component analysis to obtain an estimated independent source signal matrix and an estimated mixing matrix; a source signal automatic labeling module 340, configured to perform source signal automatic labeling on the estimated independent source signal matrix and the field environment temperature time series to obtain a source signal with a physical label; a high-fidelity deformation signal reconstruction module 350, configured to perform high-fidelity deformation signal reconstruction on the estimated mixing matrix and the observation signal matrix based on the source signal with the physical label to obtain a high-fidelity deformation signal matrix; and a trend analysis module 360, configured to perform trend analysis on the high-fidelity deformation signal matrix to obtain a deformation rate and an acceleration rate.

[0040] As described above, the geotechnical engineering slope stability real-time monitoring system 300 according to the embodiments of the present application can be implemented in various wireless terminals, such as a server with a geotechnical engineering slope stability real-time monitoring algorithm, etc. In one possible implementation, the geotechnical engineering slope stability real-time monitoring system 300 according to the embodiments of the present application can be integrated into a wireless terminal as a software module and / or a hardware module. For example, the geotechnical engineering slope stability real-time monitoring system 300 can be a software module in the operating system of the wireless terminal, or can be an application program developed for the wireless terminal; of course, the geotechnical engineering slope stability real-time monitoring system 300 can also be one of the many hardware modules of the wireless terminal.

[0041] Alternatively, in another example, the geotechnical engineering slope stability real-time monitoring system 300 and the wireless terminal can also be separate devices, and the geotechnical engineering slope stability real-time monitoring system 300 can be connected to the wireless terminal through a wired and / or wireless network, and transmit interactive information in an agreed data format.

[0042] The above has described the embodiments of the present disclosure, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes are obvious to those skilled in the art without departing from the scope and spirit of the described embodiments. The selection of the terms used herein is intended to best explain the principles, practical applications, or improvements to the technology in the market of the embodiments, or to enable other ordinary skilled persons in the art to understand the embodiments disclosed herein.

Claims

1. A method for real-time monitoring of slope stability in geotechnical engineering, characterized in that, include: Acquire the raw displacement time series and ambient temperature time series of multiple GNSS measuring points; The original displacement time series of multiple GNSS measuring points are centered and matrix constructed to obtain the observation signal matrix; The observed signal matrix is ​​separated into source signals based on independent component analysis to obtain the estimated independent source signal matrix and the estimated mixture matrix; The estimated independent source signal matrix and the on-site ambient temperature time series are used to automatically calibrate the source signals to obtain physically tagged source signals; Based on the source signal with physical labels, the estimated mixing matrix and the observed signal matrix are reconstructed using high-fidelity deformed signal to obtain the high-fidelity deformed signal matrix. Trend analysis was performed on the high-fidelity deformation signal matrix to obtain the deformation rate and acceleration rate.

2. The method for real-time monitoring of slope stability in geotechnical engineering according to claim 1, characterized in that, The observed signal matrix is ​​subjected to source signal separation based on independent component analysis to obtain the estimated independent source signal matrix and the estimated mixture matrix, including: The observed signal matrix is ​​centered and whitened to obtain the whitened signal matrix, the whitening matrix, and the dewhitening matrix; The whitened signal matrix is ​​iteratively optimized to obtain the orthogonal rotation matrix; Based on the orthogonal rotation matrix, whitening matrix, and dewhitening matrix, the observed signal matrix is ​​structurally reconstructed to obtain the estimated independent source signal matrix and the estimated mixture matrix.

3. The method for real-time monitoring of slope stability in geotechnical engineering according to claim 1, characterized in that, The estimated independent source signal matrix and the ambient temperature time series are used to automatically calibrate the source signals to obtain physically tagged source signals, including: Based on the time series of ambient temperature, the estimated independent source signal matrix is ​​calibrated with strongly correlated external physical quantities to obtain the index of temperature effect source signals and the set of uncalibrated remaining source signal indices. Internal time-series characteristic analysis and trend component calibration are performed on the unlabeled remaining source signal index set and the estimated independent source signal matrix to obtain the index of long-term creep source signals and the final remaining source signal index set.

4. The method for real-time monitoring of slope stability in geotechnical engineering according to claim 3, characterized in that, Based on the on-site ambient temperature time series, the estimated independent source signal matrix is ​​calibrated using strongly correlated external physical quantities to obtain the index of temperature effect source signals and the set of uncalibrated remaining source signal indices, including: Time-domain to frequency-domain signal transformation is performed on the on-site ambient temperature time series and the estimated independent source signal matrix to obtain the Fourier spectrum set of the independent source signals and the Fourier spectrum of the temperature signal. The cross-power spectral density matrix is ​​obtained by calculating the cross-power spectral density between the Fourier spectrum of the temperature signal and the Fourier spectrum of each independent source signal in the set of Fourier spectrum of the independent source signals. Based on the cross-power spectral density matrix, time-delay optimal correlation quantization and calibration are performed on the on-site ambient temperature time series and the estimated independent source signal matrix to obtain the index of temperature effect source signals and the set of uncalibrated remaining source signal indexes.

5. The method for real-time monitoring of slope stability in geotechnical engineering according to claim 4, characterized in that, Calculating the cross-power spectral density matrix by means of the Fourier spectrum of the temperature signal and the Fourier spectra of each independent source signal in the set of Fourier spectra of the independent source signals includes: calculating the cross-power spectral density matrix by means of the following formula: in, This indicates that the j-th source signal and the temperature signal have a time delay of... Cross-power spectral density at time; Represents the inverse Fourier transform operator; and ...

6. The method for real-time monitoring of slope stability in geotechnical engineering according to claim 1, characterized in that, Based on the physically labeled source signals, high-fidelity deformed signal reconstruction is performed on the estimated mixing matrix and the observed signal matrix to obtain a high-fidelity deformed signal matrix, including: The source signal with physical labels and the estimated mixing matrix are selectively reprojected with noise components to obtain the projected noise matrix; Differential reconstruction and clean deformation signal extraction are performed on the projection noise matrix and the observation signal matrix to obtain a high-fidelity deformation signal matrix.

7. The method for real-time monitoring of slope stability in geotechnical engineering according to claim 1, characterized in that, Trend analysis is performed on the high-fidelity deformation signal matrix to obtain the deformation rate and acceleration rate, including: taking the first and second derivatives of the high-fidelity deformation signal matrix to obtain the deformation rate and acceleration rate.

8. A real-time monitoring system for slope stability in geotechnical engineering, characterized in that, include: The data acquisition module is used to acquire the original displacement time series and ambient temperature time series of multiple GNSS measuring points; The centralization and matrix construction module is used to centralize and construct matrices from the original displacement time series of multiple GNSS measuring points to obtain the observation signal matrix; The source signal separation module is used to perform source signal separation on the observed signal matrix based on independent component analysis to obtain the estimated independent source signal matrix and the estimated mixture matrix; The automatic source signal calibration module is used to automatically calibrate the estimated independent source signal matrix and the on-site ambient temperature time series to obtain source signals with physical labels. The high-fidelity deformable signal reconstruction module is used to reconstruct the estimated mixing matrix and the observed signal matrix in high fidelity based on the source signal with physical labels to obtain the high-fidelity deformable signal matrix. The trend analysis module is used to perform trend analysis on the high-fidelity deformation signal matrix to obtain the deformation rate and acceleration rate.

Images