Method for inverting field-scale debris flow through earth sound signals

Abstract

The invention discloses a method for inverting field scale debris flow through earth sound signals, and the method comprises the following steps: S1, constructing a conversion relation between debris flow earth sound signals and vibration signals, converting the physical work of a rock-soil body into signal work, and constructing a theoretical relation between the signal amplitude epsilon'of an FBG sensor and the single compression work W of the rock-soil body; and S2, on the basis of the theoretical basis of the step S1, constructing a physical model for inverting the debris flow. According to the method, inverse operation of Fch-kinetic parameters can be completed, an inversion chain of'earth sound signals-Fch-kinetic parameters' is formed, and a physical model of earth sound inversion debris flow is successfully constructed.

Description

Technical Field

[0001] The present invention relates to the field of debris flow discharge measurement, and particularly to a method for inverting the field-scale debris flow discharge from ground sound signals. Background Art

[0002] The debris flow discharge can directly characterize the scale of the debris flow and indirectly reflect the impact and burial damage ability on the downstream. It is one of the most important dynamic parameters of the debris flow. Therefore, inverting the debris flow discharge in a scientific and reasonable manner can provide more refined early warning information for debris flow disaster prevention and mitigation.

[0003] The ground sound signal of the debris flow contains debris flow attribute parameters and dynamic parameters. Inverting the target parameters from the complex ground sound signal can provide a basis for accurate debris flow early warning. The ground sound of the debris flow mainly comes from the ground vibration generated by the collision and friction between the debris flow front and the channel. This vibration continuously spreads and propagates on the ground to form the ground sound. The slight vibration of the ground felt during the occurrence of the debris flow is the most intuitive manifestation of the ground sound. This phenomenon can be generalized into two physical processes: (1) The loading process of the debris flow front impacting the soil mass of the channel bed cross-section is an external excitation factor leading to the ground sound of the debris flow, which involves the dynamic theory research on the debris flow impacting the channel bed cross-section, aiming to establish the theoretical equation of the impact force F of the debris flow on the channel bed ch and the debris flow dynamic parameters (such as flow velocity and mud depth) and attribute parameters (such as particle size and particle gradation) as the main research objectives. The process of the soil mass of the channel bed vibrating under the impact of the debris flow front is the internal cause of the formation of the ground sound of the debris flow, which involves the dynamic theory research on the forced vibration characteristics of the channel bed cross-section, aiming to establish the theoretical equation of the impact force F of the debris flow on the channel bed ch and the theoretical equation of the surface deformation characteristics as the main research objectives. The above two physical processes form a theoretical research chain of "dynamic parameters → F ch → ground sound signal". However, the inversion chain of "ground sound signal → F ch → dynamic parameters" has not been fully established. The inversion of the ground sound signal to F ch results in a broken chain, and the ground sound signal cannot directly invert the dynamic parameters. Summary of the Invention

[0004] To solve the above problems, the purpose of the present invention is to provide a method for inverting the field-scale debris flow discharge from ground sound signals, which can complete the inverse operation of F ch → dynamic parameters, form an inversion chain of "ground sound signal → F ch → dynamic parameters", and successfully construct a physical model for inverting the debris flow discharge from ground sound.

[0005] The present invention is achieved by the following technical solutions:

[0006] Method for inverting debris flow discharge at field scale based on ground sound signals, comprising the following steps: S1. Establish the conversion relationship between debris flow ground sound signals and vibration signals, convert the physical work of the rock and soil mass into signal work, and establish the theoretical relationship between the ground sound signal amplitude ε' and the single compression work W of the rock and soil mass:

[0007]

[0008] y is the radius of the optical fiber cross-section, l is the length of the optical fiber in the sensor, i.e., the length of the simply supported beam, E0I0 is the bending stiffness of the optical fiber, dm is the mass of the microelement in the length direction of the fiber Bragg grating, E is the elastic modulus of the rock and soil mass, H is the proportional coefficient between the vibration acceleration of the rock and soil mass and the maximum deformation compression amount. In the same gully, is a parameter only related to the sensor. Let When the gully and the sensor remain unchanged, β is a fixed value, then formula 2-10 can be simplified to:

[0009]

[0010] S2. Based on the theoretical basis of step S1, establish a physical model for inverting debris flow discharge: establish the coupling relationship among the debris flow, the gully, and the ground sound sensor. Select the flow rate Q and density ρ to represent the debris flow influencing factors, select the gully roughness n to represent the gully influencing factors, and select the length l as the ground sound sensor influencing factor. According to the dimensional relationship of the parameters W, Q, ρ, n, l, establish the following equation:

[0011] W = f′(Q, ρ, n, l) 2-12

[0012] According to the dimensional analysis method, the dimensional analysis result is:

[0013]

[0014] Adjust formula 2-13 to:

[0015]

[0016] Simultaneously solve formulas 2-11 and 2-14:

[0017]

[0018] Let Simplify formula 2-15:

[0019]

[0020] Obtain the inversion model of debris flow discharge, where ε' is the ground sound signal amplitude, which is the input of the flow rate inversion model, and Q is the debris flow discharge, which is the output of the flow rate inversion model.

[0021] Construct the conversion relationship between debris flow geoacoustic signal and vibration signal, convert the physical work of rock and soil into signal work, and construct the theoretical relationship between the amplitude of geoacoustic signal ε' and the work W of single compression of rock and soil. The specific steps include: obtaining the relationship between the change of FBG central wavelength Δλ and the amplitude of geoacoustic signal ε' and the relationship between the change of FBG central wavelength Δλ and the acceleration of the bottom of the water tank. The linear relationships are as follows:

[0022]

[0023] and

[0024] Based on formula 2-5, the change in the central wavelength of the FBG Δλ is related to the vibration acceleration of the channel Linear positive correlation, when the external excitation of the channel increases, the vibration acceleration increases, and the maximum compression of the rock and soil deformation increases, that is, the following results can be obtained:

[0025]

[0026] Then the relationship between the channel vibration acceleration and the maximum deformation compression can be expressed by the following formula:

[0027]

[0028] Where H is the proportionality coefficient between the vibration acceleration of the rock mass and the maximum deformation compression; then combine formulas 2-4, 2-5, 2-9 and the single compression work W of the rock mass:

[0029] σ=EΔl2-6

[0030]

[0031] In the formula, σ is the normal stress on the rock and soil, E is the elastic modulus of the rock and soil, and Δl is the maximum compression of the rock and soil deformation; finally, formula 2-10 is obtained.

[0032] Before inverting the debris flow flow, the ground sound signal representing the discrete impact of the debris flow in the debris flow-induced ground sound needs to be smoothed and filtered.

[0033] The specific method of smoothing and filtering is to separate the spike pulses induced by the discrete impact of debris flow, and then perform smoothing filtering with different window lengths to determine the most suitable smoothing method.

[0034] The inventors found that the inversion of the ground acoustic signal to F ch The fundamental reason why the geoacoustic signal cannot directly invert the dynamic parameters is that: (1) the theoretical equation u(t) = F ch(t) The calculation parameters required for the Green's function in G(t) are complex and highly sensitive to anisotropic formations. The geoacoustic signal → F ch The robustness of inversion is low; (2) Describing F ch The theoretical calculation formula involves many variables such as debris flow particle size, instantaneous velocity, flow depth, particle gradation, etc. (Tsai et al., 2012; Zhang et al., 2020). These independent variables are coupled and affect F ch . Using F ch to invert a specific kinetic parameter among them requires a complex decoupling process and cannot complete the inverse operation target of F ch → kinetic parameters. The inventor intends to use the internal force received by the geoacoustic measurement point and the elastic compression potential energy of the measurement point as a connection, and use the methods of elastic mechanics and dimensionless theory analysis to construct the theoretical relationship between debris flow discharge and geoacoustic signal Q = f(acc); adopt the flume experiment and fiber optic sensing detection technology of the invention patent (application number: 202410552719.X) to obtain the corresponding flume experiment data, correct the theoretical relationship Q = f(acc), and construct a physical model for inverting debris flow discharge from geoacoustic signals;

[0035] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0036] The present invention can complete the inverse operation of F ch → kinetic parameters, form an inversion chain of "geoacoustic signal → F ch → kinetic parameters", and successfully construct a physical model for inverting debris flow discharge from geoacoustic signals. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, form a part of this application, and do not constitute a limitation to the embodiments of the present invention. In the drawings:

[0038] Figure 1 is the calculation result of the model coefficient K.

[0039] Figure 2 is the comparison diagram of the debris flow discharge inversion result and the experimental measured result.

[0040] Figure 3 is the comparison diagram of the model inversion result and the measured data.

[0041] Figure 4 is the processing result diagram of discrete impact with different smoothing window lengths.

[0042] Figure 5 is the field debris flow discharge inversion result diagram.

[0043] Figure 6It is a result map of small-volume debris flow inversion. Specific implementation manners

[0044] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with embodiments and the accompanying drawings. The illustrative embodiments of the present invention and their descriptions are only used to explain the present invention and do not limit the present invention.

[0045] Embodiment 1

[0046] 1.1 Monitoring of debris flow ground sound signals

[0047] The ground sound signal detection device used in the present invention is the same as that of the invention patent (application number: 202410552719.X). Based on the bending theory of a simply supported beam, the theoretical relationship between the FBG central wavelength shift and the external excitation acceleration is as follows:

[0048] According to the physical characteristics of the fiber Bragg grating, the relationship between the change in the FBG central wavelength Δλ and the ground sound signal amplitude ε' is as follows:

[0049]

[0050] In the formula, 0.0012 is the FBG strain first-order term coefficient, which is the characteristic parameter of the FBG. 0.0012 is the FBG first-order strain coefficient. The result of is με, and by adding 10 -6 Convert the unit με to the standard unit ε. M is the bending moment at the center position. As can be seen from the above, the change in the FBG central wavelength Δλ is linearly and positively correlated with the acceleration at the bottom of the water tank. Combining formulas 2-3 and 2-4, the results are as follows: Show a linear positive correlation, and the combined results of formulas 2-3 and 2-4 are as follows:

[0051]

[0052] In actual application scenarios, the initial data of the fiber Bragg grating sensor will be adjusted according to formula 2-4, and the initial FBG central wavelength data Δλ will be converted into the ground sound signal amplitude ε', so that the vibration data obtained by the sensor has an intuitive physical meaning.

[0053] 1.2 Theoretical relationship between debris flow discharge and ground sound signal Q = f(acc)

[0054] The formation process of debris flow ground sound is generalized as the repeated movement of an infinitesimal element along a certain direction. This repeated movement of the infinitesimal element is the microscopic manifestation of ground sound. However, in actual situations, this infinitesimal element is the infinitesimal rock and soil mass in the channel, and the repeated movement of the infinitesimal element is the objective manifestation of the compression and recovery of the rock and soil mass. The state change of this compression and recovery of the rock and soil mass can be analyzed using the principles of material mechanics.

[0055] During the entire process of rock and soil compression and recovery, the compression amount of the rock and soil material is positively correlated with the magnitude of the external force. Therefore, the work W done by a single compression of the rock and soil body can be used to characterize the strength of the external force.

[0056] σ=EΔl2-6

[0057]

[0058] Where σ is the normal stress of the rock and soil, E is the elastic modulus of the rock and soil, and Δl is the maximum compression of the rock and soil deformation.

[0059] During the movement of debris flow, the factors that affect the intensity of debris flow on the channel are mainly the impact intensity of debris flow and the stress state of the channel. The impact intensity of debris flow is mainly related to the density and flow rate of debris flow; the stress state of the channel is related to the ground state of the channel, and the parameter that best reflects the ground state of the channel is the channel roughness. When the channel roughness increases, the impact force on the channel in the same debris flow event is stronger. In actual situations, it is difficult to accurately measure the deformation of rock and soil, and overly complex measurement methods are not suitable for debris flow flow inversion.

[0060] Since it is difficult to directly measure the deformation of rock and soil, the work done by single compression of rock and soil can be replaced by calculating the signal work of ground sound. The ground sound signal is derived from the one-dimensional discrete data of the central wavelength change monitored by the FBG sensor during the movement of debris flow. According to formula 2-5, the change of the FBG central wavelength Δλ is related to the vibration acceleration of the channel. Linear positive correlation, when the external excitation of the channel increases, the vibration acceleration increases, and the maximum compression of the rock and soil deformation increases. In this way, the following results can be obtained:

[0061]

[0062] Where σ is the maximum normal stress of the rock mass. The relationship between the channel vibration acceleration and the maximum deformation compression can be expressed by the following formula:

[0063]

[0064] Where H is the proportional coefficient between the vibration acceleration of the rock and soil mass and the maximum deformation compression.

[0065] Combining formulas 2-4, 2-5, 2-7, and 2-9, the signal work of a single compression of the rock mass is:

[0066]

[0067] In the formula, the amplitude of the geoacoustic signal ε' after simple pre-processing establishes a theoretical bridge with the work done by the single compression of the rock and soil body W. In the same channel, It is a parameter related only to the sensor, and the determination of H in this parameter is very complicated. To simplify the complexity of the flow inversion model and improve the universality of the model, the acquisition of this parameter is more suitable to be obtained through in-situ experiments.

[0068] Let When the channel and the sensor remain unchanged, β is a fixed value, then Equation 2-10 can be simplified to:

[0069]

[0070] Through the connection of the geophone sensor, physical work is converted into signal work. As can be seen from the above, this signal work is the result of the coupling of debris flow, channel and geophone sensor. For debris flow, its contribution to signal work is mainly provided by the impact force of debris flow. Therefore, the flow rate Q and density ρ are selected to represent the debris flow influence factors. The flow rate includes the volume and velocity of debris flow passing through the channel cross-section at a certain moment, and the combination of density and volume can represent the mass of debris flow; for the channel, the channel roughness n is selected to represent the channel influence factor. The magnitude of the roughness indicates the amount of energy received by the channel from the impact force of debris flow. The higher the roughness, the higher the impact energy received by the channel; for the FBG sensor, the cross-sectional size of the optical fiber and the mass dm of the micro-element segment in the length direction are usually constants. Therefore, the length l is selected as the geophone sensor influence factor. The dimensions of the above relevant parameters are as follows:

[0071] Table 1-1 Dimensions of Related Physical Quantities

[0072]

[0073] Analyze the relevant factors by the dimensional analysis method and list the following equations:

[0074] W = f′(Q, σ, n, l) 2-12

[0075] The result of dimensional analysis is:

[0076]

[0077] Adjust Equation 2-13 to:

[0078]

[0079] Combine Equations 2-11 and 2-14:

[0080]

[0081] Let Simplify Equation 2-15:

[0082]

[0083] Through the dimensional analysis method, the functional relationship (flow inversion model) between relevant parameters and debris flow discharge is successfully established, where ε' is the amplitude of the ground sound signal, which is the input of the flow inversion model, and Q is the debris flow discharge, which is the output of the flow inversion model.

[0084] 2. Practice and verification

[0085] Apply the physical model to the Jiangjiagou debris flow gully in Dongchuan, Yunnan. Use the field-measured ground sound signal and flow data to verify the accuracy of the physical model in the field.

[0086] 2.1 Physical model for inverting debris flow discharge from ground sound signals under small-scale conditions

[0087] The flume experiment design is consistent with the invention patent (application number: 202410552719.X).

[0088] (1) First, use the flume experiment to calibrate the K coefficient in Equation 2-16:

[0089] During the flume experiment, not all parameters of the flow inversion model are known, and the model coefficient K still needs to be determined through experiments. Given that In this experiment, β is a fixed value related to the channel filling and sensors, and f(n) is an unknown quantity obtained through dimensional analysis based on Equation 2-12. Therefore, in the flume experiment, only the channel roughness may affect the value of the model coefficient K. Thus, in this study, the experimental results were inversely calculated for the K value by changing the channel roughness in the flume experiment, and the calculation method is as follows:

[0090]

[0091] The flume experiment groups are shown in Table 2-1 below:

[0092] Table 2-1 Experimental group conditions

[0093]

[0094] After collecting and processing 12 groups of variable roughness flume experiment data, the K value was inversely calculated through Equation 3-4, and the experimental results of the repeated experiments were averaged to reduce the experimental error fluctuation. The calculated results of the K value are as Figure 1 shown. The model coefficient K under different roughness conditions fluctuates little. At the same time, through the first-order fitting and second-order fitting of the average value of the experimental results of the coefficient K, the R 2 results of the two fittings are 0.49 and 0.51 respectively, and the fitting effect is poor. Therefore, the average value calculated from the flume experiment data (K = 0.9146) is selected as the value of the model coefficient K in this experimental environment.

[0095] (2)Secondly, the inverted flow process was compared with the measured flow process.

[0096] The inversion results of the flow process are as Figure 2 shown. The experimental observation results and the model calculation results have a very similar changing trend, and the correlation can reach 0.9544. When t = 3 s, the movement of the debris flow has tended to end. Since the ground vibration needs to continuously attenuate energy according to the conditions (different damping) of the material and structure at the bottom of the flume, the "tail sound" of the ground sound signal is not attenuated in time in the same frequency as the actual situation of the debris flow discharge. Therefore, in the obtained flow process line, the model inversion result is relatively flat at the end of the flow process line, while the flow process line of the experimental observation value maintains a normal change. In addition, due to the inherent defects of the video method for measuring the flow velocity, the number of balls distributed at the position of the debris flow head with the fastest flow velocity is small. Therefore, the flow process line of the experimental observation results at the debris flow head part is relatively rough, and the detail changes are slightly inferior to the calculation results of the inversion model.

[0097] (3)Then, the inverted peak discharge was compared with the measured peak discharge.

[0098] In order to understand whether the change in the debris flow volume will affect the inversion results of the peak discharge, the accuracy of the inversion results of the debris flow peak discharge model was tested through multiple groups of flume experiments with variable volumes. The test conditions are shown in Table 2-2, and the test results are as Figure 3 shown. The overall average error of the model inversion results is 32.9%.

[0099] Table 2-2 Experimental ratio schemes for different discharges

[0100]

[0101] 2.2 Precision verification of the physical model under field-scale conditions

[0102] There are four input parameters in the flow inversion model. Among them, ε' is the amplitude of the ground sound signal; K is obtained by inverse calculation through flume experiments and is a constant; E is the channel elastic modulus, which can be measured in-situ or determined according to standards; l is the sensor model parameter and is a constant; ρ is the debris flow density, which is estimated according to the source conditions of the channel.

[0103] The physical model was applied to the Jiangjiagou debris flow ditch in Dongchuan, Yunnan. Using the field-measured ground sound signals and flow data, the accuracy of the physical model in the field was verified.

[0104] When verifying according to the observation data at the Dongchuan Station, the debris flow density in the inversion model can be selected as the observed maximum density of 2300 kg / m 3 according to the observation data at the Dongchuan Station, because the density of the first 8 largest-scale (all are episodic flows) debris flows is 2300 kg / m3 In debris flow events, the determination of the maximum peak discharge of debris flow is an important parameter for the design of field monitoring and early warning schemes for debris flow.

[0105] Under field conditions, the ground sound sensor is installed on the hillside on one side of the gully. The geological conditions at this location are mainly grayish-black sandstone and slate. A concrete pile is buried and fixed on the hillside, and the ground sound sensor is connected to the concrete pile. Compared with directly installing the ground sound sensor on the hillside, installing the ground sound sensor by pouring a concrete pile makes the sensor installation more firm. At the same time, the vibration characteristics of concrete are more stable, which is beneficial to the input of gully parameters in the debris flow discharge inversion model. The elastic modulus of the selected concrete pile is 3 GPa, and the density is 2300 kg / m 3 As the gully elastic modulus and density in the discharge inversion model.

[0106] In the processing of field ground sound signals of debris flow, this study has filtered out the noise of debris flow-induced ground sound to obtain the effective ground sound signals induced by debris flow. However, if the discharge is directly inverted based on the effective ground sound signals after noise filtering, a result with a higher calculated value will inevitably be obtained. The reason for this deviation is due to the fluctuation of the physical properties of the debris flow itself during the movement process. The debris flow itself is also composed of debris particles and slurry. During the movement of the debris flow, a certain amount of debris particles will break away from the restraint of the slurry and undergo discrete impacts. The debris flow impact can be divided into discrete impacts and overall impacts.

[0107] The debris particles that break away from the slurry constraint are separated from the overall impact process and instead join the discrete impact. However, when the debris particles impact as independent individuals, a greater impact intensity will be induced compared to the overall impact. Therefore, before inverting the debris flow discharge, it is necessary to smoothly filter out the ground sound signals representing the discrete impact of the debris flow-induced ground sound.

[0108] The influence of debris flow discrete impacts on debris flow discharge inversion not only appears under field conditions but also significantly appears in flume experiments. The purpose of the flume experiment is to determine the relevant parameters in the debris flow discharge inversion model and analyze the characteristics of debris flow-induced ground sound signals and the changes in debris flow-related characteristic parameters. Therefore, when smoothly filtering the ground sound signals, it can be inferred reversely through the observed data in the experiment. However, in the verification of field data, this method cannot be used. Instead, it is necessary to separately isolate the spike pulses induced by debris flow discrete impacts and then perform smooth filtering with different window lengths to determine the most suitable smoothing method.

[0109] Such as Figure 4As shown in the figure, different smoothing window lengths are used to smooth the discrete impact signal. When the smoothing window length is greater than 90, the increase in the ability of smoothing filtering to cut the peaks of discrete impact signals has gradually tended to 0. Therefore, a smoothing window length of 90 can be selected. (Note: Strain ε corresponds to the ground acoustic amplitude signal).

[0110] Verification and Analysis of Flow Inversion Results

[0111] Taking the amplitude of the debris flow-induced ground acoustic signal after filtering out the interference signal as the input of the model formula 2-16 for flow inversion calculation, the calculation results are as Figure 5 shown.

[0112] Since the volume of the debris flow ranges from a minimum of 3.3 m 3 / s to a maximum of 278.6 m 3 / s with a very wide span, and at the same time, the time span of the sampling points of the ground acoustic signal is long and the number of samples is huge, it is very difficult to compare the actual observed values with the flow inversion results. If an extremely long smoothing window is used to smooth the ground acoustic signal on the entire time line into the overall trend line of the ground acoustic and then compare, the ground acoustic information of the debris flow with a large volume and periodicity will be lost. Therefore, in this study, the debris flows with large volumes and small volumes will be divided in advance from 78 debris flows and then compared and analyzed separately.

[0113] According to the observation of Figure 5 , a debris flow flow rate of 59 m 3 / s is selected as the division standard for large and small volumes of debris flow. When verifying the flow inversion results of small-volume debris flows, the sliding average can be used to erase the violent fluctuations in the ground acoustic inversion results without worrying about the damage to the ground acoustic induced by large-volume debris flows due to excessive sliding average. The calculation results of the small-volume debris flow flow rate are as Figure 6 shown.

[0114] Since the straight-line length between the debris flow monitoring section and the location of the ground acoustic sensor is 785 m, and the flow velocity of each debris flow is different, it is difficult to correspond the inversion results with the observation results one by one. In order to obtain a more reasonable result when evaluating the accuracy of the inversion results, the sliding average is also performed on the debris flow observation results, and then the correlation coefficient between the calculation results of the model and the smoothed results of the field observation is calculated. The accuracy is evaluated by taking the smoothed result with the maximum correlation coefficient. The calculated maximum correlation coefficient is 0.46.

[0115] The minimum error of the flow inversion results of small-volume debris flows is 3.4%, the maximum error is 77.3%, the average error is 36.0%, and the coefficient of variation of the inversion error is 0.59.

[0116] The specific embodiments described above further elaborate on the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only the specific embodiments of the present invention and is not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. Method for inverting debris flow discharge at field scale by ground sound signal, characterized in that, It includes the following steps: S1. Establish the conversion relationship between debris flow ground sound signals and the vibration acceleration of rock and soil masses, convert the physical work of rock and soil masses into signal work, and establish the theoretical relationship between the signal amplitude ε' of the FBG sensor and the single - compression work W of rock and soil masses: y is the radius of the fiber optic cross-section, l is the length of the optical fiber in the sensor, i.e., the length of the simply supported beam, E0I0 is the bending stiffness of the optical fiber, dm is the mass of the infinitesimal element in the length direction of the fiber Bragg grating, E is the elastic modulus of the rock and soil mass, H is the proportionality coefficient between the vibration acceleration of the rock and soil mass and the maximum deformation compression amount. In the same channel, is a parameter related only to the sensor. Let When β is a fixed value when the channel and the sensor remain unchanged, Equation 2-10 can be simplified to: S2. Based on the theoretical basis of step S1, establish a physical model for inverting the debris flow discharge: establish the coupling relationship among debris flow, channel and ground sound sensors. Select the discharge Q and density ρ to represent the debris flow influencing factors, select the channel roughness n to represent the channel influencing factors, and select the length l as the ground sound sensor influencing factor. According to the dimensional relationship of the parameters W, Q, ρ, n, l, establish the following equation: W = f′(Q,ρ,n,l) 2 - 12 The dimensional analysis result obtained according to the dimensional analysis method is: Adjust formula 2 - 13 to: Simultaneously solve formulas 2 - 11 and 2 - 14: Let Simplify Equation 2-15: Obtain the inversion model of debris flow discharge, where ε' is the amplitude of the ground sound signal, which is the input of the discharge inversion model, and Q is the debris flow discharge, which is the output of the discharge inversion model.

2. The method according to claim 1, wherein Construct the conversion relationship between the debris flow ground sound signal and the vibration acceleration of the rock and soil mass, convert the physical work of the rock and soil mass into signal work, and construct the theoretical relationship between the signal amplitude ε' of the FBG sensor and the single - compression work W of the rock and soil mass. The specific steps are as follows: Obtain the relationship between the change in the central wavelength Δλ of the FBG and the ground sound signal amplitude ε', and the linear relationship between the change in the central wavelength Δλ of the FBG and the acceleration at the bottom of the water tank are as follows respectively: And Based on Equation 2-5, the change in the central wavelength of the FBG, Δλ, is linearly and positively correlated with the vibration acceleration of the channel. When the external excitation received by the channel increases, the vibration acceleration increases accordingly, and the maximum compression of the deformation of the rock and soil mass increases. That is, the following results can be obtained: When the external excitation received by the channel increases, the vibration acceleration increases accordingly, and the maximum compression of the deformation of the rock and soil mass increases. That is, the following results can be obtained: Then the relationship between the channel vibration acceleration and the maximum deformation compression amount can be expressed by the following formula: In the formula, H is the proportionality coefficient between the vibration acceleration of rock and soil masses and the maximum deformation compression amount; then simultaneously solve formulas 2 - 4, 2 - 5, 2 - 9 and the formula related to the single - compression work W of rock and soil masses: σ = EΔl 2 - 6 In the formula, σ is the normal stress borne by the rock and soil mass, E is the elastic modulus of the rock and soil mass, and Δl is the maximum compression amount of the deformation of the rock and soil mass. Finally, formula 2 - 10 is obtained.

3. The method according to claim 1, characterized in that, Before inverting the debris flow discharge, it is necessary to smooth and filter the ground sound signals in the debris - flow - induced ground sound that characterize the discrete impact of debris flow.

4. The method according to claim 3, characterized in that The specific method of smoothing and filtering is: separately isolate the spike pulses induced by the discrete impact of debris flow, and then perform smoothing filtering with different window lengths to determine the most suitable smoothing method.

Images