A stress loss detection system for an anchoring system

Abstract

The application discloses a stress loss detection system of an anchoring system, comprising a multi-source acquisition unit, an edge computing unit, a cloud server and a data decision terminal, strain parameters and environmental parameters of the anchoring system are synchronously monitored through the multi-source acquisition unit, data complementation is realized, and reliability is improved; through the edge computing unit, data transmission volume is reduced and processing speed is improved, through strain multidimensional analysis, corrosion abnormal risk prediction of stress loss is realized; through the cloud server, the synergistic effect of salt spray and temperature and humidity is quantified, environmental interference dynamic compensation is realized, measurement accuracy is significantly improved, through the data decision terminal, risk grading early warning is realized, the false negative rate and the false positive rate of monitoring are reduced, through multi-source data fusion, edge-cloud collaborative calculation and a dynamic compensation model, the accuracy and efficiency of the stress loss detection of the anchoring system are significantly improved, full-link collaborative optimization is realized, and the response speed is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of stress detection, and in particular to a stress loss detection system of an anchoring system. BACKGROUND

[0002] The anchoring system is widely used in bridge, tunnel, building and other engineering, and the stability of its stress state directly affects the structural safety. When the anchoring system is in a complex natural environment, such as a building or a bridge in a coastal area, the anchoring structure will be affected by salt spray corrosion, which will lead to a decrease in the performance of the anchoring material, and further affect the accuracy of the stress loss detection result.

[0003] Among them, the chloride ion in the salt spray environment will corrode the metal materials of the anchoring system, such as anchor rods, anchor cables, etc., which will cause the metal materials to rust, reduce their effective cross-sectional area, and further reduce the mechanical properties of the materials such as strength and toughness. Due to the decrease in the material performance of the anchoring system and the decrease in the anchoring force, the deformation of the anchoring structure will increase under external load. The chloride ion in the salt spray environment has strong permeability and can penetrate the protective coating and react with the metal substrate, which will cause the coating to blister and peel off, losing its protective effect on the metal. Once the protective layer is damaged, the metal material will be directly exposed to the salt spray environment, accelerating the corrosion process.

[0004] However, the traditional stress detection method relies on a single sensor, such as a resistance strain gauge or an electromagnetic sensor, and has the limitations of environmental interference sensitivity and insufficient multi-source data fusion. Due to environmental factors such as salt spray, temperature and humidity, the sensor data may drift, affecting the measurement accuracy. Moreover, there is a lack of collaborative analysis of environmental factors such as corrosion and salt spray and stress loss, making it difficult to comprehensively assess the risks. When choosing a multi-source data fusion mode, the traditional method mainly relies on centralized cloud processing, making it difficult to realize real-time analysis and early warning, which may lead to data processing lag.

[0005] In view of the above technical defects, a solution is proposed. SUMMARY

[0006] The purpose of the present application is to solve the limitations of environmental interference sensitivity and insufficient multi-source data fusion of traditional methods, as well as the problems of insufficient environmental interference suppression, poor real-time performance and data island, providing a high-precision and high-reliability solution for anchoring system health monitoring, effectively preventing structural failure and prolonging the service life of the project.

[0007] In order to achieve the above purpose, the present application adopts the following technical scheme:

[0008] A stress loss detection system of an anchoring system, comprising a multi-source acquisition unit, an edge computing unit, a cloud server and a data decision terminal.

[0009] The multi-source acquisition unit comprises a stress detection module and an environment detection module, which respectively detect the stress loss of the anchoring system and the environment state thereof to obtain multi-source sensing data.

[0010] The edge computing unit is configured to preliminarily analyze the multi-source sensing data: the standard strain value of the anchoring system is obtained through wavelet denoising and deep analysis, the time domain feature, the frequency domain feature and the environment coupling feature of the standard strain value are obtained, and the corrosion abnormality of the anchoring system is evaluated.

[0011] The cloud server is configured to build an environment interference compensation model of the anchoring system strain: by analyzing the correlation function between the salt mist concentration and the temperature and humidity, the environment interference compensation function between the strain and the salt mist is established, the standard strain value is compensated and corrected, and the compensated strain value is generated.

[0012] The data decision terminal is configured to evaluate the stress risk degree of the anchoring system and generate a risk prompt signal for management.

[0013] Further, the specific processing process of the edge computing unit is as follows:

[0014] The standard strain value is obtained through discrete wavelet transform (DWT) of the measured strain value;

[0015] The time domain feature Sf of the standard strain value is obtained by calculating the change of the N standard strain values at adjacent time nodes;

[0016] The frequency domain feature Pf of the standard strain value is obtained by setting the frequency domain interval and performing Fourier transform on the standard strain value;

[0017] The environment coupling feature of the standard strain value is obtained by combining the standard strain value with the salt mist concentration value, and is marked as the strain-salt mist coupling feature Dfc;

[0018] The corrosion abnormality prediction model is established, and the corrosion abnormality of the anchoring system is evaluated by combining the time domain feature Sf, the frequency domain feature Pf and the strain-salt mist coupling feature Dfc of the standard strain value.

[0019] Further, the specific processing process of the cloud server is as follows:

[0020] The correlation function g1 between the salt mist concentration and the environmental temperature value is established by combining the salt mist concentration, the environmental temperature value and the corrosion reaction activation energy through the Arrhenius equation;

[0021] The correlation function g2 between the salt mist concentration and the environmental humidity value is established by combining the salt mist concentration, the environmental humidity value and the salt mist effect saturation coefficient through the salt mist effect experiment;

[0022] By combining the temperature Laplacian operator with salt spray concentration, a correlation function g3 for salt spray concentration in temperature-induced salt deposition is established.

[0023] By substituting the environmental parameters collected at the current time point into the correlation functions g1, g2, and g3, the degree of influence of environmental factors on the salt spray concentration is assessed, thereby compensating and correcting the standard strain value for environmental interference and obtaining the compensated strain value Fbc.

[0024] Furthermore, the process of acquiring multi-source sensor data is as follows:

[0025] Multi-source sensor data includes strain parameters and environmental parameters;

[0026] Strain parameters include the measured strain value Fcl and the reference strain value Fck; environmental parameters include the salt spray concentration value Cyw, the ambient temperature value Whj, and the ambient humidity value Hhj.

[0027] The stress detection module is used to detect stress loss in the anchoring system. The stress detection module includes a fiber optic strain sensor and an electromagnetic induction sensor. The fiber optic strain sensor acquires the measured strain value Fcl, and the electromagnetic induction sensor acquires the reference strain value Fck.

[0028] The environmental monitoring module is used to monitor the environmental conditions of the anchoring system. The environmental monitoring module includes an electrochemical salt spray concentration sensor and an integrated temperature and humidity sensor. The electrochemical salt spray concentration sensor collects the salt spray concentration value Cyw, and the integrated temperature and humidity sensor collects the ambient temperature value Whj and the ambient humidity value Hhj.

[0029] Furthermore, the specific process of preliminary analysis of multi-source sensor data is as follows:

[0030] Extract multi-source sensor data corresponding to N time points t;

[0031] The measured strain value Fcl was decomposed using Discrete Wavelet Transform (DWT).

[0032] Among them, the mark For low-frequency scaling coefficients, For low-frequency wavelet basis functions, These are high-frequency wavelet coefficients. Here, J is the high-frequency wavelet basis function, J is the wavelet decomposition level, and k is the index of the data index, where 0 < k ≤ N.

[0033] Extracting wavelet basis functions The corresponding signal data is labeled with the denoised strain value Fqz. 0 ;

[0034] By normalizing environmental parameters and calculating their standard deviations, the fluctuation coefficients of various environmental parameters are obtained, and then a weighted summation is used to obtain the dynamic adjustment coefficient. And obtain the wavelet threshold by combining the number of data points N. ;

[0035] Using wavelet thresholding For wavelet coefficients The signal is processed to reconstruct the denoised strain signal and labeled as the standard strain value Fqz. 1 .

[0036] Furthermore, by normalizing environmental parameters and calculating their standard deviations, the fluctuation coefficients of various environmental parameter indicators are obtained. The specific process is as follows:

[0037] Label any environmental parameter as y, and normalize y to obtain standard data for y. ;

[0038] Then, the standard data of indicator y Calculate the standard deviation to obtain the volatility coefficient of the indicator y. ;

[0039] The standard deviations of the salt spray concentration value (Cyw), ambient temperature value (Whj), and ambient humidity value (Hhj) were calculated after normalization and then labeled as salt spray fluctuation coefficients. Temperature fluctuation coefficient Humidity fluctuation coefficient .

[0040] Furthermore, a deeper analysis of the standard strain value Fqz was conducted. 1 The specific process is as follows:

[0041] By calculating N standard strain values ​​Fqz 1 The standard strain value Fqz is obtained by taking the absolute value of the difference between data points at adjacent time points and then averaging them. 1 The time-domain features Sf;

[0042] Then, by setting the frequency domain interval [f1, f2] and integrating the square of the Fourier transform amplitude of the standard strain value, the standard strain value Fqz is obtained. 1 The frequency domain characteristics Pf;

[0043] The partial derivative of strain with respect to salt spray concentration was obtained by multivariate regression fitting, and the rate of change of salt spray concentration over time was analyzed. Then, the standard strain value Fqz was used as the basis for the analysis. 1 Combined with the salt spray concentration value Cyw, the standard strain value Fqz is obtained. 1 The environmental coupling characteristics are identified and denoted as strain salt spray coupling characteristics Dfc.

[0044] Furthermore, the specific process for assessing corrosion anomalies related to stress loss is as follows:

[0045] A corrosion anomaly prediction model was established. By assigning corresponding weight coefficients to the time-domain feature Sf, the frequency-domain feature Pf, and the strain-salt spray coupling feature Dfc, the standard strain value Fqz was... 1 By combining the time-domain feature Sf, the frequency-domain feature Pf, and the strain-salt spray coupling feature Dfc, the stress corrosion anomaly index Zff is obtained to assess the corrosion anomaly risk of stress loss.

[0046] Furthermore, the specific process of analyzing the correlation function between salt spray concentration and temperature and humidity is as follows:

[0047] Under constant humidity and constant salt spray conditions, the temperature T was changed, the corrosion rate r was measured, the Arrhenius equation was fitted, and the activation energy Ea of the corrosion reaction was obtained. By combining the salt spray concentration Cyw, the ambient temperature Whj, and the activation energy Ea of the corrosion reaction, the correlation function g1 between the salt spray concentration Cyw and the ambient temperature Whj was established.

[0048] By varying the salt spray concentration Cyw under constant temperature and humidity experimental conditions, and measuring the corrosion current density I, the saturation coefficient of the salt spray effect was obtained. The salt spray concentration Cyw, ambient humidity Hhj, and salt spray effect saturation coefficient were used to determine the salt spray effect saturation coefficient. By combining these factors, a correlation function g2 is established between the salt spray concentration Cyw and the ambient humidity value Hhj.

[0049] Calculation of the temperature Laplace operator using the finite difference method Through the temperature Laplace operator By combining the salt spray concentration Cyw, a correlation function g3 was established for the effect of salt spray concentration Cyw on temperature-induced salt deposition.

[0050] The degree to which salt spray concentration is affected by environmental factors is assessed using correlation functions g1, g2, and g3.

[0051] Furthermore, the specific process for establishing the environmental disturbance compensation function between strain and salt spray is as follows:

[0052] By substituting the environmental parameters collected at the current time point into the correlation functions g1, g2, and g3, corresponding sensitivity coefficients are assigned to the correlation functions in turn. The salt spray concentration Cyw is then compared with the coefficients of temperature-induced salt deposition and the dynamic effect of salt spray. Combined, the standard strain value Fqz 1 Environmental disturbance compensation and correction are performed to obtain the compensation strain value Fbc;

[0053] The sensitivity coefficients corresponding to the correlation functions g1, g2, and g3 are integrated into an environmental sensitivity coefficient matrix R. The environmental sensitivity coefficient matrix R and the salt spray dynamic effect coefficients are obtained by minimizing the sum of squares of the differences between the reference strain value Fck and the compensated strain value Fbc. ;

[0054] By setting a risk range for the compensation strain value Fbc and comparing the ranges, the stress risk level of the anchoring system can be assessed, and corresponding risk warning signals can be generated.

[0055] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0056] This invention significantly improves the accuracy and efficiency of stress loss detection in anchoring systems through multi-source data fusion, edge-cloud collaborative computing, and a dynamic compensation model for environmental interference. The edge computing unit is responsible for real-time preprocessing, the cloud server focuses on complex model optimization, and the data decision terminal quickly generates management signals, thereby achieving end-to-end collaborative optimization and improving response speed.

[0057] Among them, the strain parameters and environmental parameters of the anchoring system are monitored synchronously by a multi-source acquisition unit, and the fiber optic strain sensor and electromagnetic induction sensor are compared and verified to achieve data complementarity and improve reliability.

[0058] By performing wavelet denoising in real time through edge computing units and combining it with environmental parameter analysis, the amount of data transmitted is reduced and the processing speed is improved. Then, by performing multi-dimensional analysis through the time domain characteristics, frequency domain characteristics and strain-salt spray coupling characteristics, the corrosion anomaly risk prediction of stress loss is realized.

[0059] By obtaining correlation functions from cloud servers to quantify the synergistic effect of salt spray and temperature and humidity, and combining them with dynamic adjustment coefficients to achieve dynamic compensation for environmental interference, the measurement accuracy is significantly improved. Furthermore, risk classification and early warning are achieved through corrosion anomaly prediction models, thereby reducing the missed and false alarm rates of monitoring. Attached Figure Description

[0060] Fig. 1 A connection diagram of the system modules of the present invention is shown;

[0061] Fig. 2 A schematic diagram illustrating the steps of the workflow of the present invention is shown. Detailed Implementation

[0062] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0063] Example 1:

[0064] like Figs. 1-2 As shown, a stress loss detection system for an anchoring system includes a multi-source acquisition unit, an edge computing unit, a cloud server, and a data decision terminal, with communication connections between the multi-source acquisition unit, the edge computing unit, the cloud server, and the data decision terminal;

[0065] The specific work steps are as follows:

[0066] S1: The multi-source acquisition unit includes a stress detection module and an environmental detection module, which respectively detect the stress loss of the anchoring system and the environmental conditions therein, and acquire multi-source sensing data.

[0067] Multi-source sensor data includes strain parameters and environmental parameters;

[0068] Strain parameters include the measured strain value Fcl and the reference strain value Fck; environmental parameters include the salt spray concentration value Cyw, the ambient temperature value Whj, and the ambient humidity value Hhj.

[0069] The stress detection module is used to detect stress loss in the anchoring system. The stress detection module includes a fiber optic strain sensor and an electromagnetic induction sensor. The fiber optic strain sensor acquires the measured strain value Fcl, and the electromagnetic induction sensor acquires the reference strain value Fck.

[0070] The environmental monitoring module is used to monitor the environmental conditions of the anchoring system. The environmental monitoring module includes an electrochemical salt spray concentration sensor and an integrated temperature and humidity sensor. The electrochemical salt spray concentration sensor collects the salt spray concentration value Cyw, and the integrated temperature and humidity sensor collects the ambient temperature value Whj and the ambient humidity value Hhj.

[0071] By acquiring and fusing multi-source data, combining fiber optic strain sensors (high precision) and electromagnetic induction sensors (anti-interference), strain parameters (measured value Fcl, reference value Fck) and environmental parameters (salt spray Cyw, temperature Whj, humidity Hhj) are acquired simultaneously, achieving data complementarity and improving reliability.

[0072] S2, the edge computing unit performs preliminary analysis of multi-source sensor data: the standard strain value of the anchoring system is obtained by wavelet denoising and in-depth analysis is performed to obtain the time domain characteristics, frequency domain characteristics and environmental coupling characteristics of the standard strain value, and to evaluate the corrosion anomaly of the anchoring system.

[0073] S2-1, The specific process of preliminary analysis of multi-source sensor data is as follows:

[0074] Extract multi-source sensor data corresponding to N time points t;

[0075] The measured strain value Fcl was decomposed using Discrete Wavelet Transform (DWT).

[0076] ;

[0077] in, For low-frequency scaling coefficients, For low-frequency wavelet basis functions, These are high-frequency wavelet coefficients. Here, J is the high-frequency wavelet basis function, J is the wavelet decomposition level, and k is the index of the data index, where 0 < k ≤ N.

[0078] Extracting wavelet basis functions The corresponding signal data is labeled with the denoised strain value Fqz. 0 :

[0079] ;

[0080] By normalizing environmental parameters and calculating their standard deviations, the fluctuation coefficients of various environmental parameters are obtained.

[0081] Label any environmental parameter as y, and normalize y to obtain standard data for y. : ;in, and These are the maximum and minimum values ​​of the index y, respectively;

[0082] Then, the standard data of indicator y Calculate the standard deviation to obtain the volatility coefficient of the indicator y. :

[0083] ,in, N standard data for indicator y The average value, i.e. ;

[0084] The standard deviations of the salt spray concentration value (Cyw), ambient temperature value (Whj), and ambient humidity value (Hhj) were calculated after normalization and then labeled as salt spray fluctuation coefficients. Temperature fluctuation coefficient Humidity fluctuation coefficient ,but ;

[0085] Then, the dynamic adjustment coefficient is obtained by weighted summation. And obtain the wavelet threshold by combining the number of data points N. :

[0086] ;

[0087] Where m is the index of any environmental parameter, and M is the number of environmental parameters. The fluctuation coefficient of any environmental parameter. Fluctuation coefficient The weighting factors are preset and obtained after verification with a large amount of data.

[0088] Using wavelet thresholding For wavelet coefficients The signal is processed to reconstruct the denoised strain signal and labeled as the standard strain value Fqz. 1 : .

[0089] S2-2, In-depth analysis of standard strain value Fqz 1 The specific process is as follows:

[0090] By calculating N standard strain values ​​Fqz 1 The standard strain value Fqz is obtained by taking the absolute value of the difference between data points at adjacent time points and then averaging them. 1 The time-domain features Sf: ;

[0091] Among them, the higher the time domain characteristic Sf, the more drastic the change of strain in the time domain, thus reflecting the average change intensity of strain in a short period of time. In the early stage of corrosion, it often manifests as a sudden change in local strain, such as the micro-area stress concentration caused by pitting corrosion.

[0092] Then, by setting the frequency domain interval [f1, f2] and integrating the square of the Fourier transform amplitude of the standard strain value, the standard strain value Fqz is obtained. 1 Frequency domain characteristics Pf: ;

[0093] Where FT is the Fourier transform, which transforms the standard strain value Fqz. 1 The signal is converted from the time domain to the frequency domain. The amplitude is obtained after Fourier transform; the higher the frequency domain characteristic Pf, the higher the energy distribution of strain in the frequency domain interval [f1, f2], thus quantifying the high-frequency vibration energy caused by corrosion, such as crack friction, corrosion product detachment, etc.

[0094] By standard strain value Fqz 1 Combined with salt fog concentration value Cyw, standard strain value Fqz is obtained 1 The environmental coupling feature of the standard strain value Fqz is obtained and is marked as strain salt fog coupling feature Dfc:

[0095] Wherein, the partial derivative of strain to salt fog concentration is obtained by multiple regression fitting The rate of change of salt fog concentration over time is analyzed by The synergistic effect between the environment and the strain, such as additional stress caused by salt crystallization, is represented by the strain salt fog coupling feature Dfc, and the higher the strain salt fog coupling feature Dfc, the higher the degree of influence of the strain on the salt fog concentration.

[0096] S2-3, the specific process of evaluating the corrosion anomaly of stress loss is as follows:

[0097] A corrosion anomaly prediction model is established, and the stress corrosion anomaly index Zff is obtained by combining the time domain feature Sf, the frequency domain feature Pf and the strain salt fog coupling feature Dfc of the standard strain value Fqz 1

[0098]

[0099] , , The weight coefficients of the time domain feature Sf, the frequency domain feature Pf and the strain salt fog coupling feature Dfc, respectively, and the weight coefficients are obtained by pre-setting the salt fog test, and the higher the stress corrosion anomaly index Zff, the more serious the risk of evaluating the corrosion anomaly of stress loss.

[0100] By wavelet denoising (DWT), high-frequency noise is dynamically removed, combined with environmental parameter normalization and fluctuation coefficient analysis, standard strain value is extracted, data transmission quantity is reduced and processing speed is improved; through multi-dimensional analysis of time domain feature, frequency domain feature and strain salt fog coupling feature, the risk of early corrosion anomaly of stress loss, such as pitting, crack friction, etc. can be predicted;

[0101] S3, the cloud server builds an environmental interference compensation model of anchor system strain: by analyzing the correlation function between salt fog concentration and temperature and humidity, an environmental interference compensation function between strain and salt fog is established, the standard strain value is compensated and corrected, and a compensated strain value is generated;

[0102] S3-1, the specific process of analyzing the correlation function between salt fog concentration and temperature and humidity is as follows:

[0103] ​​​​S3-101, change temperature T under the condition of constant humidity and constant salt mist, measure corrosion rate r, fit Arrhenius equation: , obtain corrosion reaction activation energy Ea:

[0104] Wherein, R0 is a general gas constant and R0 takes 8.314, A is a frequency factor, with as the abscissa, with as the ordinate, draw a straight line L by least square fitting, then the intercept of the straight line L is , the slope of the straight line L is , the frequency factor A is obtained by the intercept , and the corrosion reaction activation energy Ea is obtained by the slope ;

[0105] By combining salt mist concentration Cyw, environmental temperature value Whj and corrosion reaction activation energy Ea, the correlation function g1 between salt mist concentration Cyw and environmental temperature value Whj is established:

[0106] ; wherein, e is a natural constant, e takes 2.7; T0 is a thermodynamic temperature, T0 takes 273.16;

[0107] S3-102, change salt mist concentration Cyw under the experimental condition of constant temperature and humidity, measure corrosion current density I, and obtain salt mist effect saturation coefficient : ;

[0108] By combining salt mist concentration Cyw, environmental humidity value Hhj and salt mist effect saturation coefficient , the correlation function g2 between salt mist concentration Cyw and environmental humidity value Hhj is established: ;

[0109] Wherein, ;

[0110] S3-103, by combining temperature Laplace operator and salt mist concentration Cyw, the correlation function g3 of salt mist concentration Cyw in temperature-induced salt deposition is established: ;

[0111] Wherein, the temperature Laplace operator is calculated by finite difference method:

[0112] ;

[0113] represents the space grid <x-y>The temperature value at the middle coordinate (p, q), Refers to the spatial grid spacing in the x direction, and the spatial grid spacing in the x direction is the same as that in the y direction.

[0114] S3-2, the environmental parameters collected by the current time node are substituted into the correlation functions g1, g2 and g3, the influence degree of the salt fog concentration on the environmental factors is evaluated, and the salt fog concentration Cyw is combined with the temperature-induced salt deposition effect , the standard strain value Fqz 1 is compensated and corrected to obtain the compensated strain value Fbc:

[0115] ;

[0116] Wherein, The salt fog dynamic effect coefficient, The sensitive coefficient of any correlation function gr, and the sensitive coefficients corresponding to the three correlation functions are integrated into the environmental sensitive coefficient matrix R;

[0117] Then the formula is expanded as:

[0118] ;

[0119] S3-3, the environmental sensitive coefficient matrix R and the salt fog dynamic effect coefficient are obtained by minimizing the square sum of the difference between the reference strain value Fck and the compensated strain value Fbc. .

[0120] S4, the data decision terminal evaluates the stress risk degree of the anchoring system and generates a risk prompt signal for management;

[0121] By setting the risk interval of the compensated strain value Fbc and comparing the intervals, the stress risk degree of the anchoring system is evaluated, and the corresponding risk prompt signal is generated.

[0122] In summary, the present application significantly improves the accuracy and efficiency of the stress loss detection of the anchoring system through multi-source data fusion, edge-cloud collaborative calculation and dynamic compensation model. The edge computing unit is responsible for real-time preprocessing, the cloud server focuses on complex model optimization, the data decision terminal quickly generates management signals, realizes full-link collaborative optimization and improves response speed;

[0123] Wherein, the strain parameters and environmental parameters of the anchoring system are synchronously monitored by the multi-source acquisition unit, and the optical fiber strain sensor and the electromagnetic induction sensor are compared and verified to realize data complementation and improve reliability;

[0124] Wavelet denoising is processed in real time through edge computing, and environmental parameters are analyzed to reduce data transmission and improve processing speed. Multi-dimensional analysis is performed through strain time domain features, frequency domain features and strain salt fog coupling features to realize corrosion abnormal risk prediction of stress loss.

[0125] The cloud server is used to obtain the correlation function to quantify the synergistic effect of salt fog and temperature and humidity, and dynamic adjustment coefficients are combined to realize dynamic compensation of environmental interference, thereby significantly improving measurement accuracy. A corrosion anomaly prediction model is used to realize risk grading early warning, thereby reducing the false negative rate and false positive rate of monitoring.

[0126] The size of the interval and threshold is set for easy comparison. The size of the threshold depends on the number of sample data and the base number set by the person skilled in the art for each group of sample data. As long as the proportional relationship between the parameters and the quantized values is not affected, it is acceptable.

[0127] The above formulas are dimensionless values, and the formulas are obtained by software simulation of a large amount of data to obtain a formula of the most recent real situation. The preset parameters in the formula are set by the person skilled in the art according to the actual situation.

[0128] The above description is only a preferred embodiment of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can make equivalent replacements or changes within the technical scope disclosed in the present application according to the technical solution and inventive concept of the present application, which should be covered within the protection scope of the present application.

Claims

1. A stress loss detection system for an anchoring system, characterized in that: It includes a multi-source acquisition unit, an edge computing unit, a cloud server, and a data decision terminal; The multi-source acquisition unit includes a stress detection module and an environmental detection module, which respectively detect the stress loss of the anchoring system and the environmental conditions therein, and acquire multi-source sensor data. The edge computing unit is used for preliminary analysis of multi-source sensor data: the standard strain value of the anchoring system is obtained by wavelet denoising and in-depth analysis is performed to obtain the time domain characteristics, frequency domain characteristics and environmental coupling characteristics of the standard strain value, and to evaluate the corrosion anomaly of the anchoring system. The cloud server is used to build an environmental disturbance compensation model for the strain of the anchoring system: by analyzing the correlation function between salt spray concentration and temperature and humidity, an environmental disturbance compensation function between strain and salt spray is established, the standard strain value is compensated and corrected, and the compensated strain value is generated. By combining salt spray concentration, ambient temperature, and corrosion reaction activation energy using the Arrhenius equation, a correlation function g1 between salt spray concentration and ambient temperature is established. By combining salt spray concentration, ambient humidity value and salt spray effect saturation coefficient through salt spray effect experiments, a correlation function g2 between salt spray concentration and ambient humidity value was established. By combining the temperature Laplacian operator with salt spray concentration, a correlation function g3 for salt spray concentration in temperature-induced salt deposition is established. By substituting the environmental parameters collected at the current time point into the correlation functions g1, g2, and g3, the degree of influence of environmental factors on the salt spray concentration is assessed, thereby compensating for environmental disturbances on the standard strain value and obtaining the compensated strain value Fbc. Under constant humidity and constant salt spray conditions, the temperature T was changed, the corrosion rate r was measured, the Arrhenius equation was fitted, and the activation energy Ea of the corrosion reaction was obtained. By combining the salt spray concentration Cyw, the ambient temperature Whj, and the activation energy Ea of the corrosion reaction, the correlation function g1 between the salt spray concentration Cyw and the ambient temperature Whj was established. By varying the salt spray concentration Cyw under constant temperature and humidity experimental conditions, and measuring the corrosion current density I, the saturation coefficient of the salt spray effect was obtained. The salt spray concentration Cyw, ambient humidity Hhj, and salt spray effect saturation coefficient were used to determine the relationship between these parameters. By combining these factors, a correlation function g2 is established between the salt spray concentration Cyw and the ambient humidity value Hhj. Calculation of the temperature Laplace operator using the finite difference method Through the temperature Laplace operator By combining the salt spray concentration Cyw, a correlation function g3 was established for the effect of salt spray concentration Cyw on temperature-induced salt deposition. The degree to which environmental factors affect salt spray concentration is assessed using correlation functions g1, g2, and g3. By substituting the environmental parameters collected at the current time point into the correlation functions g1, g2, and g3, corresponding sensitivity coefficients are assigned to the correlation functions in turn. The salt spray concentration Cyw is then compared with the coefficients of temperature-induced salt deposition and the dynamic effect of salt spray. Combined, the standard strain value Fqz 1 Environmental disturbance compensation and correction are performed to obtain the compensation strain value Fbc; The sensitivity coefficients corresponding to the correlation functions g1, g2, and g3 are integrated into an environmental sensitivity coefficient matrix R. The environmental sensitivity coefficient matrix R and the salt spray dynamic effect coefficients are obtained by minimizing the sum of squares of the differences between the reference strain value Fck and the compensated strain value Fbc. ; By setting risk ranges for the compensation strain value Fbc and comparing these ranges, the stress risk level of the anchoring system can be assessed, and corresponding risk warning signals can be generated. The data decision terminal is used to assess the stress risk level of the anchoring system and generate risk warning signals for management.

2. The stress loss detection system for an anchoring system according to claim 1, characterized in that: The specific processing steps of the edge computing unit are as follows: The measured strain value is decomposed into a standard strain value by discrete wavelet transform (DWT). The temporal characteristics Sf of the standard strain values ​​are obtained by calculating the changes of N standard strain values ​​at adjacent time nodes. Then, by setting the frequency domain interval and performing a Fourier transform on the standard strain value, the frequency domain characteristic Pf of the standard strain value is obtained; By combining standard strain values ​​with salt spray concentration values, the environmental coupling characteristics of standard strain values ​​are obtained and labeled as strain-salt spray coupling characteristics Dfc. A corrosion anomaly prediction model was established, and the corrosion anomaly of the anchoring system was evaluated by combining the time domain feature Sf, the frequency domain feature Pf, and the strain-salt spray coupling feature Dfc of the standard strain value.

3. The stress loss detection system for an anchoring system according to claim 1, characterized in that: The process of acquiring multi-source sensor data is as follows: Multi-source sensor data includes strain parameters and environmental parameters; Strain parameters include the measured strain value Fcl and the reference strain value Fck; environmental parameters include the salt spray concentration value Cyw, the ambient temperature value Whj, and the ambient humidity value Hhj. The stress detection module is used to detect stress loss in the anchoring system. The stress detection module includes a fiber optic strain sensor and an electromagnetic induction sensor. The fiber optic strain sensor acquires the measured strain value Fcl, and the electromagnetic induction sensor acquires the reference strain value Fck. The environmental monitoring module is used to monitor the environmental conditions of the anchoring system. The environmental monitoring module includes an electrochemical salt spray concentration sensor and an integrated temperature and humidity sensor. The electrochemical salt spray concentration sensor collects the salt spray concentration value Cyw, and the integrated temperature and humidity sensor collects the ambient temperature value Whj and the ambient humidity value Hhj.

4. The stress loss detection system for an anchoring system according to claim 2, characterized in that: The preliminary analysis of multi-source sensor data is as follows: Extract multi-source sensor data corresponding to N time points t; The measured strain value Fcl was decomposed using Discrete Wavelet Transform (DWT). ; Among them, the mark For low-frequency scaling coefficients, For low-frequency wavelet basis functions, These are high-frequency wavelet coefficients. Here, J is the high-frequency wavelet basis function, J is the wavelet decomposition level, and k is the index of the data index, where 0 < k ≤ N. Extracting wavelet basis functions The corresponding signal data is labeled with the denoised strain value Fqz. 0 ; By normalizing environmental parameters and calculating their standard deviations, the fluctuation coefficients of various environmental parameters are obtained, and then a weighted summation is used to obtain the dynamic adjustment coefficient. And obtain the wavelet threshold by combining the number of data points N. ; Using wavelet thresholding For wavelet coefficients The signal is processed to reconstruct the denoised strain signal and labeled as the standard strain value Fqz. 1 .

5. The stress loss detection system for an anchoring system according to claim 4, characterized in that: The fluctuation coefficients of various environmental parameters are obtained by normalizing the environmental parameters and calculating their standard deviations. The specific process is as follows: Label any environmental parameter as y, and normalize y to obtain standard data for y. ; Then, the standard data of indicator y Calculate the standard deviation to obtain the volatility coefficient of the indicator y. ; The standard deviations of the salt spray concentration value (Cyw), ambient temperature value (Whj), and ambient humidity value (Hhj) were calculated after normalization and then labeled as salt spray fluctuation coefficients. Temperature fluctuation coefficient Humidity fluctuation coefficient .

6. The stress loss detection system for an anchoring system according to claim 5, characterized in that: In-depth analysis of standard strain value Fqz 1 The specific process is as follows: By calculating N standard strain values ​​Fqz 1 The standard strain value Fqz is obtained by taking the absolute value of the difference between data points at adjacent time points and then averaging them. 1 The time-domain features Sf; Then, by setting the frequency domain interval [f1, f2] and integrating the square of the Fourier transform amplitude of the standard strain value, the standard strain value Fqz is obtained. 1 The frequency domain characteristics Pf; The partial derivative of strain with respect to salt spray concentration was obtained by multivariate regression fitting, and the rate of change of salt spray concentration over time was analyzed. Then, the standard strain value Fqz was used as the basis for the analysis. 1 Combined with the salt spray concentration value Cyw, the standard strain value Fqz is obtained. 1 The environmental coupling characteristics are identified and denoted as strain salt spray coupling characteristics Dfc.

7. The stress loss detection system for an anchoring system according to claim 6, characterized in that: The specific process for assessing corrosion anomalies caused by stress loss is as follows: A corrosion anomaly prediction model was established. By assigning corresponding weight coefficients to the time-domain feature Sf, the frequency-domain feature Pf, and the strain-salt spray coupling feature Dfc, the standard strain value Fqz was... 1 By combining the time-domain feature Sf, the frequency-domain feature Pf, and the strain-salt spray coupling feature Dfc, the stress corrosion anomaly index Zff is obtained to assess the corrosion anomaly risk of stress loss.

Images