Tunnel advanced geological forecast monitoring method and related equipment

Abstract

The invention discloses a tunnel advanced geological forecast monitoring method and related equipment, and relates to the technical field of intelligent monitoring, and the method comprises the steps: obtaining the time domain response of a transient electromagnetic signal; processing the time domain response of the transient electromagnetic signal by using an adaptive filtering algorithm to obtain processing data; generating a three-dimensional geologic model according to the processing data by using a three-dimensional imaging algorithm; and predicting the geological condition based on an abnormal density function, a pattern recognition algorithm and the three-dimensional geological model.

Description

Technical Field

[0001] The present application relates to the field of intelligent monitoring technology, and more specifically, to a tunnel advanced geological prediction monitoring method and related equipment. Background Art

[0002] With the continuous expansion of tunnel construction and the increase in tunnel projects under complex geological conditions, how to effectively predict the geology ahead of the tunnel, reduce construction risks, and ensure project progress and safety has become a major challenge in current tunnel projects. Existing tunnel advance geological prediction technologies mainly rely on manual observation, drilling detection, and geological radar. However, these traditional methods often face problems such as insufficient accuracy, limited detection range, and low data processing efficiency under complex geological conditions, making it difficult to meet the needs of modern tunnel construction.

[0003] In order to solve at least some of the above problems, it is urgently necessary to provide a tunnel advanced geological forecasting monitoring method that can further improve the accuracy and efficiency of geological forecasting and help tunnel construction companies better cope with complex geological environments. Summary of the invention

[0004] A series of simplified concepts are introduced in the Summary of the Invention section, which will be further described in detail in the Detailed Description of the Invention section. The Summary of the Invention section of this application does not mean to attempt to define the key features and essential technical features of the claimed technical solution, nor does it mean to attempt to determine the scope of protection of the claimed technical solution.

[0005] In a first aspect, the present application provides a tunnel advanced geological prediction monitoring method, comprising:

[0006] Obtaining the time domain response of transient electromagnetic signals;

[0007] Using an adaptive filtering algorithm to process the time domain response of the transient electromagnetic signal to obtain processed data;

[0008] Generate a three-dimensional geological model based on the processed data using a three-dimensional imaging algorithm;

[0009] The geological conditions are predicted based on the anomaly density function, pattern recognition algorithm and the above three-dimensional geological model.

[0010] In a feasible implementation manner, the time domain response of the above transient electromagnetic signal is expressed as:

[0011]

[0012]

[0013] In the formula, E(t, θ2, φ) is the time domain response of the transient electromagnetic signal, A(f, θ2, φ) is the frequency domain response of the transient electromagnetic signal, t is the current moment, f is the frequency, θ2 is the angle after reflection or refraction, φ is the depression angle, j is the imaginary unit, 2πft is the phase factor, d is the propagation depth of the electromagnetic wave, θ1 is the incident angle of the electromagnetic wave, V1 is the propagation speed of the electromagnetic wave in the first medium, and V2 is the propagation speed of the electromagnetic wave in the second medium.

[0014] In a feasible implementation, it also includes:

[0015] The frequency domain response of the above transient electromagnetic signal is determined based on the following formula:

[0016]

[0017] Among them, A0(f) is the amplitude response at frequency f, α(f, θ2, φ) is the attenuation coefficient, which represents the energy loss caused by medium absorption and attenuation during the propagation of electromagnetic waves. The attenuation coefficient changes with frequency and antenna angle. β(f, θ2, φ) is the phase shift, which represents the phase change caused by different geological structures during the propagation of electromagnetic waves.

[0018] In a feasible implementation manner, the above-mentioned use of the adaptive filtering algorithm to process the time domain response of the above-mentioned transient electromagnetic signal to obtain the processed data includes:

[0019] The above adaptive filter is expressed as:

[0020]

[0021] In the formula, is the processed data after filtering, w n is the weight coefficient of the adaptive filter, N is the order of the filter, tn is the time n sampling cycles before the current time t, and E(tn, θ2, φ) is the time domain response corresponding to the transient electromagnetic signal at time tn.

[0022] In a feasible implementation, the weight coefficient w n Determined based on the following formula:

[0023]

[0024] In the formula, and are the weight values ​​of the filter at the kth and k+1th iterations, μ is the learning rate, and Loss is the loss function;

[0025] Loss = L1 + L2 + L3 + L4

[0026] Where D(t, θ2, φ) is the desired output signal, SNR(t) -1 is the inverse value of the signal-to-noise ratio, λ1 is the weight coefficient of the signal-to-noise ratio suppression term, λ2 is the weight coefficient of the sparsity penalty term, λ3 is the weight coefficient of the time domain smoothness penalty term, and T is the length of the time series.

[0027] In a feasible implementation, the above three-dimensional geological model is further determined based on the following formula:

[0028]

[0029] Where M(x, y, z) is the three-dimensional geological model, which is the geological structure at the midpoint (x, y, z) in space, L is the total number of geological layers, l is the lth geological layer, and w is the total number of geological layers. l is the weight coefficient of the lth layer, B is the total number of frequency bands, f i is the center frequency of the ith frequency band, R(t, x, y, z, f i , l) is the time inversion operator, which converts the time domain signal into the spatial geological model information, δ(·) is the impulse response of instantaneous propagation, v l is the electromagnetic wave in the geological medium layer l, frequency f i The propagation speed below.

[0030] In a feasible implementation manner, the above-mentioned prediction of geological conditions based on the anomaly density function, the pattern recognition algorithm and the above-mentioned three-dimensional geological model includes:

[0031] The abnormal area is determined based on the abnormal density function and the three-dimensional geological model, wherein the abnormal density function is:

[0032]

[0033] In the formula, ρ a (x, y, z) is the abnormal density value of the point (x, y, z) in the three-dimensional geological model, σ(·) is the activation function, is the Laplace operator, μ M is the mean density of the three-dimensional geological model;

[0034] The abnormal area is identified based on the pattern recognition algorithm to predict the geological conditions, wherein the pattern recognition algorithm includes a feature extraction operation, an abnormal point classification operation and an abnormal type optimization operation.

[0035] In the second aspect, the present application proposes a tunnel advanced geological prediction monitoring device, comprising:

[0036] An acquisition unit, used for acquiring a time domain response of a transient electromagnetic signal;

[0037] A processing unit, used to process the time domain response of the transient electromagnetic signal using an adaptive filtering algorithm to obtain processed data;

[0038] A generating unit, used for generating a three-dimensional geological model according to the above-mentioned processed data by using a three-dimensional imaging algorithm;

[0039] The identification unit is used to predict the geological conditions based on the anomaly density function, the pattern recognition algorithm and the above three-dimensional geological model.

[0040] In the third aspect, an electronic device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is used to implement the steps of the tunnel advanced geological prediction monitoring method as described in any one of the first aspects above when executing the computer program stored in the memory.

[0041] In a fourth aspect, the present application further proposes a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the tunnel advanced geological prediction monitoring method of any one of the first aspects is implemented.

[0042] In summary, the embodiment of the present application can accurately predict geological conditions by obtaining the time domain response of transient electromagnetic signals, using an adaptive filtering algorithm to process transient electromagnetic signals, and generating a three-dimensional geological model. Compared with traditional manual observation or drilling detection methods, this solution can process data under complex geological conditions, significantly improve the prediction accuracy, and reduce the errors caused by geological complexity. Through the adaptive filtering algorithm, the time domain response of transient electromagnetic signals is quickly processed, and a three-dimensional geological model is generated in combination with a three-dimensional imaging algorithm, which effectively accelerates the data processing process. Traditional geological prediction methods often lead to low processing efficiency due to the large amount of data, and this solution can achieve efficient data processing while ensuring data accuracy. This solution uses anomaly density functions and pattern recognition algorithms to effectively identify abnormal geological areas that may exist in front of the tunnel. With the assistance of a three-dimensional geological model, the complex geological environment in front of the tunnel can be accurately predicted, ensuring that the geological conditions are fully controlled during the construction process and reducing construction risks. In complex tunnel construction, geological uncertainty is an important factor leading to construction risks. This solution can effectively avoid potential geological risks, ensure safety during the construction process, and improve construction progress through early prediction of geological conditions. This solution is based on intelligent monitoring technology, combined with advanced algorithms such as adaptive filtering, three-dimensional imaging and pattern recognition, breaking through the technical bottleneck of traditional geological forecasting and providing a more intelligent and accurate forecasting method for tunnel projects. This innovative technical method can effectively intervene in the early stages of tunnel construction to ensure the smooth progress of the project. By improving the accuracy and efficiency of geological forecasting, construction companies can better cope with complex geological conditions in the early stages, reduce additional costs such as shutdowns and rework caused by geological problems, and thus reduce overall construction costs.

[0043] The tunnel advanced geological prediction monitoring method proposed by the present invention, other advantages, objectives and features of the present invention will be reflected in part through the following description, and in part will be understood by technical personnel in the field through research and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present specification. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0045] Figure 1 A schematic diagram of a tunnel advanced geological prediction monitoring method provided in an embodiment of the present application;

[0046] Figure 2 A schematic diagram of the structure of a tunnel advanced geological prediction monitoring device provided in an embodiment of the present application;

[0047] Figure 3 A schematic diagram of the structure of an electronic device for advanced geological prediction and monitoring of tunnels provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0049] See also Figure 1 , which is a flow chart of a tunnel advanced geological prediction monitoring method provided in an embodiment of the present application, which may specifically include:

[0050] S110, obtaining a time domain response of a transient electromagnetic signal;

[0051] For example, in geological exploration in front of the tunnel, transient electromagnetic method is a non-invasive and non-destructive detection technology. By measuring transient electromagnetic responses at different angles, different levels of information on geological structures can be obtained. The multi-polarized antenna array dynamically adjusts the elevation and depression angles so that electromagnetic waves can be incident on the ground in a multi-angle and multi-polarization manner, thereby improving data coverage and enhancing the resolution of geological structures, greatly increasing the detection accuracy of geological structures. A multi-polarized antenna refers to an antenna system that can operate in multiple polarization states, and the polarization state generally refers to the direction of the electric field of an electromagnetic wave. For example, the vertical polarization and horizontal polarization of electromagnetic waves can be used to detect structures in different directions underground, respectively. The elevation and depression angles of the antenna determine the propagation direction of the electromagnetic wave. Adjusting these two parameters can detect geology at different depths.

[0052] The propagation speed and mode of electromagnetic waves in different media are closely related to the geological structure. The transient electromagnetic method infers the underground structure by measuring the reflected signals at different depths underground. Under different polarization states, the response of electromagnetic waves will vary with the conductivity, dielectric constant and other properties of the underground medium. Therefore, through a multi-polarized antenna array, multi-dimensional electromagnetic signals can be captured simultaneously, thereby constructing a more accurate geological model.

[0053] S120, using an adaptive filtering algorithm to process the time domain response of the transient electromagnetic signal to obtain processed data;

[0054] Exemplarily, the adaptive filtering algorithm is used to dynamically adjust the parameters of the filter to optimize the quality of the electromagnetic signal. The algorithm can automatically adapt to different geological conditions and suppress background noise and signal interference by adjusting the filtering parameters, thereby ensuring the accuracy of the measurement data in complex tunnel geological environments.

[0055] The adaptive filter learns through a feedback mechanism, which continuously adjusts the weight coefficient of the filter according to the characteristics of the input signal, so that the filter can better adapt to the real-time input data. This adaptability can effectively handle various interferences in complex background environments, such as noise from steel frames and cables, and interference from external electromagnetic signals.

[0056] S130, generating a three-dimensional geological model based on the processed data using a three-dimensional imaging algorithm;

[0057] For example, in tunnel advance geological prediction, it is crucial to obtain the distribution of underground media through transient electromagnetic data. The three-dimensional imaging algorithm converts the measured transient electromagnetic signals into a three-dimensional geological structure model of the underground by inverting electromagnetic data, and uses the reverse time migration algorithm to invert the time domain signal to the spatial domain, thereby constructing a three-dimensional model of the underground geology. The basic idea of ​​the reverse time migration algorithm is to "reverse" the time information of the transient electromagnetic signal back to the spatial information, and use the propagation and reflection characteristics of electromagnetic waves to map the reflected signal back to the actual position underground through the inversion process. In this way, a three-dimensional image of the geological structure can be generated.

[0058] S140. Predict geological conditions based on anomaly density function, pattern recognition algorithm and three-dimensional geological model.

[0059] For example, after generating a 3D geological model, detecting potential geological anomalies (such as cavities, faults, fracture zones, etc.) is a key step in tunnel advance geological prediction. These anomalies usually manifest as discontinuities, density changes, or significant deviations of other features in the model.

[0060] Pattern recognition algorithms are used to automatically detect and classify these abnormal areas. By analyzing the local characteristics of the model and extracting features related to geological anomalies (e.g., density, gradient, etc.), pattern recognition algorithms can classify and evaluate these abnormal areas to help predict potential risks ahead of the tunnel.

[0061] The pattern recognition algorithm uses a series of signal feature extraction, data analysis and classification techniques to identify abnormal areas in the geological model. The process is as follows:

[0062] Feature extraction: Extract features from the 3D model to describe the abnormal area. Common features include density change, structural gradient, spatial position, etc. Classifier training: Train the classifier through existing geological data sets (such as historical geological exploration data). The classifier is used to classify points or areas in the 3D model into different types of geological anomalies, such as cavities, faults, etc. Abnormal area identification: Scan the 3D model with the trained classifier to identify abnormal points or abnormal areas.

[0063] The identified geological anomalies are combined into complete anomaly areas, which may represent discontinuous structures such as cavities and faults in the geology. Pattern recognition algorithms can not only identify a single type of anomaly, but also identify complex anomaly structures through the joint analysis of multiple features. Once the anomaly area is identified, further geological analysis can predict the geological conditions ahead of the tunnel through simulation and evaluation.

[0064] In summary, the embodiment of the present application can accurately predict geological conditions by obtaining the time domain response of transient electromagnetic signals, using an adaptive filtering algorithm to process transient electromagnetic signals, and generating a three-dimensional geological model. Compared with traditional manual observation or drilling detection methods, this solution can process data under complex geological conditions, significantly improve the prediction accuracy, and reduce the errors caused by geological complexity. Through the adaptive filtering algorithm, the time domain response of transient electromagnetic signals is quickly processed, and a three-dimensional geological model is generated in combination with a three-dimensional imaging algorithm, which effectively accelerates the data processing process. Traditional geological prediction methods often lead to low processing efficiency due to the large amount of data, and this solution can achieve efficient data processing while ensuring data accuracy. This solution uses anomaly density functions and pattern recognition algorithms to effectively identify abnormal geological areas that may exist in front of the tunnel. With the assistance of a three-dimensional geological model, the complex geological environment in front of the tunnel can be accurately predicted, ensuring that the geological conditions are fully controlled during the construction process and reducing construction risks. In complex tunnel construction, geological uncertainty is an important factor leading to construction risks. This solution can effectively avoid potential geological risks, ensure safety during the construction process, and improve construction progress through early prediction of geological conditions. This solution is based on intelligent monitoring technology, combined with advanced algorithms such as adaptive filtering, three-dimensional imaging and pattern recognition, breaking through the technical bottleneck of traditional geological forecasting and providing a more intelligent and accurate forecasting method for tunnel projects.

[0065] In some examples, the time domain response of the above transient electromagnetic signal is expressed as:

[0066]

[0067] d=v2tcos(θ2)cos(φ)

[0068] In the formula, E(t, θ2, φ) is the time domain response of the transient electromagnetic signal, A(f, θ2, φ) is the frequency domain response of the transient electromagnetic signal, t is the current moment, f is the frequency, θ2 is the angle after reflection or refraction, φ is the depression angle, j is the imaginary unit, 2πft is the phase factor, which describes the phase change of the signal at time t, d is the propagation depth of the electromagnetic wave, θ1 is the incident angle of the electromagnetic wave, that is, the angle at which the electromagnetic wave enters the medium interface, V1 is the propagation speed of the electromagnetic wave in the first medium, and V2 is the propagation speed of the electromagnetic wave in the second medium.

[0069] Exemplarily, the time domain response of the electromagnetic signal can be converted from the frequency domain signal by inverse Fourier transform. The frequency domain signal A(f, θ2, φ) is the response of the transient electromagnetic wave at a specific frequency f. Through inverse Fourier transform, the frequency domain signal can be converted into a time domain signal. Considering the reflection and refraction phenomenon of electromagnetic waves, electromagnetic waves will be reflected and refracted in different media due to the existence of medium interfaces. The propagation direction of the reflected electromagnetic wave can be described by Snell's law. The angle of the electromagnetic wave received by the antenna array is no longer a single elevation angle and depression angle, but is determined by the reflection or refraction angle.

[0070] Although the reflection and refraction angles of Snell's law mainly affect the propagation angle perpendicular to the interface normal vector direction (i.e., the change in the incident angle), the depression angle, as the horizontal component of the electromagnetic wave, determines another dimension of the propagation direction of the electromagnetic wave and still affects the spatial distribution of the entire signal and data acquisition.

[0071] The elevation angle mainly describes the propagation path of electromagnetic waves in the vertical direction, which affects the interaction between electromagnetic waves and underground interfaces (such as reflection and refraction). The depression angle describes the propagation angle on the horizontal plane. Even if the propagation direction of electromagnetic waves in different media has changed due to Snell's law, the distribution on the horizontal plane still determines how electromagnetic waves cover different spatial areas on the horizontal level.

[0072] The calculation formula of depth d depends on the elevation angle and depression angle. Even if the elevation angle changes due to reflection or refraction, the projection of the horizontal depression angle on the propagation depth is still valid. θ2 is the reflection or refraction angle adjusted by Snell's law, which affects the propagation in the vertical direction; φ still affects the propagation direction on the horizontal plane. By adjusting the depression angle φ, the antenna can detect the changes of electromagnetic waves in the horizontal space, which is critical to improving the lateral resolution between measurement points.

[0073] In practical applications, electromagnetic wave data collection at different depths and angles can be achieved by jointly adjusting θ2 and φ. The depression angle φ can control the horizontal propagation direction of electromagnetic waves, ensuring that the antenna array can cover a larger area in front of or on the side of the tunnel. Specifically, the detector will adjust φ point by point during measurement to make the measurement density on the horizontal plane higher, thereby increasing the lateral coverage of electromagnetic wave detection.

[0074] In some examples, this also includes:

[0075] The frequency domain response of the above transient electromagnetic signal is determined based on the following formula:

[0076]

[0077] Among them, A0(f) is the amplitude response at frequency f, which is the initial frequency domain response of the signal without the influence of reflection or refraction. α(f, θ2, φ) is the attenuation coefficient, which represents the energy loss caused by medium absorption and attenuation during the propagation of electromagnetic waves. The attenuation coefficient varies with frequency and antenna angle. β(f, θ2, φ) is the phase shift, which represents the phase change caused by different geological structures during the propagation of electromagnetic waves.

[0078] For example, the angle θ2 will change the propagation path of the signal, affecting the energy distribution and phase shift of the signal. Since the propagation speed and propagation direction of the wave are different at different angles on the reflective surface or refractive medium, the attenuation α(f, θ2, φ) and phase β(f, θ2, φ) will also be adjusted accordingly.

[0079] The attenuation coefficient describes the energy loss of a wave during reflection or refraction. It usually depends on the dielectric properties of the material, the characteristics of the reflecting surface or the refracting medium. The attenuation coefficient depends not only on the frequency f, but also on the angle of incidence θ2 and the depression angle φ, because the energy of the wave is attenuated to different degrees at different angles. For example, a larger angle of incidence will lead to increased attenuation of the wave when penetrating the medium, while at a smaller reflection angle, the reflected wave may experience less energy loss.

[0080] Phase change represents the effect of reflection or refraction on the propagation phase of the wave. Waves experience different phase shifts on different paths, and this shift usually depends on the speed of the wave in the medium and the length of the path. For example, when a wave penetrates or reflects, it may produce different phase delays due to different electromagnetic properties of the medium or the extension of the path.

[0081] In some examples, the method of processing the time domain response of the transient electromagnetic signal using an adaptive filtering algorithm to obtain processed data includes:

[0082] The above adaptive filter is expressed as:

[0083]

[0084] In the formula, is the processed data after filtering, which means the signal after noise and interference are suppressed, w n is the weight coefficient of the adaptive filter, which indicates the influence of the filter on the signal in the past n moments, N is the order of the filter, tn is the backtracking of the signal at the past n moments, and the past signal is used to adjust the output signal at the current moment during the filtering process, indicating the moment of n sampling cycles before the current moment t, and E(tn, θ2, φ) is the time domain response corresponding to the transient electromagnetic signal at moment tn.

[0085] Exemplarily, given an input signal, an adaptive filter is used to process the signal to obtain a processed output signal. The output signal represents the electromagnetic signal after filtering, noise and interference are effectively suppressed, and the signal quality is significantly improved. The core of the adaptive signal processing algorithm is that it can adaptively adjust the filter parameters according to the signal characteristics under different geological conditions. The specific adaptive adjustment strategy is as follows:

[0086] Noise detection: In complex geological environments, electromagnetic signals are often interfered by background noise (such as steel frames, cables, external environmental electromagnetic fields, etc.). Adaptive filters can identify these noises by analyzing the frequency characteristics of the signal and reduce their impact by adjusting the weights.

[0087] Signal enhancement: The filter not only suppresses noise, but also enhances the useful signal through adaptive adjustment. By optimizing the loss function, the adaptive algorithm can amplify the useful part of the electromagnetic signal to ensure that geological information can still be effectively extracted in complex environments.

[0088] Exemplarily, N represents the order of the adaptive filter, that is, the number of past time instants the filter performs backtracking and weighted processing on the signal. In an adaptive filter, the signals of the past N time instants are usually used to calculate the current output signal. It can be understood as the length of historical data used by the filter. The larger this value is, the more historical information the filter considers.

[0089] Specifically, N determines the complexity of the filter and the length of its memory of the input signal. For example, when N increases, the filter will use more past time signals to make an estimate of the current time, which helps to better suppress noise and interference, but it may also increase the computational complexity.

[0090] In some examples, this also includes:

[0091] The above weight coefficient w n Determined based on the following formula:

[0092]

[0093] In the formula, and are the weight values ​​of the filter at the kth and k+1th iterations, μ is the learning rate, and Loss is the loss function;

[0094] Loss = L1 + L2 + L3 + L4

[0095] Where D(t, θ2, φ) is the desired output signal, SNR(t) -1is the inverse value of the signal-to-noise ratio, λ1 is the weight coefficient of the signal-to-noise ratio suppression term, λ2 is the weight coefficient of the sparsity penalty term, λ3 is the weight coefficient of the time domain smoothness penalty term, and T is the length of the time series.

[0096] For example, in a complex geological environment, the signal is interfered by various noises, and the sparsity and time-frequency characteristics of the signal are helpful to improve the signal quality. The defined loss function satisfies: the gap between the processed signal and the expected signal is minimized; the noise suppression term is increased to ensure that the noise component in the filtered signal is effectively weakened; the sparsity of the signal (that is, there are fewer meaningful non-zero elements in the signal) can also be used as one of the optimization goals, so that important information can be retained and unnecessary noise components can be removed; the time-frequency distribution characteristics of the electromagnetic signal need to be maintained during the filtering process to avoid signal distortion in the time-frequency domain.

[0097] The first term, the mean square error (MSE), measures the square of the error between the filtered signal and the desired signal; by minimizing this term, the filtered signal is closer to the desired signal. In the second term, the signal-to-noise ratio suppression term, λ1 is a weight coefficient used to balance the influence of the signal-to-noise ratio suppression term in the total loss function. A larger λ1 will make the signal-to-noise ratio suppression more important; the inverse of the signal-to-noise ratio (SNR) SNR(t) -1 , the signal-to-noise ratio represents the ratio of signal strength to noise strength, and the inverse value indicates that the stronger the noise (the smaller the signal-to-noise ratio), the greater the weight of this item; therefore, in the period of strong noise, the filter will pay more attention to suppressing noise; the square term of the signal after filtering and signal-to-noise ratio SNR(t) -1 Multiplication means that the greater the signal strength, the stronger the need for noise suppression. In the third signal sparsity penalty term, λ2 is the weight coefficient that controls the sparsity penalty term. A larger λ2 value means a greater tendency to retain sparsity and make the signal as simple and clean as possible. is the L1 norm, which represents the sum of the absolute values ​​of the signal. This term is used to control the sparsity of the signal. By minimizing the L1 norm, most signal components can be made close to 0, and a few important signal components can be retained to meet the sparsity requirements. In the fourth term of the time-frequency smoothing penalty term, λ3 is the weight coefficient that controls the time domain smoothness penalty term. A larger λ3 value means that the signal is expected to change smoothly in time; the second-order derivative of the filtered signal represents the acceleration change of the signal over time. The larger the second-order derivative, the more drastic the signal changes in time. By minimizing this term, the drastic changes in the signal can be reduced, making the signal smoother in time; in order to ensure signal smoothness, the second-order derivative is squared so that drastic changes in the signal will be punished more.

[0098] In some examples, this also includes:

[0099] The above three-dimensional geological model is determined based on the following formula:

[0100]

[0101] Where M(x, y, z) is the three-dimensional geological model, which is the geological structure at the midpoint (x, y, z) in space, L is the total number of geological layers, l is the lth geological layer, and w is the total number of geological layers. l is the weight coefficient of the lth layer, B is the total number of frequency bands, f i is the center frequency of the ith frequency band, R(t, x, y, z, f i , l) is the time inversion operator, which converts the time domain signal into the spatial geological model information. This operator indicates that the electromagnetic wave propagates from time t, and through time inversion, calculates its path to the position (x, y, z) in the geological model, and combines the frequency band and the characteristics of the geological layer l for imaging. δ(·) is the impulse response of instantaneous propagation, v l is the electromagnetic wave in the geological medium layer l, frequency f i The propagation speed below.

[0102] Exemplarily, the reverse time migration algorithm is a high-precision algorithm commonly used in seismic exploration and electromagnetic data imaging. It reconstructs underground structural information by reversely propagating the electromagnetic signal measured on the ground. The core of three-dimensional imaging is to construct a three-dimensional structural model of the underground through the method of multi-band joint and multi-layer inversion of the processed transient electromagnetic signal. In the imaging process, electromagnetic waves in different frequency bands have different penetration depths. The lower frequency band is suitable for detecting deep geological structures, while the higher frequency band is suitable for detecting shallow structures. In order to improve the imaging accuracy, the electromagnetic data of multiple frequency bands are inverted to achieve multi-band joint imaging. The underground medium often has a multi-layer structure, and the conductivity and electromagnetic wave propagation characteristics of each layer are different. Therefore, when three-dimensional imaging is performed, the underground structure needs to be divided into multiple layers, and the inversion results of each layer will be superimposed on the final three-dimensional model to achieve multi-layer inversion.

[0103] In some examples, the above-mentioned prediction of geological conditions based on the anomaly density function, the pattern recognition algorithm and the above-mentioned three-dimensional geological model includes:

[0104] The abnormal area is determined based on the abnormal density function and the three-dimensional geological model, wherein the abnormal density function is:

[0105]

[0106] In the formula, ρ a (x, y, z) is the abnormal density value of the point (x, y, z) in the three-dimensional geological model, reflecting whether the point is likely to belong to the abnormal area; σ(·) is an activation function (similar to the activation function in a neural network), which is used to normalize the calculation results, and Sigmoid or Tanh functions can be used; is the Laplace operator, which is used to capture local changes in the model, especially rapidly changing areas (such as faults or holes); μ M is the mean density of the three-dimensional geological model;

[0107] The abnormal area is identified based on the pattern recognition algorithm to predict the geological conditions, wherein the pattern recognition algorithm includes a feature extraction operation, an abnormal point classification operation and an abnormal type optimization operation.

[0108] For example, the Laplacian operator It is a tool used to calculate local changes in 3D models. It can detect rapidly changing areas in the model, such as faults or holes. Its expression is:

[0109]

[0110] The above formula is the sum of the second-order derivatives of the model in three directions. If the Laplace value of a point is large, it indicates that there is a significant density change near the point, which may be a geological anomaly area.

[0111] Mean density μ M The expression is:

[0112]

[0113] Where V is the volume of the geological model.

[0114] First, the anomaly density is calculated based on the 3D geological model. The goal of this step is to identify local anomaly areas in the model. a The calculation of (x, y, z) will capture significant changes in the model by performing a Laplace operation on the second-order derivatives of the model, including:

[0115] Calculate the Laplace operator of the three-dimensional geological model M (x, y, z); calculate the mean density to remove the influence of the global uniform part; calculate the anomaly density function.

[0116] Secondly, after detecting the abnormal density, feature extraction operations are performed on these abnormal areas so that they can be further input into the pattern recognition algorithm for classification. The goal of feature extraction is to extract useful information from the abnormal area that can characterize the type of anomaly (such as faults, voids, etc.), including: extracting shape features in the abnormal area, such as the size, shape, and distribution of Laplace values ​​of the area; extracting density distribution features, especially the density gradient and local density change rate in the abnormal area; these features are used to form a feature vector F (x, y, z) as the input of the next pattern recognition algorithm.

[0117] Then, the outlier classification operation is performed, and the extracted features are classified using a pattern recognition algorithm. The specific algorithm can select a classifier such as a support vector machine (SVM), a random forest, or a neural network. Each outlier is classified according to the feature vector, including:

[0118] Train the classifier using prior data (historical geological data or simulated data) to train the classifier. I , C I}, where F I is the eigenvector of the I-th data point, C I is the corresponding geological anomaly category (such as fault, cavity, etc.). Training classifier f(F I ) enables it to learn the mapping relationship from feature vectors to anomaly types C.

[0119] Apply the classifier to classify the new 3D geological model data. Use the trained classifier f(F(x, y, z)) to classify the feature vector of each point and obtain the anomaly category (preliminary anomaly type) of each point, expressed as:

[0120] C(x,y,z)=f(F(x,y,z))

[0121] Finally, the anomaly type optimization operation is performed, and the classification results are optimized using the probability model. The points are classified into the most likely anomaly type through the maximum probability method, including:

[0122] Using a Bayesian classifier or a neural network, calculate the anomaly density ρ at each point (x, y, z) a (x, y, z) belong to different categories P k (ρ a (x, y, z)), expressed as:

[0123] P k (ρ a (x, y, z)) = P(ρ a (x,y,z)|C k )

[0124] Among them, C k represents the kth abnormal category, P k (ρ a (x, y, z)) means that the point (x, y, z) belongs to category C k probability.

[0125] According to the maximum probability classification method, the anomaly of point (x, y, z) is classified into the category with the highest probability, which is expressed as:

[0126] C(x, y, z) = argmax k P k (ρ a (x, y, z)

[0127] Where C(x, y, z) is the classification result at (x, y, z), indicating which type of geological anomaly the point belongs to; argmax k Represents finding the category k that maximizes the probability, P k (ρ a (x, y, z)) represents the abnormal density value ρ at the point (x, y, z) a The probability that (x, y, z) belongs to class k; this probability is calculated using a pattern recognition algorithm, usually using a trained classifier such as a neural network or random forest.

[0128] By calculating the anomaly density ρ at each point a (x, y, z), and classify these abnormal areas through pattern recognition algorithms. The abnormal areas in the geological model can be divided into different categories. For example, a fault may appear as a large density change, while a cavity may appear as a sudden decrease in density. The probability model is used to determine which type of geological anomaly each point belongs to.

[0129] When multiple abnormal points are distributed continuously, they together form an abnormal area (such as a cavity or fault). Therefore, the density function is first calculated at each point, and then the abnormal area is formed by the combination of multiple abnormal points. The abnormal area can be formed by density changes. The pattern recognition algorithm analyzes and classifies the characteristics of each point (such as density, shape, position, etc.) to identify whether the point belongs to a certain type of geological anomaly (such as a cavity, fault, etc.). Each point is classified independently, and then the abnormal area can be further identified through the classification results of these points. For example, when multiple consecutive points are classified as "cavities", the area composed of these points is a cavity area.

[0130] Once the anomaly is identified, geological analysis can predict the geological conditions ahead of the tunnel through simulation and evaluation, including:

[0131] First, a detailed analysis of the geometric and physical characteristics of these regions is required, including:

[0132] The shape and size of the abnormal area, for example, whether it is a long fault, a round or irregular cavity.

[0133] The spatial distribution of the region, by analyzing the location and distribution pattern of the abnormal areas, can infer whether these anomalies are isolated or have a strong association with the surrounding geological structure.

[0134] Abnormal distribution of density or conductivity, judging the type of geological material in the area and its stability based on the physical characteristics of the area in the 3D model, such as conductivity fluctuations.

[0135] Secondly, the identified anomaly areas are compared with existing geological databases or historical data. By comparing these anomalies with similar structures in history, the nature of these anomalies can be better understood, including:

[0136] Historical tunnel project failure cases are compared with the areas where collapse or other accidents have occurred in the past to determine whether these anomalies are potentially destructive.

[0137] Precedents of similar geological settings, comparing the area with the same type of geological structures, further infer the nature of the anomaly and its possible stability.

[0138] After identifying and analyzing the abnormal area, simulation using physical models is a key step in predicting the geological conditions ahead of the tunnel. By simulating the behavior of the abnormal area under different scenarios, the potential changes in the area during tunnel excavation can be predicted. The simulation methods include:

[0139] Numerical simulation, using the finite element method (FEM) or finite difference method (FDM) to simulate the abnormal area, simulate the impact of tunnel construction on the abnormal area, including stress and strain distribution.

[0140] Seismic wave simulation simulates the impact of vibration during tunnel construction on abnormal areas and assesses whether structural instability or local collapse will occur in abnormal areas during construction.

[0141] Fluid dynamics simulation: if there is an aquifer in the abnormal area, simulate the flow of water and its impact on the geological structure to determine potential problems such as seepage and dissolution in the area during construction.

[0142] Then, through the above-mentioned model simulation and historical data analysis, the risks of the identified abnormal areas to tunnel construction are further evaluated, including:

[0143] Collapse risk, assessing the likelihood of collapse of the area during tunnelling or construction, particularly for identified voids and fault areas.

[0144] Groundwater leakage: simulate and evaluate whether there are hydrological problems in abnormal areas, such as groundwater leakage, water inrush, etc. These problems may lead to unstable foundation or other hidden dangers during construction.

[0145] Explosive gas risk: In certain geological environments, abnormal areas may be rich in gas. By monitoring and evaluating the gas content in abnormal areas, it is possible to determine whether there is a risk of sudden gas leakage or explosion.

[0146] Finally, after completing the above analysis and simulation, a comprehensive prediction of the geological conditions ahead of the tunnel is carried out. The comprehensive prediction includes:

[0147] Short-term prediction, based on the structural characteristics of the current abnormal area, evaluates the geological risks that may be encountered in the short term in front of the tunnel during construction, such as the response measures when the tunnel boring machine encounters an abnormal area.

[0148] Long-term prediction considers the impact of abnormal areas on the long-term safety of the entire tunnel and evaluates whether the abnormal areas will have long-term effects after the tunnel is completed, such as the stability of the tunnel and maintenance needs.

[0149] See also Figure 2 , which is a schematic diagram of the structure of a tunnel advanced geological prediction monitoring device provided in an embodiment of the present application, comprising:

[0150] An acquisition unit 21, used to acquire a time domain response of a transient electromagnetic signal;

[0151] The processing unit 22 is used to process the time domain response of the transient electromagnetic signal using an adaptive filtering algorithm to obtain processed data;

[0152] A generating unit 23, used to generate a three-dimensional geological model according to the above-mentioned processed data using a three-dimensional imaging algorithm;

[0153] The identification unit 24 is used to predict the geological conditions based on the anomaly density function, the pattern recognition algorithm and the above three-dimensional geological model.

[0154] See also Figure 3 The embodiment of the present application also provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor. When the processor 320 executes the computer program 311, the steps of any of the above-mentioned methods for advanced tunnel geological forecasting monitoring are implemented.

[0155] Since the electronic device introduced in this embodiment is a device used to implement a tunnel advanced geological prediction monitoring device in the embodiment of the present application, based on the method introduced in the embodiment of the present application, the technical personnel in this field can understand the specific implementation mode of the electronic device of this embodiment and its various variations. Therefore, how the electronic device implements the method in the embodiment of the present application is not introduced in detail here. As long as the equipment used by the technical personnel in this field to implement the method in the embodiment of the present application is within the scope of protection of this application.

[0156] During the specific implementation process, when the computer program 311 is executed by a processor, any implementation method in the embodiments corresponding to the first aspect can be implemented.

[0157] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0158] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-readable program code.

[0159] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0160] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0161] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0162] The present application also provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device executes Figure 1The process of the tunnel advanced geological prediction monitoring method in the corresponding embodiment.

[0163] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a server or a data center that includes one or more available media integration. The available medium can be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)), etc.

[0164] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0165] In the several embodiments provided in the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0166] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0167] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0168] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program codes.

[0169] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

[0170] Although the preferred embodiments of this specification have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of this specification.

[0171] Obviously, those skilled in the art can make various changes and modifications to this specification without departing from the spirit and scope of this specification. Thus, if these modifications and variations of this specification fall within the scope of the claims of this specification and their equivalents, this specification is also intended to include these modifications and variations.

Claims

1. A tunnel advanced geological prediction monitoring method, characterized in that: The method comprises: Obtaining the time domain response of transient electromagnetic signals; Using an adaptive filtering algorithm to process the time domain response of the transient electromagnetic signal to obtain processed data; generating a three-dimensional geological model based on the processed data using a three-dimensional imaging algorithm; Based on the anomaly density function, the pattern recognition algorithm and the three-dimensional geological model, the geological conditions are predicted.

2. The tunnel advanced geological prediction monitoring method according to claim 1 is characterized in that: The time domain response of the transient electromagnetic signal is expressed as: d=v2tcos(θ2)cos(φ) In the formula, E(t, θ2, φ) is the time domain response of the transient electromagnetic signal, A(f, θ2, φ) is the frequency domain response of the transient electromagnetic signal, t is the current moment, f is the frequency, θ2 is the angle after reflection or refraction, φ is the depression angle, j is the imaginary unit, 2πft is the phase factor, d is the propagation depth of the electromagnetic wave, θ1 is the incident angle of the electromagnetic wave, V1 is the propagation speed of the electromagnetic wave in the first medium, and V2 is the propagation speed of the electromagnetic wave in the second medium.

3. The tunnel advanced geological prediction monitoring method according to claim 2 is characterized in that: Also includes: The frequency domain response of the transient electromagnetic signal is determined based on the following formula: Among them, A0(f) is the amplitude response at frequency f, a(f, θ2, φ) is the attenuation coefficient, which represents the energy loss caused by medium absorption and attenuation during the propagation of electromagnetic waves, and the attenuation coefficient varies with frequency and antenna angle, and β(f, θ2, φ) is the phase shift, which represents the phase change caused by different geological structures during the propagation of electromagnetic waves.

4. The tunnel advanced geological prediction monitoring method according to claim 2 is characterized in that: The method of using an adaptive filtering algorithm to process the time domain response of the transient electromagnetic signal to obtain processed data includes: The adaptive filter is expressed as: In the formula, is the processed data after filtering, w n is the weight coefficient of the adaptive filter, N is the order of the filter, tn is the time n sampling cycles before the current time t, and E(tn, θ2, φ) is the time domain response corresponding to the transient electromagnetic signal at time tn.

5. The tunnel advanced geological prediction monitoring method according to claim 4 is characterized in that: Also includes: The weight coefficient w n Determined based on the following formula: In the formula, and are the weight values ​​of the filter at the kth and k+1th iterations, μ is the learning rate, and Loss is the loss function; Loss = L1 + L2 + L3 + L4 Where D(t, θ2, φ) is the desired output signal, SNR(t) -1 is the inverse value of the signal-to-noise ratio, λ1 is the weight coefficient of the signal-to-noise ratio suppression term, λ2 is the weight coefficient of the sparsity penalty term, λ3 is the weight coefficient of the time domain smoothness penalty term, and T is the length of the time series.

6. The tunnel advanced geological prediction monitoring method according to claim 1 is characterized in that: Also includes: The three-dimensional geological model is determined based on the following formula: Where M(x,y,z) is the three-dimensional geological model, which is the geological structure at the midpoint (x,y,z) in space, L is the total number of geological layers, l is the lth geological layer, and w l is the weight coefficient of the lth layer, B is the total number of frequency bands, f i is the center frequency of the ith frequency band, R(t, x, y, z, f i , l) is the time inversion operator, which converts the time domain signal into the spatial geological model information, δ(·) is the impulse response of instantaneous propagation, v l is the electromagnetic wave in the geological medium layer l, frequency f i The propagation speed below.

7. The tunnel advanced geological prediction monitoring method according to claim 1 is characterized in that: The method of predicting geological conditions based on the anomaly density function, the pattern recognition algorithm and the three-dimensional geological model includes: The abnormal area is determined based on the abnormal density function and the three-dimensional geological model, wherein the abnormal density function is: In the formula, ρ a (x, y, z) is the abnormal density value of the point (x, y, z) in the three-dimensional geological model, σ(·) is the activation function, is the Laplace operator, μ M is the mean density of the three-dimensional geological model; The abnormal area is identified based on the pattern recognition algorithm to predict the geological condition, wherein the pattern recognition algorithm includes a feature extraction operation, an abnormal point classification operation and an abnormal type optimization operation.

8. A tunnel advanced geological prediction monitoring device, characterized in that: include: An acquisition unit, used for acquiring a time domain response of a transient electromagnetic signal; A processing unit, used for processing the time domain response of the transient electromagnetic signal using an adaptive filtering algorithm to obtain processed data; A generating unit, configured to generate a three-dimensional geological model according to the processed data using a three-dimensional imaging algorithm; The identification unit is used to predict geological conditions based on the anomaly density function, the pattern recognition algorithm and the three-dimensional geological model.

9. An electronic device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is used to implement the steps of the tunnel advanced geological prediction monitoring method as described in any one of claims 1 to 7 when executing the computer program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the tunnel advanced geological prediction monitoring method as described in any one of claims 1 to 7 is implemented.

Images