On-line monitoring and alarming system for settlement of power cable terminal based on multi-sensor cooperation

Abstract

This invention relates to the field of cable monitoring and early warning technology, specifically to an online monitoring and alarm system for power cable terminal settlement based on multi-sensor collaboration. The system includes a memory and a processor. The processor executes a computer program stored in the memory to: acquire current and tilt angle data of the three phases of the cable terminal at each moment within the current monitoring cycle; construct a closed-loop current-tilt-angle curve for the current monitoring cycle; analyze the slope and area distributions of the closed-loop current-tilt-angle curves for each phase to obtain settlement characteristic indicators for each phase; and provide graded early warnings for cable terminal fault conditions based on the settlement characteristic indicators of each phase in the current monitoring cycle, combined with the settlement characteristic indicators of each phase in historical monitoring cycles. This invention avoids missed reports or resource misallocation caused by single threshold judgments, thus improving the accuracy of cable terminal fault monitoring results.

Description

Technical Field

[0001] This invention relates to the field of cable monitoring and early warning technology, specifically to an online monitoring and alarm system for power cable terminal settlement based on multi-sensor collaboration. Background Technology

[0002] As a crucial link connecting underground cables and overhead lines, power cable terminals are typically installed on outdoor steel supports or concrete towers. During long-term operation, changes in geological conditions and disturbances from surrounding construction may cause uneven settlement of the terminal tower foundation, thereby altering the boundary constraints of the terminal structure and resulting in abnormal stress at the connection between the cable body and the rigid terminal.

[0003] In actual operation, cable terminals are simultaneously affected by two types of factors: First, the periodic fluctuations of load current and the diurnal variations of solar radiation drive significant thermal expansion and contraction in the cable conductor and insulation layer. This phenomenon, known as "thermal breathing," causes substantial, reversible periodic thermal displacement at the terminal. Second, changes in geological conditions or disturbances from surrounding construction may trigger uneven settlement of the tower foundation, leading to minute, irreversible settlement displacement at the terminal. The amplitude of thermal displacement caused by thermal breathing is often one or even several orders of magnitude higher than the minute displacement caused by early settlement. The two are completely superimposed in tilt angle monitoring data and are difficult to separate. Conventional monitoring methods, such as single tilt angle threshold monitoring methods, struggle to distinguish between normal thermal drift and abnormal settlement drift, easily leading to false alarms or missed alarms. Summary of the Invention

[0004] To address the technical problem that existing single tilt angle threshold monitoring methods struggle to distinguish between normal thermal drift and abnormal settlement drift, leading to inaccurate cable terminal fault monitoring results, this invention aims to provide an online monitoring and alarm system for power cable terminal settlement based on multi-sensor collaboration. This system includes a memory and a processor, wherein the processor executes a computer program stored in the memory to achieve the following steps:

[0005] Acquire the current and tilt angle data of the three phases of the cable terminal at each moment in the current monitoring cycle, and construct the current and tilt angle closed curve in the current monitoring cycle based on the time offset correlation of the current and tilt angle data of each phase in the current monitoring cycle.

[0006] Based on the slope distribution and area distribution of the current tilt angle closed curve of each phase, the slope deviation and area deviation of each phase are analyzed to obtain the settlement characteristic index of each phase.

[0007] Based on the settlement characteristic indicators of each phase in the current monitoring cycle, and combined with the settlement characteristic indicators of each phase in the historical monitoring cycles, the fault status of the cable terminal is classified and warned.

[0008] Preferably, the step of constructing a current-tilt angle closed curve for the current monitoring period based on the time offset correlation between the current data and tilt angle data of each phase in the current monitoring period specifically includes:

[0009] The optimal time length is determined based on the correlation between the current data and tilt angle data of each phase under different time displacement lengths.

[0010] The tilt data within the current monitoring period are shifted over time according to the optimal time length to obtain the reconstructed tilt angle for the current monitoring period;

[0011] Using the current data within the current monitoring period as the x-axis and the reconstructed tilt angle within the current monitoring period as the y-axis, a closed curve of the current tilt angle within the current monitoring period is constructed.

[0012] Preferably, determining the optimal time length based on the correlation between the current data and tilt angle data of each phase under different time displacement lengths specifically includes:

[0013] For any phase, the tilt angle data within the current monitoring cycle is time-shifted according to each different time displacement length to obtain the shifted tilt angle at each time displacement length; the cross-correlation function value between the absolute value of the shifted tilt angle at each time displacement length within the current monitoring cycle and the current data at the same time sequence is calculated, and the time displacement length corresponding to the maximum value of the cross-correlation function value is taken as the optimal time length for that phase.

[0014] Preferably, the step of analyzing the slope deviation and area deviation of each phase based on the slope distribution and area distribution of the current tilt angle closed curve of each phase to obtain the settlement characteristic index of each phase specifically includes:

[0015] Linear regression fitting is performed on all points on the current tilt angle closed curve of each phase, and the slope of the fitted line is used as the trend coefficient corresponding to the current tilt angle closed curve of each phase; the area formed by the current tilt angle closed curve of each phase is used as the geometric coefficient corresponding to the current tilt angle closed curve of each phase.

[0016] The trend deviation of each phase is obtained based on the deviation between the trend coefficient of each phase and the overall distribution of the trend coefficients of the three phases; the geometric deviation of each phase is obtained based on the deviation between the geometric coefficient of each phase and the overall distribution of the geometric coefficients of the three phases.

[0017] Settlement characteristic indices for each phase are determined based on the Euclidean norm of the trend deviation and geometric deviation of each phase.

[0018] Preferably, the step of obtaining the trend deviation degree of each phase based on the deviation between the trend coefficient of each phase and the overall distribution of the trend coefficients of the three phases specifically includes:

[0019] The median of the trend coefficients of the three phases is obtained as the three-phase trend characteristic value. The absolute value of the difference between the trend coefficient of each phase and the three-phase trend characteristic value is taken as the trend difference of each phase. The trend deviation of each phase is determined based on the ratio between the trend difference of each phase and the absolute value of the three-phase trend characteristic value.

[0020] Preferably, the step of obtaining the geometric deviation degree of each phase based on the deviation between the geometric coefficient of each phase and the overall distribution of the geometric coefficients of the three phases specifically includes:

[0021] The median of the geometric coefficients of the three phases is obtained as the three-phase geometric characteristic value. The difference between the geometric coefficient of each phase and the three-phase geometric characteristic value is taken as the geometric difference of each phase. Based on the trend difference of each phase and the three-phase geometric characteristic value, the geometric deviation of each phase is determined.

[0022] Preferably, for any phase, when the absolute value of the trend coefficient is greater than the preset tilt threshold, or when the maximum value of the absolute value of the tilt angle data of that phase in the current monitoring period is greater than the preset angle threshold, the highest level emergency warning signal is issued for the fault status of that phase at the cable terminal.

[0023] Preferably, the step of classifying and issuing early warnings for cable terminal faults based on the settlement characteristic indicators of each phase in the current monitoring cycle, combined with the settlement characteristic indicators of each phase in historical monitoring cycles, specifically includes:

[0024] For any phase, if the settlement characteristic index of the phase is greater than the preset first settlement threshold during the current monitoring period, a first-level early warning signal will be issued for the fault status of the phase at the cable terminal.

[0025] If the settlement characteristic index of the phase is less than or equal to the preset first settlement threshold in the current monitoring period, then based on the continuous distribution of the settlement characteristic index of the phase in the historical monitoring periods before the current monitoring period, it is determined whether the fault condition of the cable terminal should respond to the secondary early warning signal.

[0026] Preferably, determining whether a fault condition at the cable terminal should trigger a secondary early warning signal based on the continuous distribution of settlement characteristic indicators of the phase in historical monitoring cycles prior to the current monitoring cycle specifically includes:

[0027] If, within a monitoring period window consisting of a first preset number of historical monitoring periods prior to the current monitoring period, the settlement characteristic index of at least a second preset number of historical monitoring periods is greater than a preset second settlement threshold, then a level-two early warning signal will be issued for the fault status of the cable terminal in the current monitoring period.

[0028] Preferably, the second preset quantity is less than or equal to the first preset quantity, and the second settlement threshold is 50% of the first settlement threshold.

[0029] The embodiments of the present invention have at least the following beneficial effects:

[0030] This invention first collects current and tilt angle data for each of the three phases. By eliminating the time lag interference between the current thermal effect and the tilt angle response, the dynamic signal in the time domain is transformed into a stable spatial geometric trajectory, constructing a closed curve of current and tilt angle. Structural changes caused by settlement directly alter the geometric shape of this closed curve. This morphological difference is unaffected by the magnitude of thermal drift. Even with large thermal drift, as long as the three phases have inconsistent morphologies, settlement can be identified through deviation, enabling the extraction of minute settlement signals from a strong thermal drift background. This provides a reliable input for subsequent decoupling of minute settlement characteristics from a strong thermal displacement background. Finally, based on the real-time status and temporal evolution of settlement characteristic indicators, a graded and precise early warning system for cable terminal faults is achieved. By distinguishing between sudden emergency faults and gradual potential risks, clear priority guidance is provided for operation and maintenance decisions, avoiding missed reports or resource misallocation due to single threshold judgments, and improving the accuracy of cable terminal fault monitoring results. Attached Figure Description

[0031] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0032] Figure 1 This is a schematic diagram of the structure of the online monitoring and alarm system for settlement of power cable terminals based on multi-sensor collaboration provided by the present invention;

[0033] Figure 2 This is a flowchart of the steps of the online monitoring and alarm method for settlement of power cable terminals based on multi-sensor collaboration provided by the present invention. Detailed Implementation

[0034] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the online monitoring and alarm system for settlement of power cable terminals based on multi-sensor collaboration proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0035] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0036] The specific solution of the online monitoring and alarm system for settlement of power cable terminals based on multi-sensor collaboration provided by the present invention will be described in detail below with reference to the accompanying drawings. For example... Figure 1 As shown, the present invention provides an online monitoring and alarm system for settlement of power cable terminals based on multi-sensor collaboration, including a memory and a processor. The processor executes a computer program stored in the memory to implement the steps of the online monitoring and alarm method for settlement of power cable terminals based on multi-sensor collaboration.

[0037] Please see Figure 2 The diagram illustrates a flowchart of an online monitoring and alarm method for settlement of power cable terminals based on multi-sensor collaboration, according to an embodiment of the present invention. The method includes the following steps:

[0038] Step S100: Obtain the current data and tilt angle data of the three phases of the cable terminal at each moment in the current monitoring cycle. Based on the time offset correlation of the current data and tilt angle data of each phase in the current monitoring cycle, construct the current tilt angle closed curve in the current monitoring cycle.

[0039] The main purpose of this step is to construct a standardized geometric feature carrier that can truly reflect the mechanical response of the cable terminal structure. By eliminating the time lag interference between the current thermal effect and the tilt response, the dynamic signal in the time domain is transformed into a stable spatial geometric trajectory, providing a reliable input for subsequent decoupling of small settlement features from the background of strong thermal displacement.

[0040] First, preset the baseline full-load current parameter. This parameter is taken as the design rated current of the cable under test, such as 630A or 1250A. The implementer needs to set it according to the specific implementation scenario. It can be used to normalize the load current under different operating conditions to the same scale later. Set the duration of a single monitoring cycle to 24 hours, set the data sampling frequency to 1Hz, and define the set of three-phase monitoring objects. Three phases These correspond to the three-phase cable terminals that are parallel to each other on the same tower or arranged adjacently. The following data collection and filtering operations are performed for each monitoring cycle; this embodiment uses the current monitoring cycle as an example for explanation.

[0041] Then, for sets Each phase in The system collects load current and records it as current data through current transformers installed in the cable loop, and collects tilt angle and records it as tilt angle data through MEMS tilt angle sensors rigidly connected to the top of the terminal tower. The tilt angle is a tilt angle component along the line direction (longitudinal) or perpendicular to the line direction (lateral), and the specific selection depends on the terminal's preset sensitive direction.

[0042] It should be noted that during low-load operation or power outage maintenance, the cable conductor generates very little heat and the structure does not undergo significant thermal expansion and contraction. At this time, the sensor data mainly consists of environmental noise, and forced analysis can lead to feature divergence or false alarms. Therefore, this embodiment performs validity screening of the monitoring cycle before conducting feature analysis.

[0043] Specifically, the magnitude of change in current data for each phase within the current monitoring cycle is calculated. , This indicates the magnitude of change (i.e., the change in each phase within the current monitoring period). (range of all current data) This indicates each phase within the current monitoring period. The maximum value of all current data. This indicates each phase within the current monitoring period. The minimum value of all current data.

[0044] If the change range of the current monitoring period If the current monitoring period is determined to be a valid monitoring period, the subsequent monitoring steps in this embodiment can be performed. If the change in the current monitoring period... If the current monitoring period is not specified, it will be classified as a silent monitoring period, and the subsequent monitoring steps in this embodiment will not be performed for that monitoring period. To effectively incentivize the threshold, The value can be 10% of the base full-load current, and the implementer can set it according to the specific implementation scenario.

[0045] It should be understood that a silent monitoring cycle refers to a monitoring cycle in which the cable terminal experiences very low load or power outage, resulting in a weak thermal effect. The monitoring data (current, tilt angle) lacks an effective physical response, contains only environmental noise, and has no value for structural condition analysis.

[0046] The basic parameter settings are now complete. The specific method for constructing the current tilt angle closed curve for the current monitoring period is shown in the following steps.

[0047] The first step is to determine the optimal time length based on the correlation between the current data and tilt angle data of each phase under different time displacement lengths.

[0048] The tilt response of a cable terminal is an indirect mapping of the thermal effect of current. Changes in current cause the conductor to heat up, which is then conducted to the terminal structure, resulting in thermal expansion and contraction, ultimately manifesting as a change in tilt angle. This process inherently involves a time lag. Directly analyzing current and tilt angle data with misaligned timing sequences will lead to a misalignment in their response relationship, creating spurious features caused by time differences rather than structural conditions, failing to accurately reflect the correspondence between thermal cycling and structural deformation. Therefore, it is necessary to analyze the correlation between the two at different time shifts to determine the optimal time length that physically matches the excitation (current) and response (tilt angle). This is the core prerequisite for eliminating thermal hysteresis interference and ensuring data correlation, providing a fundamental guarantee for the authenticity of subsequent feature extraction.

[0049] Specifically, for any phase, the tilt angle data within the current monitoring cycle is time-shifted according to each different time displacement length to obtain the shifted tilt angle at each time displacement length; the cross-correlation function value between the absolute value of the shifted tilt angle at each time displacement length within the current monitoring cycle and the current data at the same time sequence is calculated, and the time displacement length corresponding to the maximum value of the cross-correlation function value is taken as the optimal time length for that phase.

[0050] As a concrete example, for phase A, in the time displacement length The lower cross-correlation function can be expressed as ,in, This represents the translation tilt angle at time t. This represents the current data at time t. In this embodiment, the maximum value of the time displacement length can be 4 hours, which can be set by the implementer according to the specific implementation scenario.

[0051] Furthermore, for phase A, the time displacement length corresponding to the maximum value of the cross-correlation function is obtained as the optimal time length of phase A.

[0052] It should be noted that, in order to prevent the calculation of incorrect time shifts when there is strong wind interference or unclear load fluctuation characteristics, the optimal time length of each phase in the current monitoring cycle is determined based on the value of the cross-correlation function corresponding to the optimal time length.

[0053] Specifically, the normalized value of the cross-correlation function corresponding to the optimal time length of each phase in the current monitoring cycle is obtained as the correlation coefficient. Taking any phase as an example, for phase A, if the correlation coefficient corresponding to the optimal time length is less than the preset confidence threshold, it indicates that the correlation between current and tilt angle is weak in the current monitoring cycle, and the calculated time displacement length is unreliable. Therefore, it is determined that the time alignment operation using the optimal time length is invalid, and the optimal time length of the adjacent previous historical monitoring cycle can be used as the optimal time length of the current monitoring cycle. That is, the time offset calculated by the previous valid monitoring cycle is still used. If the current monitoring cycle is the first monitoring cycle and there is no adjacent previous historical monitoring cycle, an empirical value, such as 2 hours, can be used.

[0054] For phase A, if the correlation coefficient corresponding to the optimal time length is greater than or equal to the preset confidence threshold, it indicates that the correlation between the current and the tilt angle is strong in the current monitoring period. Therefore, the optimal time length in the current monitoring period is effective and no additional operation is required.

[0055] The second step is to perform time shifting on the tilt data within the current monitoring period according to the optimal time length to obtain the reconstructed tilt angle for the current monitoring period.

[0056] Time-shifting the tilt angle data based on the optimal time length essentially aligns the tilt angle response with the current excitation in a physical sense. This ensures that the reconstructed tilt angle data and the corresponding current data accurately map the structural deformation characteristics under the same thermal state. The original tilt angle data, without shifting, cannot match the peak and trough times with the current changes, causing the subsequently constructed trajectory to fail to accurately reflect the mechanical behavior of the structure under thermal drive. The reconstruction operation, by compensating for the heat conduction delay, establishes a one-to-one physical correlation between the current and tilt angle changes, ensuring that the subsequently constructed closed curve objectively characterizes the thermo-mechanical response of the structure and avoids feature distortion caused by timing misalignment.

[0057] The third step is to construct a closed curve of the current tilt angle within the current monitoring period, using the current data within the current monitoring period as the horizontal axis and the reconstructed tilt angle within the current monitoring period as the vertical axis.

[0058] The raw current and tilt angle data in the time domain are highly periodic and volatile, making it difficult to directly extract minute settlement features due to factors such as load fluctuations and environmental noise. However, by using aligned current data and reconstructed tilt angle data as the horizontal and vertical axes respectively to construct a closed curve, the dynamic thermal cycling response can be solidified into a static geometric trajectory. The geometric shape of this curve (such as slope and enclosed area) can stably map the elastic constraint stiffness and hysteresis dissipation state of the structure, achieving a dimensionality reduction transformation from "dynamic time-domain signals" to "stable geometric features." This highlights the changes in structural mechanical properties caused by settlement and provides a feasible path for separating minute settlement signals from a strong thermal displacement background.

[0059] It should be noted that, in order to eliminate the differences in rated parameters of different specifications of equipment and the deviation of the initial installation angle of the sensor, this embodiment can also perform preprocessing operations on the physical parameters after time-series alignment. Specifically, the ratio of the original current data of each time series in the current monitoring cycle to the reference full-load current is used as the standardized current data. This processing ensures that cables with different rated currents have the same horizontal axis scale under full-load conditions.

[0060] Meanwhile, for the reconstructed tilt angle after time offset, the tilt angle value at the starting moment is used as the zero point reference to eliminate the static installation error of the sensor. Specifically, the difference between the reconstructed tilt angle of each time series in the current monitoring cycle and the reconstructed tilt angle of the first time series is used as the standardized tilt angle data for each time series. Furthermore, linear detrending can be performed on the standardized tilt angle data to remove the DC drift component of the tilt angle data.

[0061] Furthermore, using standardized current data within the current monitoring period as the x-axis and standardized tilt angle data within the current monitoring period as the y-axis, discrete data points are connected in chronological order to construct a closed-loop current-tilt angle curve for the current monitoring period. This closed-loop current-tilt angle curve is a closed polygonal geometric object. Its geometric shape (such as principal axis slope and enclosing area) directly maps the thermal expansion stiffness and overall hysteresis state of the cable terminal in the current thermal cycle.

[0062] It should be noted that each phase is analyzed separately within the current monitoring cycle, and the current tilt angle closed curve of each phase can be obtained by following the same method.

[0063] This step, through a logical chain of time offset correlation, tilt angle data reconstruction, and closed curve construction, solves the problems of thermal hysteresis signal misalignment and time domain signal fluctuation interference layer by layer. It not only conforms to the physical laws of thermal-mechanical response of power cable terminals, but also accurately meets the needs of subsequent feature decoupling and settlement identification. It is a key preliminary step for achieving low-cost, high-precision settlement monitoring.

[0064] Step S200: Based on the slope distribution and area distribution of the current tilt angle closed curve of each phase, analyze the slope deviation and area deviation of each phase to obtain the settlement characteristic index of each phase.

[0065] The purpose of this step is to transform the geometric shape of the current tilt angle closed curve into a quantifiable settlement characteristic index. By decoupling the abnormal deviation between structural stiffness and hysteresis dissipation state, it is possible to accurately capture minute settlement signals and provide a scientific basis for subsequent graded alarms.

[0066] Therefore, the method for obtaining the sedimentation characteristic indicators of each phase within the current monitoring period can be achieved through steps S201 to S203.

[0067] Step S201: Perform linear regression fitting on all points on the current tilt angle closed curve of each phase, and use the slope of the fitted line as the trend coefficient corresponding to the current tilt angle closed curve of each phase; use the area formed by the current tilt angle closed curve of each phase as the geometric coefficient corresponding to the current tilt angle closed curve of each phase.

[0068] The geometric shape of the current tilt angle closed curve is directly related to the structural mechanical state of the cable terminal: the curve slope reflects the deformation rate of the structure under thermal load, which is essentially a quantitative mapping of the elastic constraint stiffness of the structure. The support tilt caused by tower base settlement will change the stress boundary of the structure, which is directly manifested as abnormal fluctuations in the slope; the area enclosed by the curve characterizes the combined state of hysteresis and dissipation of the structure in thermal cycling. Problems such as component jamming and bolt loosening caused by settlement will aggravate the nonlinear response, resulting in a significant increase in area. Therefore, extracting the trend coefficient (slope) through linear regression fitting and extracting the geometric coefficient (area) through geometric calculation are the key steps to transform the abstract geometric trajectory into a concrete mechanical feature, providing a quantitative basis with clear physical meaning for subsequent anomaly identification and avoiding subjective judgment bias on the curve shape.

[0069] As a specific example, in order to suppress the influence of measurement noise on the local slope, the least squares method is used to perform linear regression fitting on the data points on the current tilt angle closed curve. For example, the slope of the fitted line corresponding to the current tilt angle closed curve of phase A is the trend coefficient of phase A. The slope of the fitted line reflects the average tilt angle change (structural deformation) caused by unit current change (heat load), that is, it reflects the deformation rate of the structure under heat load and characterizes the elastic constraint stiffness of the structure.

[0070] As a concrete example, using the discrete form of Green's theorem (shoelace formula), the directed area of ​​the closed polygon enclosed by the current tilt angle closed curve is calculated. It should be noted that this embodiment calculates the geometric area formed by standardized current data and standardized tilt angle data as the geometric coefficient. The specific calculation method is a well-known technique and will not be elaborated further here. The value of the geometric coefficient is directly proportional to the overall lag level of the system. For example, the geometric area corresponding to the A-phase current tilt angle closed curve is the geometric coefficient of phase A.

[0071] The geometric coefficient quantifies the combined hysteresis-dissipation state of the structure during a single thermal cycle. This area characteristic comprehensively reflects the thermal conduction delay and the nonlinear response caused by loosening or jamming at structural connections. Although thermal hysteresis is the main component, under the premise of a consistent three-phase thermal environment within the same tower, a relative abnormal increase in the area of ​​a single phase can sensitively indicate the deterioration of the mechanical constraint state of that phase. In other words, the geometric coefficient characterizes the geometric feature of the cable terminal's comprehensive hysteresis energy dissipation level during a single thermal cycle. Its core physical mapping is the result of the combined effect of thermal conduction delay and the structural mechanical constraint state (loosening, jamming, etc.), and it is highly sensitive to structural connection anomalies caused by settlement.

[0072] It should be noted that the area enclosed by the closed curve is due to the asynchronous nature of the current excitation (thermal load) and the tilt angle response (structural deformation). That is, during the current rise and fall, the thermal expansion and contraction response of the structure has a hysteresis effect, which causes the trajectory of the current tilt angle closed curve to form a non-overlapping closed loop. The area enclosed by the closed loop directly quantifies the energy dissipation corresponding to this hysteresis.

[0073] Furthermore, in this embodiment, before performing subsequent three-phase relative characteristic differences, the absolute attitude and stiffness state of the three-phase cable terminals are verified by using a preset physical safety limit threshold. If any phase exceeds the safety red line (such as the entire tower collapsing or the foundation sinking), the regular analysis process is immediately stopped, and the highest level emergency alarm is directly triggered. This is essentially to avoid the fatal risk of relative difference analysis failure caused by common mode extreme faults.

[0074] Prioritize intercepting extreme faults that threaten equipment safety, rather than monitoring minor settlements. Such faults (e.g., entire tower tilting, support breakage) are beyond the scope of minor settlement monitoring and are critical risks that require immediate repair and rapid response. Ensuring system safety is the highest priority to avoid subsequent deviation analysis (designed for minor settlement) from being missed due to the failure of extreme faults, while providing the fastest emergency response channel for operation and maintenance.

[0075] Specifically, the maximum absolute tilt angle of each phase within the current monitoring period is recorded as the maximum absolute tilt angle of that phase, and the median of the geometric coefficients among the three phases within the current monitoring period is recorded as the geometric median of that period. Three structural safety limit thresholds are set for the maximum absolute tilt angle, geometric median, and absolute value of the trend coefficient of each phase within the current monitoring period.

[0076] More specifically, if any phase's index exceeds the physical limit, it indicates a severe attitude anomaly at the phase terminal or the entire tower foundation, potentially corresponding to foundation collapse or external force damage. The system then triggers the highest-level "emergency warning" and sends a fault snapshot containing the original data to the operations and maintenance center. If all indices for all phases do not exceed the physical limits, the system determines that the current structure is within a relatively safe range, and the fault (if present) is a subtle, concealed settlement or early-stage defect, proceeding with subsequent characteristic analysis.

[0077] Based on this, for any phase, when the absolute value of the trend coefficient is greater than the preset tilt threshold, or when the maximum value of the absolute value of the tilt angle data of that phase in the current monitoring period is greater than the preset angle threshold, the highest level emergency warning signal will be issued for the fault status of that phase at the cable terminal.

[0078] As a specific example, if the maximum absolute tilt angle of any one of the three phases is greater than the angle threshold, or the absolute value of the trend coefficient of any one of the three phases is greater than the tilt threshold, or the geometric median is greater than the area threshold, it indicates that the physical limit has been exceeded, suggesting that a serious attitude anomaly has occurred at the end of that phase or the entire tower base.

[0079] Specifically, the angle threshold can be 5°, the tilt threshold can be 1.5-3 times the absolute value of the maximum trend coefficient within the historical health monitoring period, and the area threshold can be 1.5-3 times the maximum geometric median within the historical health monitoring period. Implementers can set these values ​​according to specific implementation scenarios or through extensive experimental data.

[0080] Step S202: Based on the deviation between the trend coefficient of each phase and the overall distribution of the trend coefficients of the three phases, obtain the trend deviation degree of each phase; based on the deviation between the geometric coefficient of each phase and the overall distribution of the geometric coefficients of the three phases, obtain the geometric deviation degree of each phase.

[0081] Three-phase cable terminals mounted on the same tower are exposed to the same environmental conditions, such as temperature and wind load. These common-mode interferences synchronously affect the curve coefficients of the three phases, causing a systematic drift in the absolute values ​​of the coefficients. Settlement, as a single-phase or local anomaly, manifests as a differential-mode signal. Directly using the absolute values ​​of the coefficients to determine anomalies can easily lead to false alarms or missed alarms due to common-mode interference. Therefore, by comparing the deviations of single-phase coefficients with the overall distribution of the three phases, and calculating the trend deviation and geometric deviation, common-mode interference caused by environmental factors can be automatically offset, focusing on the unique anomalies of single phases. This deviation analysis based on the statistical consistency of the three phases does not require external meteorological data correction, reducing system complexity and significantly improving the signal-to-noise ratio of minute settlement signals. It is the core logic for achieving low-cost, high-precision monitoring.

[0082] Specifically, the first step is to obtain the median of the trend coefficients of the three phases as the three-phase trend characteristic value, and to obtain the median of the geometric coefficients of the three phases as the three-phase geometric characteristic value.

[0083] The three-phase median harmonic reference is a dynamic, robust, and self-adaptive reference system constructed based on the micro-environmental consistency and response statistical consistency of three-phase cable terminals on the same tower. It takes the median of the three-phase trend coefficient and the median of the geometric coefficient as the core. Its core characteristics are synchronously following common-mode interference and isolating differential-mode settling signals. In essence, it is a common-mode suppression carrier that does not require external meteorological parameters, providing a reliable comparison benchmark for subsequent deviation analysis.

[0084] It should be noted that three-phase cable terminals mounted on the same tower or arranged adjacent to each other are in the same solar radiation, wind load, and ambient temperature field, and the three-phase loop currents have a high statistical correlation. This makes the three-phase response to environmental disturbances (heat and wind) synchronous and consistent under normal operating conditions (i.e., common-mode response), laying the physical foundation for the benchmark cooperative following characteristics.

[0085] The median was chosen to avoid outlier interference caused by single-phase settling. If single-phase settling occurs, its characteristic value will deviate significantly from the healthy level. If the mean is used as the benchmark, outliers will pull the benchmark off track, leading to misjudgment of the healthy phase. The median has a natural anti-outlier characteristic and can accurately reflect the normal response level of the three phases as a whole, ensuring that the benchmark is not contaminated by a single faulty phase.

[0086] The construction of the three-phase trend characteristic values ​​and three-phase geometric characteristic values ​​is based entirely on the horizontal spatial comparison of the current monitoring cycle, without relying on historical data for training. This avoids the "no benchmark" problem during the cold start phase of the monitoring system and achieves real-time adaptation. Common-mode interference factors such as environmental wind load and solar radiation will simultaneously affect the three-phase characteristics, causing the three-phase characteristics to rise and fall synchronously. However, the median benchmark will also rise and fall accordingly, thus being automatically offset in subsequent deviation characteristic analysis.

[0087] The second step is to take the absolute value of the difference between the trend coefficient of each phase and the trend characteristic value of the three phases as the trend difference of each phase, and determine the trend deviation of each phase based on the ratio between the trend difference of each phase and the absolute value of the trend characteristic value of the three phases.

[0088] As a concrete example, taking phase A as an example, the method for obtaining the trend deviation of phase A within the current monitoring period can be expressed by the formula:

[0089]

[0090] in, This indicates the trend deviation of phase A within the current monitoring period. This represents the trend coefficient of phase A within the current monitoring period. Represents the three-phase trend characteristic value. The preset minimum positive number is used to prevent division by zero. An example value is taken as follows: .

[0091] use and The inclusion of these values ​​in the calculation is to prevent issues arising during the cold start phase or extremely low response period due to the three-phase trend characteristic value. Negative numbers or numbers close to zero can cause calculation results to have a negative sign, diverge, or become meaningless. The trend difference of phase A is reflected in the trend deviation, which directly reflects the variation ratio of single-phase stiffness relative to the overall three-phase stiffness.

[0092] It should be noted that the stiffness anomaly caused by settlement is essentially the difference in stiffness between the single phase and the normal three-phase level, regardless of the direction of the difference. For example, whether the tower base tilts to the left or to the right, it is a stiffness anomaly, only the direction is different, and the degree of anomaly is equally important.

[0093] The third step is to take the difference between the geometric coefficient of each phase and the geometric characteristic value of the three phases as the geometric difference of each phase, and determine the geometric deviation of each phase based on the trend difference of each phase and the geometric characteristic value of the three phases.

[0094] As a concrete example, taking phase A as an example, the method for obtaining the geometric deviation of phase A within the current monitoring cycle can be expressed by the formula:

[0095]

[0096] in, This indicates the geometric deviation of phase A within the current monitoring period. This represents the geometric coefficient of phase A within the current monitoring period. Represents the three-phase geometric eigenvalues. The preset minimum positive number is used to prevent division by zero. An example value is taken as follows: , This is a function to find the maximum value, ensuring that only the current value is retained. Abnormal deviations that occur during the process. The geometric difference of phase A is represented by the geometric deviation, which reflects the positive increment ratio of single-phase dissipation relative to the overall three-phase dissipation.

[0097] It should be noted that settlement only leads to structural loosening and jamming, thereby reducing the area of ​​the closed curve. Increase (increase in dissipation); area of ​​closed curve A reduction typically corresponds to structural tightening or decreased thermal response sensitivity, and is unrelated to settlement risk; therefore, deviation only needs to be captured. Positive anomalies at that time.

[0098] Step S203: Determine the settlement characteristic index of each phase based on the Euclidean norm of the trend deviation and geometric deviation of each phase.

[0099] Structural anomalies caused by settlement are diverse, potentially manifesting only as stiffness changes (e.g., simple tower base tilting), only as hysteresis dissipation anomalies (e.g., component jamming), or both. By synthesizing trend deviation and geometric deviation using Euclidean norm, anomaly information from two orthogonal dimensions can be integrated into a unified dimensionless index. This synthesis logic ensures comprehensive sensitivity to different settlement patterns while eliminating dimensional differences in coefficients across dimensions through normalization, providing a unified evaluation standard for the index and facilitating the setting of general thresholds for tiered alerts. Furthermore, the mathematical properties of Euclidean norm ensure a monotonic response of the index to anomalies in both dimensions; the more significant the anomaly, the larger the index value, providing an intuitive and reliable basis for the quantitative grading of settlement risk.

[0100] As a concrete example, taking phase A as an example, the method for obtaining the settlement characteristic indicators of phase A within the current monitoring period can be expressed by the formula:

[0101]

[0102] in, This indicates the settlement characteristic index of phase A within the current monitoring period. This indicates the trend deviation of phase A within the current monitoring period. This indicates the geometric deviation of phase A within the current monitoring period. The settlement characteristic index is expressed as a percentage, intuitively quantifying the overall degree to which the current mechanical constraint state of the phase deviates from the normal level of the population. For example, This means that the overall state of this phase deviates by 15% relative to the other two phases.

[0103] Step S300: Based on the settlement characteristic indicators of each phase in the current monitoring cycle and combined with the settlement characteristic indicators of each phase in the historical monitoring cycles, classify and issue early warnings for the fault conditions of the cable terminals.

[0104] The main purpose of this step is to achieve graded and accurate early warning of cable terminal faults based on the real-time status and temporal evolution of settlement characteristic indicators. By distinguishing between sudden emergency faults and gradual potential risks, it provides clear priority guidance for operation and maintenance decisions and avoids missed reports or resource misallocation caused by a single threshold judgment.

[0105] The first step is to issue a first-level early warning signal for any phase if the settlement characteristic index of that phase is greater than the preset first settlement threshold during the current monitoring period.

[0106] Settlement faults at cable terminals can occur in extreme scenarios such as "sudden collapse," including instantaneous foundation instability and support structure fracture. These faults develop rapidly and are extremely dangerous, requiring immediate emergency repairs to prevent cascading accidents such as cable overload and insulation damage. The settlement characteristic indicators of the current monitoring period directly reflect the current degree of mechanical anomaly in the structure. If the indicators exceed the preset settlement threshold, it indicates that the structure is in a significantly abnormal state, far exceeding the mechanical constraints for safe operation.

[0107] Therefore, setting up a first-level early warning and immediate response mechanism is a key line of defense for ensuring the safe and stable operation of power lines. By using clear threshold triggering logic, it ensures that emergency risks are identified as soon as possible and the highest priority operation and maintenance instructions are pushed out, avoiding delays in handling the situation.

[0108] It should be noted that, to ensure the scientific validity of the threshold, its specific value can be determined empirically by statistically analyzing the settlement characteristic indicators of the same phase during historical monitoring periods without faults. For example, for phase A, 30 monitoring periods are selected after the equipment is first put into operation or during maintenance confirming no faults, serving as a healthy sample set. The statistical distribution of the settlement characteristic indicators of phase A in the healthy sample set is calculated to obtain the mean. and standard deviation ,in accordance with In principle, the first settlement threshold is set as follows: This is to ensure that the alarm is statistically significant. In this embodiment, the first settlement threshold is exemplarily set at 20%.

[0109] At this point, the Level 1 early warning signal response is the highest priority immediate response for "instantaneous fatal faults." When a "sudden settlement" is detected at a cable terminal due to foundation collapse, support structure failure, or external impact, causing a rapid deterioration in the structural constraint state, the emergency repair response triggered by the system is essentially a fault interception action to prevent the accident from escalating as quickly as possible.

[0110] More specifically, the Level 1 early warning signal response involves sending an emergency interruption request or immediate repair instruction to the remote control center via the system communication module. The work order specifies "sudden settlement risk" and includes fault attribute diagnosis results (such as "it is recommended to prioritize checking the tower base levelness" or "it is recommended to prioritize checking the support jamming").

[0111] The second step is to determine whether the cable terminal fault condition should respond to a secondary warning signal based on the continuous distribution of the settlement characteristic index of the phase in the historical monitoring cycles before the current monitoring cycle, if the settlement characteristic index of the phase in the current monitoring cycle is less than or equal to the preset first settlement threshold.

[0112] Most settlement faults do not occur instantaneously, but rather exhibit a gradual evolution, such as soil consolidation and slow changes in groundwater levels. The settlement characteristic indicators for the current period may not have exceeded the safety threshold, but they may show a persistently high trend. Judging solely based on current indicators can easily overlook such potential risks, allowing small settlements to gradually accumulate into serious faults. Therefore, it is necessary to consider the assumption that current indicators do not exceed the threshold and to trace the continuous distribution of indicators over historical monitoring periods. By analyzing whether indicators have been in the critical range for a long time or show a monotonically increasing trend, the evolutionary pattern of gradual settlement can be captured. This time-series analysis logic can distinguish between normal trends in a single period and abnormal trends in continuous periods, enabling early prediction of potential risks and allowing maintenance personnel sufficient time for investigation and rectification, thus preventing the escalation of faults at the source.

[0113] Specifically, within a monitoring period window consisting of a first preset number of historical monitoring periods prior to the current monitoring period, if the settlement characteristic index of at least a second preset number of historical monitoring periods is greater than a preset second settlement threshold, then a level-two early warning signal will be issued for the cable terminal fault situation in the current monitoring period. The second preset number is less than or equal to the first preset number, and the second settlement threshold is 50% of the first settlement threshold.

[0114] As a specific example, in this embodiment, the first preset quantity is 7, and the second preset quantity can be 5, which can be set by the implementer according to the specific implementation scenario. That is, the monitoring period window consists of the seven consecutive historical monitoring periods preceding the current monitoring period. Within this monitoring period window, if the settlement characteristic index of at least five historical monitoring periods is greater than the preset second settlement threshold, it indicates that there is a slow-accumulating monotonic settlement trend at the end of the phase, corresponding to foundation settlement caused by soil consolidation or groundwater level changes. At this time, a secondary early warning signal is triggered. In this embodiment, the second settlement threshold is 10%.

[0115] Thus, the Level 2 early warning signal response is a medium-priority tracking and intervention response for "potential slow-progressing defects". When the system detects that the cable terminal is experiencing "progressive creep" caused by soil consolidation, slow changes in groundwater level, and long-term disturbance from surrounding construction, and the settlement shows a monotonous cumulative trend, the early warning tracking action triggered by the system is essentially a preventive response to intervene in advance and prevent slow-progressing defects from deteriorating into sudden failures.

[0116] More specifically, the Level 1 warning signal response is that the system automatically increases the sampling frequency of the phase in step S100 (e.g., from 1Hz to 10Hz) and includes the device in the key inspection list for the next manual inspection, prompting maintenance personnel to pay attention to the settlement trend of the phase.

[0117] At this point, power line operation and maintenance faces a real-world scenario involving multiple devices and a wide range of areas. Applying the same response level to all anomalies would lead to delays in handling emergency faults or excessive resource allocation to minor potential risks. A Level 1 early warning system focuses on immediate, severe anomalies, corresponding to the highest priority repairs; a Level 2 early warning system focuses on potential trend anomalies, corresponding to key areas of focus during routine inspections. This hierarchical design achieves precise matching of early warning systems with operation and maintenance resources. Simultaneously, the hierarchical logic covers both sudden and progressive fault evolution modes, avoiding missed reports of emergency faults and preventing the neglect of potential risks. This ensures the comprehensiveness and practicality of the early warning system, providing clear action guidelines for operation and maintenance decisions and improving overall operation and maintenance efficiency and line safety.

[0118] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A multi-sensor collaborative online monitoring and alarm system for settlement of power cable terminals, comprising a memory and a processor, characterized in that, The processor executes the computer program stored in the memory to perform the following steps: Acquire the current and tilt angle data of the three phases of the cable terminal at each moment in the current monitoring cycle, and construct the current and tilt angle closed curve in the current monitoring cycle based on the time offset correlation of the current and tilt angle data of each phase in the current monitoring cycle. Based on the slope distribution and area distribution of the current tilt angle closed curve of each phase, the slope deviation and area deviation of each phase are analyzed to obtain the settlement characteristic index of each phase. Based on the settlement characteristic indicators of each phase in the current monitoring cycle, and combined with the settlement characteristic indicators of each phase in the historical monitoring cycles, the fault status of the cable terminal is classified and warned.

2. The online monitoring and alarm system for settlement of power cable terminals based on multi-sensor collaboration as described in claim 1, characterized in that, The construction of a current-tilt angle closed curve for the current monitoring cycle based on the time offset correlation between the current data and tilt angle data of each phase within the current monitoring cycle specifically includes: The optimal time length is determined based on the correlation between the current data and tilt angle data of each phase under different time displacement lengths. The tilt data within the current monitoring period are shifted over time according to the optimal time length to obtain the reconstructed tilt angle for the current monitoring period; Using the current data within the current monitoring period as the horizontal axis and the reconstructed tilt angle within the current monitoring period as the vertical axis, a closed curve of the current tilt angle within the current monitoring period is constructed.

3. The online monitoring and alarm system for settlement of power cable terminals based on multi-sensor collaboration according to claim 2, characterized in that, The determination of the optimal time length based on the correlation between the current data and tilt angle data of each phase under different time displacement lengths specifically includes: For any phase, the tilt angle data within the current monitoring cycle is time-shifted according to each different time displacement length to obtain the shifted tilt angle at each time displacement length; the cross-correlation function value between the absolute value of the shifted tilt angle at each time displacement length within the current monitoring cycle and the current data at the same time sequence is calculated, and the time displacement length corresponding to the maximum value of the cross-correlation function value is taken as the optimal time length for that phase.

4. The online monitoring and alarm system for settlement of power cable terminals based on multi-sensor collaboration according to claim 1, characterized in that, The method involves analyzing the slope and area distribution of the current tilt angle closed curve for each phase, and then analyzing the slope and area deviations of each phase to obtain the settlement characteristic indicators for each phase. Specifically, this includes: Linear regression fitting is performed on all points on the current tilt angle closed curve of each phase, and the slope of the fitted line is used as the trend coefficient corresponding to the current tilt angle closed curve of each phase; the area formed by the current tilt angle closed curve of each phase is used as the geometric coefficient corresponding to the current tilt angle closed curve of each phase. The trend deviation of each phase is obtained based on the deviation between the trend coefficient of each phase and the overall distribution of the trend coefficients of the three phases; the geometric deviation of each phase is obtained based on the deviation between the geometric coefficient of each phase and the overall distribution of the geometric coefficients of the three phases. Settlement characteristic indices for each phase are determined based on the Euclidean norm of the trend deviation and geometric deviation of each phase.

5. The online monitoring and alarm system for settlement of power cable terminals based on multi-sensor collaboration according to claim 4, characterized in that, The process of obtaining the trend deviation degree of each phase based on the deviation between the trend coefficient of each phase and the overall distribution of the trend coefficients of the three phases specifically includes: The median of the trend coefficients of the three phases is obtained as the three-phase trend characteristic value. The absolute value of the difference between the trend coefficient of each phase and the three-phase trend characteristic value is taken as the trend difference of each phase. The trend deviation of each phase is determined based on the ratio between the trend difference of each phase and the absolute value of the three-phase trend characteristic value.

6. The online monitoring and alarm system for settlement of power cable terminals based on multi-sensor collaboration according to claim 4, characterized in that, The geometric deviation of each phase is obtained based on the deviation between the geometric coefficient of each phase and the overall distribution of the geometric coefficients of the three phases. Specifically, this includes: The median of the geometric coefficients of the three phases is obtained as the three-phase geometric characteristic value. The difference between the geometric coefficient of each phase and the three-phase geometric characteristic value is taken as the geometric difference of each phase. Based on the trend difference of each phase and the three-phase geometric characteristic value, the geometric deviation of each phase is determined.

7. The online monitoring and alarm system for settlement of power cable terminals based on multi-sensor collaboration according to claim 4, characterized in that, For any phase, when the absolute value of the trend coefficient is greater than the preset tilt threshold, or when the maximum value of the absolute value of the tilt angle data of that phase in the current monitoring period is greater than the preset angle threshold, the highest level emergency warning signal will be issued for the fault status of that phase at the cable terminal.

8. The online monitoring and alarm system for settlement of power cable terminals based on multi-sensor collaboration according to claim 1, characterized in that, The method involves classifying and issuing early warnings for cable terminal faults based on the settlement characteristic indicators of each phase in the current monitoring cycle, combined with the settlement characteristic indicators of each phase in historical monitoring cycles. Specifically, this includes: For any phase, if the settlement characteristic index of the phase is greater than the preset first settlement threshold during the current monitoring period, a first-level early warning signal will be issued for the fault status of the phase at the cable terminal. If the settlement characteristic index of the phase is less than or equal to the preset first settlement threshold in the current monitoring period, then based on the continuous distribution of the settlement characteristic index of the phase in the historical monitoring periods before the current monitoring period, it is determined whether the fault condition of the cable terminal should respond to the secondary early warning signal.

9. The online monitoring and alarm system for settlement of power cable terminals based on multi-sensor collaboration according to claim 8, characterized in that, The determination of whether a cable terminal fault should trigger a secondary early warning signal based on the continuous distribution of settlement characteristic indicators of the phase in historical monitoring cycles prior to the current monitoring cycle includes: If, within a monitoring period window consisting of a first preset number of historical monitoring periods prior to the current monitoring period, the settlement characteristic index of at least a second preset number of historical monitoring periods is greater than a preset second settlement threshold, then a level-two early warning signal will be issued for the fault status of the cable terminal in the current monitoring period.

10. The online monitoring and alarm system for settlement of power cable terminals based on multi-sensor collaboration according to claim 9, characterized in that, The second preset quantity is less than or equal to the first preset quantity, and the second settlement threshold is 50% of the first settlement threshold.

Images