Intelligent control method and system for anchoring force of complex stratum freight cableway

Abstract

The application discloses a kind of complex stratum freight cableway anchoring force intelligent regulation and control method and system, relating to cableway engineering technical field, comprising the following steps: temperature decoupling is carried out to wavelength shift by self-calibration temperature decoupling method, and actual strain value and strain value confidence are obtained, and the wavelength shift is extracted from the obtained anchor rod multi-source data set;Based on actual strain value and anchor rod multi-source data set, with axial force distribution as solving target, axial force data of anchor rod is obtained by physical constraint inversion method;Real-time prediction of axial force data is carried out by Bayesian algorithm, and compensation control instruction is generated according to the evolution trend of predicted axial force data.The application is used to solve the problem that the prediction result of axial force is not accurate due to temperature interference and the anchoring force regulation means is lagging.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cableway engineering, and more particularly, to an intelligent regulation method and system for anchoring force of freight cableway in complex stratum. BACKGROUND

[0002] As an important transportation mode in mountainous and complex stratum environments, freight cableway plays an irreplaceable role in mineral exploitation, material transportation in mountainous areas and infrastructure construction. Its operation safety highly depends on the stability of the anchoring system. As a key force component of the cableway anchoring system, the stress state of the anchor rod directly determines the anchoring performance and structural safety. However, in the stratum environment of soft surrounding rock, high stress area and complex hydrogeological conditions, the anchoring force evolution process is often affected by the coupling of multiple factors, including rock mass heterogeneity, pore water pressure change and vibration disturbance. This makes the anchoring system show nonlinear, non-stationary and uncertain characteristics in long-term service, which brings challenges to monitoring and regulation.

[0003] In recent years, multi-source sensing means have been introduced into the monitoring of the anchoring system, realizing real-time acquisition of strain, temperature, pore pressure and dynamic disturbance, providing data support for axial force distribution inversion and anchoring performance evaluation. However, the existing methods generally lack adaptive decoupling process, resulting in the mixing of temperature, noise and cross interference in the sensor signal, which reduces the conversion accuracy of strain-axial force. Moreover, most prediction models are based on empirical regression or single numerical simulation, without systematically introducing mechanical equilibrium conditions and shear action models, making it difficult to reflect the real anchoring stress process.

[0004] For example, the invention patent announcement No. CN112095596B based on cloud platform's slope pre-stressed anchor rod intelligent monitoring and early warning system and method, relates to the technical field of anchor rod intelligent early warning, the method comprises the following steps: S1, based on a plurality of stress sheets, real-time acquisition of strain values of each point on the anchor rod, and uploading the strain values to the cloud server through the field communication gateway; S2, according to the strain values of each point, the average shear stress prediction model of anchor rod is constructed; S3, the void stress function relationship between the anchor rod and the anchor hole is calculated; S4, according to the void stress function relationship in S3, the average shear stress prediction model of anchor rod is corrected to obtain the anchor rod shear stress prediction model; S5, according to the time sequence, the absolute value D of the difference between the anchor rod shear stress value output by the anchor rod shear stress prediction model and the anchor rod shear stress value measured by the magnetic flux sensor is measured; S6, according to the anchor rod accident type, the expert knowledge base is used to judge the authenticity of the accident, and the absolute value D is divided to obtain the divided absolute value d1, d2, d3, d4, … dn; S7, based on deep learning, and according to the divided absolute value d1, d2, d3, d4, … dn, the neural network model of accident warning is constructed, and the output accident warning information is transmitted to the management personnel APP and the field construction APP through the accident warning neural network model.

[0005] In the above disclosed technical solution, at least the following technical problems exist:

[0006] Mainly rely on stress sheet and magnetic flux sensor, it is difficult to fully characterize the anchoring stress state under complex stratum, the existing model only considers the average shear stress and pore stress function, does not carry out distributed load inversion, can not obtain the fine distribution of the full length axial force of anchoring body, only stays at the level of "accident warning", does not propose the active compensation and intelligent control mechanism based on the prediction result, and can not realize the prospective adjustment of anchoring force.

[0007] In view of the above problems, the present application provides a solution. SUMMARY

[0008] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a complex stratum and freight cableway anchoring force intelligent control method and system, which generates compensation control instructions by temperature decoupling and inversion of predicted axial force evolution trend, to solve the problems of inaccurate axial force prediction results disturbed by temperature and lag of anchoring force control means.

[0009] To achieve the above object, the present application provides the following technical scheme:

[0010] The application discloses an intelligent regulation and control method for anchoring force of a complex stratum freight cableway, and comprises the following steps: temperature decoupling is performed on a wavelength offset by a self-calibration temperature decoupling method to obtain an actual strain value and a strain value confidence, wherein the wavelength offset is extracted from an obtained anchor rod multi-source data set; based on the actual strain value and the anchor rod multi-source data set, an axial force distribution is taken as a solving target, and an axial force data of the anchor rod is obtained by a physical constraint inversion method; the axial force data is predicted in real time by a Bayesian algorithm, and a compensation regulation instruction is generated according to an evolution trend of the predicted axial force data.

[0011] In a preferred embodiment, the anchor rod multi-source data set comprises a wavelength offset, a temperature, a pore water pressure and a vibration acceleration measured by a multi-source sensor; the wavelength offset comprises a strain-induced wavelength offset and a temperature-induced wavelength offset.

[0012] In a preferred embodiment, the temperature decoupling of the wavelength offset by the self-calibration temperature decoupling method to obtain the actual strain value specifically comprises the following steps: a first sensor coefficient is obtained and a strain coupling equation is constructed; a first wavelength offset is obtained, the strain coupling equation is decoupled once by a double-wavelength difference method to obtain a decoupled strain reference value; based on a real-time feature vector and the first wavelength offset, the first sensor coefficient is updated by a recursive least square method to obtain a second sensor coefficient, wherein the real-time feature vector comprises the strain reference value and a temperature change amount; and the strain coupling equation is decoupled twice based on the second sensor coefficient to obtain the actual strain value.

[0013] In a preferred embodiment, the method for obtaining the strain value confidence specifically comprises the following steps: a second wavelength offset is calculated by the strain coupling equation according to the second sensor coefficient; a Mahalanobis distance is calculated according to a residual error between the second wavelength offset and the first wavelength offset; if the Mahalanobis distance result is greater than a first threshold value, it is determined that the corresponding sensor is abnormal, and the strain value confidence is calculated based on an inverse function of the Mahalanobis distance; and if the Mahalanobis distance result is greater than a second threshold value, it is determined that the corresponding sensor is invalid, and the strain value confidence is set to zero.

[0014] In a preferred embodiment, the axial force data of the anchor rod is obtained by taking the axial force distribution as the solving target based on the actual strain value and the anchor rod multi-source data set by the physical constraint inversion method, and specifically comprises the following steps: a distributed load on the anchor rod is calculated by linear coupling based on the pore water pressure and the vibration acceleration; an axial force balance equation is established based on the distributed load; an inversion model with the axial force distribution as the solving target is constructed by a Bayesian weighted least square method with the axial force balance equation as a physical constraint, and the axial force data is output.

[0015] In a preferred embodiment, the construction is an inversion model with axial force distribution as the solving target, specifically: a plurality of actual strain values are spliced to form an actual strain vector, and a weight matrix is generated according to the corresponding strain value confidence; the to-be-solved axial force data is converted into an approximate strain vector through a strain-axial force conversion matrix; under the weighting of the weight matrix, an inversion model with axial force distribution as the solving target is constructed by taking the error between the actual strain vector and the approximate strain vector as the target and introducing a regularization term.

[0016] In a preferred embodiment, the axial force data is predicted in real time through a Bayesian algorithm, and compensation control instructions are generated according to the evolution trend of the predicted axial force data, specifically: a Bayesian linear model is constructed through a Bayesian algorithm based on historical axial force data, the Bayesian linear model is composed of a state equation and an axial force observation equation, and the axial force observation equation is obtained based on the state equation; the Bayesian linear model is recursively solved through a Kalman filtering algorithm to obtain predicted axial force data; when the predicted axial force data is lower than a warning threshold, corresponding compensation control instructions are generated according to the descending mode of the axial force curve.

[0017] In a preferred embodiment, the Bayesian linear model is recursively solved through a Kalman filtering algorithm to obtain predicted axial force data, specifically: the prior distribution at the current time is calculated through the state equation based on the posterior distribution at the previous time, and the axial force prediction mean at the current time is calculated through the axial force observation equation based on the prior distribution; the error between the axial force data at the current time and the axial force prediction mean is calculated, and the Kalman gain matrix is calculated according to the error; the prior distribution at the current time is modified using the Kalman gain matrix to obtain the posterior distribution at the current time; the state equation is recursively solved based on the posterior distribution at the current time to obtain the prior distribution of multiple states, and the predicted axial force data is calculated through the axial force observation equation.

[0018] In a preferred embodiment, the compensation control instructions are generated according to the evolution trend of the predicted axial force data, specifically: the relative local steepness index is calculated according to the first derivative of the predicted axial force curve; the maximum persistence ratio of the minimum point of the axial force curve is obtained through persistence analysis, and the persistence of the minimum point is determined by the difference in function values between the maximum point and the minimum point; according to the relative local steepness index and the maximum persistence ratio, it is determined whether the descending mode of the curve is local sudden drop or uniform slow drop; if the axial force curve presents a local sudden drop state, a grouting compensation instruction is generated; if the axial force curve presents a uniform slow drop state, a tension compensation instruction is generated.

[0019] The application discloses an intelligent regulation and control system for anchoring force of a freight cableway in a complex stratum, and relates to the technical field of cableway anchoring force regulation and control.

[0020] The application discloses an intelligent regulation and control method and system for anchoring force of a freight cableway in a complex stratum.

[0021] The application discloses an intelligent regulation and control method and system for anchoring force of a freight cableway in a complex stratum. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 The application discloses an intelligent regulation and control method and system for anchoring force of a freight cableway in a complex stratum.

[0023] Figure 2 The application discloses an intelligent regulation and control method and system for anchoring force of a freight cableway in a complex stratum.

[0024] Figure 3 The application discloses an intelligent regulation and control method and system for anchoring force of a freight cableway in a complex stratum. DETAILED DESCRIPTION

[0025] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the application.

[0026] Embodiment 1, Figure 1 The application discloses an intelligent regulation and control method and system for anchoring force of a freight cableway in a complex stratum.

[0027] S1, temperature decoupling is performed on the wavelength offset by using a self-calibration temperature decoupling method, actual strain values and strain value confidence levels are obtained, and the wavelength offset is extracted from the obtained anchor rod multi-source data set.

[0028] S2, based on the actual strain value and the anchor rod multi-source data set, taking the axial force distribution as a solving target, obtaining the axial force data of the anchor rod through a physical constraint inversion method;

[0029] S3, predicting the axial force data in real time through a Bayesian algorithm, generating a compensation control instruction according to an evolution trend of the predicted axial force data.

[0030] The application eliminates the influence of temperature on the strain signal by introducing a self-calibration and temperature decoupling method; combines a physical constraint inversion method to establish an inversion model taking the axial force distribution as a solving target, thereby ensuring the consistency of the predicted and actual stress states; further adopts a dynamic linear model to predict the evolution trend of the future effective anchoring force, and generates a compensation control instruction based on the prediction result, thereby realizing real-time and reliable monitoring and intelligent dynamic control of the anchoring force.

[0031] S1, performing temperature decoupling on the wavelength shift through a self-calibration temperature decoupling method to obtain the actual strain value and the strain value confidence, wherein the wavelength shift is extracted from the obtained anchor rod multi-source data set.

[0032] In this embodiment, the anchor rod multi-source data set includes the wavelength shift, temperature, pore water pressure and vibration acceleration measured by the multi-source sensors.

[0033] In this embodiment, the fiber Bragg grating (FBG) sensor array, pore water pressure sensor and vibration sensor are arranged in the full length range of the anchor rod. The FBG sensor represents the strain and temperature change in the form of wavelength shift, the pore water pressure sensor outputs the pore water pressure, and the vibration sensor outputs the acceleration signal, and the data acquisition frequency is not less than 10Hz. The wavelength shift, temperature, pore water pressure and vibration acceleration measured by the multi-source sensors are spliced to form the anchor rod multi-source data set.

[0034] In this embodiment, the wavelength shift is decoupled through the self-calibration temperature decoupling method to obtain the actual strain value, specifically as follows:

[0035] Obtaining the first sensor coefficient and constructing the strain coupling equation;

[0036] Obtaining the first wavelength shift, decoupling the strain coupling equation once through the dual-wavelength difference method to obtain the decoupled strain reference value;

[0037] Based on the real-time feature vector and the first wavelength shift, updating the first sensor coefficient through the recursive least squares method to obtain the second sensor coefficient, wherein the real-time feature vector includes the strain reference value and the temperature change;

[0038] Based on the second sensor coefficient, decoupling the strain coupling equation twice to obtain the actual strain value.

[0039] In the embodiment, the wavelength shift includes a strain-induced wavelength shift and a temperature-induced wavelength shift.

[0040] In the embodiment, according to the wavelength shift and the temperature in the anchor rod multi-source data set and the first sensor coefficient of the sensor, a strain coupling equation is established by the wavelength shift principle of the fiber Bragg grating (FBG) sensor, and the first sensor coefficient includes a strain sensitivity coefficient and a temperature sensitivity coefficient.

[0041] The first wavelength shift is observed from the fiber Bragg grating (FBG) sensor, a double-wavelength differential method is used for decoupling once, two strain coupling equations with different strain sensitivity coefficients are set, the two equations are subtracted, and since the temperature change is the same, it is assumed that the temperature sensitivity coefficients are similar, so the temperature term is greatly offset, thereby obtaining a rough estimated strain reference value which eliminates most of the temperature influence.

[0042] On this basis, the real-time feature vector and the first wavelength shift are taken as inputs, the first sensor coefficient is updated by using the recursive least square method, and a second sensor coefficient is obtained, and the real-time feature vector includes the strain reference value and the temperature change.

[0043] On the basis of the second sensor coefficient, the double-wavelength differential method is used to decouple the strain coupling equation again, and an actual strain value is obtained.

[0044] It should be noted that the first sensor coefficient uses the initial nominal value provided by the manufacturer or the estimated value obtained by the last recursive least square method iteration.

[0045] The strain coupling equation is as follows:

[0046]

[0047] In the formula, is a wavelength shift, is a strain sensitivity coefficient, is a strain value, is a temperature sensitivity coefficient, is a temperature change.

[0048] The double-wavelength differential method formula is as follows:

[0049]

[0050] In the formula, is a wavelength shift of the first FBG sensor grating, is a wavelength shift of the second FBG sensor grating, The strain sensitivity coefficient of the first FBG sensor grating, The strain sensitivity coefficient of the second FBG sensor grating, The decoupled strain reference value.

[0051] The recursive least square formula is as follows:

[0052]

[0053] In the formula, The parameter vector to be estimated is The updated strain sensitivity coefficient is The updated temperature sensitivity coefficient is The real-time feature vector is The strain value at the moment is The temperature change amount at the moment is The wavelength change amount observed at the moment is The parameter estimation error variance matrix is The forgetting factor is

[0054] Figure 3 The wave length offset decoupling and the wave length offset after decoupling are given in the wave length offset decoupling and the wave length offset after decoupling.

[0055] In the embodiment, the strain value confidence acquisition method is specifically:

[0056] The second wave length offset is calculated through the strain coupling equation according to the second sensor coefficient;

[0057] The Mahalanobis distance is calculated according to the residual between the second wave length offset and the first wave length offset;

[0058] If the Mahalanobis distance result is greater than the first threshold value, it is determined that the corresponding sensor is abnormal, and the strain value confidence is calculated based on the inverse function of the Mahalanobis distance;

[0059] If the Mahalanobis distance result is greater than the second threshold value, it is determined that the corresponding sensor is invalid, and the strain value confidence is zero.

[0060] In the embodiment, the second sensor coefficient obtained by the recursive least square method is re-input into the strain coupling equation of the wave length drift, and the second wave length offset is calculated;

[0061] The residual of the first wave length offset and the second wave length offset actually observed is calculated, and the Mahalanobis distance is calculated under the constraint of the variance matrix according to the residual; ​​​

[0062] If the calculated Mahalanobis distance is greater than a set first threshold value, it is determined that there is an abnormal offset of the corresponding sensor, and the system adjusts the strain value confidence of the sensor in inverse proportion to the Mahalanobis distance;

[0063] If the calculated Mahalanobis distance is greater than a set second threshold value, it is determined that the sensor is in a serious failure state, and the system directly sets the strain value confidence of the sensor to zero and excludes the observation results of the sensor in subsequent calculation and inversion.

[0064] The calculation formula of the Mahalanobis distance is as follows:

[0065]

[0066] In the formula, is the Mahalanobis distance, is the residual, is the residual variance matrix.

[0067] S2, based on the actual strain value and the anchor rod multi-source data set, taking the axial force distribution as the solving target, the axial force data of the anchor rod is obtained by the physical constraint inversion method.

[0068] In the embodiment, the axial force data of the anchor rod is obtained based on the actual strain value and the anchor rod multi-source data set, taking the axial force distribution as the solving target, by the physical constraint inversion method, specifically:

[0069] Based on the pore water pressure and the vibration acceleration, the distributed load on the anchor rod is calculated through linear coupling;

[0070] The axial force balance equation is established based on the distributed load;

[0071] Taking the axial force balance equation as the physical constraint, an inversion model taking the axial force distribution as the solving target is constructed by the Bayesian weighted least squares method, and the axial force data is output.

[0072] In the embodiment, the inversion model taking the axial force distribution as the solving target is constructed, specifically:

[0073] A plurality of actual strain values are spliced to form an actual strain vector, and a weight matrix is generated according to the corresponding strain value confidence;

[0074] The to-be-solved axial force data is converted into an approximate strain vector through a strain-axial force conversion matrix;

[0075] Under the weighting of the weight matrix, an inversion model taking the axial force distribution as the solving target is constructed by taking the minimum error between the actual strain vector and the approximate strain vector as the target and introducing a regularization term.

[0076] In this embodiment, the target inversion model is solved using the Markov chain Monte Carlo method to obtain the axial force data over the entire length of the anchor bolt.

[0077] It should be noted that the Markov chain Monte Carlo method is existing technology and will not be described in detail in this embodiment.

[0078] The formula for distributed load is as follows:

[0079]

[0080] in, For the anchor bolt in position and time Distributed load, , The coupling coefficient is... For the anchor bolt in position and time pore water pressure, For the anchor bolt in position and time The vibration acceleration.

[0081] The axial force equilibrium equations are as follows:

[0082]

[0083] In the formula, For the anchor bolt in position and time axial force.

[0084] The specific formula for constructing the inversion model using the Bayesian weighted least squares method is as follows:

[0085]

[0086]

[0087] In the formula, This is the actual strain vector. This represents the actual strain value; This is the weight matrix. Confidence level for strain values; This is the strain-axial force transformation matrix. For elastic modulus, Let be the cross-sectional area of ​​the anchor rod perpendicular to the axis. This is an approximate strain vector. For the differential operator matrix, For regularization parameters, This represents the axial force distribution vector of the entire anchor bolt. To minimize the dependent variable value.

[0088] S3, predicting the axle force data in real time through a Bayesian algorithm, and generating compensation control instructions according to the evolution trend of the predicted axle force data.

[0089] In the embodiment, the predicting the axle force data in real time through the Bayesian algorithm, and generating the compensation control instructions according to the evolution trend of the predicted axle force data specifically comprises:

[0090] Based on historical axle force data, a Bayesian linear model is constructed through a Bayesian algorithm, the Bayesian linear model is composed of a state equation and an axle force observation equation, and the axle force observation equation is obtained based on the state equation;

[0091] The Bayesian linear model is recursively solved through a Kalman filtering algorithm to obtain predicted axle force data;

[0092] When the predicted axle force data is lower than a warning threshold, corresponding compensation control instructions are generated according to a descending mode of an axle force curve.

[0093] In the embodiment, the axle force observation equation specifically comprises:

[0094]

[0095] In the formula, is an axle force observation value at a current time, is a design matrix, is a state vector, is an observation noise, is an observation noise variance.

[0096] The state equation specifically comprises:

[0097]

[0098] is a state vector, is a state transition matrix, is a state noise, is a state noise variance.

[0099] It should be noted that the design matrix is a matrix that maps the state vector to the observation space; and the state transition matrix is a matrix that automatically evolves the state vector from the previous time to the current time.

[0100] In the embodiment, the recursively solving the Bayesian linear model through the Kalman filtering algorithm to obtain the predicted axle force data specifically comprises:

[0101] Based on a posterior distribution at the previous time, a prior distribution at the current time is calculated through the state equation, and an axle force prediction mean at the current time is calculated through the axle force observation equation according to the prior distribution.

[0102] Calculate the error between the current axial force data and the predicted mean axial force, and calculate the Kalman gain matrix based on the error;

[0103] The prior distribution at the current time step is corrected using the Kalman gain matrix to obtain the posterior distribution at the current time step.

[0104] Based on the posterior distribution of the current moment, the recursive state equation is derived to obtain the prior distribution of the multi-step state, and the predicted axial force data is calculated through the axial force observation equation.

[0105] In this embodiment, it is assumed that at time The posterior distribution is as follows:

[0106]

[0107] in, for The state vector at time t, For time 1 to All data at any given time, The mean of the state estimates. Let be the variance matrix.

[0108] The prior distribution at the current moment is calculated using the state equation as follows:

[0109]

[0110] In the formula, The mean of the state predictions. This represents the variance of the state prediction.

[0111] In this embodiment, the predicted mean value of the axial force at the current moment is calculated based on the prior distribution using the axial force observation equation, as follows:

[0112]

[0113] In the formula, To One-step prediction value, This represents the variance of the one-step prediction.

[0114] In this embodiment, the error between the actual axial force data and the predicted average axial force is calculated as follows:

[0115]

[0116] In the formula, This represents the prediction error.

[0117] In this embodiment, the formula for calculating the Kalman gain matrix is ​​as follows:

[0118]

[0119] wherein, is the Kalman gain matrix.

[0120] In this embodiment, the posterior distribution at the current time is as follows:

[0121]

[0122] wherein, is the updated mean of the state estimate, is the updated variance of the state estimate.

[0123] In this embodiment, the state equation is recursively applied based on the posterior distribution at the current time to obtain a multi-step state prior distribution, which is as follows:

[0124]

[0125] wherein, is the mean of the state prediction of the th step, is the variance of the state prediction of the th step, is the state transition matrix of the th step, is the state noise variance of the th step.

[0126] In this embodiment, the predicted axial force data is calculated by the observation equation, which is as follows:

[0127]

[0128] wherein, is the predicted value of the axial force point of the th step in the future, is the variance of the predicted distribution, is the design matrix, is the observation noise variance of the th step.

[0129] In this embodiment, when the predicted axial force data is lower than the early warning threshold, the corresponding compensation control instruction is generated according to the descending pattern of the axial force curve.

[0130] In this embodiment, the compensation control instruction is generated according to the evolution trend of the predicted axial force data, which is as follows:

[0131] The relative local steepness index is calculated according to the first derivative of the predicted axial force curve;

[0132] The maximum durability percentage of the minimum point of the axial force curve is obtained by durability analysis. The durability of the minimum point is determined by the difference in function values ​​between the maximum and the minimum point.

[0133] Based on the relative local steepness index and the percentage of maximum persistence, the curve descent pattern is determined to be either a local sudden drop or a uniform gradual drop.

[0134] If the axial force curve shows a local sudden drop, a grouting compensation command will be generated.

[0135] If the axial force curve shows a uniform and gradual decrease, a tension compensation command is generated.

[0136] It should be noted that the flatter and longer the axial force curve declines, the more the load is distributed among the strata, and the stronger the overall capacity (anchoring force) of the anchoring system. Conversely, the steeper and shorter the axial force curve declines, the weaker the strata, the shorter the effective anchoring section, and the weaker the overall anchoring force. When the axial force curve declines evenly and gradually, it indicates that the anchoring system is generally robust, but the tension of the anchor bolt itself needs to be fine-tuned to ensure that the entire anchor bolt section is evenly stressed and maintains the designed prestressed state. When the axial force curve shows a sudden drop in a local area, it indicates that the local bearing capacity is insufficient and needs to be quickly reinforced. In this case, for the section with a sudden drop in axial force, grouting is used to enhance the bond and bearing capacity between the anchor bolt and the surrounding strata, thereby improving the anchoring capacity of the local section.

[0137] In this embodiment, the formula for calculating the relative local steepness index is as follows:

[0138]

[0139] In the formula, This is a relative local steepness index. For the first The first derivative of each sampling point To prevent small constants with a denominator of zero.

[0140] The formula for calculating the maximum durability percentage is as follows:

[0141]

[0142]

[0143] In the formula, The percentage of maximum durability at the smallest point. The maximum durability at the minimum point of the curve. The sum of the durability of all minima. To prevent small constants with a denominator of zero, For minimal point durability, This represents the axial force value at a local maximum point. This represents the axial force value at a local minimum point.

[0144] In this embodiment, according to the joint results of the relative local steepness index and the maximum duration ratio, it is determined that the curve is locally dropped or uniformly dropped, specifically as follows:

[0145] If And , it is determined to be a local drop;

[0146] If And , and the average derivative , it is determined to be a uniform drop;

[0147] Other cases are marked as ambiguous state and need to be reviewed for a short time.

[0148] In this embodiment, if the axial force curve presents a local drop state, a grouting compensation instruction is generated;

[0149] If the axial force curve presents a uniform drop state, a tension compensation instruction is generated.

[0150] Embodiment 2, Figure 2 The application discloses an intelligent regulation and control system for anchoring force of a complex stratum freight cableway, which comprises:

[0151] A temperature decoupling module is configured to perform temperature decoupling on the wavelength shift by using a self-calibration temperature decoupling method, so as to obtain actual strain values and strain value confidence levels, wherein the wavelength shift is extracted from the obtained anchor rod multi-source data set.

[0152] A constraint inversion module is configured to obtain the axial force data of the anchor rod by using a physical constraint inversion method, with the actual strain values and the anchor rod multi-source data set as the solving target and the axial force distribution as the solving target.

[0153] A prediction and control module is configured to predict the axial force data in real time by using a Bayesian algorithm, and to generate a compensation and control instruction according to the evolution trend of the predicted axial force data.

[0154] The above formulas are all dimensionless numerical calculations, and the formulas are obtained by software simulation of a large amount of data to obtain a formula of the nearest real situation, and the preset parameters in the formula are set by a person skilled in the art according to the actual situation.

[0155] The above embodiments can be realized by software, hardware, firmware or any combination thereof, in whole or in part. When realized by software, the above embodiments can be realized in the form of a computer program product in whole or in part.

[0156] Those skilled in the art can understand that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0157] In addition, each functional module in each embodiment of the present application can be integrated in one processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.

[0158] The above is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0159] Finally: the above is only the preferred embodiment of the present application, and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for intelligent regulation of anchoring force of a complex terrain freight cableway, characterized in that, The method comprises the following steps: temperature decoupling is performed on the wavelength shift extracted from the obtained anchor multi-source data set by a self-calibration temperature decoupling method to obtain actual strain values and strain value confidence levels; the temperature decoupling is performed to obtain the actual strain values, specifically as follows: a first sensor coefficient is obtained and a strain coupling equation is constructed; a first wavelength shift is obtained, and the strain coupling equation is decoupled once by a dual-wavelength difference method to obtain a decoupled strain reference value; based on a real-time feature vector and the first wavelength shift, the first sensor coefficient is updated by a recursive least squares method to obtain a second sensor coefficient, wherein the real-time feature vector comprises the strain reference value and a temperature change amount; the strain coupling equation is decoupled twice based on the second sensor coefficient to obtain the actual strain values; based on the actual strain values and the anchor multi-source data set, a physical constraint inversion method is used to obtain the axial force data of the anchor with the axial force distribution as a solving target; the axial force data is predicted in real time by a Bayesian algorithm, and compensation control instructions are generated according to the evolution trend of the predicted axial force data.

2. The method for intelligent regulation of anchoring force of complex terrain freight cableway according to claim 1, characterized in that, The anchor multi-source data set comprises wavelength shifts, temperatures, pore water pressures and vibration accelerations measured by multi-source sensors; the wavelength shifts comprise strain-induced wavelength shifts and temperature-induced wavelength shifts.

3. The method of claim 2, wherein the method further comprises: The method for obtaining the strain value confidence level is specifically as follows: a second wavelength shift is calculated by the strain coupling equation according to the second sensor coefficient; a Mahalanobis distance is calculated according to the residual error between the second wavelength shift and the first wavelength shift; if the Mahalanobis distance result is greater than a first threshold value, it is determined that the corresponding sensor is abnormal, and the strain value confidence level is calculated based on the inverse function of the Mahalanobis distance; if the Mahalanobis distance result is greater than a second threshold value, it is determined that the corresponding sensor is invalid, and the strain value confidence level is set to zero.

4. The method of claim 3, wherein the method further comprises: The physical constraint inversion method is used to obtain the axial force data of the anchor with the axial force distribution as a solving target based on the actual strain values and the anchor multi-source data set, specifically as follows: a distributed load on the anchor is calculated by linear coupling based on the pore water pressure and the vibration acceleration; an axial force balance equation is established based on the distributed load; an inversion model with the axial force distribution as a solving target is constructed by a Bayesian weighted least squares method with the axial force balance equation as a physical constraint, and the axial force data is output.

5. The method for intelligent regulation of the anchoring force of a complex terrain freight cableway according to claim 4, characterized in that, The inversion model with the axial force distribution as a solving target is constructed, specifically as follows: a plurality of actual strain values are spliced to form an actual strain vector, and a weight matrix is generated according to the corresponding strain value confidence levels; the to-be-solved axial force data is converted into an approximate strain vector by a strain-axial force conversion matrix; an inversion model with the axial force distribution as a solving target is constructed by minimizing the error between the actual strain vector and the approximate strain vector under the weighting of the weight matrix and by introducing a regularization term.

6. The method for intelligent regulation of anchoring force of complex terrain freight cableway according to claim 5, characterized in that, The Bayesian algorithm is used to predict the axial force data in real time, and compensation control instructions are generated according to the evolution trend of the predicted axial force data, specifically as follows: a Bayesian linear model is constructed by a Bayesian algorithm based on historical axial force data, wherein the Bayesian linear model is composed of a state equation and an axial force observation equation, and the axial force observation equation is obtained based on the state equation; The predicted axial force data is obtained by recursively solving the Bayesian linear model through a Kalman filtering algorithm; When the predicted axial force data is lower than the early warning threshold, corresponding compensation control instructions are generated according to the descending mode of the axial force curve.

7. The method for intelligent regulation of the anchoring force of a complex terrain freight cableway according to claim 6, characterized in that, The predicted axial force data is obtained by recursively solving the Bayesian linear model through a Kalman filtering algorithm, specifically as follows: Based on the posterior distribution at the previous time, the prior distribution at the current time is calculated through the state equation, and the axial force prediction mean at the current time is calculated according to the prior distribution through the axial force observation equation; The error between the axial force data at the current time and the axial force prediction mean is calculated, and the Kalman gain matrix is calculated according to the error; The prior distribution at the current time is modified using the Kalman gain matrix to obtain the posterior distribution at the current time; The state equation is recursively solved based on the posterior distribution at the current time to obtain the prior distribution of multiple states, and the predicted axial force data is calculated through the axial force observation equation.

8. The method for intelligent regulation of the anchoring force of a complex terrain freight cableway according to claim 7, characterized in that, The compensation control instructions are generated according to the evolution trend of the predicted axial force data, specifically as follows: The relative local steepness index is calculated according to the first derivative of the predicted axial force curve; The maximum persistence ratio of the minimum point of the axial force curve is obtained through persistence analysis, and the persistence of the minimum point is determined by the difference between the function values of the maximum point and the minimum point; According to the relative local steepness index and the maximum persistence ratio, it is determined whether the curve descending mode is local sudden drop or uniform slow drop; If the axial force curve presents a local sudden drop state, a grouting compensation instruction is generated; If the axial force curve presents a uniform slow drop state, a tension compensation instruction is generated.

9. A system for intelligent control of anchoring force of a complex stratum freight cableway using the method according to any one of claims 1-8, comprising: A temperature decoupling module is configured to decouple the wavelength shift from the obtained anchor rod multi-source data set by a self-calibration temperature decoupling method to obtain actual strain values and strain value confidence, and the actual strain values are obtained by decoupling the strain coupling equation through the first sensor coefficient. The first wavelength shift is obtained, and the strain reference value after decoupling is obtained by decoupling the strain coupling equation once through a dual-wavelength difference method. Based on the real-time feature vector and the first wavelength shift, the second sensor coefficient is obtained by updating the first sensor coefficient through a recursive least squares method, and the real-time feature vector includes the strain reference value and the temperature change amount; and the actual strain value is obtained by decoupling the strain coupling equation twice based on the second sensor coefficient. A constraint inversion module is configured to obtain the axial force data of the anchor rod by a physical constraint inversion method based on the actual strain value and the anchor rod multi-source data set, with the axial force distribution as the solving target. A prediction and control module is configured to predict the axial force data in real time through a Bayesian algorithm, and generate compensation control instructions according to the evolution trend of the predicted axial force data.

Images