Cable abnormity identification and positioning device fusing current and voltage data

Abstract

The invention relates to the field of cable monitoring, in particular to a cable abnormity identifying and positioning device fusing current and voltage data. Comprising a data acquisition module, an edge calculation module, a positioning module, a communication module and the like, current and voltage signals on a cable are synchronously acquired, and the edge calculation module performs multi-dimensional feature fusion calculation including transient apparent power, transient impedance and fusion high-frequency distortion features; and the characteristic quantities are compared with a dynamic self-learning normal operation reference model, so that accurate identification of cable abnormity is realized, and the problem of high false alarm rate of single parameter monitoring is effectively solved. Once the identification is abnormal, the positioning module accurately calculates the position of an abnormal point by using a traveling wave distance measurement method. By adopting the method, the identification capability on the complex disturbance of the power grid can be greatly enhanced, and passive alarm is converted into intelligent identification, so that the reliability of a monitoring result is remarkably improved fundamentally, and the false alarm rate is greatly reduced.

Description

Technical Field

[0001] This invention relates to the field of cable monitoring, and in particular to a cable anomaly identification and location device that integrates current and voltage data. Background Technology

[0002] As power generation systems develop towards intelligence and high density, medium-voltage cables, as the core carrier of power transmission, are widely used in urban distribution networks, industrial parks, and new energy power plants. Operating long-term in underground cable trenches, tunnels, or complex outdoor environments, they are susceptible to insulation aging, mechanical damage, partial discharge, and environmental corrosion, leading to short circuits, grounding faults, and in severe cases, large-scale power outages and significant economic losses. Therefore, achieving early identification and accurate location of cable anomalies is a key requirement for ensuring the safe and stable operation of power systems.

[0003] Existing devices typically rely on a single monitoring parameter, making them highly susceptible to normal load fluctuations or transient disturbances in the power grid, resulting in numerous false alarms. This severely interferes with the judgment of maintenance personnel and also represents a significant waste of maintenance resources. Summary of the Invention

[0004] The present invention aims to provide a cable anomaly identification and location device that integrates current and voltage data to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A cable anomaly identification and location device that integrates current and voltage data, comprising: The data acquisition module is used to synchronously acquire current and voltage signals on the cable line; An edge computing module, connected to the data acquisition module, is used to perform feature fusion calculations based on the current signal and voltage signal, and compare them with a pre-stored normal operation benchmark model to identify cable anomalies. The feature fusion calculation includes: calculating a first feature quantity reflecting the power transmission state of the cable based on the current signal and voltage signal, calculating a second feature quantity reflecting the impedance characteristic state of the cable, and calculating a third feature quantity reflecting the distortion state of the current or voltage waveform. The positioning module, connected to the edge computing module, is used to calculate the location of the abnormal point using the traveling wave ranging method after an anomaly is detected. The communication module, connected to the edge computing module and the positioning module, is used to upload abnormal information, early warning information and positioning results.

[0006] Preferably, the logic for the edge computing module to identify cable anomalies is as follows: an anomaly is determined when any one of the first feature, the second feature, or the third feature meets the following condition: or or in, This represents the transient apparent power of the first characteristic quantity. and These are the mean and standard deviation of its normal operating benchmark model, respectively; This represents the transient impedance, which is the second characteristic quantity. and These are the mean and standard deviation of its normal operating benchmark model, respectively; This represents the high-frequency distortion feature fused with the third feature quantity. It is the mean of its normal operating benchmark model; , , This is a configurable sensitivity coefficient.

[0007] Preferably, the formula for calculating the transient apparent power of the first characteristic quantity is: in, and These are the instantaneous voltage and current signals acquired synchronously, respectively.

[0008] Preferably, the formula for calculating the second characteristic quantity, transient impedance, is as follows: in, and These are the instantaneous voltage and current signals acquired synchronously, respectively.

[0009] Preferably, the calculation formula for the third feature quantity fused with the high-frequency distortion feature is as follows: in, and These are the total harmonic distortion rates of current and voltage, respectively. and The weighting coefficients are and satisfy the following conditions: .

[0010] Preferably, the normal operation benchmark model is a dynamic model that uses a sliding window mechanism for online adaptive updates.

[0011] Preferably, the positioning formula of the traveling wave ranging method is: in, Distance to outlier points For traveling wave speed, This refers to the time difference between the arrival of the traveling wave signal at different monitoring points, or the propagation time from the point of anomaly to the monitoring point.

[0012] Preferably, the edge computing module further includes an early warning function: when the feature quantity continuously deviates from the normal operation benchmark model but does not reach the anomaly judgment threshold, the edge computing module generates and reports early warning information to achieve predictive maintenance. The edge computing module executes a self-learning process during the initialization phase to establish the normal operation benchmark model during the period when the cable is operating without abnormalities.

[0013] Preferably, the data acquisition module includes: A current sensor is used to acquire the current signal; A voltage sensor is used to acquire the voltage signal; A synchronous acquisition card is connected to the current sensor and voltage sensor respectively, and is used to convert analog signals into synchronous digital signals.

[0014] Preferably, the abnormal information uploaded by the communication module includes the abnormality type, trigger time, and characteristic data that caused the abnormality.

[0015] The beneficial effects of this technical solution compared to existing technologies are as follows: (1) This solution solves the problem of high false alarm rate caused by existing technologies relying on a single parameter by setting up a data acquisition module to synchronously acquire current and voltage signals, and then using an edge computing module to fuse and calculate transient apparent power, transient impedance, and high-frequency distortion characteristics. This fundamentally solves the problem of high false alarm rate caused by existing technologies relying on a single parameter, and avoids wasting maintenance resources. By intelligently analyzing the coordinated change patterns and consistency of different characteristic quantities under specific operating conditions, it can accurately distinguish between characteristic deviations that represent real anomalies and synchronous changes of characteristics caused by normal load fluctuations. This greatly enhances the ability to identify complex disturbances in the power grid, realizing the transformation from passive alarm to intelligent identification, thereby significantly improving the reliability of monitoring results and greatly reducing the false alarm rate at the root.

[0016] (2) By setting a dynamic benchmark model, the device is equipped with self-learning and self-adaptation capabilities. This model can automatically track and adapt to the slow drift of cable load caused by seasonal and diurnal changes, avoiding the problem of decreased sensitivity or surge in false alarms caused by environmental changes in the fixed threshold method, and ensuring the accuracy and environmental adaptability of the monitoring system in long-term operation.

[0017] (3) By setting up a positioning module and using the traveling wave ranging method, the industry pain points of difficulty in finding fault points of long-distance cables and long investigation time are solved. This can greatly shorten the fault repair time, reduce the economic losses caused by power outages, and improve the reliability of power supply.

[0018] (4) By setting up the early warning function and initial self-learning process in the edge computing module, the system has achieved a leap from "fault alarm" to "predictive maintenance". The system can not only detect the anomalies that have occurred, but also capture the weak and early performance degradation trends, and provide early warnings when the characteristic value continues to deviate from the normal range, providing valuable decision support window for operation and maintenance personnel, thereby realizing predictive maintenance and preventing problems before they occur. Attached Figure Description

[0019] Figure 1 A structural framework diagram of the device provided by the present invention; Detailed Implementation The present invention will now be described in further detail with reference to the accompanying drawings and embodiments: like Figure 1 The cable anomaly identification and location device shown includes: The data acquisition module is used to synchronously acquire current and voltage signals on the cable line, providing a high-quality data foundation for subsequent anomaly identification and location algorithms. The data acquisition module includes: The current sensor is used to collect current signals. It employs a high-frequency current transformer with a bandwidth of no less than 100kHz to ensure accurate capture of high-frequency transient current components in the nanosecond to microsecond range caused by anomalies such as partial discharge.

[0020] The voltage sensor is used to collect voltage signals. It adopts a voltage divider capacitor voltage divider suitable for high-voltage line measurement and has sufficient insulation strength and measurement accuracy to obtain reliable instantaneous voltage values. The synchronous acquisition card connects to both the current sensor and the voltage sensor. Its core feature is its multi-channel synchronous acquisition capability, ensuring that each set of instantaneous current values ​​is acquired. With instantaneous voltage value All have strict timing alignment, and this synchronization is crucial for subsequent calculations of transient apparent power. With transient impedance A necessary condition for equal characteristic quantities.

[0021] It should be noted that those skilled in the art will understand that current sensors are not limited to high-frequency current transformers, but include, but are not limited to, devices that can achieve the same function, such as Rogowski coils and optical current transformers; voltage sensors are not limited to capacitive voltage dividers, but include, but are not limited to resistive voltage dividers and inductive voltage dividers; synchronous acquisition function can be implemented through conventional technical means in the art, such as dedicated synchronous acquisition cards or customized acquisition circuits based on FPGAs.

[0022] The edge computing module, connected to the data acquisition module, performs feature fusion calculations based on current and voltage signals and compares them with a pre-stored normal operation benchmark model to identify cable anomalies. The normal operation benchmark model is a dynamic model that uses a sliding window mechanism for online adaptive updates. Its update strategy is as follows: the system maintains a fixed-duration (e.g., 24-hour) window of recent normal data; when new data is determined to be normal, it is added to this window, and the oldest data in the window is removed; subsequently, based on all the data within the updated window, the mean is recalculated. with standard deviation To dynamically adjust the baseline model.

[0023] The feature fusion calculation includes: calculating a first feature quantity reflecting the power transmission state of the cable based on current and voltage signals, calculating a second feature quantity reflecting the impedance characteristics of the cable, and calculating a third feature quantity reflecting the distortion state of the current or voltage waveform. First characteristic quantity: transient apparent power It can detect abnormal fluctuations in the instantaneous energy transmission of cables and is sensitive to power surges caused by poor contact, partial discharge, etc.

[0024] The calculation formula is as follows: in, and These are the instantaneous voltage and current signals acquired synchronously. This characteristic quantity is extremely sensitive to instantaneous energy fluctuations in the cable and can effectively capture abnormal power changes caused by poor contact, partial discharge, etc.

[0025] In an ideal situation, It should change smoothly with the load. When partial discharge occurs in the cable, it will generate a transient current pulse, which may be accompanied by a voltage drop, leading to... Electric spikes may appear. Poor contact or loose connections can cause intermittent arcing, which can also lead to... High-frequency oscillations or spikes.

[0026] Second characteristic quantity: transient impedance Monitor changes in the electrical characteristics of cable lines to accurately identify impedance anomalies caused by insulation aging, moisture, breakage, joint corrosion, etc.

[0027] The calculation formula is as follows: in, and These are the instantaneous voltage and current signals acquired synchronously. This characteristic directly reflects the impedance characteristics of the cable line and has a unique advantage in identifying impedance changes caused by insulation aging, moisture, or breakage.

[0028] Impedance is an inherent property of cables. A healthy cable has a relatively stable impedance under the same load conditions. When cable insulation ages or becomes damp, its insulation performance deteriorates, leakage current increases, and this manifests as decreased impedance. It shows a decreasing trend. When cable joints oxidize / corrode, it leads to an increase in contact resistance, which manifests as... Increase. This can occur when the cable is short-circuited or overloaded. It dropped sharply to near zero.

[0029] Third characteristic: Fusion of high-frequency distortion characteristics It quantifies the degree of distortion of current and voltage waveforms, focuses on high-frequency distortion caused by abnormalities in the cable itself (such as partial discharge), and suppresses system background interference.

[0030] The calculation formula is as follows: in, and These are the total harmonic distortion rates of current and voltage, respectively. and The weighting coefficients are and satisfy the following conditions: . It is a dimensionless index that comprehensively quantifies the degree of distortion of current and voltage waveforms.

[0031] Weighting coefficient and The configuration is based on and Sensitivity to characterizing cable anomalies. In a preferred embodiment, given that current signals are more directly and sensitive to high-frequency events such as partial discharge, while voltage signals are more susceptible to system background interference, therefore, a sensitivity setting is provided. > This configuration can effectively improve the monitoring signal-to-noise ratio and reliability.

[0032] Fusion of high frequency distortion features The calculation is based on the total harmonic distortion rate of current and voltage. .

[0033] Total Harmonic Distortion It is a well-known indicator for measuring the degree to which a waveform deviates from a sine wave. Its calculation formula is as follows: in, , These are the effective values ​​of the fundamental components of the current and voltage, respectively. and Representing the The effective values ​​of subharmonic current and voltage; For the highest harmonic under consideration. Summation term. , The squares of the harmonic component intensities are then summed to amplify the influence of larger harmonics while ensuring a positive contribution. The square root of the sum is then taken to restore the dimensions, yielding a single value representing the total harmonic content that is physically comparable to the effective value of the fundamental frequency. This can be understood as the "Total Harmonic Effective Value".

[0034] single and The increase may originate from non-cable anomalies (such as equipment startup or background harmonics in the power grid); while cable-related anomalies (such as partial discharge) can also cause this. and Significantly increased, fused by weights It can accurately focus on this type of dual distortion caused by cable faults, thereby effectively suppressing background interference and improving the targeting of identification.

[0035] The logic for edge computing modules to identify cable anomalies is as follows: an anomaly is determined when any one of the first, second, or third feature quantities meets the following conditions: (Capturing the instantaneous changes in energy) or (Capturing steady-state shifts in the electrical characteristics of the line) or (Capturing high-frequency distortion of the waveform) in, The first characteristic quantity is the transient apparent power. and These are the mean and standard deviation of its normal operating benchmark model, respectively; Represents the transient impedance, which is the second characteristic quantity. and These are the mean and standard deviation of its normal operating benchmark model, respectively; The third characteristic quantity represents the fusion of high-frequency distortion features. It is the mean of its normal operating benchmark model; , , This is a configurable sensitivity coefficient.

[0036] (1) Scenario 1: High-power motor starts (not due to cable abnormality) Phenomenon: Electric current A sharp increase, leading to A significant increase could trigger the first condition. Simultaneously, the voltage... Due to the impact of the launch, it fell. calculate, increase, reduce, The fluctuation range is small. Normal motor startup does not involve high-frequency harmonic injection and will not cause [problems]. Dramatic changes.

[0037] Judgment result: Only condition 1 can be triggered, the "cable abnormality" judgment logic is not met, and the system does not alarm (as expected).

[0038] (2) Scenario 2: Partial discharge in cable (real anomaly) Phenomenon: A momentary spike occurs (trigger condition 1), and the discharge point is equivalent to a parallel high impedance, leading to Tiny fluctuations (potentially triggering condition 2) inject a large number of high-frequency harmonics into partial discharge, simultaneously causing... and rise, Significantly exceeded the limit (trigger condition 3).

[0039] Judgment result: If at least two conditions are met, the system determines that the cable is abnormal (as expected).

[0040] The positioning module, connected to the edge computing module, is used to calculate the location of the anomaly point after an anomaly is detected, using the traveling wave ranging method. The positioning formula for the traveling wave ranging method is: in, Distance to outlier points For traveling wave speed, This refers to the time difference between the arrival of the traveling wave signal at different monitoring points, or the propagation time from the point of anomaly to the monitoring point.

[0041] The edge computing module also includes an early warning function: when a feature value continuously deviates from the normal operating baseline model but does not reach the anomaly detection threshold, the edge computing module generates and reports an early warning message to achieve predictive maintenance. To achieve predictive maintenance, the early warning function includes early warning judgment logic. This logic performs independent judgments on each feature value in parallel.

[0042] The early warning trigger condition involves the edge computing module calculating the instantaneous offset of each feature quantity from its normal operating baseline in real time. When the offset of any feature quantity meets the following condition, it is considered to enter the early warning candidate state: (1) For the first characteristic quantity (transient apparent power) ) and the second characteristic quantity (transient impedance) ): When the absolute value deviation of its characteristic quantity exceeds a preset multiple of its baseline fluctuation range, a warning candidate state is triggered. The mathematical determination condition is as follows: in, The monitored feature quantity at time 10:00 The instantaneous value. In specific calculations, Substitute the first characteristic value into each Or the second characteristic quantity . Representative characteristic The arithmetic mean of its normal operating baseline model. It characterizes the central level of this characteristic under normal conditions. Representative characteristic The standard deviation within its normal operating baseline model. It characterizes the range of fluctuation of this characteristic quantity under normal conditions. The preset warning sensitivity coefficient (a dimensionless constant) is used to set the tightness of the warning threshold. This is the absolute value operator. Its function is to ensure that the characteristic value is maintained regardless of whether it deviates from its normal central value in a positive or negative direction. As long as the deviation is large enough, it can be triggered.

[0043] (2) For the third feature quantity (fusion high-frequency distortion feature) ): This characteristic quantity represents the degree of waveform distortion; its value is always positive and is generally considered better the smaller it is (closer to a pure sine wave). Therefore, its warning logic aims to detect abnormally rising trends. Its judgment condition is: Represents the third characteristic quantity at time t. The instantaneous value. Represents the third characteristic quantity The arithmetic mean in its normal operating benchmark model. This is a pre-set warning sensitivity coefficient specifically for the third characteristic quantity.

[0044] Early warning sensitivity coefficient and The value of must be less than its corresponding anomaly detection sensitivity coefficient. This key setting establishes a gradient alarm mechanism of "normal < early warning < anomaly," which is the core logic for achieving predictive maintenance.

[0045] Duration determination: The module starts an internal timer for each feature that enters the early warning candidate state and accumulates the timer value; when the feature exits the early warning candidate state, its corresponding timer is reset to zero. The module only generates and reports early warning information for a feature when the accumulated value of its timer exceeds its preset duration threshold.

[0046] The edge computing module executes a self-learning process during the initialization phase to establish a normal operation baseline model for the first time during periods of normal cable operation. Within a preset initialization learning period, the system collects data from normal cable operation and establishes normal operation baseline models for the first characteristic (transient apparent power), the second characteristic (transient impedance), and the third characteristic (integrated high-frequency distortion characteristics). For the first and second characteristic, the baseline model includes the mathematical expectation of their values ​​(…). ) and standard deviation ( ), to characterize the central position and dispersion of its normal fluctuations; for the third characteristic, the benchmark model is the mathematical expectation of its value ( Subsequently, to adapt to the long-term, slow changes in cable operating conditions, the system switched to an online adaptive update mode based on a sliding window mechanism to continuously optimize the baseline model.

[0047] The communication module, connected to the edge computing module and the positioning module, is used to upload anomaly information, early warning information, and positioning results to an external remote monitoring center. The anomaly information uploaded by the communication module includes the anomaly type, trigger time, and characteristic data that caused the anomaly.

[0048] The specific implementation process is as follows: After the device is powered on, the edge computing module executes a self-learning process. Within the preset initialization learning period, the system collects data on the cable during normal operation and establishes normal operation benchmark models (including mean values) for the first characteristic (transient apparent power), the second characteristic (transient impedance), and the third characteristic (fused high-frequency distortion characteristics). and standard deviation Subsequently, the system automatically switches to an online adaptive update mode based on a sliding window mechanism to continuously optimize the baseline model.

[0049] Meanwhile, the data acquisition module continues to operate. Current and voltage sensors respectively acquire the current signals from the cable. With voltage signal After the two are strictly synchronized by the synchronous acquisition card, the data is sent to the edge computing module.

[0050] The edge computing module receives synchronized data. and Then, the three features are computed in parallel: Calculate the transient apparent power of the first characteristic: Calculate the transient impedance of the second characteristic quantity: Calculate the third feature quantity to fuse high-frequency distortion features: Subsequently, the edge computing module compares the calculated real-time values ​​of the features with the baseline model and makes judgments according to the following logic: 1. Early Warning Judgment: Each feature quantity is checked in parallel. If any feature quantity meets its early warning trigger condition ( or If the duration of this state exceeds a preset threshold, an early warning message will be generated and reported.

[0051] 2. Anomaly Detection: Each feature is checked in parallel. If any feature meets its anomaly detection criteria ( or or If the cable is faulty, it is immediately identified as a cable malfunction.

[0052] When the edge computing module detects an anomaly, it immediately activates the positioning module. The positioning module uses traveling wave ranging to capture high-frequency traveling wave signals and calculate their propagation time. Through formula Calculate the precise location of the anomaly. .

[0053] Finally, the communication module packages the final results and uploads them to an external remote monitoring center.

[0054] If an anomaly occurs, report the anomaly information (including anomaly type, time, and relevant feature data) and the precise location result.

[0055] If an early warning is issued, the early warning information (including the warning type, time, relevant characteristic quantities and their trends) shall be reported.

[0056] After completing one judgment and reporting, the system immediately returns to the next data acquisition and processing cycle, realizing continuous and uninterrupted monitoring of the cable status.

[0057] The above descriptions are merely embodiments of the present invention, and common knowledge such as specific technical solutions and / or characteristics are not described in detail here. It should be noted that those skilled in the art can make various modifications and improvements without departing from the technical solutions of the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A cable anomaly identification and location device that integrates current and voltage data, characterized in that, include: The data acquisition module is used to synchronously acquire current and voltage signals on the cable line; An edge computing module, connected to the data acquisition module, is used to perform feature fusion calculations based on the current signal and voltage signal, and compare them with a pre-stored normal operation benchmark model to identify cable anomalies. The feature fusion calculation includes: calculating a first feature quantity reflecting the power transmission state of the cable based on the current signal and voltage signal, calculating a second feature quantity reflecting the impedance characteristic state of the cable, and calculating a third feature quantity reflecting the distortion state of the current or voltage waveform. The positioning module, connected to the edge computing module, is used to calculate the location of the abnormal point using the traveling wave ranging method after an anomaly is detected. The communication module, connected to the edge computing module and the positioning module, is used to upload abnormal information, early warning information and positioning results.

2. The cable anomaly identification and location device that integrates current and voltage data as described in claim 1, characterized in that, The logic for the edge computing module to identify cable anomalies is as follows: an anomaly is determined when any one of the first, second, or third feature quantities meets the following conditions: or or in, This represents the transient apparent power of the first characteristic quantity. and These are the mean and standard deviation of its normal operating benchmark model, respectively; This represents the transient impedance, which is the second characteristic quantity. and These are the mean and standard deviation of its normal operating benchmark model, respectively; This represents the high-frequency distortion feature fused with the third feature quantity. It is the mean of its normal operating benchmark model; , , This is a configurable sensitivity coefficient.

3. The cable anomaly identification and location device that integrates current and voltage data as described in claim 1, characterized in that, The formula for calculating the transient apparent power of the first characteristic quantity is: in, and These are the instantaneous voltage and current signals acquired synchronously, respectively.

4. The cable anomaly identification and location device that integrates current and voltage data as described in claim 1, characterized in that, The formula for calculating the second characteristic quantity, transient impedance, is as follows: in, and These are the instantaneous voltage and current signals acquired synchronously, respectively.

5. The cable anomaly identification and location device that integrates current and voltage data as described in claim 1, characterized in that, The calculation formula for the third feature quantity fused with high-frequency distortion features is as follows: in, and These are the total harmonic distortion rates of current and voltage, respectively. and The weighting coefficients are and satisfy the following conditions: .

6. The cable anomaly identification and location device that integrates current and voltage data as described in claim 1, characterized in that: The normal operating baseline model is a dynamic model that uses a sliding window mechanism for online adaptive updates.

7. The cable anomaly identification and location device that integrates current and voltage data as described in claim 1, characterized in that, The positioning formula for the traveling wave ranging method is as follows: in, Distance to outlier points For traveling wave speed, This refers to the time difference between the arrival of the traveling wave signal at different monitoring points, or the propagation time from the point of anomaly to the monitoring point.

8. The cable anomaly identification and location device that integrates current and voltage data as described in claim 1, characterized in that, The edge computing module also includes an early warning function: when the feature quantity continuously deviates from the normal operation benchmark model but does not reach the anomaly judgment threshold, the edge computing module generates and reports early warning information to achieve predictive maintenance. The edge computing module executes a self-learning process during the initialization phase to establish the normal operation benchmark model during the period when the cable is operating without abnormalities.

9. A cable anomaly identification and location device integrating current and voltage data as described in claim 1, characterized in that, The data acquisition module includes: A current sensor is used to acquire the current signal; A voltage sensor is used to acquire the voltage signal; A synchronous acquisition card is connected to the current sensor and voltage sensor respectively, and is used to convert analog signals into synchronous digital signals.

10. A cable anomaly identification and location device that integrates current and voltage data as described in claim 1, characterized in that: The abnormal information uploaded by the communication module includes the abnormality type, trigger time, and characteristic data that caused the abnormality.

Images