Spiral anchor pile instability early warning method based on seabed slot evolution and six-degree-of-freedom response

Abstract

The application provides a spiral anchor pile instability early warning method for seabed slot evolution and six-degree-of-freedom response, comprising the following steps: obtaining seabed monitoring data, six-degree-of-freedom response data, load data and health benchmark data in a judgment window cycle, identifying a slot area caused by cyclic loading and determining slot geometric characteristics, weighting the soil loss amount by a spatial attenuation weight to obtain effective soil loss characteristics, soil loss asymmetry degree and soil loss main direction, extracting residual six-degree-of-freedom response under a unified reference phase to determine residual response characteristics and residual response growth characteristics, identifying a generalized flexibility matrix to determine flexibility drift rate, determining slot severity and response severity according to the flexibility drift rate, determining an instability risk index and outputting a warning level after direction consistency discrimination and time consistency discrimination. The application realizes a cause-and-effect confirmation type graded early warning for slot causes and anchor pile responses, and outputs a remaining safe cycle number while suppressing false positives.

Description

Technical Field

[0001] This application relates to the field of marine engineering anchorage structure monitoring technology, and in particular to a method for early warning of instability of helical anchor piles based on seabed trench evolution and six-degree-of-freedom response. Background Technology

[0002] Helical anchor piles are commonly used anchoring foundations in floating structure mooring systems such as offshore floating wind turbines and floating platforms, offering advantages such as high anchoring efficiency and convenient construction in soft clay seabeds. During long-term service, helical anchor piles bear cyclic mooring loads, with the anchor chain repeatedly contacting the seabed as the load changes, resulting in continuous cutting and erosion between the anchor pile and the seabed, gradually forming asymmetric trenches around the anchor pile. The formation and expansion of these trenches cause soil loss around the anchor, weakening the pull-out and load-bearing support provided by the anchor disc. This leads to residual responses in the helical anchor pile that are coupled with residual lateral displacement, pull-out, tilting, yaw, and torsion, potentially developing into anchor foundation instability and threatening the safety of the floating structure.

[0003] In existing monitoring and early warning methods for helical anchor pile instability, seabed scanning and structural monitoring are usually conducted independently. Seabed scanning only reports the geometry of the trench, while structural monitoring only reports whether the inclination or displacement exceeds limits. These two types of information are disconnected, making it difficult to determine whether the trench is causing the root cause of anchor pile instability. Alarms triggered solely by changes in seabed topography or attitude and displacement thresholds cannot distinguish between pseudo-topographic changes caused by natural seabed erosion and deposition, nor can they differentiate between recoverable elastic cyclic motion and permanent plastic cumulative deformation, resulting in both false alarms and missed alarms. Existing criteria are mostly limited to single-point displacement or inclination, lacking indicators that reflect system-level stiffness degradation, making it difficult to detect earlier instability precursors such as lateral displacement, uplift, tilting, yaw, and enhanced torsional coupling.

[0004] Existing methods lack health benchmarks and causal verification mechanisms. They fail to verify the spatial consistency between the trenching direction and the anchor pile response direction, nor do they verify the temporal causal relationship between trenching evolution precedence and attitude response lag. Furthermore, existing methods only provide alarms without offering lifetime-related information such as the remaining safe cycles, making it difficult to support quantitative decision-making regarding maintenance windows and emergency plans. Summary of the Invention

[0005] To address the aforementioned issues, this application provides a method for early warning of helical anchor pile instability based on seabed trench evolution and six-degree-of-freedom response. This method uses seabed trench evolution as the inducing factor, weighted effective soil loss as the mechanical mediator, six-degree-of-freedom residual response and generalized flexibility drift under a unified load phase as evidence on the anchor pile side, and spatial direction consistency and temporal lag correlation to complete causal confirmation. The above information is then integrated into a dimensionless instability risk index, enabling a causal confirmation-based hierarchical early warning system for helical anchor pile instability. This system suppresses false alarms while outputting the remaining safe cycle number.

[0006] The first aspect of this application proposes a method for early warning of instability of helical anchor piles based on seabed trenching evolution and six-degree-of-freedom response, including the following steps:

[0007] Seabed monitoring data of the seabed area where the helical anchor is located is acquired periodically with a discrimination window as the period. The six-degree-of-freedom response data of the helical anchor under cyclic load in each discrimination window and the load data corresponding to the cyclic load are acquired. The health baseline data of the helical anchor is also acquired. The health baseline data includes the baseline seabed surface, the baseline residual six-degree-of-freedom response and the baseline generalized compliance matrix.

[0008] Based on seabed monitoring data and the baseline seabed surface, the grooving areas caused by cyclic loading in the seabed region are identified, and the grooving geometry of the grooving areas is determined.

[0009] A spatial attenuation weight is set, which decreases as the horizontal and vertical distances of the soil loss unit from the center of the anchor plate of the helical anchor pile increase. The spatial attenuation weight is used to weight and measure the soil loss in the trench area to determine the effective soil loss characteristics, the degree of soil loss asymmetry, and the main direction of soil loss.

[0010] Within each discrimination window, the residual six degrees of freedom response when the cyclic load is in a unified reference phase is extracted from the six degrees of freedom response data. The residual response characteristics are determined based on the residual six degrees of freedom response, and the residual response growth characteristics are determined based on the changing trend of the residual response characteristics.

[0011] Based on the residual six-degree-of-freedom response and load data, the generalized compliance matrix of the helical anchor pile is identified, and the compliance drift rate of the generalized compliance matrix relative to the reference generalized compliance matrix is ​​determined.

[0012] The severity of trenching is determined based on the trenching geometry, effective soil loss characteristics, and the degree of soil loss asymmetry; the severity of response is determined based on the residual response characteristics and the flexibility drift rate.

[0013] The directional consistency coefficient is determined by classifying the main direction of soil loss and the response direction corresponding to the residual response characteristics; the temporal consistency coefficient is determined by classifying the trenching time series of trenching severity and the response time series of response severity for each discrimination window.

[0014] When both the severity of the trenching and the severity of the response meet the preset abnormal conditions, the instability risk index of the helical anchor pile is determined based on the severity of the trenching, the severity of the response, the growth characteristics of the residual response, the directional consistency coefficient, and the time consistency coefficient. The warning level is then determined and output based on the instability risk index.

[0015] The method for early warning of instability of helical anchor piles based on seabed trenching evolution and six-degree-of-freedom response according to the first aspect of the embodiments of this application has at least the following beneficial effects:

[0016] This application combines the analysis of seabed trenching evolution and the six-degree-of-freedom response of helical anchor piles within the same causal verification framework. By using spatial attenuation weights to weight the soil loss within the trenching area, the topographical trenching volume is transformed into a mechanically relevant quantity that truly affects the anchor plate's stress. Based on this, the determined effective soil loss characteristics, the degree of soil loss asymmetry, and the principal direction of soil loss can distinguish between pseudo-topographical changes caused by natural seabed erosion and deposition and soil losses that genuinely weaken the anchor plate's bearing capacity, thus suppressing false alarms at their source.

[0017] By extracting the residual six-degree-of-freedom response of cyclic load under a unified reference phase, the recoverable elastic cyclic motion is stripped away, and only the residual state component representing damage accumulation is retained. The flexibility drift ratio reflects the system-level stiffness degradation caused by softening of the anchorage soil and the enhanced cross-coupling between load channels. This can capture the precursors of instability earlier before the displacement or tilt angle has obviously exceeded the limit, making up for the shortcomings of existing criteria that only rely on single-point displacement or tilt angle thresholds.

[0018] By judging the spatial consistency between the main direction of soil loss and the direction of residual response, and by judging the time lag correlation between the severity of trenching and the severity of response, the causal relationship between trenching and anchor pile response is confirmed, avoiding misjudging other incidental factors occurring simultaneously with trenching as causes of instability. Furthermore, the severity of trenching, the severity of response, the growth characteristics of residual response, the directional consistency coefficient, and the time consistency coefficient are integrated into a dimensionless instability risk index, achieving causal confirmation-based graded early warning. When the risk index continues to rise, the remaining safety cycles are output, providing quantitative decision-making basis for maintenance windows and emergency plans, balancing the timeliness and stability of early warning.

[0019] In one possible implementation, the health baseline data is calibrated over multiple consecutive stable load cycles when the helical anchor piles are installed and no significant trenching has formed in the seabed area, or when the seabed monitoring data is first acquired. When the instability risk index is lower than the preset baseline update threshold, the health baseline data is updated in a rolling manner using the currently acquired seabed monitoring data, the residual six-degree-of-freedom response, and the generalized compliance matrix.

[0020] In one possible implementation, the seabed monitoring data of each discrimination window is subtracted from the reference seabed surface point by point to obtain the soil loss depth of each seabed surface sampling point in the seabed area, and the candidate areas with soil loss depth not less than a preset depth threshold are determined as trenching areas; the preset depth threshold is the larger of the product of the measurement noise statistics of the seabed monitoring data and a preset multiple, and the product of the outer diameter of the helical anchor plate and a preset coefficient; the trenching geometric features are extracted from the trenching area; the trenching geometric features include trenching depth, trenching width, trenching length, trenching volume, trenching eccentricity and trenching main axis direction, and the trenching depth, trenching width, trenching length and trenching eccentricity are dimensionless with respect to the outer diameter of the anchor plate.

[0021] In one possible implementation, each soil loss unit is weighted and accumulated using spatial attenuation weights to obtain the effective soil loss volume; the influence height is determined based on the anchor plate outer diameter, and the reference soil volume is determined based on the anchor plate outer diameter and the influence height; the effective soil loss volume is normalized relative to the reference soil volume to obtain the effective soil loss ratio, which is used as the effective soil loss characteristic; with the anchor plate center as the origin, the horizontal position vector of each soil loss unit is multiplied by the corresponding soil loss amount, and then weighted and accumulated using spatial attenuation weights to obtain the first-order directional moment; the influence radius is determined based on the anchor plate outer diameter, and the first-order directional moment is normalized using the product of the influence radius and the effective soil loss volume as the scale to obtain the slotting asymmetry vector; the direction of the slotting asymmetry vector is used as the main direction of soil loss, and the modulus of the slotting asymmetry vector is used as the degree of soil loss asymmetry.

[0022] In one possible implementation, the unified reference phase is the position where the cyclic load reaches minimum tension within a single cycle, or the position where the cyclic load reaches a preset reference tension; the residual six-degree-of-freedom response is the value or average value of the six-degree-of-freedom response data within a phase interval that does not exceed a preset proportion of a single cycle period before and after the unified reference phase at the unified reference phase; the residual response characteristics include the residual horizontal displacement ratio, the residual upward displacement ratio, the total residual tilt angle, and the residual yaw angle. The residual horizontal displacement ratio and the residual upward displacement ratio are determined based on the normalization of the horizontal residual displacement component and the vertical residual displacement component in the residual six-degree-of-freedom response relative to the outer diameter of the anchor plate, respectively; the total residual tilt angle is determined by constructing an attitude rotation matrix based on the attitude components in the residual six-degree-of-freedom response, applying the attitude rotation matrix to the initial axis unit vector of the helical anchor pile to obtain the current axis unit vector, and then determining the angle between the initial axis unit vector and the current axis unit vector.

[0023] In one possible implementation, the row channels of the generalized compliance matrix correspond to the response channels of the residual six-degree-of-freedom response, and the column channels of the generalized compliance matrix correspond to the load channels of the load data. When identifying the generalized compliance matrix, the load data and the residual six-degree-of-freedom response are dimensionless using the characteristic bearing capacity of the helical anchor pile, the outer diameter of the anchor plate, and a preset reference angle, respectively. Within the sliding identification window, the dimensionless load increment and response increment are fitted using least squares estimation with a regularization term to obtain the generalized compliance matrix. The compliance drift rate is determined based on the ratio of the first matrix norm of the difference between the generalized compliance matrix and the reference generalized compliance matrix to the second matrix norm of the reference generalized compliance matrix.

[0024] In one possible implementation, the residual horizontal displacement direction is determined based on the horizontal residual displacement component, and the tilted horizontal projection direction is determined based on the attitude component; the response direction includes the residual horizontal displacement direction and the tilted horizontal projection direction; the first cosine value of the first directional deviation angle between the principal direction of soil loss and the residual horizontal displacement direction is determined, and the second cosine value of the second directional deviation angle between the principal direction of soil loss and the tilted horizontal projection direction is determined, and the larger of the first cosine value and the second cosine value is used as the directional consistency coefficient; based on the trenching time series and the response time series, the hysteresis cross-correlation is calculated within a preset maximum hysteresis range in which the trenching severity leads the response severity, and the maximum hysteresis cross-correlation value is used as the time consistency coefficient.

[0025] In one possible implementation, the warning level is divided into four preset levels. When the instability risk index falls into different preset grading intervals, the corresponding warning level is determined. When the severity of the slotting and the severity of the response do not simultaneously meet the preset abnormal conditions, only a general abnormality is recorded, and no warning level is output. When at least one of the slotting geometric features, residual response features, and flexibility drift rate reaches the corresponding mandatory warning threshold, the warning level is raised to no less than the corresponding preset mandatory level. The upgrade of the warning level requires confirmation after a first preset number of consecutive discrimination windows meet the target level judgment conditions, and the downgrade of the warning level requires confirmation after a second preset number of consecutive discrimination windows meet the lower level judgment conditions. The second preset number is greater than the first preset number.

[0026] In one possible implementation, when the instability risk index continues to rise within a third preset number of discrimination windows without reaching the preset maximum risk threshold, the remaining safe number of cycles required for the helical anchor pile to reach the maximum risk threshold is calculated based on the rate of increase of the instability risk index. Based on the comparison between the remaining safe number of cycles and the preset cycle number threshold, the lower limit of the warning level is set to be no lower than the corresponding preset lower limit level. When the cyclic load is a non-strictly periodic load, the non-periodic load sequence is converted into an equivalent number of cycles by rainflow counting or segmentation according to the main cycle, and each discrimination window is determined based on the equivalent number of cycles.

[0027] The second aspect of this application proposes a helical anchor pile instability early warning device based on seabed trenching evolution and six-degree-of-freedom response, comprising:

[0028] The data acquisition module is configured to acquire seabed monitoring data of the seabed area where the helical anchor is located in a periodic manner with a discrimination window, acquire the six-degree-of-freedom response data of the helical anchor under cyclic load in each discrimination window and the load data corresponding to the cyclic load, and acquire the health baseline data of the helical anchor; the health baseline data includes the baseline seabed surface, the baseline residual six-degree-of-freedom response and the baseline generalized compliance matrix;

[0029] The slotting identification module is configured to identify slotted areas caused by cyclic loading in the seabed region based on seabed monitoring data and a reference seabed surface, and to determine the slotting geometry of the slotted areas.

[0030] The soil loss measurement module is configured to set a spatial attenuation weight, which decreases as the horizontal and vertical distances of the soil loss unit from the anchor center of the helical anchor pile increase. The spatial attenuation weight is used to weight and measure the soil loss in the trench area to determine the effective soil loss characteristics, the degree of soil loss asymmetry, and the main direction of soil loss.

[0031] The residual response extraction module is configured to extract the residual six degrees of freedom response from the six degrees of freedom response data when the cyclic load is in a unified reference phase within each discrimination window, determine the residual response characteristics based on the residual six degrees of freedom response, and determine the residual response growth characteristics based on the changing trend of the residual response characteristics.

[0032] The compliance identification module is configured to identify the generalized compliance matrix of the helical anchor based on the residual six-degree-of-freedom response and load data, and determine the compliance drift rate of the generalized compliance matrix relative to the reference generalized compliance matrix.

[0033] The severity determination module is configured to determine the severity of the trenching based on the trenching geometry, the effective soil loss characteristics, and the degree of soil loss asymmetry, and to determine the severity of the response based on the residual response characteristics and the compliance drift rate.

[0034] The consistency discrimination module is configured to perform directional consistency discrimination between the main direction of soil loss and the response direction corresponding to the residual response characteristics in order to determine the directional consistency coefficient; and to perform temporal consistency discrimination between the trenching time series of trenching severity and the response time series of response severity in each discrimination window in order to determine the temporal consistency coefficient.

[0035] The early warning output module is configured to determine the instability risk index of the helical anchor pile based on the instability risk index, the response severity, the residual response growth characteristics, the directional consistency coefficient, and the time consistency coefficient when both the severity of the slotting and the severity of the response meet the preset abnormal conditions, and to determine and output the early warning level based on the instability risk index. Attached Figure Description

[0036] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a flowchart illustrating an embodiment of the helical anchor pile instability early warning method based on seabed trenching evolution and six-degree-of-freedom response provided in this application.

[0038] Figure 2 This is a schematic diagram of the structure of a helical anchor pile instability early warning device for seabed trenching evolution and six-degree-of-freedom response provided in an embodiment of this application. Detailed Implementation

[0039] The embodiments of this implementation are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this implementation, and should not be construed as limiting this implementation.

[0040] In the description of this embodiment, it should be understood that the orientation descriptions, such as up, down, front, back, left, right, etc., are based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this embodiment and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this embodiment.

[0041] In the description of this embodiment, "several" means one or more, "multiple" means two or more, "greater than," "less than," and "exceeding" are understood to exclude the stated number, while "above," "below," and "within" are understood to include the stated number. The use of "first" and "second" in the description is merely for distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.

[0042] In the description of this embodiment, unless otherwise explicitly limited, terms such as setting, installing, and connecting should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this embodiment in conjunction with the specific content of the technical solution.

[0043] Please refer to Figure 1 This document illustrates a flowchart of a helical anchor pile instability early warning method based on seabed trenching evolution and six-degree-of-freedom response, according to an embodiment of this application. In this embodiment, the helical anchor pile instability early warning method based on seabed trenching evolution and six-degree-of-freedom response includes the following steps:

[0044] Step 101: Acquire seabed monitoring data of the seabed area where the helical anchor is located in a periodic manner with a discrimination window as the period. Acquire the six-degree-of-freedom response data of the helical anchor under cyclic load in each discrimination window and the load data corresponding to the cyclic load. Also acquire the health baseline data of the helical anchor. The health baseline data includes the baseline seabed surface, the baseline residual six-degree-of-freedom response and the baseline generalized compliance matrix.

[0045] To provide a unified description of the spatial response of the helical anchor pile, a local coordinate system is first established with the initial central axis of the helical anchor pile and the direction of the main mooring load. The vertical axis of the local coordinate system is set along the initial axis of the helical anchor pile, the first horizontal axis is set along the direction of the main horizontal mooring load, and the second horizontal axis is determined according to the right-hand rule. This constitutes a set of spatial reference benchmarks bound to the installation state of the helical anchor pile. All subsequent displacements, angles, and directions are expressed in this local coordinate system.

[0046] In this local coordinate system, the six-degree-of-freedom state vector of the helical anchor pile at any given time is defined as:

[0047]

[0048] in, and These represent the displacement components of the helical anchor pile along the two horizontal axes, respectively. This represents the displacement component of the helical anchor pile along the vertical axis. This indicates the roll angle of the helical anchor pile. This indicates the pitch angle of the helical anchor pile. The yaw angle represents the helical anchor pile. The first three components describe the positional change of the helical anchor pile, and the last three components describe the attitude change of the helical anchor pile. The six components together constitute the six-degree-of-freedom response data of the helical anchor pile at any time.

[0049] Correspondingly, the load data acting on the helical anchor pile is represented by a combined load vector:

[0050]

[0051] in, This represents the vertical load component in the load data corresponding to cyclic loads. This represents the horizontal load component in the load data corresponding to cyclic loads. This represents the torque load component in the load data corresponding to the cyclic load. The three components together constitute the load data acting on the helical anchor pile.

[0052] The outer diameter of the anchor disc of a spiral anchor pile is denoted as The initial burial depth of the anchor plate center of the helical anchor pile is denoted as . These two parameters are fundamental parameters reflecting the geometric dimensions of the helical anchor pile, and are used for the dimensionless processing of various geometric and response quantities in the subsequent process.

[0053] Seabed monitoring data can be collected through multibeam bathymetry, laser scanning, or acoustic imaging scanning. Six-degree-of-freedom response data can be measured by a combination of inertial measurement unit, attitude sensor, and underwater acoustic positioning. Load data can be collected by tension sensors installed at the connection points of anchor chains or helical anchor piles. All types of data are calibrated using the same time reference during the acquisition process to ensure time alignment of data within each discrimination window.

[0054] The health baseline data is calibrated over multiple consecutive stable load cycles when the helical anchor piles are installed and no significant trenching has formed in the seabed area, or when the seabed monitoring data is acquired for the first time. When the instability risk index is lower than the preset baseline update threshold, the health baseline data is updated on a rolling basis using the currently acquired seabed monitoring data, the residual six-degree-of-freedom response, and the generalized compliance matrix.

[0055] The number of continuous stable load cycles used to calibrate health baseline data is denoted as The value range is 20 to 200 cycles, preferably 50 to 100 cycles. Within this calibration window, the seabed surface elevation, residual six-degree-of-freedom response, and generalized compliance matrix are collected and calculated respectively. The calculation results are used as the reference seabed surface, reference residual six-degree-of-freedom response, and reference generalized compliance matrix.

[0056] In practical engineering, if the seabed is already trenched when the helical anchor piles are installed, or if the monitoring equipment is deployed too late to obtain data in a completely trenchless state, the calculation results corresponding to the moment when the seabed monitoring data, six-degree-of-freedom response data and load data are first fully acquired can be used as the engineering baseline.

[0057] During the operation of the helical anchor pile, when the instability risk index is continuously lower than the preset benchmark update threshold and is determined to be in a low-risk operation phase, the health benchmark data is updated on a rolling basis using the seabed monitoring data, residual six-degree-of-freedom response and generalized compliance matrix obtained in the current discrimination window. This allows the health benchmark data to be dynamically corrected as the helical anchor pile is in long-term service, avoiding subsequent discrimination distortion due to the accumulation of initial benchmark deviations.

[0058] Step 102: Based on the seabed monitoring data and the reference seabed surface, identify the grooving areas in the seabed region caused by cyclic loading, and determine the grooving geometry of the grooving areas.

[0059] The seabed monitoring data of each discrimination window is subtracted from the reference seabed surface point by point to obtain the soil loss depth of each seabed surface sampling point in the seabed area, and the candidate area with soil loss depth not less than the preset depth threshold is determined as the trenching area. The preset depth threshold is the larger of the product of the measurement noise statistics of the seabed monitoring data and the preset multiple, and the product of the outer diameter of the helical anchor plate and the preset coefficient.

[0060] Let the seabed surface elevation under healthy baseline conditions be... The seabed surface elevation corresponding to a certain current discrimination window is The soil loss depth at each seabed surface sampling point is obtained by subtracting the two values ​​point by point.

[0061]

[0062] in, The horizontal coordinates of any seabed surface sampling point within the seabed area. For the time identifier of the corresponding discrimination window, The seabed surface elevation is the reference seabed surface. The seabed surface elevation is the value of the seabed monitoring data within the current discrimination window. This represents the soil loss depth at the seabed surface sampling point. When the seabed surface elevation of a certain seabed surface sampling point is higher than the seabed surface elevation of the reference seabed surface, the soil loss depth is set to zero to eliminate the interference of local siltation on trench identification.

[0063] Candidate regions are defined as the set of sampling points where the soil loss depth is not less than a preset depth threshold.

[0064]

[0065] in, Candidate region A preset depth threshold is defined. This preset depth threshold is determined by jointly controlling the measurement noise of the seabed monitoring data and the outer diameter of the helical anchor disc.

[0066]

[0067] in, This refers to the statistical values ​​of measurement noise in seabed monitoring data, specifically the standard deviation of measurement elevation in the seabed monitoring data. This is a preset coefficient, with a value ranging from 0.005 to 0.020, preferably 0.010. This refers to the outer diameter of the anchor disc of the helical anchor pile.

[0068] The larger of the following two factors is taken as the preset depth threshold: the product of the seabed monitoring data measurement noise statistics and a preset multiple, and the product of the helical anchor plate outer diameter and a preset coefficient. This avoids the preset depth threshold being too low and being overwhelmed by measurement noise, or the preset depth threshold being too high and missing the real trench area.

[0069] Spatial scanning resolution of seabed monitoring data Ideally, the value should be controlled at 0.01. Up to 0.10 Between these values, a further optimization is preferred, with a value not exceeding 0.05. This ensures that the identification accuracy of the candidate area matches the outer diameter of the anchor plate of the helical anchor pile.

[0070] Extract the grooving geometric features of the grooving area; the grooving geometric features include grooving depth, grooving width, grooving length, grooving volume, grooving eccentricity and grooving main axis direction, and make the grooving depth, grooving width, grooving length and grooving eccentricity dimensionless with the outer diameter of the anchor plate respectively.

[0071] The trench depth is taken as the maximum value of the soil loss depth within the candidate area:

[0072]

[0073] The trench volume is obtained by integrating the soil loss depth within the candidate region:

[0074]

[0075] Slot width With the length of the slot The slot eccentricity is determined based on the maximum extension scale of the candidate region in two orthogonal directions. The direction of the slotting main axis is determined based on the horizontal distance between the geometric center of the candidate area and the central axis of the helical anchor pile. Determined based on the direction of the principal inertial axis of the candidate region.

[0076] The four length parameters—groove depth, groove width, groove length, and groove eccentricity—are respectively measured using the outer diameter of the anchor disc of the helical anchor pile. Dimensionlessization:

[0077]

[0078] Among them, the slot volume Maintain the original dimensions, do not divide by length, and perform slotting spindle direction. As an angular quantity, it is not involved in the dimensionless length conversion. In this embodiment, the conventional identification ranges of the above-mentioned dimensionless quantities are set to 0 to 1.0, 0 to 2.0, 0 to 10.0, and 0 to 5.0, respectively. The above ranges are only used as the initial algorithm search domain and do not constitute engineering limits.

[0079] Seabed scan data may have incomplete coverage or local missing measurements. When the effective coverage of seabed scan data is less than 70%, the asymmetric vector of the trench area will not be updated temporarily to avoid distortion of direction discrimination due to missing data. When the effective coverage is between 70% and 90%, scalar features such as trench volume are allowed to be updated, while the confidence weight of the direction consistency index is reduced. When the effective coverage is greater than 90%, all trench geometric features and effective soil loss features are calculated according to the standard procedure.

[0080] Step 103: Set the spatial attenuation weight. The spatial attenuation weight decreases as the horizontal and vertical distances of the soil loss unit from the center of the anchor plate of the helical anchor pile increase. The spatial attenuation weight is used to weight and measure the soil loss in the trench area to determine the effective soil loss characteristics, the degree of soil loss asymmetry, and the main direction of soil loss.

[0081] Not all soil losses within the trench area have an equal impact on the stress state of the anchor plate of the helical anchor pile; only the soil losses within the trench that may significantly affect the stress on the anchor plate are given a higher weight. The horizontal radius of influence and the vertical height of influence are defined with the center of the anchor plate of the helical anchor pile as the reference point.

[0082]

[0083]

[0084] in, The horizontal radius of influence is determined based on the outer diameter of the anchor disc of the helical anchor pile. The vertical influence height is determined based on the outer diameter of the anchor plate of the helical anchor pile. The horizontal influence radius coefficient ranges from 1.0 to 3.0, with 2.0 being preferred. The vertical influence height coefficient has a value ranging from 0.5 to 2.0, with 1.0 being preferred.

[0085] For any soil loss unit within the trench area, the horizontal distance from the soil loss unit to the center of the anchor plate of the helical anchor pile is used as the basis for determining the soil loss unit. Vertical distance Set spatial decay weight:

[0086]

[0087] in, The horizontal distance from the soil loss unit to the center of the anchor plate of the helical anchor pile is denoted as . This represents the vertical distance of the soil loss unit from the center of the anchor plate of the helical anchor pile. The horizontal attenuation coefficient ranges from 0.5 to 1.5, with 1.0 being preferred. The vertical attenuation coefficient ranges from 0.5 to 1.5, with 1.0 being preferred. This is the spatial decay weight, with a value between 0 and 1.

[0088] The spatial attenuation weight decreases exponentially with increasing horizontal and vertical distance. Soil loss units that are closer to the center of the anchor plate of the helical anchor pile and have a greater impact on the stress on the anchor plate receive higher weights. Soil loss units that are far from the center of the anchor plate of the helical anchor pile have weights that approach zero. This transforms the trench volume in the sense of topography into a mechanically relevant quantity that truly affects the stress on the anchor plate of the helical anchor pile, which is different from the measurement based solely on geometric volume.

[0089] The effective soil loss volume is obtained by weighting and summing each soil loss unit with spatial attenuation weight.

[0090] The effective soil loss volume is obtained by weighting and summing the soil loss units within the trench area using spatial attenuation weights.

[0091]

[0092] Compared to the trench volume in the trench geometry, the effective soil loss volume better reflects the true impact of soil loss on the stress state of the anchor plate of the helical anchor pile. This avoids misjudgment caused by the equal inclusion of trench soil loss in anchor plates far from the helical anchor pile or those with limited contribution to the anchor plate bearing capacity.

[0093] The influence height is determined based on the outer diameter of the anchor plate. The reference soil volume is determined based on the outer diameter of the anchor plate and the influence height. The effective soil loss volume is normalized relative to the reference soil volume to obtain the effective soil loss ratio. The effective soil loss ratio is used as the characteristic of effective soil loss.

[0094] The influence height is determined based on the outer diameter of the anchor plate of the helical anchor pile, and then the reference soil volume is determined based on the outer diameter of the anchor plate and the influence height of the helical anchor pile.

[0095]

[0096] in, For reference to the soil volume, only the outer diameter of the anchor disc of the helical anchor pile is considered. Vertical influence height Sure.

[0097] Normalizing the effective soil loss volume relative to the reference soil volume yields the effective soil loss ratio:

[0098]

[0099] The effective soil loss ratio is used as the characteristic of effective soil loss. In this embodiment, the key monitoring range of the effective soil loss ratio is 0 to 0.50, where a ratio less than about 0.02 is temporarily treated as a minor loss, about 0.02 to 0.08 is temporarily treated as an observable loss, about 0.08 to 0.20 is temporarily treated as a significant loss, and a ratio greater than about 0.20 is temporarily treated as a severe loss.

[0100] With the anchor plate center as the origin, the horizontal position vector of each soil loss unit is multiplied by the corresponding soil loss amount, and then weighted by spatial attenuation weight and accumulated to obtain the first-order directional moment. The influence radius is determined according to the outer diameter of the anchor plate, and the first-order directional moment is normalized by the product of the influence radius and the effective soil loss volume to obtain the slotting asymmetric vector. The direction of the slotting asymmetric vector is taken as the main direction of soil loss, and the modulus of the slotting asymmetric vector is taken as the degree of soil loss asymmetry.

[0101] Taking the center of the anchor plate of the helical anchor pile as the origin, multiply the horizontal position vector of each soil loss element by the corresponding soil loss amount, then weight them by spatial attenuation weight and sum them up to obtain the first-order directional moment:

[0102]

[0103] in, It is the first-order direction moment. and For the horizontal coordinate components of the anchor plate center of the helical anchor pile, Let be the horizontal position vector of the soil loss element relative to the center of the anchor plate of the helical anchor pile. The first-order directional moment characterizes the directional offset of the weighted soil loss relative to the center of the anchor plate of the helical anchor pile.

[0104] The influence radius is determined based on the outer diameter of the anchor plate of the helical anchor pile. The first-order directional moment is normalized using the product of the influence radius and the effective soil loss volume as the scale to obtain the slotting asymmetric vector:

[0105]

[0106] in, A positive number is set to prevent the denominator from being zero; its value range is 10⁻ 8 Up to 10⁻ 4 .

[0107] The degree of soil loss asymmetry is taken as the smaller value between the trench asymmetry vector magnitude and 1:

[0108]

[0109] The principal direction of soil loss is determined by the two horizontal components of the trenching asymmetric vector:

[0110]

[0111] in, and For the two horizontal components of the slotted asymmetric vector, This is the arctangent function in the four quadrants. The introduction of the asymmetric vector of the trench and the principal direction of soil loss allows the trench area to not only represent the magnitude of the loss as a scalar, but also to represent the directionality of the loss distribution as a vector, providing directly comparable directional quantities for subsequent spatial consistency determination with the residual response direction.

[0112] Step 104: Within each discrimination window, extract the residual six-degree-of-freedom response from the six-degree-of-freedom response data when the cyclic load is in a unified reference phase, determine the residual response characteristics based on the residual six-degree-of-freedom response, and determine the residual response growth characteristics based on the changing trend of the residual response characteristics.

[0113] The unified reference phase is the position where the cyclic load reaches the minimum tension within a single cycle, or the position where the cyclic load reaches the preset reference tension.

[0114] The six-degree-of-freedom response data and load data are synchronously acquired using the same time reference. For each main load cycle, at least 20 valid sampling points are acquired per cycle, preferably 50 to 200. The time synchronization error between different sensor data does not exceed 2% of one main load cycle to ensure the accuracy of phase alignment during residual response extraction.

[0115] The purpose of introducing a unified reference phase is to eliminate the interference of the recoverable elastic motion of the helical anchor pile under cyclic loading on the instability judgment, retaining only the residual state component representing damage accumulation. The residual six-degree-of-freedom response corresponding to the Nth load cycle is denoted as:

[0116]

[0117] in, and These are the two horizontal residual displacement components in the residual six-degree-of-freedom response. The vertical residual displacement component is in the residual six-degree-of-freedom response. , and These are the three attitude components in the residual six-degree-of-freedom response, corresponding to the roll, pitch, and yaw directions, respectively.

[0118] The residual six-degree-of-freedom response is the value or average value of the six-degree-of-freedom response data when the cyclic load is at a unified reference phase, or within a phase interval that does not exceed a preset proportion of a single cycle period before or after the unified reference phase.

[0119] A phase averaging window can be set near the unified reference phase. The window width is 2% to 5% of a single load cycle, preferably 2.5%. Within this window, the average value of the six-degree-of-freedom response data is taken as the residual six-degree-of-freedom response of the corresponding cycle, so as to reduce the impact of single-point sampling noise on the extraction accuracy.

[0120] The residual response characteristics include the residual horizontal displacement ratio, the residual upward displacement ratio, the total residual tilt angle, and the residual yaw angle. The residual horizontal displacement ratio and the residual upward displacement ratio are determined by normalizing the horizontal residual displacement component and the vertical residual displacement component in the residual six-degree-of-freedom response relative to the outer diameter of the anchor plate, respectively.

[0121] The residual horizontal displacement ratio is determined based on the normalization of the horizontal residual displacement components relative to the outer diameter of the anchor plate:

[0122]

[0123] in, and For the first Two horizontal residual displacement components in the residual six-degree-of-freedom response of each cycle. The outer diameter of the anchor disc of the helical anchor pile. The residual horizontal displacement ratio reflects the degree of horizontal lateral displacement of the helical anchor pile relative to the outer diameter of the anchor disc.

[0124] The residual upward displacement ratio is determined based on the normalization of the vertical residual displacement component relative to the outer diameter of the anchor plate:

[0125]

[0126] in, For the first The vertical residual displacement components in the residual six-degree-of-freedom response of each cycle are normalized after taking their absolute values. The residual upward displacement ratio reflects the degree of upward pull-out of the helical anchor relative to the outer diameter of the anchor disc.

[0127] The residual yaw angle is directly taken from the yaw attitude component in the residual six-degree-of-freedom response. .

[0128] In this embodiment, a residual horizontal displacement ratio less than 0.003 is considered minor, 0.003 to 0.010 is considered observable, 0.010 to 0.025 is considered significant, and greater than 0.025 is considered severe; a residual upward displacement ratio less than 0.002 is considered minor, 0.002 to 0.006 is considered observable, 0.006 to 0.015 is considered significant, and greater than 0.015 is considered severe.

[0129] The total residual tilt angle is determined by constructing an attitude rotation matrix based on the attitude components in the residual six-degree-of-freedom response, applying the attitude rotation matrix to the initial axis unit vector of the helical anchor pile to obtain the current axis unit vector, and then determining the angle between the initial axis unit vector and the current axis unit vector.

[0130] The orientation of the helical anchor pile is calculated using an attitude rotation matrix instead of directly using Euler angles to avoid potential ambiguity in Euler angle expressions when the helical anchor pile undergoes significant rotation. This is based on the attitude components in the residual six-degree-of-freedom response. , , Construct the attitude rotation matrix Apply the attitude rotation matrix to the initial axis unit vector of the helical anchor pile. This yields the current axis unit vector:

[0131]

[0132] The total residual tilt angle is determined by the angle between the current axis unit vector and the vertical direction unit vector:

[0133]

[0134] in, It is an inverse cosine function. Let be the initial unit vector of the helical anchor pile's axis, i.e., the unit vector in the vertical direction. The current axis unit vector, The total residual tilt angle is the angle between the initial unit vector of the helical anchor pile and the current unit vector of the helical anchor pile.

[0135] The tilted horizontal projection direction is determined by the horizontal component of the current axis unit vector:

[0136]

[0137] in, The direction of the inclined horizontal projection. and These are the two horizontal components of the current axis unit vector.

[0138] In this embodiment, a total residual tilt angle less than 0.3° is considered minor, 0.3° to 1.0° is considered observable, 1.0° to 2.5° is considered significant, and greater than 2.5° is considered severe; the absolute value of the residual yaw angle less than 0.5° is considered minor, 0.5° to 1.5° is considered observable, 1.5° to 4.0° is considered significant, and greater than 4.0° is considered severe.

[0139] The residual response growth characteristics were calculated using a sliding cyclic window. The sliding window length... The value ranges from 10 to 200 cycles, with 50 cycles being preferred.

[0140] The increase in residual horizontal displacement per hundred cycles is determined by the following formula:

[0141]

[0142] The total increase in tilt angle per hundred cycles is determined by the following formula:

[0143]

[0144] in, The current loop number. The length of the sliding window. The starting point of the window is the cycle number. The two growth values ​​reflect the cumulative growth rate of the residual horizontal displacement ratio and the total residual tilt angle for every 100 cycles.

[0145] In this embodiment, the increase in residual horizontal displacement per hundred cycles is less than 3 × 10⁻ 4 Determined as minor, 3 x 10⁻ 4 An increase of 1×10⁻³ was considered observable, 1×10⁻³ to 3×10⁻³ was considered significant, and greater than 3×10⁻³ was considered severe; the increase in total tilt angle per hundred cycles was less than 0.03° and considered slight, 0.03° to 0.10° and considered observable, 0.10° to 0.30° and considered significant, and greater than 0.30° and considered severe.

[0146] To further characterize whether the deterioration trend is accelerating, the acceleration ratio of the recent growth rate relative to the healthy baseline growth rate is defined:

[0147]

[0148] in, This represents the growth rate within the recent sliding window. The growth rate corresponding to the healthy baseline stage. To prevent positive numbers with a denominator of zero, The acceleration ratio is used to indicate acceleration. An acceleration ratio less than 1.5 is considered as no significant acceleration, 1.5 to 2.5 is considered as slight acceleration, 2.5 to 5.0 is considered as significant acceleration, and a ratio of not less than 5.0 is considered as rapid acceleration.

[0149] The proportion of effective six-degree-of-freedom data within a single discrimination window should be no less than 80%, preferably no less than 95%. When a certain attitude channel experiences a short-term data gap, the remaining displacement and attitude channels can be used to continue forming general anomaly indicators, but the direction consistency indicator dependent on this missing channel is not calculated to ensure that the discrimination results remain reliable even in the case of partial data gaps.

[0150] Step 105: Based on the residual six-degree-of-freedom response and load data, identify the generalized flexibility matrix of the helical anchor pile and determine the flexibility drift rate of the generalized flexibility matrix relative to the reference generalized flexibility matrix.

[0151] The identification of the generalized compliance matrix is ​​based on dimensionless loads and dimensionless responses. The dimensionless load vector is defined as:

[0152]

[0153] in, , , These represent the vertical load component, horizontal load component, and torque load component in the load data corresponding to the cyclic load acting on the helical anchor pile. , , These represent the characteristic bearing capacities of the helical anchor piles in the vertical, horizontal, and torsional directions, respectively. The corresponding unidirectional ultimate bearing capacity, design bearing capacity, or characteristic load amplitude at the healthy reference stage can be taken.

[0154] The dimensionless six-degree-of-freedom response is defined as:

[0155]

[0156] in, , , These are the two horizontal and vertical displacement components in the residual six-degree-of-freedom response, respectively, divided by the outer diameter of the anchor disc of the helical anchor pile. Normalization , , The three attitude components in the residual six-degree-of-freedom response are each divided by a preset reference angle. , Normalization and For example, all are set to 5°. By making the load data and the residual six-degree-of-freedom response dimensionless, the load components and response components with different physical dimensions can be compared and regressed on the same numerical scale.

[0157] The row channels of the generalized flexibility matrix correspond to the response channels of the residual six degrees of freedom response, and the column channels of the generalized flexibility matrix correspond to the load channels of the load data. When identifying the generalized flexibility matrix, the load data and the residual six degrees of freedom response are dimensionless using the characteristic bearing capacity of the helical anchor pile, the outer diameter of the anchor plate, and the preset reference angle, respectively. Within the sliding identification window, the dimensionless load increment and response increment are fitted using least squares estimation with regularization terms to obtain the generalized flexibility matrix.

[0158] Within each sliding recognition window, the generalized compliance matrix satisfies the following relationship with the dimensionless load increment and the dimensionless response increment:

[0159]

[0160] in, The generalized compliance matrix is ​​a 6x3 matrix. Each row corresponds to one response channel in the residual six-degree-of-freedom response, and each column corresponds to one load channel in the load data. The diagonal elements reflect the relationship between loads and responses in the same direction, and the off-diagonal elements reflect the cross-coupling relationship between different load components and different response components. Its specific form is as follows:

[0161]

[0162] In the Within a sliding recognition window, the generalized compliance matrix is ​​obtained through least-squares estimation with regularization terms:

[0163]

[0164] in, For the first The generalized compliance matrix obtained within a sliding recognition window Therefore, the matrix is ​​composed of the dimensionless response increments within the sliding recognition window. Therefore, the matrix formed by the dimensionless load increments within the sliding recognition window is used. This is the regularization coefficient, with a value range of 10⁻ 6 Up to 10⁻², preferably 10⁻ 4 , This is the identity matrix. The purpose of introducing a regularization term is to avoid ill-conditioned solutions in least squares estimation when there is correlation between load channels. The length of the compliance identification window ranges from 20 to 200 cycles, preferably 50 to 100 cycles.

[0165] The flexibility drift rate is determined by the ratio of the first matrix norm of the difference between the generalized flexibility matrix and the reference generalized flexibility matrix to the second matrix norm of the reference generalized flexibility matrix.

[0166] The generalized compliance matrix obtained from the health baseline stage is denoted as... The compliance drift ratio is defined as the ratio of the matrix norm of the difference between the generalized compliance matrix and the reference generalized compliance matrix to the matrix norm of the reference generalized compliance matrix.

[0167]

[0168] in, The generalized compliance matrix obtained by identifying the current window. As the baseline generalized compliance matrix, Let Frobenius be the Frobenius norm of the matrix, where the numerator and denominator both use the same norm definition. To prevent positive numbers with a denominator of zero, This represents the flexibility drift rate.

[0169] In this embodiment, a flexibility drift ratio less than 0.10 is considered stable, 0.10 to 0.25 is considered slightly drifted, 0.25 to 0.50 is considered significantly drifted, and not less than 0.50 is considered severely drifted. When the flexibility drift ratio is greater than 0.80, it can be set as one of the mandatory severe warning trigger conditions, that is, it is considered that the overall stiffness of the helical anchor pile and soil system has substantially deteriorated.

[0170] Compared to simple displacement or tilt threshold criteria, the flexibility drift ratio can more deeply reflect the system-level stiffness degradation caused by the softening of the soil around the helical anchor pile and the enhanced cross-coupling relationship between load channels, thus capturing the instability trend in advance before the response is obviously abnormal.

[0171] Step 106: Determine the severity of trenching based on the trenching geometry, effective soil loss characteristics, and asymmetry of soil loss; determine the severity of response based on residual response characteristics and flexibility drift rate.

[0172] To facilitate a unified comparison of indicators with different dimensions, a standardized linear cutoff scoring function is adopted:

[0173]

[0174] in, To assign values ​​to the specific indicators to be scored, This is the lower limit for when risk begins to be factored in. This is the upper limit of the score, which is 1. This scoring function maps any indicator to the range of 0 to 1, with 0 points for scores below the lower limit and 1 point for scores above the upper limit, and linear interpolation in the middle.

[0175] The severity of trenching is obtained by weighted fusion of three factors: dimensionless trenching depth, dimensionless effective soil loss ratio, and the degree of asymmetry in soil loss.

[0176]

[0177] in, The dimensionless groove depth is obtained by dimensionlessly converting the groove depth in the groove geometry into the outer diameter of the anchor plate of the helical anchor pile. The dimensionless effective soil loss ratio corresponding to the effective soil loss characteristics. To represent the degree of asymmetry in soil loss, three weights are 0.40, 0.35, and 0.25. For example, the allowable ranges are 0.30 to 0.50, 0.25 to 0.45, and 0.15 to 0.35, respectively, with a constraint that the sum of the three weights is 1. The severity of the trenching reflects both the geometric scale of the trenching geometry and the actual impact of soil loss on the anchor plate stress of the helical anchor pile, as well as its spatial asymmetry.

[0178] The severity of the response is determined jointly by the scores corresponding to the residual response features and the compliance drift rate. The score corresponding to the residual response features is the maximum value among the scores of the main residual response indicators.

[0179]

[0180] in, , , , These are the residual horizontal displacement ratio, residual upward displacement ratio, total residual tilt angle, and residual yaw angle, respectively, which are the residual response characteristics.

[0181] The score corresponding to the softness drift rate is defined as follows:

[0182]

[0183] The severity of the response is determined by the larger value between the residual response characteristic corresponding score and the compliance score. That is, the one that reflects the abnormality of the helical anchor pile response more significantly is taken as the severity of the response, which is used to form the basis for determining the type of instability together with the severity of the trenching.

[0184] Step 107: Perform directional consistency judgment on the main direction of soil loss and the response direction corresponding to the residual response characteristics to determine the directional consistency coefficient; perform time consistency judgment on the trenching time series of trenching severity and the response time series of response severity in each judgment window to determine the time consistency coefficient.

[0185] The direction of residual horizontal displacement is determined based on the horizontal residual displacement component, and the direction of tilted horizontal projection is determined based on the attitude component; the response direction includes the direction of residual horizontal displacement and the direction of tilted horizontal projection; the first cosine value of the first directional deviation angle between the principal direction of soil loss and the direction of residual horizontal displacement is determined, and the second cosine value of the second directional deviation angle between the principal direction of soil loss and the direction of tilted horizontal projection is determined, and the larger of the first cosine value and the second cosine value is taken as the directional consistency coefficient.

[0186] The direction of the residual horizontal displacement is determined by the horizontal residual displacement component in the residual six-degree-of-freedom response:

[0187]

[0188] in, and For the first Two horizontal residual displacement components in the residual six-degree-of-freedom response of each cycle.

[0189] The directional consistency coefficient between the principal direction of soil loss and the direction of residual horizontal displacement is determined based on the cosine of the first directional deflection angle:

[0190]

[0191] The directional consistency coefficient between the principal direction of soil loss and the inclined horizontal projection direction is determined based on the cosine value of the second direction deflection angle:

[0192]

[0193] in, The direction of the residual horizontal displacement. The direction of the inclined horizontal projection. The main direction of soil loss. The function is a cosine function, and the first direction deflection angle is... and The included angle between them, and the second direction deflection angle is and The angle between the two directional angles. The cosine value between the two directional angles is mapped to the interval 0 to 1 through affine normalization. The coefficient is 1 when the directions are completely consistent and 0 when they are completely opposite.

[0194] The directional consistency coefficient is the larger of the two directional consistency coefficients mentioned above:

[0195]

[0196] The directional consistency coefficient ranges from 0 to 1. In this embodiment, a value less than 0.50 is considered weak directional correlation, 0.50 to 0.70 is considered medium directional correlation, 0.70 to 0.85 is considered strong directional correlation, and a value not less than 0.85 is considered strong directional correlation.

[0197] The reason for taking the larger value is that the helical anchor pile may have both lateral displacement and tilting responses at the same time. The one with a higher degree of directional agreement is taken as the criterion to avoid missing the real causal relationship due to the single response direction not matching the slotting direction.

[0198] Based on the slotting time series and the response time series, the lag cross-correlation is calculated within a preset maximum lag range where the severity of slotting precedes the severity of response, and the maximum lag cross-correlation value is used as the time consistency coefficient.

[0199] The time consistency coefficient is achieved by searching for the correlation coefficient between the slotting severity time series and the response severity time series within a preset maximum lag range. The maximum search lag is... The value ranges from 0 to 50 loops, with 20 loops being preferred.

[0200]

[0201] in, The time series of grooving severity The response time series is based on the severity of the response. It is a lag quantity. For maximum search lag, For the correlation coefficient function, This is the time consistency coefficient. The calculation result is limited to the range of 0 to 1.

[0202] In this embodiment, a time consistency coefficient less than 0.40 is judged as weak time correlation, 0.40 to 0.60 is judged as medium time correlation, 0.60 to 0.80 is judged as strong time correlation, and not less than 0.80 is judged as strong time correlation.

[0203] By verifying the lagging correlation between the severity of the slotting and the severity of the response, it is confirmed that the abnormal response of the helical anchor pile is indeed driven by the slotting evolution process, rather than by other accidental factors that occur simultaneously with the slotting, thus improving the causal credibility of the instability risk assessment.

[0204] Step 108: When both the severity of the trenching and the severity of the response meet the preset abnormal conditions, determine the instability risk index of the helical anchor pile based on the severity of the trenching, the severity of the response, the growth characteristics of the residual response, the directional consistency coefficient, and the time consistency coefficient, and determine and output the early warning level based on the instability risk index.

[0205] The score corresponding to the residual response growth characteristic is the maximum value among the scores for growth per 100 cycles and speedup:

[0206]

[0207] in, This represents the percentage increase in residual horizontal displacement per hundred cycles in the residual response growth characteristics. This represents the increase in total tilt angle per hundred cycles in the residual response growth characteristic. This represents the acceleration ratio in the residual response growth characteristic.

[0208] The comprehensive instability risk index is obtained by weighted fusion of six indicators:

[0209]

[0210] in, Depending on the severity of the grooving, The residual response score corresponds to the severity of the response. The growth rate score corresponds to the growth characteristics of the residual response. The softness score corresponds to the softness drift rate. The directional consistency coefficient. This is the time consistency coefficient.

[0211] In this embodiment, the ranges and preferred values ​​of the six weights are as follows: weight for grooving severity 0.20 to 0.35, preferably 0.25; weight for residual response score 0.20 to 0.35, preferably 0.25; weight for growth rate score 0.15 to 0.25, preferably 0.20; weight for flexibility score 0.10 to 0.20, preferably 0.15; weight for directional consistency 0.05 to 0.15, preferably 0.10; and weight for temporal consistency 0.02 to 0.10, preferably 0.05. The sum of each weight is constrained to 1.

[0212] The warning level is divided into four preset levels. When the instability risk index falls into different preset grading intervals, the corresponding warning level is determined. When the severity of the slotting and the severity of the response do not simultaneously meet the preset abnormal conditions, only a general abnormality is recorded and no warning level is output. When at least one of the slotting geometric features, residual response features, and flexibility drift rate reaches the corresponding mandatory warning threshold, the warning level is raised to no less than the corresponding preset mandatory level. The upgrade of the warning level requires confirmation after the first preset number of consecutive discrimination windows meet the target level judgment conditions, and the downgrade of the warning level requires confirmation after the second preset number of consecutive discrimination windows meet the lower level judgment conditions. The second preset number is greater than the first preset number.

[0213] To avoid false alarms caused by simple seabed topographic changes or short-term attitude fluctuations, anomalies must simultaneously exist in both the seabed domain and the helical anchor response domain to determine whether a trench-induced instability is induced. In this embodiment, the dual-domain triggering condition is that both the trench severity and the response severity are not less than 0.20. If neither condition is met simultaneously, only a general anomaly is recorded, no warning level is output, and it is not directly determined as trench-induced instability.

[0214] Under the premise of meeting the dual-domain triggering conditions, the early warning level is divided into four preset levels according to the degree of instability risk from low to high. The criteria set in this embodiment are as follows: when the instability risk index is less than 0.35, it is judged as Level I, corresponding to a stable or slightly slotted state, and routine monitoring is maintained; when the instability risk index is between 0.35 and 0.55, it is judged as Level II, corresponding to a state of continuous change in slotting or residual response, and the monitoring frequency is increased; when the instability risk index is between 0.55 and 0.75, it is judged as Level III, corresponding to a state of obvious coupling between slotting and attitude response, and entering the instability early warning state; when the instability risk index is not less than 0.75, it is judged as Level IV, corresponding to a state of rapid deterioration of residual response and system flexibility, and entering the severe early warning state.

[0215] To ensure safety, several mandatory early warning thresholds are set. When any one of the following conditions is met: dimensionless groove depth not less than 0.50, total residual tilt angle not less than 3.5°, or residual horizontal displacement ratio not less than 0.035, the early warning level is forcibly raised to level III or higher; when any one of the following conditions is met: total residual tilt angle not less than 5.0°, residual horizontal displacement ratio not less than 0.050, or compliance drift rate not less than 0.80, the early warning level is directly raised to level IV.

[0216] To suppress level jumps caused by a single anomaly, an upgrade of the warning level requires confirmation after a first preset number of consecutive discrimination windows meet the target level determination criteria, while a downgrade requires confirmation after a second preset number of consecutive discrimination windows meet the criteria for the next lower level, with the second preset number being greater than the first preset number. The first preset number ranges from 3 to 10 discrimination windows, preferably 5; the second preset number ranges from 5 to 20 discrimination windows, preferably 10. This forms an asymmetric upgrade / downgrade hysteresis mechanism, allowing the warning level to be upgraded more quickly when an abnormal state occurs, and then downgraded after more thorough and continuous observation when the state improves, balancing the timeliness and stability of the warning.

[0217] When the instability risk index continues to rise within the third preset number of discrimination windows without reaching the preset maximum risk threshold, the remaining safe number of cycles required for the helical anchor pile to reach the maximum risk threshold is calculated based on the rate of increase of the instability risk index. Based on the comparison between the remaining safe number of cycles and the preset cycle number threshold, the lower limit of the warning level is set to be no lower than the corresponding preset lower limit level. When the cyclic load is a non-strictly periodic load, the non-periodic load sequence is converted into an equivalent number of cycles by rainflow counting or segmentation according to the main cycle, and each discrimination window is determined based on the equivalent number of cycles.

[0218] The rate of increase of the instability risk index is determined by the following formula:

[0219]

[0220] in, This is the current cycle's instability risk index. The instability risk index for the starting point cycle of the window. The sliding window length is also used as the third preset number to determine the continuous rise of the instability risk index. The value ranges from 10 to 200 cycles, with 50 cycles being preferred.

[0221] When the rate of increase is greater than zero and has not yet reached the preset maximum risk threshold, the remaining number of safe cycles is estimated using the following formula:

[0222]

[0223] in, The preset maximum risk threshold is set to 0.75. This represents the current risk index of instability. The rate of increase of the instability risk index. This represents the remaining safe cycle number.

[0224] The additional criterion set in this embodiment is that when the remaining safe cycles are less than 500, the lower limit of the warning level is set to no less than Level III; when the remaining safe cycles are less than 100, the lower limit of the warning level is set to no less than Level IV. The remaining safe cycles serve as both a predictive output reference and a hard constraint on the warning level, avoiding delays in warning response due to a clear upward trend in the risk index that has not yet reached the classification threshold.

[0225] When the cyclic load is a non-strictly periodic load, the non-periodic load sequence is converted into an equivalent cycle number through rainflow counting or segmentation according to the main cycle period, and each discrimination window is determined based on the equivalent cycle number. For load histories exhibiting non-strictly periodic characteristics under the combined action of waves and wind waves, the rainflow counting method or segmentation according to the dominant cycle can be used to convert the actual time history into an equivalent cycle number, making this method feasible under irregular sea state load conditions.

[0226] The algorithm operation process of this embodiment is illustrated using a specific discrimination window as an example. It is assumed that this discrimination window obtains the dimensionless slot depth. The dimensionless effective soil loss ratio is 0.18. The asymmetric degree of soil loss is 0.10. The value is 0.65. Substituting this into the formula for calculating the severity of the grooving, the severity of the grooving is approximately 0.57. The residual horizontal displacement ratio measured at the same time is 0.012, the residual upward displacement ratio is 0.006, the total residual tilt angle is 1.4°, and the absolute value of the residual yaw angle is 1.2°. Substituting these values ​​into the formula for calculating the residual response score, the residual response score is approximately 0.50.

[0227] If the measured increase in residual horizontal displacement per 100 cycles is 1.5 × 10⁻³, the increase in total tilt angle per 100 cycles is 0.12°, and the acceleration ratio is 2.6, substituting these values ​​into the growth rate scoring formula yields a growth rate score of approximately 0.44. If the compliance drift rate is 0.32, substituting these values ​​into the compliance scoring formula yields a compliance score of approximately 0.55. If the directional consistency coefficient is 0.84 and the temporal consistency coefficient is 0.72, substituting these values ​​into the comprehensive instability risk index calculation formula yields an instability risk index of approximately 0.56.

[0228] In this state, both the slotting severity and response severity exceed 0.20, simultaneously meeting the dual-domain triggering condition. According to the preset threshold in this embodiment, it is determined to be a Level III instability warning state. If the subsequent five consecutive judgment windows still meet the Level III condition, an upgrade is confirmed; if the instability risk index continues to rise, the remaining safe cycles are further calculated.

[0229] Each discrimination window should output at least the current geometric state of the seabed trench and its positional relationship relative to the helical anchor piles, the effective soil loss ratio and the degree of soil loss asymmetry and the principal direction of soil loss, the current six-degree-of-freedom residual response and its growth per hundred cycles, the current generalized compliance matrix and compliance drift rate, the coupling coefficient and time consistency coefficient between the trenching direction and the residual displacement direction and tilt direction, the trenching severity and response severity, the comprehensive instability risk index, the corresponding warning level, and the data quality status when data is insufficient. The above outputs provide maintenance personnel with relatively complete discrimination criteria and status backtracking capabilities.

[0230] The formula given in this embodiment above provides a directly implementable technical path. In practical applications, it is permissible to calibrate the spatial weight function form, the weights of various aspects of trenching severity, the residual response threshold, the flexibility drift threshold, the directional consistency threshold, the time lag window, the comprehensive risk weight, the four-level early warning threshold, and the remaining cycle number prediction model, without altering the technical chain of seabed trenching evolution and six-degree-of-freedom response coupling discrimination. All provisional thresholds are preferentially calibrated through joint calibration using model trench experiments, large deformation numerical simulations, and field monitoring data. During the implementation phase when data volume is insufficient, the dimensionless range given in this application is used as the initial value, and it is gradually updated by adding new data.

[0231] like Figure 2 The diagram illustrates a schematic of the helical anchor pile instability early warning device based on seabed trenching evolution and six-degree-of-freedom response provided in this application. In this embodiment, the helical anchor pile instability early warning device based on seabed trenching evolution and six-degree-of-freedom response includes:

[0232] The data acquisition module 201 is configured to acquire seabed monitoring data of the seabed area where the helical anchor is located in a periodic manner with a discrimination window, acquire the six-degree-of-freedom response data of the helical anchor under cyclic load in each discrimination window and the load data corresponding to the cyclic load, and acquire the health baseline data of the helical anchor; the health baseline data includes the baseline seabed surface, the baseline residual six-degree-of-freedom response and the baseline generalized compliance matrix;

[0233] The slotting identification module 202 is configured to identify slotted areas caused by cyclic loading in the seabed region based on seabed monitoring data and a reference seabed surface, and to determine the slotting geometry of the slotted areas.

[0234] The soil loss measurement module 203 is configured to set a spatial attenuation weight, which decreases as the horizontal and vertical distances of the soil loss unit from the anchor center of the helical anchor pile increase; the spatial attenuation weight is used to weight and measure the soil loss in the trench area to determine the effective soil loss characteristics, the degree of soil loss asymmetry, and the main direction of soil loss.

[0235] The residual response extraction module 204 is configured to extract the residual six degrees of freedom response from the six degrees of freedom response data when the cyclic load is in a unified reference phase within each discrimination window, determine the residual response characteristics based on the residual six degrees of freedom response, and determine the residual response growth characteristics based on the changing trend of the residual response characteristics.

[0236] The compliance identification module 205 is configured to identify the generalized compliance matrix of the helical anchor based on the residual six-degree-of-freedom response and load data, and determine the compliance drift rate of the generalized compliance matrix relative to the reference generalized compliance matrix.

[0237] The severity determination module 206 is configured to determine the severity of the trenching based on the trenching geometry, the effective soil loss characteristics and the degree of soil loss asymmetry, and to determine the severity of the response based on the residual response characteristics and the compliance drift rate.

[0238] The consistency discrimination module 207 is configured to perform directional consistency discrimination on the main direction of soil loss and the response direction corresponding to the residual response characteristics to determine the directional consistency coefficient; and to perform time consistency discrimination on the trenching time series of trenching severity and the response time series of response severity in each discrimination window to determine the time consistency coefficient.

[0239] The early warning output module 208 is configured to determine the instability risk index of the helical anchor pile based on the instability risk index, the response severity, the residual response growth characteristics, the directional consistency coefficient, and the time consistency coefficient when both the severity of the slotting and the severity of the response meet the preset abnormal conditions, and to determine and output the early warning level based on the instability risk index.

[0240] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments 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 unit can be implemented in hardware or as a software functional unit.

[0241] If an integrated module / unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program may include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. A computer-readable medium may include: any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0242] It should be noted that the content contained in computer-readable media may be appropriately added to or subtracted from the requirements of legislation and patent practice in a jurisdiction. For example, in some jurisdictions, computer-readable media may not include electrical carrier signals and telecommunication signals, in accordance with legislation and patent practice.

[0243] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this implementation. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0244] Although embodiments of this implementation have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of this implementation, the scope of which is defined by the claims and their equivalents.

Claims

1. A method for early warning of instability of helical anchor piles based on seabed trenching evolution and six-degree-of-freedom response, characterized in that, include: Seabed monitoring data of the seabed area where the helical anchor is located is acquired periodically with a discrimination window as the period. The six-degree-of-freedom response data of the helical anchor under cyclic load in each discrimination window and the load data corresponding to the cyclic load are acquired. The health baseline data of the helical anchor is also acquired. The health baseline data includes the baseline seabed surface, the baseline residual six-degree-of-freedom response, and the baseline generalized compliance matrix; Based on the seabed monitoring data and the reference seabed surface, identify the grooving area in the seabed region caused by the cyclic load, and determine the grooving geometry of the grooving area; A spatial attenuation weight is set, which decreases as the horizontal and vertical distances of the soil loss unit from the center of the anchor plate of the helical anchor pile increase; the spatial attenuation weight is used to weight and measure the soil loss in the trench area to determine the effective soil loss characteristics, the degree of soil loss asymmetry, and the main direction of soil loss. Within each of the discrimination windows, the residual six degrees of freedom response when the cyclic load is in a unified reference phase is extracted from the six degrees of freedom response data. The residual response characteristics are determined based on the residual six degrees of freedom response, and the residual response growth characteristics are determined based on the changing trend of the residual response characteristics. Based on the residual six-degree-of-freedom response and the load data, the generalized compliance matrix of the helical anchor is identified, and the compliance drift rate of the generalized compliance matrix relative to the reference generalized compliance matrix is ​​determined. The severity of trenching is determined based on the trenching geometry, the effective soil loss characteristics, and the degree of soil loss asymmetry; the severity of response is determined based on the residual response characteristics and the compliance drift rate. The direction consistency between the main direction of soil loss and the response direction corresponding to the residual response characteristics is determined to establish a direction consistency coefficient; the time consistency between the trenching time series of the trenching severity and the response time series of the response severity in each discrimination window is determined to establish a time consistency coefficient. When both the severity of the slotting and the severity of the response meet the preset abnormal conditions, the instability risk index of the helical anchor pile is determined based on the severity of the slotting, the severity of the response, the residual response growth characteristics, the directional consistency coefficient, and the time consistency coefficient, and the warning level is determined and output based on the instability risk index.

2. The method according to claim 1, characterized in that, The health baseline data is calibrated over multiple consecutive stable load cycles when the helical anchor piles are installed and the seabed area has not yet formed significant trenches, or when the seabed monitoring data is first acquired. When the instability risk index is lower than the preset benchmark update threshold, the health benchmark data is updated in a rolling manner using the currently acquired seabed monitoring data, the residual six-degree-of-freedom response, and the generalized compliance matrix.

3. The method according to claim 1, characterized in that, The step of identifying the grooving region caused by the cyclic load in the seabed area based on the seabed monitoring data and the reference seabed surface, and determining the grooving geometry of the grooving region, includes: The seabed monitoring data of each discrimination window is subtracted from the reference seabed surface point by point to obtain the soil loss depth of each seabed surface sampling point in the seabed area, and the candidate area with the soil loss depth not less than a preset depth threshold is determined as the trenching area; the preset depth threshold is the larger value of the product of the measurement noise statistics of the seabed monitoring data and a preset multiple, and the product of the outer diameter of the helical anchor plate and a preset coefficient; Extract the grooved geometric features from the grooved area; the grooved geometric features include groove depth, groove width, groove length, groove volume, groove eccentricity and groove main axis direction, and make the groove depth, groove width, groove length and groove eccentricity dimensionless with respect to the outer diameter of the anchor plate.

4. The method according to claim 1, characterized in that, The step of weighting the soil loss within the trench area using the spatial attenuation weight to determine the effective soil loss characteristics, the degree of soil loss asymmetry, and the principal direction of soil loss includes: The effective soil loss volume is obtained by weighting and summing each soil loss unit using the spatial attenuation weight. The influence height is determined based on the outer diameter of the anchor plate, and the reference soil volume is determined based on the outer diameter of the anchor plate and the influence height. The effective soil loss volume is normalized relative to the reference soil volume to obtain the effective soil loss ratio. The effective soil loss ratio is used as the effective soil loss characteristic. Taking the center of the anchor plate as the origin, the horizontal position vector of each soil loss unit is multiplied by the corresponding soil loss amount, and then weighted by the spatial attenuation weight and accumulated to obtain the first-order directional moment; the influence radius is determined according to the outer diameter of the anchor plate, and the first-order directional moment is normalized using the product of the influence radius and the effective soil loss volume as the scale to obtain the slotting asymmetric vector; the direction of the slotting asymmetric vector is taken as the main direction of soil loss, and the modulus of the slotting asymmetric vector is taken as the degree of soil loss asymmetry.

5. The method according to claim 1, characterized in that, The unified reference phase is the position where the cyclic load reaches the minimum tension in a single cycle, or the position where the cyclic load reaches the preset reference tension. The residual six-degree-of-freedom response is the value or average value of the six-degree-of-freedom response data within a phase interval that is at the unified reference phase or within a preset proportion of a single cycle before and after the unified reference phase. The residual response characteristics include residual horizontal displacement ratio, residual upward displacement ratio, total residual tilt angle and residual yaw angle. The residual horizontal displacement ratio and the residual upward displacement ratio are determined based on the normalization of the horizontal residual displacement component and the vertical residual displacement component in the residual six-degree-of-freedom response relative to the outer diameter of the anchor plate, respectively. The total residual tilt angle is determined by constructing an attitude rotation matrix based on the attitude components in the residual six-degree-of-freedom response, applying the attitude rotation matrix to the initial axis unit vector of the helical anchor pile to obtain the current axis unit vector, and then determining the angle between the initial axis unit vector and the current axis unit vector.

6. The method according to claim 1, characterized in that, The row channels of the generalized flexibility matrix correspond to each response channel of the residual six-degree-of-freedom response, and the column channels of the generalized flexibility matrix correspond to each load channel of the load data. When identifying the generalized compliance matrix, the load data and the residual six-degree-of-freedom response are dimensionless using the characteristic bearing capacity of the helical anchor pile, the outer diameter of the anchor plate, and the preset reference angle, respectively. The dimensionless load increment and response increment are fitted with least squares estimation with regularization term within the sliding identification window to obtain the generalized compliance matrix. The flexibility drift rate is determined based on the ratio of the first matrix norm of the difference between the generalized flexibility matrix and the reference generalized flexibility matrix to the second matrix norm of the reference generalized flexibility matrix.

7. The method according to claim 5, characterized in that, The step of determining the directional consistency coefficient by classifying the principal direction of soil loss and the response direction corresponding to the residual response characteristics; and the step of determining the temporal consistency coefficient by classifying the trenching severity time series and the response severity time series of each discrimination window, including: The residual horizontal displacement direction is determined based on the horizontal residual displacement component, and the tilted horizontal projection direction is determined based on the attitude component; the response direction includes the residual horizontal displacement direction and the tilted horizontal projection direction. Determine the first cosine value of the first directional deviation angle between the principal direction of soil loss and the direction of residual horizontal displacement, and determine the second cosine value of the second directional deviation angle between the principal direction of soil loss and the direction of inclined horizontal projection. The larger of the first cosine value and the second cosine value is taken as the directional consistency coefficient. Based on the slotting time series and the response time series, a lag cross-correlation is calculated within a preset maximum lag range in which the slotting severity precedes the response severity, and the maximum lag cross-correlation value is used as the time consistency coefficient.

8. The method according to claim 1, characterized in that, The warning level is divided into four preset levels. When the instability risk index falls into different preset grading ranges, the corresponding warning level is determined. When the severity of the slotting and the severity of the response do not simultaneously meet the preset abnormal conditions, only a general abnormality is recorded and the warning level is not output. When at least one of the slotting geometry, the residual response feature, and the compliance drift rate reaches the corresponding mandatory warning threshold, the warning level will be raised to no less than the corresponding preset mandatory level. The upgrade of the warning level requires confirmation after a first preset number of the discrimination windows meet the target level judgment conditions. The downgrade of the warning level requires confirmation after a second preset number of the discrimination windows meet the lower level judgment conditions. The second preset number is greater than the first preset number.

9. The method according to claim 1, characterized in that, When the instability risk index continues to rise within the third preset number of discrimination windows and does not reach the preset maximum risk threshold, the remaining safe cycle number required for the helical anchor pile to reach the maximum risk threshold is calculated based on the rate of increase of the instability risk index, and the lower limit of the warning level is set to be no lower than the corresponding preset lower limit level based on the comparison result between the remaining safe cycle number and the preset cycle number threshold. When the cyclic load is a non-strictly periodic load, the non-periodic load sequence is converted into an equivalent cycle number by rainflow counting or segmentation according to the main cycle period, and each discrimination window is determined based on the equivalent cycle number.

10. A helical anchor pile instability early warning device based on seabed trenching evolution and six-degree-of-freedom response, characterized in that, include: The data acquisition module is configured to acquire seabed monitoring data of the seabed area where the helical anchor is located in a periodic manner with a discrimination window, acquire the six-degree-of-freedom response data of the helical anchor under cyclic load in each discrimination window and the load data corresponding to the cyclic load, and acquire the health benchmark data of the helical anchor. The health baseline data includes the baseline seabed surface, the baseline residual six-degree-of-freedom response, and the baseline generalized compliance matrix; The slotting identification module is configured to identify slotted areas caused by the cyclic load in the seabed region based on the seabed monitoring data and the reference seabed surface, and to determine the slotting geometry of the slotted areas; The soil loss measurement module is configured to set a spatial attenuation weight, which decreases as the horizontal and vertical distances of the soil loss unit from the center of the anchor plate of the helical anchor pile increase; the spatial attenuation weight is used to weight and measure the soil loss in the trench area to determine the effective soil loss characteristics, the degree of soil loss asymmetry, and the main direction of soil loss. The residual response extraction module is configured to extract the residual six degrees of freedom response of the cyclic load when it is in a unified reference phase from the six degrees of freedom response data within each of the discrimination windows, determine the residual response characteristics based on the residual six degrees of freedom response, and determine the residual response growth characteristics based on the changing trend of the residual response characteristics. The compliance identification module is configured to identify the generalized compliance matrix of the helical anchor based on the residual six-degree-of-freedom response and the load data, and determine the compliance drift rate of the generalized compliance matrix relative to the reference generalized compliance matrix. The severity determination module is configured to determine the severity of the trenching based on the trenching geometry, the effective soil loss characteristics, and the degree of soil loss asymmetry, and to determine the severity of the response based on the residual response characteristics and the compliance drift rate. The consistency discrimination module is configured to perform directional consistency discrimination between the main direction of soil loss and the response direction corresponding to the residual response characteristics to determine the directional consistency coefficient; and to perform temporal consistency discrimination between the trenching time series of the trenching severity and the response time series of the response severity in each discrimination window to determine the temporal consistency coefficient. The early warning output module is configured to determine the instability risk index of the helical anchor pile based on the instability risk index, the response severity, the residual response growth characteristics, the directional consistency coefficient, and the time consistency coefficient when both the instability severity and the response severity meet preset abnormal conditions, and to determine and output the early warning level based on the instability risk index.

Images