A method, device, equipment and medium for compensating stress loss of anchor cables on steep slopes

Abstract

This invention provides a method, apparatus, equipment, and medium for compensating stress loss in anchor cables on steep slopes, relating to the field of stress loss compensation technology. The method includes acquiring multi-source sensor data of a steep slope; performing full-segment stress calculation on the multi-source sensor data to obtain full-segment stress distribution data; performing effective stress calculation and correction to obtain the effective prestress value at the anchor end and determining the total stress loss data at the anchor end; performing time-by-time full-segment stress calculation and time-by-time effective stress calculation on the multi-source sensor data of the steep slope to generate a time-by-time stress dataset; based on the multi-source sensor data of the steep slope, the linear total stress loss superposition model, and the time-by-time stress dataset, decoupling and separating the causes of total stress loss at the anchor end to obtain the loss causes; and based on the full-segment stress distribution data and the total stress loss data at the anchor end, making a compensation determination and performing stress compensation. This invention can effectively compensate for stress loss in anchor cables on steep slopes.

Description

Technical Field

[0001] This invention relates to the field of stress loss compensation technology, and more specifically, to a method, apparatus, equipment, and medium for stress loss compensation of anchor cables on steep slopes. Background Technology

[0002] Steep slopes are widely distributed in various geotechnical engineering projects, including highway and railway cuttings, water conservancy hub slopes, open-pit mines, municipal high-cut slopes, and wind turbine tower foundation slopes, and are typical high-risk structures. Anchor cables, as the core active support method, effectively resist slope slippage and instability by applying prestress to constrain rock mass deformation. However, during long-term service, due to the coupled effects of multiple factors such as rock mass creep, steel strand stress relaxation, temperature fluctuations, blasting vibrations, and anchorage slippage, the prestress of the anchor cables will continuously decrease, leading to a reduction in the support bearing capacity, increased slope deformation, and even inducing landslide disasters.

[0003] To meet the needs of safety management and long-term operation and maintenance of high and steep slopes throughout their entire life cycle, and to ensure that the support system maintains the design stress level and guarantees overall stability and service life, it is necessary to use dynamic compensation to offset the prestress loss caused by factors such as rock creep and steel strand relaxation, thereby achieving real-time correction and long-term guarantee of slope support performance.

[0004] Among related technologies, the stress loss compensation technology for prestressed anchor cables on steep slopes still faces significant bottlenecks: In terms of monitoring dimensions, traditional anchor head single-point force gauges cannot reveal the axial stress distribution characteristics of the anchor cable from the free section to the anchoring section, making it difficult to accurately locate the loss points and predict the evolution trend; in terms of operation mode, it relies on manual high-altitude tensioning, which requires the cooperation of heavy equipment, resulting in drawbacks such as high operation risk, low implementation efficiency, and poor control accuracy. Moreover, the compensation amount is mostly determined by experience, which easily leads to under-compensation or over-compensation; in terms of control mechanism, existing technologies are mostly passive compensation after the fact, lacking real-time perception and dynamic response capabilities, and are difficult to adapt to continuous disturbances caused by rock creep, temperature fluctuations, and blasting vibrations. Summary of the Invention

[0005] The present invention aims to solve at least one of the above-mentioned problems.

[0006] To address the aforementioned problems, this invention provides a method, apparatus, equipment, and medium for compensating stress loss in anchor cables on steep slopes.

[0007] In a first aspect, the present invention provides a method for compensating for stress loss in anchor cables on steep slopes, comprising: Multi-source sensing data of steep slopes is acquired through a pre-set multi-source sensing device. The distributed strain data in the multi-source sensing data of steep slopes is preprocessed, validated in advance, and corrected in sequence to obtain corrected distributed strain data. Based on the strain-stress conversion model, the stress of the corrected distributed strain data is calculated for the whole section to obtain the stress distribution data of the whole section. The parameters of the strain-stress conversion model are determined by laboratory calibration data and field tension calibration data. Based on the binary linear temperature correction model, the real-time data of the anchor pressure sensor in the multi-source sensing data of the high and steep slope are effectively calculated and corrected to obtain the effective prestress value of the anchor end. Based on the effective prestress value of the anchor end and the prestress reference value, the total stress loss data of the anchor end is determined. The time-series data in the multi-source sensing data of the steep slope are subjected to time-by-time full-segment stress calculation and time-by-time effective stress calculation respectively to generate time-by-time stress dataset, and damage analysis is performed to generate damage analysis surface plot. Based on the multi-source sensor data of the steep slope, the linear total stress loss superposition model, and the key feature parameters in the time-by-time stress data, the causes of loss are decoupled and separated from the total stress loss data at the anchor end to obtain the causes of loss. Based on the stress distribution data of the entire section and the total stress loss data of the anchor end, a compensation determination is made according to the preset three-level stress compensation threshold, a compensation level is generated, and stress compensation is performed based on the compensation level and the loss cause.

[0008] Optionally, the distributed strain data in the multi-source sensing data of the steep slope is sequentially preprocessed, validated, and corrected to obtain corrected distributed strain data. Based on the strain-stress conversion model, the corrected distributed strain data is used to calculate the stress distribution data of the entire slope, including: The distributed strain data is sequentially spatiotemporally aligned, subjected to wavelet denoising, moving average filtering, and normalization to obtain standard distributed strain data. Based on the validity pre-verification mechanism, the standard distributed strain data is verified and corrected to obtain pre-verified distributed strain data. The validity pre-verification mechanism includes integrity verification, outlier secondary verification, and boundary consistency verification. Based on the dual correction method, temperature strain interference terms are separated and corrected on the pre-verification distributed strain data to obtain dual-corrected distributed strain data. The dual correction method includes a distributed temperature self-compensation method and an ambient temperature calibration method. Based on the double-corrected distributed strain data, the strain-stress conversion model is used to calculate the stress distribution across the entire segment, and preliminary stress distribution data across the entire segment is obtained. Based on the anchor pressure sensor data and anchor cable displacement sensor data in the multi-source sensing data of the steep slope, the preliminary full-section stress distribution data is verified and corrected using a dual-data source cross-validation method to obtain the full-section stress distribution data. The full-section stress distribution data is then smoothed and visualized to generate stress spatial distribution curves, which are stored in a local database and a cloud platform, respectively.

[0009] Optionally, the effective stress calculation and correction of the real-time data from the anchor pressure sensor in the multi-source sensing data of the steep slope based on the binary linear temperature correction model is performed to obtain the effective prestress value at the anchor end. Based on the effective prestress value at the anchor end and the prestress reference value, the total stress loss data at the anchor end is determined, including: The real-time data of the anchor pressure sensor is preprocessed to obtain processed real-time data of the anchor pressure sensor. The preprocessing includes outlier removal, moving average filtering, and data validity determination. Using the binary linear temperature correction model, the effective stress calculation, sensitivity correction and zero-point temperature drift correction are performed on the real-time data of the processed anchor pressure sensor to obtain the initial effective prestress value at the anchor end. The effective prestress value at the anchor end is obtained by performing a double cross-validation based on the stress distribution data of the entire section and the anchor cable axial displacement sensor data in the multi-source sensing data of the steep slope. The effective prestress value at the anchor end is compared with the prestress reference value to determine the total stress loss data at the anchor end.

[0010] Optionally, the step of performing time-by-time full-segment stress calculation and time-by-time effective stress calculation on the time-series data in the multi-source sensing data of the steep slope to generate a time-by-time stress dataset, and performing damage analysis to generate a damage analysis surface plot, includes: The timing data is subjected to global clock synchronization and sampling frequency normalization to obtain unified reference timing data; Based on a three-level detection and repair mechanism, the unified benchmark time series data is repaired to obtain repaired time series data. The three-level detection and repair mechanism includes a point anomaly detection method, a segment anomaly detection method, and a data repair method. The repaired time series data is subjected to wavelet threshold denoising and moving average filtering to obtain clean time series data; For each time point in the cleaning time series data, perform time-by-time full-segment stress calculation and time-by-time effective stress calculation to generate the time-by-time stress dataset, and perform damage analysis to generate the damage analysis surface plot.

[0011] Optionally, after performing stress compensation based on the compensation level and the loss cause, the method further includes: Obtain data on the spatial distribution of stress, average effective prestress, and stress uniformity of the entire section after stress compensation, and construct a stress compensation verification dataset. Acquire data on deep deformation of the slope rock mass, slope displacement and axial displacement of anchor cables after stress compensation, determine the rock mass deformation response and anchor cable structure displacement data corresponding to the compensation action, and construct a structural safety verification dataset. Obtain the actual tension stroke, final locking position, self-locking state parameters and action response timing data of the stress-compensated self-locking anchor, and construct an actuator verification dataset; The stress compensation verification dataset, the structural safety verification dataset, and the actuator verification dataset are compared with their respective safety verification thresholds to obtain the verification results.

[0012] Optionally, before acquiring multi-source sensing data of steep slopes through a preset multi-source sensing device, and sequentially performing preprocessing, validity pre-verification, and correction to obtain multi-source sensing data of steep slopes, the method further includes: Based on slope geological survey data and rock mechanics test data, the parameters of the anchor cable support foundation were determined. Based on the anchor cable support foundation parameters, the sensor and anchorage design scheme and the preset three-level stress compensation threshold are determined.

[0013] Optionally, after determining the sensing and anchorage design scheme and the preset three-level stress compensation threshold based on the anchor cable support foundation parameters, the method further includes: Based on the aforementioned sensing and anchor design scheme, steel strand cutting and anchor cable bundle fabrication are carried out, and distributed optical fiber sensing units are deployed along the cable body to implement anchor cable hole forming, hole cleaning, lowering and bottom grouting operations. After the grouting body has been cured to the required standard, the anchor plate and magnetorheological intelligent self-locking anchor are installed, and the pre-calibrated integrated anchor pressure sensor and displacement sensor are installed on the exposed steel strand of the anchor cable in the design sequence.

[0014] Secondly, the present invention provides a stress loss compensation device for anchor cables on steep slopes, comprising: The full-segment stress module is used to acquire multi-source sensing data of steep slopes through preset multi-source sensing devices, and to sequentially preprocess, verify the validity of and correct the distributed strain data in the multi-source sensing data of steep slopes to obtain corrected distributed strain data. Based on the strain-stress conversion model, the module performs full-segment stress calculation on the corrected distributed strain data to obtain full-segment stress distribution data. The parameters of the strain-stress conversion model are determined by laboratory calibration data and field tension calibration data. The effective stress module is used to calculate and correct the effective stress of the real-time data of the anchor pressure sensor in the multi-source sensing data of the high and steep slope based on the binary linear temperature correction model, so as to obtain the effective prestress value of the anchor end, and determine the total stress loss data of the anchor end based on the effective prestress value of the anchor end and the prestress reference value. The time-series module is used to perform time-by-time full-segment stress calculation and time-by-time effective stress calculation on the time-series data in the multi-source sensing data of the steep slope, generate time-by-time stress dataset, perform damage analysis, and generate damage analysis surface plot. The cause decoupling module is used to decouple and separate the total stress loss data at the anchor end based on the multi-source sensing data of the steep slope, the linear total stress loss superposition model, and the key feature parameters in the time-by-time stress data set to obtain the cause of loss. The compensation module is used to determine the compensation level based on the stress distribution data of the entire section and the total stress loss data of the anchor end, according to the preset three-level stress compensation threshold, generate the compensation level, and perform stress compensation based on the compensation level and the loss cause.

[0015] Thirdly, the present invention provides an electronic device, including a memory and a processor; The memory is used to store computer programs; The processor is configured to implement the method for compensating for stress loss of anchor cables on steep slopes as described in the first aspect when executing the computer program.

[0016] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for compensating for stress loss of anchor cables on steep slopes as described in the first aspect.

[0017] The beneficial effects of the stress loss compensation method, device, equipment, and medium for anchor cables on steep slopes of the present invention are: By acquiring multi-source sensing data of steep slopes using pre-set multi-source sensing devices and sequentially preprocessing, validating, and correcting the distributed strain data, abnormal data and fundamental interference factors can be effectively eliminated, resulting in reliable distributed strain data. This provides a high-quality data foundation for subsequent full-process stress calculation and loss analysis, avoiding invalid data from interfering with subsequent judgment results. Based on this, a strain-stress conversion model with parameters determined by laboratory calibration data and on-site tension calibration data is used to perform full-section stress calculation. This not only obtains stress distribution data for the entire anchor cable section but also overcomes the coverage blind spots of traditional single-point anchor head monitoring. The system locates stress loss sections and ensures accurate strain-to-stress conversion through dual calibration parameters, reducing calculation deviations caused by differences in site environment and deployment. Simultaneously, a binary linear temperature correction model is used to effectively calculate and correct real-time data from anchor pressure sensors, eliminating temperature interference with anchor end pressure monitoring and obtaining accurate effective prestress values ​​at the anchor end. This data, combined with prestress benchmark values, quantifies the total stress loss at the anchor end and complements the distributed stress calculation results across the entire section, further improving the reliability of loss quantification. Subsequently, time-series data are processed for full-section stress analysis at each time step. The system calculates and generates time-by-time effective stress datasets, performs damage analysis to generate damage analysis surface diagrams, and enables continuous dynamic tracking of anchor cable stress state and loss development. It intuitively presents the spatiotemporal evolution of stress loss, grasps the rate and trend of loss development, and overcomes the limitations of traditional static discontinuous monitoring, providing complete temporal data for dynamic compensation. Furthermore, based on multi-source sensor data of steep slopes, a linear total stress loss superposition model, and key characteristic parameters in the time-by-time stress dataset, the system decouples and separates the causes of loss in the total stress loss data at the anchor end to identify the specific effects of different causes. The system identifies the corresponding loss components and contribution ratios, overcoming the traditional blind approach that fails to distinguish the causes of loss. This provides a basis for targeted compensation, preventing under-compensation or over-compensation during the compensation process. Ultimately, by combining the stress distribution data of the entire section with the total stress loss data at the anchor end, compensation is determined and a compensation level is generated based on a preset three-level stress compensation threshold. Stress compensation is then implemented by matching the compensation level with the loss causes. This enables precise, tiered stress loss control, allowing for appropriate compensation strategies tailored to the degree of loss. This replaces the traditional manual, reactive compensation model, improving the accuracy, timeliness, and adaptability of compensation, and effectively ensuring the stability of anchor cable support stress. Attached Figure Description

[0018] Figure 1 A schematic flowchart of the method for compensating for stress loss of anchor cables on steep slopes provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of the high and steep slope anchor cable stress loss compensation device provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0019] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the accompanying drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0020] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0021] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to"; the term "based on" means "at least partially based on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments"; and the term "optionally" means "optional embodiments". Definitions of other terms will be given in the following description. It should be noted that the concepts of "first," "second," etc., mentioned in this invention are used only to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0022] It should be noted that the terms "a" and "a plurality of" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0023] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0024] like Figure 1 As shown in the figure, an embodiment of the present invention provides a method for compensating for stress loss in anchor cables on steep slopes, comprising: Multi-source sensing data of steep slopes is acquired through a pre-set multi-source sensing device. The distributed strain data in the multi-source sensing data of steep slopes is preprocessed, validated, and corrected in sequence to obtain corrected distributed strain data. Based on the strain-stress conversion model, the stress of the corrected distributed strain data is calculated for the entire section to obtain the stress distribution data of the entire section. The parameters of the strain-stress conversion model are determined by laboratory calibration data and field tension calibration data.

[0025] Specifically, this embodiment is applied to a steep slope engineering project, employing prestressed anchor cables for active support. To address the issue that traditional single-point monitoring cannot capture the stress distribution across the entire slope, distributed fiber optic sensing units are deployed along the entire length of the anchor cables, simultaneously complemented by multi-source sensing devices to form a complete monitoring system. During operation, the pre-set multi-source sensing devices continuously collect comprehensive slope monitoring data, covering data such as distributed strain along the entire anchor cable length, anchor pressure, anchor axial displacement, ambient temperature and humidity, deep slope deformation, and slope surface displacement. For the core distributed strain data, preprocessing is first performed to remove random noise generated during acquisition, data loss during transmission, and abnormal jumps caused by hardware failures. The data format and units are standardized, converting the original non-standardized data into a reliable basic dataset. After preprocessing, the data undergoes a pre-validation check, verifying data usability from three dimensions: data integrity, outlier rationality, and physical boundary consistency. Missing data is supplemented, and out-of-limit data is corrected to prevent invalid data from entering subsequent calculation processes, ensuring data quality. After successful verification, interference correction was applied to the strain data to eliminate strain measurement errors caused by factors such as ambient temperature and transmission loss, resulting in accurate corrected distributed strain data. Subsequently, stress calculations were performed across the entire length based on a strain-stress conversion model. The key parameters of this model were determined through a dual process of "laboratory calibration + on-site tensioning calibration": first, steel strands of the same specification were cut in the laboratory, and optical fibers were fixed according to the engineering layout. Initial strain-stress conversion parameters were obtained through graded tensioning using a universal testing machine; then, during the initial tensioning stage of the anchor cable on-site, the parameters were calibrated a second time using force data from a standard tensioning device to ensure that the model parameters matched the actual on-site working conditions and to eliminate transmission errors caused by the layout process. The axial stress values ​​at each measuring point along the entire length of the anchor cable were calculated point-by-point using the conversion model, ultimately obtaining stress distribution data covering the free section, anchored section, and exposed section. This data visually presents the stress levels and distribution characteristics at different locations, providing a spatial data foundation for subsequent stress loss assessment, cause identification, and compensation control. This embodiment can achieve full-dimensional state perception of anchor cables through multi-source sensing devices. The accuracy of strain data is ensured through preprocessing, validity verification and correction. Combined with the dual-calibrated conversion model to solve the stress of the entire section, it not only eliminates the blind spots of traditional single-point monitoring, but also ensures the adaptability of the solution results to the on-site working conditions, providing reliable data support for subsequent stress loss control.

[0026] Based on the binary linear temperature correction model, the real-time data of the anchor pressure sensor in the multi-source sensing data of the high and steep slope are effectively calculated and corrected to obtain the effective prestress value of the anchor end. Based on the effective prestress value of the anchor end and the prestress reference value, the total stress loss data of the anchor end is determined.

[0027] Specifically, anchor pressure sensors are installed at the anchor head of each anchor cable to directly monitor the prestress state at the anchor end. Because the slope is in an open-air environment with significant diurnal and annual temperature differences, temperature drift significantly affects the measurement accuracy of the pressure sensors. Traditional single-zero-point correction methods have large errors; therefore, a binary linear temperature correction model is used for accurate calculation. First, real-time data from the anchor pressure sensors is acquired. Based on the binary linear temperature correction model, both the sensor's sensitivity temperature drift and zero-point temperature drift are corrected simultaneously, rather than the traditional method of only correcting the zero point. This model obtains the sensitivity temperature coefficient and zero-point temperature coefficient through laboratory high and low temperature calibration. Combined with the real-time temperature data of the sensor body, it simultaneously corrects the temperature effects in both dimensions, effectively eliminating measurement errors caused by large-scale temperature changes and obtaining the original pressure value after temperature correction. Based on this, a secondary correction is made in conjunction with the structural characteristics of the anchorage: considering the pressure loss caused by the compression deformation of the anchor plate and sensor itself, a structural compression deformation correction coefficient is introduced for adjustment; simultaneously, combining the self-locking characteristics of the magnetorheological intelligent anchorage, the influence of stress fallback during the locking process is corrected, ultimately obtaining an accurate effective prestress value at the anchor end, truly reflecting the actual pretension level of the anchor cable. After obtaining the real-time effective prestress value at the anchor end, it is compared with a preset prestress benchmark value for calculation. This benchmark value is the effective prestress value obtained after complete calibration and correction after the initial tensioning and locking of the anchor cable is completed and the slope is in a undisturbed stable state, serving as the calculation benchmark for stress loss throughout the service life. By comparing the difference between the real-time value and the benchmark value, the total stress loss data at the anchor end is calculated, including two core indicators: total stress loss amount and total stress loss rate, intuitively presenting the overall degree of prestress loss in the anchor cable, providing a direct quantitative basis for subsequent compensation level determination. This embodiment can effectively eliminate the measurement interference of large temperature difference environment in the field on pressure sensor through binary linear temperature correction model. Combined with secondary correction of structure and anchor characteristics, it can further get closer to the real working condition, greatly improve the calculation accuracy of effective prestress at anchor end. The total loss data obtained by benchmarking with the benchmark value can quickly reflect the anchor cable support status, and provide a reliable quantitative basis for subsequent compensation triggering judgment.

[0028] The time-series data in the multi-source sensing data of the steep slope are subjected to time-by-time full-segment stress calculation and time-by-time effective stress calculation respectively to generate time-by-time stress dataset, and damage analysis is performed to generate damage analysis surface diagram.

[0029] Specifically, due to differences in sampling frequency, clock reference, and transmission delay among different sensing devices, the time scale of the original time-series data cannot be aligned, making it difficult to directly use for continuous stress evolution analysis. Therefore, the first step is to unify the time reference of the multi-source time-series data: by synchronizing the global clock, the time reference of all devices is unified to the high-precision clock of the edge terminal, eliminating clock deviation; by normalizing the sampling frequency, data with different sampling rates are interpolated to a unified time step, balancing monitoring accuracy and data volume, ultimately forming a unified reference time-series data with consistent time scale and aligned multi-source data. Based on this, time-by-time full-segment stress calculation and time-by-time anchor-end effective stress calculation are performed on the monitoring data at each time point: for the distributed strain data of each time section, the stress distribution of the anchor cable at the corresponding time point is calculated; for the anchor pressure data of each time section, the effective prestress value of the anchor end at the corresponding time point is calculated. The calculation results of all time points are integrated in chronological order to generate a structured time-by-time stress dataset. This dataset simultaneously contains the full-segment stress distribution in the spatial dimension and the continuous stress change in the temporal dimension, achieving full coverage of the spatiotemporal dimensions and providing a complete data foundation for damage analysis. Long-term damage evolution analysis is conducted based on time-series stress datasets, extracting damage characteristic parameters such as stress loss rate, stress fluctuation amplitude, cumulative loss, segment loss ratio, and stress distribution non-uniformity coefficient. This analysis examines the damage development patterns in different segments and identifies inflection points and abnormal periods of damage acceleration. Simultaneously, stress data, spatial location, and time series are fused in three dimensions to generate a damage analysis surface map. This visually presents the dynamic evolution of anchor cable stress over time and space, clearly demonstrating the differences in damage development between anchored and free segments, as well as the influence of external factors such as temperature and vibration on stress. This assists maintenance personnel in quickly locating areas of concentrated damage and their development trends. This embodiment, through unified time-series benchmarks and precise time-series calculations, forms a full-cycle, comprehensive stress time-series dataset, completely reconstructing the dynamic evolution of anchor cable stress. Combined with damage feature extraction and three-dimensional surface visualization, it provides an intuitive understanding of damage development trends and spatial distribution characteristics, offering comprehensive data support for subsequent stress loss cause separation, long-term performance evaluation, and early warning.

[0030] Based on the multi-source sensing data of the steep slope, the linear total stress loss superposition model, and the key feature parameters in the time-by-time stress data, the causes of loss are decoupled and separated from the total stress loss data at the anchor end to obtain the causes of loss.

[0031] Specifically, this embodiment performs stress loss cause decoupling. First, it reads the total stress loss data at the anchor end, the stress distribution data across the entire section, and the time-series stress dataset. Combined with time-series monitoring data from deep inclinometer measurements, slip surface displacement, ambient temperature, and blasting vibration collected by multi-source slope sensors, it uses a linear total stress loss superposition model to complete the decoupling and separation of multiple causes, accurately separating four independent and coupled stress loss components. The formula for the linear total stress loss superposition model used in this invention is: ; in, Let be the total stress loss at time t. Stress loss due to rock mass deformation, including rock mass creep loss. and potential slip surface misalignment loss , Stress loss due to the deformation of the anchor cable itself, including relaxation loss of the steel strand. Bond slip loss of anchorage section , Stress loss due to temperature deformation, including stress loss due to temperature fluctuations. and permanent temperature loss ; Stress loss caused by vibration and impact, including instantaneous stress fluctuation loss. and permanent stress loss ; This is the model residual term, which includes unconsidered secondary contributing factors and measurement errors.

[0032] The overall decoupling process is executed in layers from high to low feature recognition. Periodic temperature-related losses are separated first, followed by vibration and impact abrupt change losses, then slowly and monotonically changing rock creep losses, and finally the remaining components are determined to be the deformation losses of the anchor cable itself and the bonding interface. After each round of calculation, residual iterative correction is carried out. When the residual accounts for more than 5% of the total stress loss, the residual is allocated to each inducing module according to the variance ratio of the four types of loss components. The process is iterated until the residual ratio is less than 5%.

[0033] To address the separation of rock mass deformation causes, the rock mass deformation field data collected using a deep inclinometer includes: deep inclinometer data of the slope, i.e., horizontal displacement at different depths. (z represents depth); Potential slip surface displacement gauge data, i.e., the relative displacement between the upper and lower plates of the slip surface. Slope surface displacement monitoring data, i.e., the three-dimensional displacement of each point on the slope. Regarding the creep loss of the rock mass, the improved Burgers creep model is used to fit the creep strain of the rock mass. The model formula is as follows: ,in, This represents the axial creep strain of the rock mass. The initial stress of the rock mass under the action of the anchor cable. , These are the elastic modulus and viscosity coefficient of Maxwell volume, respectively. , These are the elastic modulus and viscosity coefficient of the Kelvin mass, respectively. The separation step includes: extracting the average creep displacement of the rock mass within the anchorage section of the anchor cable from the deep inclinometer data. This is converted into axial creep strain of the rock mass: ,in, Let the anchorage length be denoted as ; substitute it into the Burgers model, and obtain the model parameters by least squares fitting. Calculate the stress loss of anchor cables caused by rock mass creep. ,in, The elastic modulus of the steel strand. The cross-sectional area of ​​the steel strand is... The length of the free section of the anchor cable. The creep-stress transfer coefficient (values ​​range from 0.8 to 0.95, determined by field tests).

[0034] Regarding potential slip surface displacement losses, abrupt change points in the slip surface displacement time series data are detected using the sliding window t-test method. ,in, The test statistic is used to determine whether there is a significant difference between two sets of data within a window. , These are the average displacements of the first and last sub-samples divided by the sliding window, respectively. , These represent the number of data points contained in the first and second subsamples, respectively. , These are the displacement variances of the first and second subsamples, respectively.

[0035] when At that time, it was determined to be a mutation point. The significance level was set at [value missing]. The slip surface displacement time series data were divided into multiple stages based on abrupt change points, and linear fitting was applied to each stage: ; in, Let be the relative displacement between the upper and lower plates on the potential slip surface at time t. , ,..., The moments of abrupt displacement changes identified through the t-test are used as nodes for stage division. , ,..., Let be the slope of the straight line fitted to the displacement at each stage, and represent the slip rate of the slip surface at the corresponding stage. , ,..., The intercept of the straight line fitted to the displacement at each stage is used to calculate the stress loss of the anchor cable caused by the slip surface displacement at each stage. ,in, The angle between the anchor cable and the slip surface is denoted by , and 'n' represents the number of anchor cables in a single row, determined based on the slope width and cable spacing. Total rock mass deformation loss. .

[0036] To address the separation of deformation loss within the anchor cable itself, the strain distribution data across the entire anchor cable section used includes axial strain data from distributed fiber optic monitoring. (x is the distance from the bottom of the hole), data from the anchor pressure sensor. and displacement data of exposed sections of anchor cables Regarding the relaxation loss of steel strands, based on the full-strain data of distributed optical fibers, we distinguish between uniform relaxation loss in the free section and local bond slip loss in the anchorage. Steel strand relaxation is a stress attenuation phenomenon in steel under constant strain, characterized by uniform distribution throughout the cable and long-term monotonic attenuation. The ISO 2090 standard relaxation model is adopted for steel strand relaxation. ; in, Let t be the relaxation stress loss of the steel strand at time t (MPa). The initial tensile stress of the steel strand is (MPa). Here, m represents the standard value of the tensile strength of the steel strand (MPa), m is the relaxation index (0.15~0.2 for low-relaxation steel strands), and A and B are empirical coefficients (A=0.025, B=0.18 for low-relaxation steel strands). The separation steps include: Extract the average strain of the free segment from the strain distribution data of the entire segment. ; Calculate the total stress variation in the free section Subtract the creep and temperature loss of the separated rock mass from the total stress change in the free section. The relaxation loss of the steel strand is obtained as follows: ; Converted to total relaxation loss force: ; Regarding the bond slip loss in the anchorage section, it is a phenomenon of relative displacement at the bond interface between the grout and the steel strand or rock mass. It is characterized by locality and stress redistribution (manifested as the stress peak in the anchorage section shifting inwards into the hole). The bond slip constitutive model and stress distribution inversion method are used to separate it. The separation steps are as follows: Establish a bond-slip constitutive model for the anchorage section: ; in, The bonding stress (MPa) at the interface between the steel strand and the grouting body. This refers to the relative slippage between the steel strand and the grouting body. The ultimate bond strength (MPa, measured by interfacial bonding performance test). The limit slip is measured in mm, usually 1~3 mm. The bonding stress softening coefficient (mm⁻¹, determined by experimental fitting). The natural constant is approximately 2.71828; stress distribution in the anchorage section based on distributed optical fiber monitoring. Inverse calculation of bond stress at each point Where U is the perimeter of the steel strand. Calculate the total bond slip loss of the anchorage section. The total anchor cable deformation loss is .

[0037] For the separation of temperature-induced deformation losses, the temperature data used includes ambient atmospheric temperature. Temperature inside the anchor cable hole (Simultaneous monitoring of temperature via distributed optical fiber) and the internal temperature of the rock mass Regarding the temperature fluctuation loss, the stress changes caused by temperature fluctuations exhibit obvious periodicity (daily cycle, annual cycle) and reversibility. It can be separated using wavelet transform and periodic component extraction methods. The separation steps include: Five-level db6 wavelet decomposition was performed on the ambient temperature time series data to obtain components at different scales. ,in, As a low-frequency trend component, For high-frequency detail components, extract daily periodic components. ( (24-hour cycle) and annual cycle components ( (365-day cycle) Calculate the stress change of anchor cables caused by temperature fluctuations. ,in, The coefficient of thermal expansion of the steel strand is _____. The coefficient of thermal expansion of the rock mass. This represents the temperature fluctuation (relative to the reference temperature). The temperature-stress transfer coefficient (values ​​range from 0.7 to 0.9, determined by field tests).

[0038] Permanent temperature loss is caused by irreversible damage such as bond interface damage and steel strand fatigue due to repeated temperature changes. It is cumulative and positively correlated with the amplitude of temperature fluctuations, and can be described using a cumulative damage model. ,in, The temperature damage coefficient, Let be the temperature change amplitude during the i-th temperature fluctuation cycle. Let be the duration of the i-th temperature fluctuation cycle. This represents the total number of temperature fluctuation cycles within the statistical period. Total temperature deformation loss. .

[0039] For the separation of vibration-induced losses, the vibration data used includes the triaxial vibration acceleration at the anchor cable location. Vibration velocity and displacement The occurrence time and duration of vibration events. Regarding instantaneous stress fluctuations, those caused by vibration are characterized by short duration (milliseconds), large amplitude, and rapid attenuation. They can be separated using event triggering and bandpass filtering methods. The separation steps include: Set vibration acceleration threshold ,when The event was marked as a vibration event. Stress data for 10 seconds before and after the vibration event were extracted, and the high-frequency stress fluctuation components caused by the vibration were extracted using a Butterworth bandpass filter (10~100Hz). The peak value and duration of the instantaneous stress fluctuation were calculated. ,in, This represents the maximum value of instantaneous stress fluctuation loss. This represents the average effective prestress before vibration.

[0040] Regarding permanent stress loss, the permanent loss caused by vibration is due to irreversible deformation such as damage to the anchorage bond interface and slippage of the steel strand caused by vibration, which can be described using a vibration intensity-loss regression model: ,in, Let be the amount of permanent loss caused by the i-th vibration event. The vibration damage coefficient is... The peak particle vibration velocity of the i-th vibration event, The duration (s) of the i-th vibration event. Total vibration impact loss. ,in, This represents the total number of vibration events that have occurred up to time t.

[0041] After all the contributing factors were separated, the decoupling accuracy was cross-checked using independent monitoring data. The relative error between rock mass deformation loss and loss directly calculated by displacement gauges was controlled within 8%, the error between anchor cable loss and indoor tensile test results did not exceed 10%, and the relative error between total loss and measured values ​​from sensors under the anchor was less than 5%. The final output included the percentage and real-time loss of eight sub-categories of losses: rock mass creep, slippage, steel strand relaxation, anchorage slippage, temperature fluctuation, permanent temperature damage, instantaneous vibration, and permanent vibration damage. The spatial segment and temporal variation characteristics corresponding to each type of loss were fully labeled and simultaneously pushed to the three-level compensation judgment module as the basis for formulating differentiated compensation strategies. More specifically, the residual term was obtained by subtracting the four separated contributing factor components from the total stress loss. ; If the residual is greater than 5% of the total loss, then the residual is allocated to each component according to the variance proportion of each contributing factor: ,in, Loss after allocation Loss before allocation The variance of the i-th contributing factor is used, and the iteration is repeated until the residual is less than 5% of the total loss. This embodiment can accurately decompose various prestressing loss contributing factors under multi-field coupling by relying on the hierarchical decoupling and residual iteration correction algorithm. It solves the shortcomings of traditional monitoring that can only count the total loss and cannot locate the root cause of the problem. Based on multi-dimensional measured data of slope rock mass, anchor cable, and environment, it realizes the quantification of each loss component, providing quantitative data support for gradient-based and targeted intelligent compensation, and avoiding engineering problems such as over-compensation, under-compensation, and anchor cable overload caused by indiscriminate uniform tensioning.

[0042] Based on the stress distribution data of the entire section and the total stress loss data of the anchor end, a compensation determination is made according to the preset three-level stress compensation threshold, a compensation level is generated, and stress compensation is performed based on the compensation level and the loss cause.

[0043] Specifically, in this embodiment, the loss type, full-segment stress distribution data, and anchor end total stress loss data output by the cause decoupling are received. The edge terminal's built-in intelligent compensation control program executes the three-level stress compensation judgment and automated tensioning control. The system pre-stores a complete set of three-level compensation control rules in the edge terminal. The rule library includes preset three-level stress loss thresholds, dual judgment conditions for triggering each level of compensation, tensioning step lengths corresponding to different loss causes, pressure holding time, magnetorheological anchor self-locking logic, and emergency locking rules for severe losses. The first-level compensation threshold is 5%~8% of the total stress loss rate, the second-level compensation threshold is 8%~15%, and the third-level warning threshold is when the loss rate exceeds 15% or a sudden change in stress loss occurs. The judgment process consists of two steps: initial judgment and secondary review. In the initial judgment stage, the real-time total stress loss rate at the anchor end is directly retrieved and matched with the three-level threshold range. The stress distribution curve of the entire section is read simultaneously to identify whether there are local understress or stress concentration areas in the anchorage section, and the stress loss development rate parameter is marked as an auxiliary judgment indicator. In the secondary review stage, the risk verification is completed by combining the loss cause results output by the previous steps. If the initial judgment is a level one minor loss, the review conditions are a stable loss rate, no slip surface displacement, and no damage to the anchorage bond. Only a small amount of adaptive supplementary tensioning is performed. If the initial judgment is a level two moderate loss, the review confirms that there is no risk of slope instability and that the anchor cable strands have not exceeded the stress limit. A graded pressure stabilization and supplementary tensioning operation is performed. If the level three warning threshold is reached or a sudden change in the slip surface, large-area anchorage slippage, or permanent loss due to strong vibration is detected, all automatic tensioning execution permissions are immediately locked, and only manual on-site operation permissions and emergency self-locking permissions are opened. At the same time, audible and visual warnings and SMS alarms are pushed to the PC maintenance terminal and mobile APP. Differentiated compensation processes are implemented for different loss causes. For primary reversible losses caused by simple temperature fluctuations, the system only corrects the data and does not initiate physical tensioning. For long-term losses due to rock mass creep and uniform losses due to steel strand relaxation, a three-stage loading process is adopted, with stabilization for 5 minutes after each loading stage. For moderate losses caused by local slippage in the anchorage section, the tensioning rate is reduced and the stabilization time is extended to avoid further damage to the bonding interface. For sudden severe losses due to vibration, the self-locking protection program is directly triggered, and automatic compensation is prohibited. After the tensioning operation is completed, the magnetorheological intelligent self-locking anchor is controlled to complete the reverse locking without retraction. Relying on its extremely low self-locking loss rate of less than 0.5%, it eliminates the 3%~5% retraction prestress loss of traditional wedge anchors. The system synchronously and completely records the compensation start time, tensioning load at each stage, stabilization time, final locking prestress, stress recovery data of each section, loss cause type, and execution permission status. All time-series execution logs are structured and stored in the local database and synchronized to the cloud platform, forming a complete compensation execution closed loop.This embodiment can distinguish different damage risk levels through a dual-core three-level threshold control mechanism, combine stress loss induction factors to match differentiated tensioning and stabilization processes, and use low-retraction magnetorheological intelligent anchors to achieve high-precision automatic compensation. The entire process does not require personnel to erect heavy tensioning equipment at high altitudes, which greatly reduces the safety risks of high-altitude construction on steep slopes. At the same time, through graded control, it avoids frequent micro-tensioning that damages anchor cables and prevents excessive tensioning that leads to over-limit damage to steel strands.

[0044] For example, the edge terminal is a core device for local data processing and intelligent control deployed at the top of steep slopes or at stable locations on platforms at various levels. A single unit can manage no more than 50 anchor cables and no more than 200 sensors of various types, with a data processing latency of no more than 100ms. It undertakes the responsibility of data aggregation at the front-end sensing layer, and can collect multi-source monitoring data such as distributed optical fiber, anchor pressure, displacement, deep inclinometer, temperature and humidity, and vibration. It performs local data preprocessing, full-section stress calculation, loss cause separation, and three-level compensation threshold determination, and directly controls the magnetorheological intelligent self-locking anchor to perform automated tensioning, realizing local closed-loop control of stress compensation. At the transmission level, it adopts a dual-redundancy configuration of a dedicated optical fiber main link plus a 4G / 5G wireless backup link, which can automatically switch in case of failure. Data is stored locally and simultaneously synchronized to the cloud operation and maintenance platform. The equipment is equipped with an outdoor control cabinet with IP65 protection rating, installed on a concrete foundation 0.3m above the ground. It can adapt to harsh working conditions such as high and low temperatures, humidity, and dust in the field. It is the core hub connecting the front-end sensing equipment and the back-end cloud platform, which not only ensures the real-time performance of compensation control, but also improves the operational reliability of the field system.

[0045] In this embodiment, multi-source sensing data of steep slopes is acquired through pre-set multi-source sensing devices, and pre-processing, validity verification, and correction are performed sequentially. This effectively eliminates abnormal data and basic interference factors, resulting in reliable multi-source sensing data for steep slopes. This provides a high-quality data foundation for subsequent full-process stress calculation and loss analysis, avoiding invalid data from interfering with subsequent judgment results. Based on this, a strain-stress conversion model with parameters determined by laboratory calibration data and on-site tension calibration data is used to perform full-section stress calculation. This not only acquires stress distribution data for the entire anchor cable section but also overcomes the coverage blindness of traditional single-point monitoring of anchor heads. The system locates stress loss sections and ensures accurate strain-to-stress conversion through dual calibration parameters, reducing calculation deviations caused by differences in site environment and deployment. Simultaneously, a binary linear temperature correction model is used to effectively calculate and correct real-time data from anchor pressure sensors, eliminating temperature interference with anchor end pressure monitoring and obtaining accurate effective prestress values ​​at the anchor end. This data, combined with prestress benchmark values, quantifies the total stress loss at the anchor end and complements the distributed stress calculation results across the entire section, further improving the reliability of loss quantification. Subsequently, time-series data are processed for full-section stress analysis at each time step. The system calculates and generates time-by-time effective stress datasets, performs damage analysis to generate damage analysis surface diagrams, and enables continuous dynamic tracking of anchor cable stress state and loss development. It intuitively presents the spatiotemporal evolution of stress loss, grasps the rate and trend of loss development, and overcomes the limitations of traditional static discontinuous monitoring, providing complete temporal data for dynamic compensation. Furthermore, based on multi-source sensor data of steep slopes, a linear total stress loss superposition model, and key characteristic parameters in the time-by-time stress dataset, the system decouples and separates the causes of loss in the total stress loss data at the anchor end to identify the specific effects of different causes. The system identifies the corresponding loss components and contribution ratios, overcoming the traditional blind approach that fails to distinguish the causes of loss. This provides a basis for targeted compensation, preventing under-compensation or over-compensation during the compensation process. Ultimately, by combining the stress distribution data of the entire section with the total stress loss data at the anchor end, compensation is determined and a compensation level is generated based on a preset three-level stress compensation threshold. Stress compensation is then implemented by matching the compensation level with the loss causes. This enables precise, tiered stress loss control, allowing for appropriate compensation strategies tailored to the degree of loss. This replaces the traditional manual, reactive compensation model, improving the accuracy, timeliness, and adaptability of compensation, and effectively ensuring the stability of anchor cable support stress.

[0046] Optionally, the distributed strain data in the multi-source sensing data of the steep slope is sequentially preprocessed, validated, and corrected to obtain corrected distributed strain data. Based on the strain-stress conversion model, the corrected distributed strain data is used to calculate the stress distribution data of the entire slope, including: The distributed strain data is sequentially spatiotemporally aligned, subjected to wavelet denoising, moving average filtering, and normalization to obtain standard distributed strain data. Based on the validity pre-verification mechanism, the standard distributed strain data is verified and corrected to obtain pre-verified distributed strain data. The validity pre-verification mechanism includes integrity verification, outlier secondary verification, and boundary consistency verification. Based on the dual correction method, temperature strain interference terms are separated and corrected on the pre-verification distributed strain data to obtain dual-corrected distributed strain data. The dual correction method includes a distributed temperature self-compensation method and an ambient temperature calibration method. Based on the double-corrected distributed strain data, the strain-stress conversion model is used to calculate the stress distribution across the entire segment, and preliminary stress distribution data across the entire segment is obtained. Based on the anchor pressure sensor data and anchor cable displacement sensor data in the multi-source sensing data of the steep slope, the preliminary full-section stress distribution data is verified and corrected using a dual-data source cross-validation method to obtain the full-section stress distribution data. The full-section stress distribution data is then smoothed and visualized to generate stress spatial distribution curves, which are stored in a local database and a cloud platform, respectively.

[0047] Specifically, the raw distributed strain data acquired by distributed optical fibers undergoes four preprocessing operations: spatiotemporal alignment, wavelet denoising, moving average filtering, and normalization, to obtain standard distributed strain numbers. These numbers include: a spatial dimension (coordinates of continuous measuring points along the entire length of the anchor cable); a temporal dimension (time-series data with synchronization timestamps); and a data dimension (the raw strain value at each measuring point, the corresponding ambient temperature value, and data quality identifier). After the standard distributed strain numbers are generated, three validity pre-checks are initiated. The integrity check statistically analyzes the missing length and total missing rate of the entire anchor cable data. When continuous missing data exceeds 0.5m or the total missing rate exceeds 1%, linear interpolation of adjacent measuring points is used to fill in the missing data and the interpolation points are marked. The outlier secondary check uses the 3σ criterion to remove distorted strains exceeding the elastic limit of the steel strand of 2000με, and replaces the outlier values ​​with the average of the previous and next 5 measuring points. The boundary consistency check verifies that the strain at the bottom of the hole is close to 0 and that the strain at the anchor end matches the trend of the anchor pressure data. If the boundary deviation exceeds the limit, the optical fiber sensing acquisition parameters are recalibrated. The pre-verification distributed strain data that has passed verification employs a dual correction method of distributed temperature self-compensation and environmental calibration to eliminate temperature strain interference, relying on coupled equations. The additional strain caused by the separation temperature, among which, To couple the data, For the Brillouin temperature shift coefficient, we can take... , Let be the strain coefficient, which can be taken as . , For temperature changes, The structural strain is calculated directly from the frequency shift of the anchorage section under no external load. After the grout solidifies, the steel strands are firmly bonded to the rock mass. Under no external load, the structural strain is... =0, at which point the temperature change at each measuring point can be directly calculated. The free section utilizes the measured values ​​from temperature sensors at both ends of the anchor cable as boundary conditions. Temperature distribution at each measuring point in the free section is obtained through linear interpolation, and temperature strain is calculated using the temperature coefficient. Subtracting the temperature strain from the original total strain yields the effective structural strain caused solely by the anchor cable stress. ,in, Let be the effective structural strain at time t, at a distance x from the bottom of the hole. Let x be the original total strain at time t. Let x be the temperature strain at time t. A time period where the slope is in a stable state and no anchor cable tensioning is performed is selected to verify that the corrected effective strain should approach the initial reference strain. If the fluctuation exceeds the range, the temperature coefficient and strain coefficient of the optical fiber are recalibrated. After the double correction is completed, the strain transfer coefficient k is obtained through laboratory and field dual calibration to construct a strain-stress conversion model. ,in, Let be the axial stress of the steel strand at a distance x from the bottom of the hole at time t. Let ρ be the elastic modulus of the steel strand, and k be the strain transfer coefficient. To determine the effective structural strain at point x at time t, the anchor cable is differentially calculated in three segments: the free segment (x∈[La,L-Le]), without grout bonding constraints, has uniform stress distribution in the steel strands, and the calculation result is the average of five adjacent measuring points, with random noise smoothed; the anchored segment (x∈[0,La]), where stress gradually decreases from the bottom of the hole towards the free segment, exhibiting a non-linear distribution, retains the original measuring point resolution without excessive smoothing to accurately capture stress concentration areas; and the exposed segment (x∈[L-Le,L]), which is only used as a stress transmission segment, and the calculation result is used for comparison with data from the anchor pressure sensor, but is not included in the stress loss calculation. The calculated stress values ​​are converted from MPa to the commonly used kN (single steel strand) in engineering using the following conversion formula: ,in, This refers to the cross-sectional area of ​​the steel strand; Generate a formatted stress dataset, i.e. preliminary stress distribution data for the entire section, including the spatial coordinates, timestamp, stress value, and data source identifier (measured and interpolated) for each measuring point.

[0048] After obtaining preliminary stress distribution data for the entire section, cross-validation with dual data sources is initiated. First, cross-validation with data from the anchor pressure sensor is performed to calculate the total tension force corresponding to the average stress of the free section calculated by the distributed optical fiber. ,in, The length of the free section of the anchor cable. The length of the anchorage section of the anchor cable. L is the length of the exposed section of the anchor cable, n is the total length of the anchor cable, and n is the number of steel strands in the anchor cable bundle. The total tension calculated by the optical fiber is compared with the measured value of the pressure sensor under the anchor. In comparison, when the relative error exceeds 3%, the strain transfer coefficient k and temperature correction results are re-examined, and on-site secondary calibration is performed if necessary. Then, the data is cross-validated with the anchor cable displacement sensor data. Based on Hooke's law, the theoretical elongation of the anchor cable is calculated from the solved average stress of the free segment. The calculation results were compared with the elongation measured by the anchor cable axial displacement sensor. For comparison, the relative error should be ≤4%. If the error exceeds this range, check for abnormalities such as steel strand slippage or loose anchorages. Finally, calculate the correction factor based on the cross-validation results. The stress calculation results for the entire section are corrected as follows: The corrected effective full-segment stress data is used to generate stress spatial distribution curves with 10cm intervals through cubic spline interpolation. The curves are then smoothed using a 5-point moving average to eliminate high-frequency noise and retain the overall trend and local characteristics of the stress distribution. Key information is marked on the generated stress distribution curves, including: segment divisions to clearly identify the boundaries of anchored, free, and exposed segments; a baseline stress line to draw the design prestress baseline after initial tensioning and locking, serving as a reference for stress loss comparison; key locations to mark the anchor cable positions corresponding to potential slip surfaces, joints, fissures, and weak interlayers; and abnormal areas to mark stress concentration areas (stress ≥ 1.2 times the design value), understressed areas (stress ≤ 0.8 times the design value), and areas with abnormal stress loss using different colors. A static distribution curve is then plotted to display the stress distribution across the entire section at a specific moment, used for current state assessment. A dynamic time-series curve is generated to create animations of stress distribution curves at different times, showing the stress change trend over time. A loss rate distribution curve is used to calculate the stress loss rate at each measuring point based on the benchmark stress, generating a spatial distribution curve of the loss rate to intuitively locate areas of concentrated loss. Finally, the stress distribution data across the entire section, cross-validation results, and error correction coefficients are synchronously stored in a structured format in the edge local database and uploaded to the cloud platform for real-time use in subsequent decoupling steps. This embodiment utilizes a multi-stage quality control process, including standardized preprocessing, triple validity verification, dual temperature correction, and dual data source cross-validation, to comprehensively eliminate measurement errors caused by fiber optic acquisition, ambient temperature, and equipment transmission. This significantly improves the accuracy of calculating the spatial distribution of stress across the entire anchor cable section, fully realizing accurate characterization of stress zoning in the anchored and free sections, and completely solving the monitoring blind spots caused by traditional single-point monitoring that cannot identify local understress and stress concentration.

[0049] For example, key parameters of the strain-stress transformation model include the elastic modulus of the steel strand. Cross-sectional area of ​​steel strand The strain transfer coefficient k of fiber-steel strand was determined using a dual calibration method combining laboratory calibration and on-site tension calibration. Key parameters of the strain-stress conversion model were also determined, including: Steel strands of the same specifications as those used in the project were cut and fixed to the distributed optical fibers. Graded tension tests were then conducted on a universal testing machine with a loading level of 0.2. 0.4 0.6 0.8 , To determine the ultimate tensile strength of the steel strand, strain was measured simultaneously using optical fiber. The true strain of the steel strand calculated by the displacement of the testing machine The strain transfer coefficient k is obtained through linear regression as follows: ; During the initial tensioning and locking stage of the anchor cable, the theoretical average stress of the steel strand is calculated using the tension force P measured by the anchor pressure sensor. Simultaneously, the average strain of the free segment was measured using distributed optical fiber. Calculate the measured average stress By comparing the theoretical stress with the measured stress, the strain transfer coefficient k is optimized.

[0050] Optionally, the effective stress calculation and correction of the real-time data from the anchor pressure sensor in the multi-source sensing data of the steep slope based on the binary linear temperature correction model is performed to obtain the effective prestress value at the anchor end. Based on the effective prestress value at the anchor end and the prestress reference value, the total stress loss data at the anchor end is determined, including: The real-time data of the anchor pressure sensor is preprocessed to obtain processed real-time data of the anchor pressure sensor. The preprocessing includes outlier removal, moving average filtering, and data validity determination. Using the binary linear temperature correction model, the effective stress calculation, sensitivity correction and zero-point temperature drift correction are performed on the real-time data of the processed anchor pressure sensor to obtain the initial effective prestress value at the anchor end. The effective prestress value at the anchor end is obtained by performing a double cross-validation based on the stress distribution data of the entire section and the anchor cable axial displacement sensor data in the multi-source sensing data of the steep slope. The effective prestress value at the anchor end is compared with the prestress reference value to determine the total stress loss data at the anchor end.

[0051] Specifically, the anchor pressure sensor undergoes three preprocessing operations: outlier removal, 5-point moving average filtering, and data validity assessment. Outlier removal employs a sliding window 3σ criterion to eliminate voltage spikes. The 5-point moving average filtering formula is as follows: ,in, Let be the output voltage (i.e., the smoothed value) after passing through the moving average filter at time t. For the raw sampled data at time point t+i in the original signal, when determining the validity of the data, if 10 consecutive sampling points exceed the normal operating range of the sensor, or the data fluctuation exceeds ±5%FS, it is determined that the sensor is abnormal, triggering a system warning and automatically switching to distributed fiber optic data as the main monitoring source. The preprocessed filtered voltage data is substituted into a binary linear temperature correction model to simultaneously correct the sensitivity and zero-point temperature drift parameters. The correction formula is: ; ; ; in, This is the sensitivity coefficient after temperature correction. This is the temperature-corrected zero-point offset value. The reference temperature calibrated on-site. The temperature coefficient of sensitivity. It is the zero-point temperature coefficient; After the anchor cable is tensioned, the anchor plate and the sensor itself will undergo compressive deformation, causing the pressure measured by the sensor to be slightly less than the actual effective prestress of the anchor cable. Therefore, a structural compressive deformation correction factor is introduced. The expression is modified as follows: ; in, This is the structural compression deformation correction factor, with a value ranging from 0.005 to 0.015, which is related to the anchor plate material, thickness, and sensor stiffness. The magnetorheological intelligent anchor used has a stress-free self-locking characteristic, and adjusts the correction term for the shrinkage loss of traditional anchors. For traditional wedge anchors: ,in The anchorage retraction loss rate is typically taken as 0.03~0.05; for this magnetorheological smart anchorage: ,in The self-locking loss rate of the magnetorheological anchor is ≤0.005, which can be approximated as 0.

[0052] After the above multi-factor correction, the initial effective prestress at the anchor end is obtained. ,in, Let t be the overall effective prestress at the anchor end of the anchor cable. This is the sensitivity coefficient after temperature correction. This is the filtered sensor output voltage. This is the temperature-corrected zero-point offset value. This is the structural compressive deformation correction factor. This represents the self-locking loss rate of the magnetorheological anchor. Subsequently, a double cross-validation correction was performed. The first cross-validation was compared with the total prestress of the free segment calculated using the distributed fiber optic solution. First, the average effective prestress of the free segment calculated using the distributed fiber optic solution was calculated. ,in, The axial stress at various points on the free segment of the distributed optical fiber is calculated. The cross-sectional area of ​​a single steel strand. The number of steel strands in the anchor cable bundle; calculate the relative error between the two. ,like The solution is considered reliable; if The average of the two values ​​is taken as the final effective prestress value; if The system alarm is triggered, requiring manual troubleshooting of sensor or fiber optic monitoring system malfunctions; the second method relies on the measured elongation of the anchor cable's free section to calculate the effective prestress using Hooke's Law. ,in, The measured elongation of the free section of the anchor cable is given; then the relative error between the two is calculated: ,Require If the value exceeds the range, it is necessary to check for abnormalities such as steel strand slippage or anchorage loosening. The value obtained after cross-verification shall be used as the final effective prestress value at the anchor end. Compare this value with the preset prestress benchmark value. In comparison, the total stress loss rate was calculated. The subsequent judgment steps are continuously output at a frequency of 1Hz. This embodiment can comprehensively eliminate the prestress measurement deviation caused by outdoor temperature difference, equipment installation gap, tensioning and retraction, and component compression through a complete sensor calibration system, binary linear two-dimensional temperature correction, multiple loss compensation of structure and anchorage, and fiber-optic-displacement dual cross-verification. Compared with the traditional single pressure sensor monitoring method, the measurement accuracy is significantly improved, providing a stable and reliable quantitative basis for the overall prestress of the anchor end for compensation classification judgment.

[0053] For example, before acquiring real-time data from the anchor pressure sensor, the anchor pressure sensor needs to be calibrated, including: First, a static calibration was performed in the laboratory. The anchor pressure sensor to be calibrated was installed on a standard pressure testing machine, and a preload was applied to 5% of the sensor's rated range. After holding the load for 5 minutes, it was unloaded to zero. This process was repeated three times to eliminate the initial gap in the sensor. Then, a graded loading method was used, starting from 0% and gradually increasing the load to 100% of the sensor's rated range. The loading levels were: 0% → 20% → 40% → 60% → 80% → 100%, with each level held for 3 minutes. The corresponding loading force was recorded. and sensor output voltage Unloading was performed using the same grading method, and corresponding data were recorded during the unloading process. The average of the loading and unloading data was taken as the final calibration data. The linear fitting equation of the sensor was obtained through least squares linear regression. ,in, For the sensor's original measured pressure, To calibrate the sensitivity coefficient of the sensor in the laboratory. For the sensor to output voltage in real time, To calibrate the zero-point offset value of the sensor in the laboratory; to calculate the nonlinear error of the sensor. ,Require (FS stands for full scale): ,in, The rated full-scale pressure of the sensor.

[0054] Secondly, a secondary on-site calibration was performed after installation. During the initial tensioning stage of the anchor cables, a calibrated through-hole jack was used for graded tensioning, with the tension levels consistent with the laboratory calibration. Each tensioning level was held for 3 minutes, and the actual output force of the jack was recorded simultaneously. and the output voltage of the anchor pressure sensor Then, by performing linear regression using the least squares method, the corrected equation after on-site calibration was obtained: ; in, For measuring pressure using sensors that have been calibrated on-site, To calibrate the sensitivity coefficient on site, To determine the zero-point offset value for on-site calibration, which is mainly caused by installation preload and contact surface gap, calculate the deviation rate between on-site calibration and laboratory calibration. ,like If the failure rate is >5%, the sensor installation quality needs to be re-inspected, and the sensor should be replaced and reinstalled if necessary.

[0055] Finally, data acquisition is performed using a unified synchronous clock with the distributed fiber optic demodulator and the ambient temperature and humidity sensor. The sampling frequency is consistent with the fiber optic monitoring. The acquired data includes: sensor output voltage. Sensor body temperature Ambient temperature Data collection timestamp t.

[0056] Optionally, the step of performing time-by-time full-segment stress calculation and time-by-time effective stress calculation on the time-series data in the multi-source sensing data of the steep slope to generate a time-by-time stress dataset, and performing damage analysis to generate a damage analysis surface plot, includes: The timing data is subjected to global clock synchronization and sampling frequency normalization to obtain unified reference timing data; Based on a three-level detection and repair mechanism, the unified benchmark time series data is repaired to obtain repaired time series data. The three-level detection and repair mechanism includes a point anomaly detection method, a segment anomaly detection method, and a data repair method. The repaired time series data is subjected to wavelet threshold denoising and moving average filtering to obtain clean time series data; For each time point in the cleaning time series data, perform time-by-time full-segment stress calculation and time-by-time effective stress calculation to generate the time-by-time stress dataset, and perform damage analysis to generate the damage analysis surface plot.

[0057] Specifically, firstly, global clock synchronization is implemented. All acquisition devices synchronize to the high-precision clock of the edge terminal via the NTP protocol, with the synchronization error controlled within 10ms. Then, linear interpolation is used to normalize the sampling frequency. The time step is uniformly 1 minute for routine maintenance, and switched to 1 second for high-frequency monitoring periods such as blasting and construction, based on the edge terminal's reception time. Based on this, the transmission delay of each sensor, calibrated using multiple round-trip tests, is corrected. Correcting the unified timestamp ,in, The average transmission delay of the i-th type of sensor is calibrated through multiple round-trip tests. The obtained unified time-series data initiates a three-level anomaly detection and repair mechanism, relying on a sliding window 3σ algorithm to identify single-point anomalies, as shown in the formula: ,in, and Let be the mean and standard deviation of the data within a sliding window centered at t and with width w, respectively; the abrupt change segment is determined by exceeding the threshold at three consecutive points using the difference method, with the formula as follows: ,in, This represents the original sampled data value (or signal amplitude) at the current time t. For the previous time point t The original sampled data value of 1, For the change, The anomaly detection threshold is used; the data repair method is as follows: for single-point confirmation or anomalies, linear interpolation of adjacent 3 points is used for repair; for short-term missing data within 30 minutes, cubic spline interpolation is used for completion; for long-term missing data exceeding 30 days, the LSTM historical prediction model is used to generate alternative data and all data are marked with prediction labels; the repaired clean time series data is subjected to 3-level db4 wavelet decomposition to obtain low-frequency approximation coefficients and high-frequency detail coefficients; a soft thresholding function is used to threshold the high-frequency detail coefficients. ; Among them, the general threshold The formula is ,in For the thresholded k-th wavelet detail coefficients of the j-th layer, For the original wavelet detail coefficients of the j-th layer and k-th wavelet, ( ) represents the sign function. For general thresholds, The standard deviation of noise. The total length of the time series data is given. Wavelet reconstruction is performed on the processed coefficients to obtain preliminary denoised data. A 5-point moving average filter is then used for secondary smoothing to finally obtain clean time series data. .

[0058] After processing, the clean time series dataset is subjected to full-segment stress calculation and effective stress calculation under the anchor at each time section to form a time-by-time stress dataset. The steps for performing time-by-time stress calculations across the entire anchor cable section include: First, temperature strain correction is performed, extracted from the temperature time series data of the cleaning time series data. Temperature distribution at time T Calculate temperature strain To obtain effective structural strain Then, strain-stress conversion is performed based on the calibrated strain transfer coefficient k and the elastic modulus of the steel strand. Calculate the axial stress at each point along the entire section. Next, the average stress of the region is calculated, and the average stress of the anchorage section is calculated separately. With the average stress of the free segment Finally, cross-validation was performed to correct the results. Cross-validation was performed on the effective stress under the anchor at time t, and the correction factor was calculated. The stress across the entire segment is corrected, and the corrected time-series stress data is as follows: ,in, , for Effective prestress under the anchor at any given time.

[0059] For each time point The steps for calculating the effective stress under the anchor include: First, sensor output extraction is performed, extracting data from the voltage time series data of the cleaning time series data. Filtered voltage at time 1 and sensor temperature Then, temperature correction is performed, and the sensitivity coefficient after temperature correction is calculated as follows: ; The zero offset value is: ; Then, structural deformation and anchorage characteristics are corrected: corrected time-series effective stress data. Finally, a second check for outliers is performed. If the deviation from the previous time value exceeds 5% and there is no corresponding tensioning or compensation event record, it is judged as abnormal, and the average value of the previous 3 time values ​​is used instead.

[0060] The time-series full-range stress data and the time-series effective stress data together constitute a time-by-time stress dataset. Key feature parameters are extracted from the time-by-time stress dataset for stress loss analysis and compensation trigger determination, generating the damage analysis surface plot. The steps include: Extract fundamental feature parameters, including total stress loss rate. ,in, For designing prestress reference values; stress change rate The unit is MPa / h, used to distinguish between slow creep loss and sudden vibration loss; stress fluctuation amplitude. Where w is the sliding window width, used to assess the impact of periodic factors such as temperature changes; cumulative stress loss. , where N is the number of valid data points within the statistical period.

[0061] Extracting segment characteristic parameters, including stress characteristic parameters of the anchored and free sections, is used to locate the sections where losses occur. The average stress loss rate of the anchored section is: ; The average stress loss rate of the free section is: ; The stress distribution non-uniformity coefficient is: ; Where M is the total number of measuring points in the entire section. The larger the value, the more uneven the stress distribution, and the more likely it is to cause local stress concentration or anchorage failure.

[0062] Extracting anomalous event feature parameters, including the magnitude of stress mutation. ,in, and The time points before and after the abrupt change are used to assess the impact of instantaneous loads such as blasting and construction; stress recovery rate. ,in, This is the stable point after the event ends, used to assess the self-recovery capability and permanent loss of the anchor cable.

[0063] By fusing stress time history data from the time-by-time stress dataset with time-series data such as environmental temperature and humidity, vibration, and slope displacement, real-time separation of stress loss causes is achieved. These causes include separation of temperature deformation causes, rock mass creep and steel strand relaxation causes, and vibration and impact causes. The steps include: For temperature-induced deformation, the stress fluctuations caused by temperature changes exhibit significant periodicity and correlation. Using linear regression to separate these factors, the established linear regression model between stress and temperature is as follows: ; in, The data consists of time-series ambient temperature data, where a and b are regression coefficients. This is the residual term.

[0064] The stress fluctuation component caused by temperature is calculated as follows: ; The non-temperature-induced stress change component is obtained as follows: ; Regarding the causes of rock mass creep and steel strand relaxation, the stress loss caused by rock mass creep and steel strand relaxation exhibits a long-term monotonically decreasing trend. An exponential function is used for separation, specifically: [The text abruptly ends here, so the translation stops as well.] Perform an exponential fit, and the fitting formula is: ,in, Where k is the initial prestress, and k is the attenuation coefficient. To ensure long-term stable prestress, the cumulative losses caused by creep and relaxation are then calculated. .

[0065] For vibration-induced shocks, stress changes caused by vibrations from blasting, construction, etc., exhibit instantaneous and abrupt changes. An event-triggered method is used for separation, with the following steps: when the vibration sensor detects that the vibration acceleration exceeds a threshold... When the event occurs, it is marked as a vibration event; stress data for one hour before and after the event are extracted, and the magnitude of stress mutation and permanent loss are calculated; the correspondence between vibration intensity and stress loss is established as follows: ,in Let be the peak acceleration of the vibration, and c and d be empirical coefficients.

[0066] Finally, the damage analysis surface plot is generated, including: the time history curve of the effective prestress under the anchor, with time as the horizontal axis and the effective prestress under the anchor. A continuous curve is plotted with the x-axis as the vertical axis, simultaneously marking the design prestressing baseline, the third-level compensation threshold line, and key events (tensioning, compensation, blasting, etc.). A full-section stress time history surface is displayed as a 3D surface plot, with the x-axis representing anchor cable length, the y-axis representing time, and the z-axis representing stress value, visually showing the stress variation trend and spatial distribution differences over time. A section average stress comparison curve is plotted, showing the average stress time history curves of the anchored section and the free section in the same coordinate system, comparing the loss patterns of different sections. A stress loss rate time history curve is plotted with time as the horizontal axis, showing the total stress loss rate. A curve is plotted with time on the vertical axis to visually assess the extent of loss development; the stress change rate time history curve has time on the horizontal axis, representing the stress change rate. The vertical axis is used to plot curves for rapid identification of abnormal changes; the cause separation and superimposed curve decomposes the total stress change into components of different causes such as temperature, creep, and vibration, and plots them as superimposed bar charts or area charts to clearly show the contribution ratio of each cause; the time history curve adopts a real-time rolling update mode, every The data points are updated once every t time interval; when the stress loss rate reaches the first-level compensation threshold (5%~8%), the corresponding section is marked in yellow; when it reaches the second-level compensation threshold (8%~15%), it is marked in orange; when it reaches the third-level warning threshold (>15%) or a stress change occurs, it is marked in red and an audible and visual warning is triggered.

[0067] In addition, the time-by-time stress dataset, characteristic parameters, and cause separation results are stored in a structured format on the edge terminal and cloud database; daily, weekly, and monthly reports are automatically generated, including stress change trends, loss statistics, cause analysis, and compensation execution records; the stress loss rate and change rate are pushed to the three-level compensation control module in real time, and the corresponding level of compensation program is automatically started when the trigger conditions are met; based on years of time history data, a long-term performance degradation model of anchor cables is established to predict the remaining service life and provide a basis for slope operation and maintenance decisions.

[0068] This embodiment can ensure the integrity and smoothness of long-term monitoring data through clock synchronization, anomaly hierarchical repair, and combined noise reduction algorithms. It can synchronously calculate the stress of the entire space and anchor end at each time moment to realize spatiotemporal integrated damage characterization. The three-dimensional damage surface visualization intuitively shows the long-term deterioration trend of the slope anchor cable, providing long-term data support for predicting anchor cable failure in advance and formulating preventive compensation plans.

[0069] Optionally, after performing stress compensation based on the compensation level and the loss cause, the method further includes: Obtain data on the spatial distribution of stress, average effective prestress, and stress uniformity of the entire section after stress compensation, and construct a stress compensation verification dataset. Acquire data on deep deformation of the slope rock mass, slope displacement and axial displacement of anchor cables after stress compensation, determine the rock mass deformation response and anchor cable structure displacement data corresponding to the compensation action, and construct a structural safety verification dataset. Obtain the actual tension stroke, final locking position, self-locking state parameters and action response timing data of the stress-compensated self-locking anchor, and construct an actuator verification dataset; The stress compensation verification dataset, the structural safety verification dataset, and the actuator verification dataset are compared with their respective safety verification thresholds to obtain the verification results.

[0070] Specifically, based on the preset verification scheme, after the compensation execution and anchor self-locking are completed, multi-dimensional data acquisition and fusion analysis are carried out simultaneously: continuously acquiring the distributed strain and effective prestress time series data of the entire anchor cable section, extracting the stress spatial distribution, average effective prestress and stress uniformity parameters of the entire section after stabilization, and constructing a stress compensation verification dataset; simultaneously acquiring deep deformation of the slope rock mass, slope displacement and anchor cable axial displacement data, identifying the rock mass deformation response and structural displacement characteristics induced by the compensation action, verifying the risks of abnormal rock mass deformation, anchor cable slippage and local stress concentration, and forming a structural safety verification dataset; simultaneously recording the actual tensioning stroke, final locking position, self-locking state and action response time series of the self-locking anchor, comparing the preset control parameters to obtain the control accuracy, stroke deviation and self-locking reliability indicators of the actuator, and establishing an actuator verification dataset. Based on this, by comparing the measured effective prestress of the stabilized anchor cable with the design benchmark value and the compensation target value, the compensation deviation rate and stress loss recovery rate are calculated to determine whether they meet the design allowable deviation. Local stress concentration and under-compensated sections are identified based on the uniformity of stress distribution throughout the entire section. By comparing the rock mass and anchor cable displacement data with safety limits, a comprehensive assessment is made to determine whether the compensation action induces abnormal rock mass deformation, potential slip surface displacement, anchor cable slippage, and anchor loosening risks. This embodiment establishes a complete closed-loop verification system from three dimensions: compensation accuracy, slope structure safety, and intelligent anchor equipment reliability. It quantifies the compensation effect and traces various abnormal deviations, avoiding safety hazards such as substandard prestress, slope disturbance, and equipment failure after compensation tensioning. It achieves full-process verifiability and traceability for each compensation operation, improving the long-term operation and maintenance safety management level of anchor cables on steep slopes.

[0071] Optionally, before acquiring multi-source sensing data of steep slopes through a preset multi-source sensing device, and sequentially performing preprocessing, validity pre-verification, and correction to obtain multi-source sensing data of steep slopes, the method further includes: Based on slope geological survey data and rock mechanics test data, the parameters of the anchor cable support foundation were determined. Based on the anchor cable support foundation parameters, the sensor and anchorage design scheme and the preset three-level stress compensation threshold are determined.

[0072] Specifically, geological mapping, drilling exploration, and rock mechanics testing were conducted on the entire steep slope. Data on slope gradient, free face height, and topography were collected. Drilling revealed the slope lithology, rock mass integrity, joint and fissure development, distribution of weak interlayers, spatial location of potential slip surfaces, and groundwater transport patterns. Based on rock mechanics and long-term creep tests, mechanical parameters were obtained, including the slope rock mass elastic modulus, Poisson's ratio, cohesion, internal friction angle, long-term creep characteristics, and relaxation characteristics. Simultaneously, interfacial bonding performance parameters between the anchor cable strands and the grouting body, and between the grouting body and the borehole wall rock mass, were tested. Based on the obtained geological, mechanical, and interfacial bonding performance parameters, anchor cable planar layout parameters, spatial layout parameters, borehole diameter, and total anchor cable length were designed. Anchor cable prestress values, over-tension coefficients, and locking values ​​were calculated. Grouting body mix proportions, strength grades, and grouting process parameters were designed, completing the prestressed anchor cable benchmark support design. More specifically, based on the survey and test data, a refined calculation of all anchor cable parameters is carried out, and the formula for the total anchor cable length is given. ,in, Given the length of the free segment of the anchor cable, and introducing the spatial curvature and inclination angle of the potential slip surface, the calculation formula is as follows: H represents the potential depth of the slip surface. The anchor cable inclination angle is 1.5m, which is the safety margin for crossing the slip surface. The formula for calculating the anchorage length of the anchor cable, after introducing the interfacial bonding performance test coefficient, is as follows: , The long-term bond strength reduction factor is 0.75~0.85, based on long-term creep tests, where D is the pore size. The interfacial bond strength between the grout and the borehole wall rock mass; The exposed section length of the anchor cable is 0.8~1.2m, which is taken to accommodate the installation space of the magnetorheological smart anchor. To reserve length for long-term creep deformation of the rock mass, based on long-term creep test data of the rock mass, the formula is as follows: , This represents the cumulative creep strain of the free section of the rock mass during the service life of the anchor cable. For the distributed fiber redundancy length, 0.3m is reserved at each end of the anchor cable and 0.2m is reserved at the boundary between the anchored section and the free section, totaling 0.8m, for fiber splicing and protection. Based on the instantaneous sliding force distribution of the slope, and considering the long-term stress loss from multi-field coupling, the three-level compensation reserve, and the uniformity of stress distribution throughout the entire section, in order to solve the problem of insufficient prestress or local overload in the later stage, the formula for calculating the design prestress value of a single anchor cable is as follows: ; in, The prestress value for a single anchor cable is taken as the final design value. The residual sliding force per unit length of the slope is calculated based on the limit equilibrium method, taking into account groundwater and seismic action. The slope safety factor is set at 1.2 to 1.5, depending on the engineering grade. The number of anchor cables in a single row is determined based on the slope width and the spacing between the cables. The anchor effect coefficient for the anchor cable group is taken as 0.85~0.95, with a lower value taken when the anchor cable spacing is less than 3m; The angle between the anchor cable and the horizontal plane is set at 15° to 30°, taking into account both construction convenience and support efficiency. The long-term stress loss coefficient due to multi-field coupling is calculated using a multi-field coupling model, and the formula is as follows: , For rock mass creep loss rate, For the relaxation loss rate of steel strand, For permanent loss rate due to temperature deformation, Permanent loss rate due to vibration and shock; A three-level compensation reserve coefficient is set at 1.08 to 1.12, corresponding to the preset three-level stress compensation threshold system in subsequent steps, to ensure that even if a first-level loss of 5% to 8% occurs, immediate compensation is not required. The stress uniformity correction coefficient for the entire section is set to 0.92~1.05 based on the stress distribution characteristics monitored by distributed optical fibers. A lower value is used when stress is concentrated in the anchorage section, and a higher value is used when stress is uneven in the free section. Generally, the over-tensioning coefficient only considers anchor retraction and instantaneous elastic loss of the steel strand, and a fixed value of 1.05~1.1 is adopted. Now, utilizing the stress-free fall-off characteristic of magnetorheological intelligent anchors to solve the problem of excessive over-tensioning or insufficient compensation, the formula for calculating the over-tensioning coefficient is as follows: ; in, The over-tension factor for the anchor cable is taken as 1.01~1.05, which is the final design value. The instantaneous elastic deformation loss of the steel strand is calculated using the following formula: , The elastic modulus of the steel strand. This represents the cross-sectional area of ​​the steel strand; The elastic compression loss of the anchor cable bundle is calculated using the following formula: , The elastic modulus of the grout. The cross-sectional area of ​​the grouting body; The self-locking efficiency coefficient of the magnetorheological intelligent anchor is 0.995~1.0, with almost no stress drop.

[0073] Anchor lock value = over-tension value - anchor recoil loss. A preset three-level stress compensation threshold constraint needs to be introduced to address the issues of excessively low lock values ​​leading to frequent compensation, or excessively high lock values ​​causing anchor overload. The formula for calculating the anchor lock value is as follows: ; in, For anchor cable locking values, take the final design value; The ultimate tensile bearing capacity of the anchor cable strand is calculated using the following formula: , This refers to the standard value of the tensile strength of steel strand; This is the upper limit coefficient for anchor cable prestress; The threshold for triggering Level 1 compensation is set at 5% to 8% to ensure that the locked value is not lower than the lower limit of Level 1 compensation triggering and to reduce unnecessary compensation actions.

[0074] Based on the anchor cable bundle diameter and protective layer thickness, and considering the space requirements and interface bonding strength optimization of the distributed optical fiber anti-corrosion sheath, the formula for calculating the hole diameter is as follows: ; Where D is the hole diameter, which is the final design value; The equivalent diameter of the anchor cable bundle is calculated using the following formula: n is the number of steel strands, and d is the diameter of a single steel strand; The thickness of the protective layer for the grouting body is 25~30mm; The thickness of the anti-corrosion sheath for distributed optical fibers is 3~5mm. To optimize the incremental bonding strength of the interface, when measured... When the pressure is less than 1.0 MPa, a diameter of 10-20 mm is used to increase the bonding area between the grout and the hole wall.

[0075] In related technologies, the plane and spatial layout parameters of anchor cables are generally calculated using an evenly spaced arrangement, without considering the spatial distribution differences of the slope stress field. To address the issues of insufficient support in high-stress areas and waste in low-stress areas, this embodiment uses the following formula for calculating the plane and spatial layout parameters of anchor cables: ; in, The anchor spacing in the i-th region is used to achieve non-uniform distribution; The stress concentration factor for the slope in the i-th region is determined based on finite element numerical simulation and potential slip surface distribution. It is taken as 1.2~1.5 in the high stress zone and 0.7~0.9 in the low stress zone. The sliding force per unit length in the i-th region is calculated by considering the influence of weak interlayers and groundwater distribution.

[0076] After completing the design of the anchor cable body, a distributed optical fiber sensing deployment scheme was designed. Based on the total length and segment characteristics of a single anchor cable, the deployment path and fixed points of the sensing optical fiber along the steel strand were designed to cover the free section and anchored section of the anchor cable. Anchor pressure sensors, anchor displacement sensors, environmental temperature and humidity sensors, and deep slope inclination sensors were deployed. The deployment locations of surface displacement monitoring points, line transmission architecture, and edge data acquisition terminals were designed.

[0077] More specifically, a three-stage variable path layout scheme is adopted to address the different mechanical characteristics and functional requirements of the anchorage section, free section, and exposed section. This involves segmented variable pitch spiral layout in the anchorage section, straight parallel layout in the free section, and protective channel layout in the exposed section. The segmented variable pitch spiral layout in the anchorage section is based on the stress distribution characteristic of the anchorage section being high at the bottom of the hole and low near the free section; variable pitch enables high-precision monitoring of high-stress areas. The straight parallel layout in the free section is based on the characteristic that the steel strand in the free section is only subjected to axial tension; straight layout ensures uniform strain transmission and avoids local stress concentration. The protective channel layout in the exposed section is based on the characteristic that the exposed section needs to connect to the demodulator and withstand anchor installation and tensioning operations; reserved channels protect the optical fiber from mechanical damage.

[0078] Furthermore, since the maximum spacing between the fixed points of the optical fiber is determined by the strain transfer accuracy requirements, the derived formula is as follows: ; in, This represents the maximum allowable spacing between fixed points. To allow for strain transmission errors, The elastic modulus of optical fiber, The cross-sectional area of ​​the optical fiber. The interfacial bonding strength between the optical fiber and the steel strand is 0.2~0.3 MPa, measured when fixed with nylon clips. This refers to the outer diameter of the optical fiber.

[0079] More specifically, in addition to setting conventional fixing points according to the calculated spacing, the locations where forced fixing points must be set include the bottom end of the anchor cable (at the guide cap), the top end of the anchor cable (at the boundary between the exposed section and the free section), the boundary between the anchored section and the free section (one additional fixing point on each side, spaced 100mm apart), the anchor cable position corresponding to the potential slip surface (two additional fixing points on each side, spaced 150mm apart), the anchor cable position corresponding to joints, fissures, and weak interlayers, and both sides of the fiber optic splice point; combining the calculation model and engineering practice, the fixing spacing for each area is set as follows: the fixing spacing for the anchored section is 300~500mm (consistent with the helical pitch, one fixing point for each pitch); the fixing spacing for the free section is 500~800mm (taking the maximum value of 800mm from the calculation model, with denser fixing at key locations); and the fixing spacing for the exposed section is 200~300mm (dense fixing to prevent fiber optic swaying during tensioning).

[0080] More specifically, a three-tiered network and zoned densification scheme is adopted to design the location of surface displacement monitoring points, achieving comprehensive coverage of slope surface deformation and high-precision monitoring of key areas. Data is integrated with data from deep inclinometer sensors and anchor cable stress monitoring. The three-tiered monitoring network scheme includes: On stable bedrock outside the influence range of the slope, benchmark points are set up at a distance of ≥50m from the top of the slope and at an elevation higher than the highest point of the slope, with ≥3 benchmark points for each slope, forming a closed triangular network; At stable locations on the slope top, slope toe, and each level of platform, work base points are set up at a distance of ≤30m from the monitoring point, with ≥2 points on each level of platform and ≥3 points on the slope top and slope toe. Monitoring points should be deployed in sensitive areas of slope surface deformation according to the principle of zoned densification. This includes one monitoring point every 20-30m along the slope top line, slope toe line, and edges of each level of platform; one monitoring point every 5-10m along the boundary line of potential slip surfaces, exposed weak interlayers, and areas with dense joints and fissures; one monitoring point every 10-15m in areas with dense anchor cable deployment and stress concentration areas, corresponding one-to-one with the location of deep inclinometer sensors; monitoring points must be deployed at slope corners and protruding parts of the free face; additional monitoring points should be set up near construction access roads and temporary facilities, with a spacing of 10-15m. It adopts a three-layer distributed transmission architecture of perception layer-edge layer-cloud layer, which is divided into perception layer transmission (sensor → edge terminal), edge layer transmission (edge ​​terminal → cloud platform) and cloud layer transmission (cloud platform → operation and maintenance management terminal).

[0081] The perception layer transmission scheme includes: Distributed fiber optic data is transmitted using single-mode fiber optic patch cords at a transmission rate of 1Gbps and a maximum transmission distance of 10km, directly connected to the distributed fiber optic demodulator built into the edge terminal; low-frequency sensor data (collecting anchor pressure, anchor cable displacement, temperature and humidity, and rainfall) is transmitted via RS485 bus at a transmission distance ≤1000m and a baud rate of 9600bps, using shielded twisted-pair cables protected by galvanized steel pipes; high-frequency sensor data (collecting vibration and shock) is transmitted via industrial Ethernet at a transmission rate of 100Mbps and a transmission distance ≤100m; GNSS monitoring points for surface displacement data use 4G / 5G wireless transmission or LoRa wireless transmission (transmission distance ≤5km, low power consumption).

[0082] Edge layer transmission schemes include: Dual-link redundancy backup is adopted. The main link is a dedicated fiber optic line with a transmission rate of 100Mbps. The backup link is 4G / 5G wireless transmission. Automatic switching occurs when the main link fails, with a switching time of ≤30s. Edge terminals are connected by industrial Ethernet switches to form a local area network, enabling data sharing and device redundancy. Data transmission uses the MQTT protocol.

[0083] Cloud-based transmission solutions include: It uses HTTPS encrypted transmission, supports real-time access on PC and mobile APP, and provides functions such as data query, curve display, and early warning push. It also supports third-party platforms.

[0084] More specifically, based on the slope length and the number of anchor cables, the slope is divided into several monitoring and control zones. Each zone is equipped with one edge terminal, responsible for collecting data from all sensors within that zone and performing anchor cable compensation control within that zone. Each edge terminal manages ≤50 anchor cables and ≤200 sensors, ensuring a data processing latency of ≤100ms. The deployment location requirements for the edge data acquisition terminals are as follows: The RS485 sensor located at a stable position on the top of the slope or on each platform should be ≤500m away from the farthest sensor in this zone. Installed in an outdoor control cabinet with IP65 protection rating, it is adapted to harsh outdoor environments such as high temperature, low temperature, humidity, and dust. The bottom of the control cabinet is 0.3m above the ground and has a concrete foundation to prevent water accumulation and soaking. There are no tall vegetation or obstacles within a 2-meter radius around the control cabinet to ensure smooth wireless signal transmission.

[0085] More specifically, based on the design prestress value of the anchor cable, the matching interface between the anchor and the anchor plate and the limiting plate is optimized, and the installation cavity of the integrated pressure and displacement sensing module is designed. Taking the design prestress value of the anchor cable as a benchmark, combined with the rock mechanics parameters and the long-term service characteristics of the anchor cable, a three-level stress compensation threshold is preset, an adaptive intelligent compensation control algorithm is designed, and the tensioning step length, holding time, pressure stabilization parameters and reverse locking logic under different loss conditions are preset, as well as the early warning triggering rules under severe loss conditions.

[0086] This embodiment can incorporate the long-term prestress loss due to rock mass creep and temperature vibration, as well as the low shrinkage characteristics of intelligent anchors, into a multi-factor coupled, refined calculation model of anchor cables, sensing, and transmission systems. It reserves three levels of compensation safety redundancy in the design stage, and the non-uniform anchor cable layout takes into account the support needs of different areas of the slope. It can build a complete integrated sensing and compensation hardware solution, reducing the probability of engineering problems such as insufficient prestress, monitoring blind spots, and frequent compensation in the later stage from the source.

[0087] Optionally, after determining the sensing and anchorage design scheme and the preset three-level stress compensation threshold based on the anchor cable support foundation parameters, the method further includes: Based on the aforementioned sensing and anchor design scheme, steel strand cutting and anchor cable bundle fabrication are carried out, and distributed optical fiber sensing units are deployed along the cable body to implement anchor cable hole forming, hole cleaning, lowering and bottom grouting operations. After the grouting body has been cured to the required standard, the anchor plate and magnetorheological intelligent self-locking anchor are installed, and the pre-calibrated integrated anchor pressure sensor and displacement sensor are installed on the exposed steel strand of the anchor cable in the design sequence.

[0088] Specifically, the steel strands are first cut and the anchor cable bundles are assembled on a flat processing platform. The steel strands are cut to the designed length using an abrasive wheel. Distributed optical fibers are laid along the central axis of the bundle. Nylon clips are used to fix the segments according to the calculated fixed spacing. Redundant protective rings for optical fibers are reserved at both ends of the anchor cable and at the free boundary of the anchorage. High-density polyethylene anti-corrosion sleeves are installed throughout the line, and the joints are sealed with waterproof tape and heated shrink tubing. A centering isolation frame is installed in the anchorage section to ensure the thickness of the grouting protective layer. The free section is fitted with a sealed anti-corrosion sleeve. A guide cap is installed at the bottom of the anchor cable to protect the optical fibers and steel strands. After assembly, the optical fiber attenuation is tested and found to be qualified before it can be transported to the hole position. The drilling rig was positioned according to the designed borehole location, inclination angle, and azimuth angle. The borehole was formed using the casing drilling process. Changes in rock strata and groundwater were recorded throughout the drilling process. After reaching the designed depth, high-pressure air was used to remove rock powder and sediment from the borehole. Anchor cable bundles were then steadily lowered to the designed elevation using hoisting equipment. The anchor cables were fixed in the center of the borehole opening to prevent deflection. The grouting slurry was mixed according to the proportions. After the consistency and fluidity were tested and found to be up to standard, the bottom return pressure grouting process was adopted. The grouting pipe was buried inside the grout throughout the process. After the thick slurry overflowed from the borehole opening, the pressure was stabilized to complete the grouting. Simultaneously, strength test blocks were made and standard curing was performed. Subsequent procedures were carried out after the strength of the grouting body reached or exceeded the design threshold. The rock surface at the orifice was leveled with high-strength mortar. Anchor plates, limit plates, pre-calibrated integrated anchor pressure sensors, and displacement sensors were then installed sequentially. Sensor control lines were neatly arranged to allow for tensioning movement. Finally, magnetorheological intelligent self-locking anchors were installed and the clamps were evenly tapped into place. After equipment installation, graded on-site tensioning calibration was conducted. Prestress was applied in stages according to the design, with each stage held for 3 minutes. Simultaneously, standard force values ​​from the jacks and data from the fiber optic and pressure sensors were collected. The strain transfer coefficient and sensor sensitivity zero-point parameters were recalibrated. After tensioning to exceed the design tension value, the magnetorheological anchor self-locking program was initiated, and the jacks were unloaded to complete the initial locking. After all hardware, wiring, and tensioning calibration were completed, system integration and debugging were carried out. Sensors, demodulators, edge terminals, and intelligent anchor control units were individually debugged. The three-layer transmission link dual-link automatic switching, data storage, three-level compensation logic, and audible and visual early warning functions were tested. After continuous, trouble-free trial operation, the system was officially handed over to the long-term slope monitoring and compensation system for use. All construction records, sensor calibration reports, and fiber optic testing data were simultaneously archived to form a complete construction archive. This embodiment can strictly implement the detailed design parameters in the early stage through standardized anchor cable processing, hole drilling and grouting, and sensor anchor installation construction procedures. It can simultaneously complete the parameter deviation between the theoretical design and the on-site installation of the dual calibration correction of the on-site equipment, and ensure the installation accuracy and long-term working stability of the entire hardware system of distributed optical fiber full-segment monitoring and magnetorheological intelligent automatic compensation. This provides a reliable hardware foundation for subsequent full-cycle stress monitoring, cause decoupling, and graded automatic compensation.

[0089] like Figure 2 As shown in the figure, an embodiment of the present invention provides a stress loss compensation device for anchor cables on steep slopes, comprising: The full-segment stress module is used to acquire multi-source sensing data of steep slopes through preset multi-source sensing devices, and to sequentially preprocess, verify the validity of and correct the distributed strain data in the multi-source sensing data of steep slopes to obtain corrected distributed strain data. Based on the strain-stress conversion model, the module performs full-segment stress calculation on the corrected distributed strain data to obtain full-segment stress distribution data. The parameters of the strain-stress conversion model are determined by laboratory calibration data and field tension calibration data. The effective stress module is used to calculate and correct the effective stress of the real-time data of the anchor pressure sensor in the multi-source sensing data of the high and steep slope based on the binary linear temperature correction model, so as to obtain the effective prestress value of the anchor end, and determine the total stress loss data of the anchor end based on the effective prestress value of the anchor end and the prestress reference value. The time-series module is used to perform time-by-time full-segment stress calculation and time-by-time effective stress calculation on the time-series data in the multi-source sensing data of the steep slope, generate time-by-time stress dataset, perform damage analysis, and generate damage analysis surface plot. The cause decoupling module is used to decouple and separate the total stress loss data at the anchor end based on the multi-source sensing data of the steep slope, the linear total stress loss superposition model, and the key feature parameters in the time-by-time stress data set to obtain the cause of loss. The compensation module is used to determine the compensation level based on the stress distribution data of the entire section and the total stress loss data of the anchor end, according to the preset three-level stress compensation threshold, generate the compensation level, and perform stress compensation based on the compensation level and the loss cause.

[0090] like Figure 3 As shown, an electronic device 300 provided in this embodiment of the invention includes a memory 310 and a processor 320; the memory 310 is used to store a computer program; the processor 320 is used to implement the stress loss compensation method for anchor cables on steep slopes as described above when the computer program is executed.

[0091] Alternatively, an electronic device 300 includes a memory 310 and a processor 320 coupled to the memory 310; the memory 310 is configured to store a computer program; and the processor 320 is configured to perform the following operations when the computer program is executed: Multi-source sensing data of steep slopes is acquired through a pre-set multi-source sensing device. The distributed strain data in the multi-source sensing data of steep slopes is preprocessed, validated in advance, and corrected in sequence to obtain corrected distributed strain data. Based on the strain-stress conversion model, the stress of the corrected distributed strain data is calculated for the whole section to obtain the stress distribution data of the whole section. The parameters of the strain-stress conversion model are determined by laboratory calibration data and field tension calibration data. Based on the binary linear temperature correction model, the real-time data of the anchor pressure sensor in the multi-source sensing data of the high and steep slope are effectively calculated and corrected to obtain the effective prestress value of the anchor end. Based on the effective prestress value of the anchor end and the prestress reference value, the total stress loss data of the anchor end is determined. The time-series data in the multi-source sensing data of the steep slope are subjected to time-by-time full-segment stress calculation and time-by-time effective stress calculation respectively to generate time-by-time stress dataset, and damage analysis is performed to generate damage analysis surface plot. Based on the multi-source sensor data of the steep slope, the linear total stress loss superposition model, and the key feature parameters in the time-by-time stress data, the causes of loss are decoupled and separated from the total stress loss data at the anchor end to obtain the causes of loss. Based on the stress distribution data of the entire section and the total stress loss data of the anchor end, a compensation determination is made according to the preset three-level stress compensation threshold, a compensation level is generated, and stress compensation is performed based on the compensation level and the loss cause.

[0092] This invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the stress loss compensation method for anchor cables on steep slopes as described above.

[0093] Alternatively, a non-volatile computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the following operations: Multi-source sensing data of steep slopes is acquired through a pre-set multi-source sensing device. The distributed strain data in the multi-source sensing data of steep slopes is preprocessed, validated in advance, and corrected in sequence to obtain corrected distributed strain data. Based on the strain-stress conversion model, the stress of the corrected distributed strain data is calculated for the whole section to obtain the stress distribution data of the whole section. The parameters of the strain-stress conversion model are determined by laboratory calibration data and field tension calibration data. Based on the binary linear temperature correction model, the real-time data of the anchor pressure sensor in the multi-source sensing data of the high and steep slope are effectively calculated and corrected to obtain the effective prestress value of the anchor end. Based on the effective prestress value of the anchor end and the prestress reference value, the total stress loss data of the anchor end is determined. The time-series data in the multi-source sensing data of the steep slope are subjected to time-by-time full-segment stress calculation and time-by-time effective stress calculation respectively to generate time-by-time stress dataset, and damage analysis is performed to generate damage analysis surface plot. Based on the multi-source sensor data of the steep slope, the linear total stress loss superposition model, and the key feature parameters in the time-by-time stress data, the causes of loss are decoupled and separated from the total stress loss data at the anchor end to obtain the causes of loss. Based on the stress distribution data of the entire section and the total stress loss data of the anchor end, a compensation determination is made according to the preset three-level stress compensation threshold, a compensation level is generated, and stress compensation is performed based on the compensation level and the loss cause.

[0094] The present invention will now be described an electronic device 300 that can serve as a server or client of the present invention, which is an example of a hardware device that can be applied to various aspects of the present invention. Electronic device 300 is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device 300 can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0095] Electronic device 300 includes a computing unit that can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) or a computer program loaded from a storage unit into random access memory (RAM). The RAM may also store various programs and data required for device operation. The computing unit, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0096] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc. In this application, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention according to actual needs. Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units can be implemented in hardware or as software functional units.

[0097] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.

Claims

1. A method for compensating stress loss in anchor cables on steep slopes, characterized in that, include: Multi-source sensing data of steep slopes is acquired through a pre-set multi-source sensing device. The distributed strain data in the multi-source sensing data of steep slopes is preprocessed, validated in advance, and corrected in sequence to obtain corrected distributed strain data. Based on the strain-stress conversion model, the stress of the corrected distributed strain data is calculated for the whole section to obtain the stress distribution data of the whole section. The parameters of the strain-stress conversion model are determined by laboratory calibration data and field tension calibration data. Based on the binary linear temperature correction model, the real-time data of the anchor pressure sensor in the multi-source sensing data of the high and steep slope are effectively calculated and corrected to obtain the effective prestress value of the anchor end. Based on the effective prestress value of the anchor end and the prestress reference value, the total stress loss data of the anchor end is determined. The time-series data in the multi-source sensing data of the steep slope are subjected to time-by-time full-segment stress calculation and time-by-time effective stress calculation respectively to generate time-by-time stress dataset, and damage analysis is performed to generate damage analysis surface plot. Based on the multi-source sensor data of the steep slope, the linear total stress loss superposition model, and the key feature parameters in the time-by-time stress data, the causes of loss are decoupled and separated from the total stress loss data at the anchor end to obtain the causes of loss. Based on the stress distribution data of the entire section and the total stress loss data of the anchor end, a compensation determination is made according to the preset three-level stress compensation threshold, a compensation level is generated, and stress compensation is performed based on the compensation level and the loss cause.

2. The method for compensating for stress loss in anchor cables on steep slopes according to claim 1, characterized in that, The distributed strain data from the multi-source sensing data of the steep slope is sequentially preprocessed, validated, and corrected to obtain corrected distributed strain data. Based on the strain-stress conversion model, the stress distribution data of the corrected distributed strain data is calculated for the entire section to obtain stress distribution data for the entire section, including: The distributed strain data is sequentially spatiotemporally aligned, subjected to wavelet denoising, moving average filtering, and normalization to obtain standard distributed strain data. Based on the validity pre-verification mechanism, the standard distributed strain data is verified and corrected to obtain pre-verified distributed strain data. The validity pre-verification mechanism includes integrity verification, outlier secondary verification, and boundary consistency verification. Based on the dual correction method, temperature strain interference terms are separated and corrected on the pre-verification distributed strain data to obtain dual-corrected distributed strain data. The dual correction method includes a distributed temperature self-compensation method and an ambient temperature calibration method. Based on the double-corrected distributed strain data, the strain-stress conversion model is used to calculate the stress distribution across the entire segment, and preliminary stress distribution data across the entire segment is obtained. Based on the anchor pressure sensor data and anchor cable displacement sensor data in the multi-source sensing data of the steep slope, the preliminary full-section stress distribution data is verified and corrected using a dual-data source cross-validation method to obtain the full-section stress distribution data. The full-section stress distribution data is then smoothed and visualized to generate stress spatial distribution curves, which are stored in a local database and a cloud platform, respectively.

3. The method for compensating for stress loss in anchor cables on steep slopes according to claim 1, characterized in that, The binary linear temperature correction model is used to effectively calculate and correct the real-time data of the anchor pressure sensor in the multi-source sensing data of the steep slope, to obtain the effective prestress value at the anchor end. Based on the effective prestress value and the prestress reference value at the anchor end, the total stress loss data at the anchor end is determined, including: The real-time data of the anchor pressure sensor is preprocessed to obtain processed real-time data of the anchor pressure sensor. The preprocessing includes outlier removal, moving average filtering, and data validity determination. Using the binary linear temperature correction model, the effective stress calculation, sensitivity correction and zero-point temperature drift correction are performed on the real-time data of the processed anchor pressure sensor to obtain the initial effective prestress value at the anchor end. The effective prestress value at the anchor end is obtained by performing a double cross-validation based on the stress distribution data of the entire section and the anchor cable axial displacement sensor data in the multi-source sensing data of the steep slope. The effective prestress value at the anchor end is compared with the prestress reference value to determine the total stress loss data at the anchor end.

4. The method for compensating for stress loss in anchor cables on steep slopes according to claim 1, characterized in that, The time-series data from the multi-source sensing data of the steep slope are subjected to time-by-time full-segment stress calculation and time-by-time effective stress calculation respectively to generate a time-by-time stress dataset, and damage analysis is performed to generate a damage analysis surface plot, including: The timing data is subjected to global clock synchronization and sampling frequency normalization to obtain unified reference timing data; Based on a three-level detection and repair mechanism, the unified benchmark time series data is repaired to obtain repaired time series data. The three-level detection and repair mechanism includes a point anomaly detection method, a segment anomaly detection method, and a data repair method. The repaired time series data is subjected to wavelet threshold denoising and moving average filtering to obtain clean time series data; For each time point in the cleaning time series data, perform time-by-time full-segment stress calculation and time-by-time effective stress calculation to generate the time-by-time stress dataset, and perform damage analysis to generate the damage analysis surface plot.

5. The method for compensating for stress loss in anchor cables on steep slopes according to claim 1, characterized in that, After performing stress compensation based on the compensation level and the loss cause, the method further includes: Obtain data on the spatial distribution of stress, average effective prestress, and stress uniformity of the entire section after stress compensation, and construct a stress compensation verification dataset. Acquire data on deep deformation of the slope rock mass, slope displacement and axial displacement of anchor cables after stress compensation, determine the rock mass deformation response and anchor cable structure displacement data corresponding to the compensation action, and construct a structural safety verification dataset. Obtain the actual tension stroke, final locking position, self-locking state parameters and action response timing data of the stress-compensated self-locking anchor, and construct an actuator verification dataset; The stress compensation verification dataset, the structural safety verification dataset, and the actuator verification dataset are compared with their respective safety verification thresholds to obtain the verification results.

6. The method for compensating for stress loss in anchor cables on steep slopes according to claim 1, characterized in that, Before acquiring multi-source sensing data of steep slopes through preset multi-source sensing devices, and sequentially performing preprocessing, validity pre-verification, and correction to obtain multi-source sensing data of steep slopes, the process further includes: Based on slope geological survey data and rock mechanics test data, the parameters of the anchor cable support foundation were determined. Based on the anchor cable support foundation parameters, the sensor and anchorage design scheme and the preset three-level stress compensation threshold are determined.

7. The method for compensating for stress loss in anchor cables on steep slopes according to claim 6, characterized in that, After determining the sensing and anchorage design scheme and the preset three-level stress compensation threshold based on the anchor cable support foundation parameters, the method further includes: Based on the aforementioned sensing and anchor design scheme, steel strand cutting and anchor cable bundle fabrication are carried out, and distributed optical fiber sensing units are deployed along the cable body to implement anchor cable hole forming, hole cleaning, lowering and bottom grouting operations. After the grouting body has been cured to the required standard, the anchor plate and magnetorheological intelligent self-locking anchor are installed, and the pre-calibrated integrated anchor pressure sensor and displacement sensor are installed on the exposed steel strand of the anchor cable in the design sequence.

8. A stress loss compensation device for anchor cables on steep slopes, characterized in that, include: The full-segment stress module is used to acquire multi-source sensing data of steep slopes through preset multi-source sensing devices, and to sequentially preprocess, verify the validity of and correct the distributed strain data in the multi-source sensing data of steep slopes to obtain corrected distributed strain data. Based on the strain-stress conversion model, the module performs full-segment stress calculation on the corrected distributed strain data to obtain full-segment stress distribution data. The parameters of the strain-stress conversion model are determined by laboratory calibration data and field tension calibration data. The effective stress module is used to calculate and correct the effective stress of the real-time data of the anchor pressure sensor in the multi-source sensing data of the high and steep slope based on the binary linear temperature correction model, so as to obtain the effective prestress value of the anchor end, and determine the total stress loss data of the anchor end based on the effective prestress value of the anchor end and the prestress reference value. The time-series module is used to perform time-by-time full-segment stress calculation and time-by-time effective stress calculation on the time-series data in the multi-source sensing data of the steep slope, generate time-by-time stress dataset, perform damage analysis, and generate damage analysis surface plot. The cause decoupling module is used to decouple and separate the total stress loss data at the anchor end based on the multi-source sensing data of the steep slope, the linear total stress loss superposition model, and the key feature parameters in the time-by-time stress data set to obtain the cause of loss. The compensation module is used to determine the compensation level based on the stress distribution data of the entire section and the total stress loss data of the anchor end, according to the preset three-level stress compensation threshold, generate the compensation level, and perform stress compensation based on the compensation level and the loss cause.

9. An electronic device, characterized in that, Including memory and processor; The memory is used to store computer programs; The processor is configured to implement the method for compensating for stress loss of anchor cables on steep slopes as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the method for compensating for stress loss of anchor cables on steep slopes as described in any one of claims 1 to 7.

Images