SPD state real-time monitoring method and system

Abstract

The invention discloses an SPD state real-time monitoring method and system, and relates to the technical field of SPD monitoring, a dynamic action critical window is constructed according to an SPD conduction threshold, and all disturbances falling into the dynamic action critical window are recorded as virtual action events; extracting virtual action events in a preset time window, and constructing a virtual action event sequence; constructing an action disuse index of each virtual action event by extracting action disuse sign information after each virtual action of each virtual action event in the virtual action event sequence; calculating a mean value of the action disusing indexes of the virtual action event sequence, and outputting an SPD state monitoring result according to the mean value of the action disusing indexes; therefore, whether the SPD performance is really influenced by high-frequency harmonic waves, PWM edge disturbance, grounding potential fluctuation and the like can be judged according to actual conditions, and a monitoring result is output in time, so that the balance between effective early warning and misjudgment reduction is realized, and the misjudgment rate of the real-time state of the SPD is reduced.

Description

Technical Field

[0001] This invention relates to the field of SPD monitoring technology, specifically to a method and system for real-time monitoring of SPD status. Background Technology

[0002] Surge protectors (SPDs) are key components in power and low-voltage systems to prevent lightning strikes and power surges, undertaking important tasks of overvoltage discharge and circuit protection. To ensure that SPDs are in an effective protective state during their operating cycle, current engineering projects generally configure SPD real-time status monitoring modules to provide timely warnings and maintenance / replacement when they degrade, fail, or show signs of deterioration. Existing SPD monitoring methods typically include the following two categories: (1) Remote signaling contact monitoring: determining whether the SPD has completely failed by monitoring whether the thermal fuse or thermal tripping mechanism of the SPD has operated; (2) Analog quantity monitoring: assessing the state life of the SPD by changes in physical quantities such as voltage, current, leakage current trend, and temperature rise. These methods have high accuracy when facing "obvious damage" and can capture major fault phenomena such as thermal breakdown and failure disconnection.

[0003] However, in some practical applications, there may be a large number of high-frequency harmonics, PWM edge disturbances, and ground potential fluctuations. These can cause SPDs to exhibit numerous "false alarms" or "failed actions" that are close to the conduction threshold. On the one hand, these disturbances may cause chronic degradation and sub-optimal health of the SPD device over a long period. On the other hand, in some cases, they may not substantially affect the SPD's performance but could be misjudged as device deterioration, leading to false alarms or excessive maintenance. Existing monitoring systems generally rely on "conduction" or "failure" as the basis for judgment, lacking the ability to fine-grainedly discern the true state after such micro-disturbances. This makes it difficult to simultaneously achieve a balance between "effective early warning" and "reduced false alarms," ​​resulting in a high false alarm rate. Summary of the Invention

[0004] The purpose of this invention is to solve the problems mentioned above and provide a method and system for real-time monitoring of SPD status.

[0005] In a first aspect of this invention, a method for real-time monitoring of SPD status is first proposed, the method comprising: A dynamic action critical window is constructed based on the SPD conduction threshold, and all disturbances falling into the dynamic action critical window are recorded as virtual action events. Extract virtual action events within a preset time window and construct a sequence of virtual action events; Extract the action failure indication information after each virtual action in the virtual action event sequence, and construct the action failure index for each virtual action event based on the action failure indication information; Calculate the mean of the action failure index of the dummy action event sequence, and output the SPD status monitoring results based on the mean of the action failure index.

[0006] Optionally, the action failure indication information includes a high-frequency electric field hysteresis out-of-phase index and a transient coupling offset index. The step of constructing the action failure index for each dummy action event based on the action failure indication information is as follows: The high-frequency electric field hysteresis out-of-phase index and transient coupling offset index are normalized and the units are removed. The normalized high-frequency electric field hysteresis out-of-phase index and transient coupling offset index are then weighted and summed to obtain the action failure index of each virtual action event.

[0007] Optionally, the calculation steps for the high-frequency electric field hysteresis out-of-phase index are as follows: For each dummy action event, obtain the peak voltage time corresponding to each dummy action event, and set the sampling interval with the peak voltage time as the center; The voltage change rate is calculated using the differential method based on the voltage across the SPD collected in the sampling interval. Remove the units from the voltage and voltage change rate across the SPD, and plot the voltage values ​​on the x-axis. The value corresponding to the rate of change of the electric field is on the ordinate. Construct a time series trajectory; plot the trajectory as a two-dimensional path to obtain the response loop trajectory during the disturbance process, and calculate the geometric center of the response loop trajectory; Calculate the distance between the geometric center of the response loop's trajectory and the starting point of the disturbance; The median voltage value in response to the circular trajectory As a dividing line, all points are divided into the left trajectory, i.e. The trajectory on the right, i.e. Calculate the area of ​​the trajectory on the left side respectively. and the area of ​​the trajectory on the right Area of ​​the trajectory on the left The calculation formula is: In the formula, This indicates the total number of points on the left trajectory. Indicates the first The x-coordinates of the trajectory points Indicates the first The ordinates of the trajectory points The left side shows the index of the points on the trajectory; the right side shows the area of ​​the trajectory. The calculation formula is: In the formula, This indicates the total number of points on the right-hand track. Indicates the first The x-coordinates of the trajectory points Indicates the first The ordinates of the trajectory points The index of the trajectory points on the right; The formula for calculating the area asymmetry index is as follows: In the formula, It is the area asymmetry index; The distance between the geometric center of the response ring trajectory and the starting point of the disturbance is normalized and mapped to the numerical range of 0-1. The normalized distance is then multiplied by the area asymmetry index to obtain the high-frequency electric field hysteresis out-of-phase index.

[0008] Optionally, the calculation steps for the transient coupling offset index are as follows: For each dummy action event, obtain the peak voltage time corresponding to each dummy action event, and set the sampling interval with the peak voltage time as the center; The voltages across the SPD corresponding to the sampling interval are extracted to obtain a voltage sequence, and the current corresponding to each voltage in the voltage sequence is extracted to obtain a current sequence. The voltage and current sequences are then normalized to the range of 0-1 to obtain normalized voltage and current sequences. The first voltage in the normalized voltage sequence Each sample value, The first one in the normalized current sequence Each sample value; The normalized voltage and current are treated as a two-dimensional trajectory point. ,calculate The change in vector direction between each pair of adjacent trajectory points yields a sequence of trajectory angle deviations. Specifically... In the formula, To minimize the value and prevent the denominator from being zero, it is generally set to a value of [value missing]. ; Represents the first in the trajectory angle deviation sequence One trajectory angle deviation, ; Indicates the total number of sampling points in the sampling interval; The absolute difference between two adjacent trajectory angle deviations in the trajectory angle deviation sequence is calculated to obtain the trajectory angle jump. Calculate the mean of all trajectory angle jumps and normalize them to map them to the range of 0-1. Use the normalized mean of trajectory angle jumps as the transient coupling offset index.

[0009] Optionally, the steps for outputting the SPD status monitoring results based on the mean of the action failure index are as follows: Compare the mean of the action failure index of the virtual action event sequence with the preset mean threshold. If the mean is not less than the preset mean threshold, it means that the SPD is in a sub-healthy state and its internal performance has deteriorated, and it needs to be replaced in time. If the mean is less than the preset mean threshold, it means that the SPD is currently performing well and has not deteriorated, so it does not need to be replaced for the time being.

[0010] In a second aspect of this invention, a real-time monitoring system for SPD status is provided, the system comprising: Virtual Action Module: Constructs a dynamic action critical window based on the SPD conduction threshold, and records all disturbances falling into the dynamic action critical window as virtual action events; Sequence module: Extracts virtual action events within a preset time window and constructs a sequence of virtual action events; Analysis module: Extracts action failure indication information after each virtual action in the virtual action event sequence, and constructs the action failure index for each virtual action event based on the action failure indication information; Monitoring module: Calculates the mean of the action failure index of the dummy action event sequence, and outputs the SPD status monitoring results based on the mean of the action failure index.

[0011] Optionally, the application of the analysis module includes: The information on failed actions includes the high-frequency electric field hysteresis out-of-phase index and the transient coupling offset index. The high-frequency electric field hysteresis out-of-phase index and the transient coupling offset index are normalized and the units are removed. The normalized high-frequency electric field hysteresis out-of-phase index and the transient coupling offset index are then weighted and summed to obtain the failed action index for each virtual action event.

[0012] Optionally, the analysis module further includes: Sampling interval module: For each dummy action event, obtain the peak voltage time corresponding to each dummy action event, and set the sampling interval with the peak voltage time as the center; Differential calculation module: Calculates the voltage change rate using the differential method based on the voltage across the SPD collected in the sampling interval; The response loop trajectory module removes units from the voltage and rate of change at both ends of the SPD, and uses the corresponding voltage values ​​as the horizontal axis. The value corresponding to the rate of change of the electric field is on the ordinate. Construct a time series trajectory; plot the trajectory as a two-dimensional path to obtain the response loop trajectory during the disturbance process, and calculate the geometric center of the response loop trajectory; Distance module: Calculates the distance between the geometric center of the response loop's trajectory and the starting point of the disturbance; Area calculation module: Calculates the median voltage value in response to the circular trajectory. As a dividing line, all points are divided into the left trajectory, i.e. The trajectory on the right, i.e. Calculate the area of ​​the trajectory on the left side respectively. and the area of ​​the trajectory on the right Area of ​​the trajectory on the left The calculation formula is: In the formula, This indicates the total number of points on the left trajectory. Indicates the first The x-coordinates of the trajectory points Indicates the first The ordinates of the trajectory points The left side shows the index of the points on the trajectory; the right side shows the area of ​​the trajectory. The calculation formula is: In the formula, This indicates the total number of points on the right-hand track. Indicates the first The x-coordinates of the trajectory points Indicates the first The ordinates of the trajectory points The index of the trajectory points on the right; Area asymmetry module: Calculates the area asymmetry index. The formula is as follows: In the formula, It is the area asymmetry index; High-frequency electric field hysteresis out-of-phase index module: The distance between the geometric center of the response ring trajectory and the starting point of the disturbance is normalized and mapped to the numerical range of 0-1. The normalized distance is then multiplied by the area asymmetry index to obtain the high-frequency electric field hysteresis out-of-phase index.

[0013] Optionally, the analysis module further includes: Sampling interval module: For each dummy action event, obtain the peak voltage time corresponding to each dummy action event, and set the sampling interval with the peak voltage time as the center; Normalization module: Extracts the voltage across the SPD corresponding to the sampling interval to obtain a voltage sequence, and extracts the current corresponding to each voltage in the voltage sequence to obtain a current sequence; normalizes the voltage sequence and current sequence to the range of 0-1 respectively to obtain normalized voltage and current sequences; where, The first voltage in the normalized voltage sequence Each sample value, The first one in the normalized current sequence Each sample value; Trajectory angle deviation module: Treats the normalized voltage and current as a two-dimensional trajectory point. ,calculate The change in vector direction between each pair of adjacent trajectory points yields a sequence of trajectory angle deviations. Specifically... In the formula, To minimize the value and prevent the denominator from being zero, it is generally set to a value of [value missing]. ; Represents the first in the trajectory angle deviation sequence One trajectory angle deviation, ; Indicates the total number of sampling points in the sampling interval; The trajectory angle jump module calculates the absolute difference between two adjacent trajectory angle deviations in the trajectory angle deviation sequence to obtain the trajectory angle jump. Transient Coupling Offset Index Module: Calculates the mean of all trajectory angle jumps, normalizes them, maps them to the range of 0-1, and uses the normalized mean of trajectory angle jumps as the transient coupling offset index.

[0014] Optionally, the monitoring module includes: The first comparison monitoring module compares the mean of the action failure index of the virtual action event sequence with the preset mean threshold. If the mean is not less than the preset mean threshold, it means that the SPD is in a sub-healthy state and its internal performance has deteriorated, requiring timely replacement. The second comparison monitoring module: If the mean value is less than the preset mean value threshold, it means that the SPD is currently performing well and has not deteriorated, and does not need to be replaced for the time being.

[0015] The beneficial effects of this invention are: This invention proposes a real-time monitoring method and system for SPD status. A dynamic action critical window is constructed based on the SPD conduction threshold, and all disturbances falling within this window are recorded as dummy action events. Dummy action events within a preset time window are extracted to construct a sequence. Attempt failure information is extracted from each dummy action event in the sequence, and an attempt failure index is constructed for each event. The mean of the attempt failure indices for the dummy action event sequence is calculated, and the SPD status monitoring result is output based on this mean. This allows for the determination of whether high-frequency harmonics, PWM edge disturbances, and ground potential fluctuations truly affect SPD performance, and timely output of monitoring results. This achieves a balance between effective early warning and reduced misjudgment, thereby reducing the misjudgment rate of the SPD's real-time status. Attached Figure Description

[0016] The invention will now be further described with reference to the accompanying drawings.

[0017] Figure 1 A flowchart of a method for real-time monitoring of SPD status; Figure 2 This is a framework diagram of a real-time SPD status monitoring system. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] This invention provides a method for real-time monitoring of SPD status. See also... Figure 1 , Figure 1 A flowchart illustrating a real-time monitoring method for SPD status provided in an embodiment of the present invention. The method includes the following steps: S1: Construct a dynamic action critical window based on the SPD conduction threshold, and record all disturbances falling into the dynamic action critical window as virtual action events; S2: Extract virtual action events within a preset time window and construct a sequence of virtual action events; S3: Extract the action failure indication information after each virtual action in the virtual action event sequence, and construct the action failure index of each virtual action event based on the action failure indication information; S4: Calculate the mean of the action failure index of the virtual action event sequence, and output the SPD status monitoring results based on the mean of the action failure index.

[0020] Based on the SPD status real-time monitoring method provided by the embodiments of the present invention, the above method can determine whether high-frequency harmonics, PWM edge disturbances and ground potential fluctuations have really affected the SPD performance according to the actual situation, and output the monitoring results in a timely manner, thereby achieving a balance between effective early warning and reducing false judgments and reducing the false judgment rate of SPD real-time status.

[0021] In one embodiment, a dynamic action critical window is constructed based on the SPD conduction threshold, and all disturbances falling into the dynamic action critical window are recorded as dummy action events. It should be noted that the SPD turn-on threshold refers to the electrical critical point at which the device (surge protector) begins to exhibit significant conductive (conduction) behavior. Generally, it is the maximum voltage value that the SPD can withstand. When this threshold is exceeded, the SPD will rapidly transition from a high-resistance state to a low-resistance state, initiating significant conductive behavior, thereby dissipating surge energy into the grounding system and suppressing overvoltage. This threshold is usually related to the characteristics of the varistor device (such as MOV, TVS, gas discharge tube, etc.) used inside the SPD. The turn-on threshold of common MOV-type SPDs is generally set at about 1.5 times the nominal operating voltage. For example, for an SPD with a rated operating voltage of 220V, its turn-on threshold is usually around 350V. The specific value of the SPD turn-on threshold depends on the actual SPD application scenario and is not limited.

[0022] Furthermore, in practical applications, when SPDs operate in specific electrical environments, such as rail transit power supply systems, high-end semiconductor manufacturing workshops, data center UPS parallel systems, or large industrial frequency converter drive environments, there may be numerous complex disturbance sources, including high-frequency harmonics, PWM edge disturbances, and ground potential fluctuations. Although these disturbances may not exceed the SPD's conduction threshold in a single instance, their high frequency, dense occurrence, and frequent approach to the conduction threshold often trigger repetitive small-amplitude conductance responses, heat dissipation accumulation, or a tendency for local capacitance structure breakdown within the SPD, manifesting as a series of "false actions" or "attempted actions." This state is usually not identified as abnormal in conventional monitoring systems because it neither triggers remote signaling nor enters a protection state, but in reality, it may erode the SPD's performance over a long period, leading to a "sub-healthy" or even "latent degradation" state. Therefore, by constructing a dynamic action critical window, extracting and analyzing false actions, and conducting trend tracking and disturbance structure feature analysis, we can use these as key inputs for subsequent health status evolution judgment, so as to achieve early perception and intervention of potential degradation risks of SPD.

[0023] In one implementation, the steps for constructing a dynamic action critical window based on the SPD conduction threshold are as follows: Let the SPD conduction threshold be denoted as Construct a dynamic action critical window as In the formula, This is the offset relative to the conduction threshold, and its value ranges from 10 to 35V. It should be noted that the general offset The offset cannot be too small (less than 10 volts) because interference from high-frequency harmonics, PWM edge disturbances, and ground potential fluctuations in the power grid often causes the voltage to briefly approach the conduction threshold. If the offset is too small, the dynamic window will be too narrow, making it impossible to effectively capture those micro-disturbances that approach but do not reach the conduction threshold. This results in a lack of awareness of the potential subconducting state of the SPD and misses early signals that may lead to performance degradation. On the other hand, the offset cannot be too large (e.g., more than 35 volts) because as the distance from the conduction threshold increases, the impact of voltage disturbances on the SPD gradually weakens. An excessively large offset will include many low-amplitude fluctuations that will not actually cause SPD degradation in the "virtual action," thereby reducing the accuracy and specificity of monitoring. In other words, if the dynamic window is too wide, it is easy to include many normal voltage fluctuations in the monitoring range, resulting in misjudgments and increased monitoring noise, which reduces the ability to identify real anomalies. For example, assuming the SPD's on-threshold A is 300 volts, and the offset B is set to 30 volts, the dynamic action critical window is (270 volts, 300 volts). This range covers the voltage range where the SPD is not yet on but is under high-voltage stress, enabling the capture of frequent critical disturbances without misjudging normal fluctuations of lower amplitude as abnormal. When the voltage across the SPD is within the dynamic action critical window, the time point and the short-term voltage waveform before and after it are considered as a single "disturbance unit," and this disturbance unit is marked as a dummy event. This disturbance unit includes not only the voltage at the peak moment but also the fluctuation process before and after the disturbance, used to capture the response behavior of the electric field. For example, assuming the SPD's on-threshold is 300V, the system's dynamic action critical window is set to 280V~299V. If, during a monitoring session, the voltage rises to 287V at t1 = 12.485mus and remains within this range for approximately 3 microseconds before falling back to a stable value, then the voltage waveform within a 5-microsecond range before and after t1 (i.e., [t1-5mus, t1+5mus]) will be automatically selected as a disturbance unit. This unit fully covers the voltage peak and recovery process corresponding to the entire disturbance and is used for subsequent dummy operation analysis. This disturbance unit can reflect how the voltage enters the critical window, at what rate it rises / falls, and whether it is accompanied by obvious hysteresis or signs of energy accumulation. The "disturbance unit" constructed in this way can serve as the basic analytical unit for dummy operation events, further extracting its high-frequency characteristics, phase evolution, asymmetry, and other indicators to assess whether the SPD has already generated microstructural responses or potential fatigue when it is not conducting. This setting helps to achieve sensitive detection of early performance degradation of the SPD without inducing actual discharge. By combining the physical characteristics of the SPD with the voltage fluctuation characteristics of the power system, this approach can ensure monitoring sensitivity while avoiding excessive false alarms, thereby achieving effective early warning and risk management of the SPD's health status.

[0024] In one embodiment, virtual action events within a preset time window are extracted, and a sequence of virtual action events is constructed. It should be noted that within a pre-set time range (e.g., two to five minutes), the monitored voltage disturbance data is continuously analyzed, and all disturbance events falling within the dynamic action critical window (i.e., the interval close to the SPD conduction threshold but not actually conducting) are identified and defined as "phantom action events." These phantom action events are then arranged in chronological order to form a continuous "phantom action event sequence," used for subsequent trend analysis and action failure index assessment. The two- to five-minute window length is chosen because PWM high-frequency disturbances or ground potential fluctuations generated in most industrial systems, such as frequency converters, welding machines, and rail transit power supplies, often form relatively continuous or periodic disturbance clusters within this time period. If the window is set too short (e.g., tens of seconds), only sporadic fluctuations may be captured, mistakenly indicating no obvious trend; conversely, if the time window is too long, multiple unrelated disturbances may be chained together, reducing monitoring accuracy. Therefore, a two- to five-minute window ensures both the correlation of disturbances and reflects sub-optimal or potentially abnormal cumulative trends.

[0025] In one embodiment, action failure indication information after each virtual action in the virtual action event sequence is extracted, and an action failure index for each virtual action event is constructed based on the action failure indication information. Specifically, the steps for constructing the action failure index for each dummy action event based on the information on action failure indications are as follows: The information on attempted actions includes the high-frequency electric field hysteresis out-of-phase index and the transient coupling offset index. These indices are normalized, units are removed, and the values ​​are mapped to the 0-1 interval. The normalized indices are then weighted and summed to obtain the attempted action index for each virtual action event. The calculation formula is as follows: In the formula This is the attempt failure index. These are the normalized high-frequency electric field hysteresis out-of-phase index and transient coupling offset index, respectively. These represent the preset weighting coefficients for the high-frequency electric field hysteresis out-of-phase index and the transient coupling offset index after multi-normalization processing, respectively. All are greater than 0; It should be noted that commonly used methods for removing dimensions include Min-Max normalization and Z-Score standardization, which will not be elaborated here. Settings should be set according to the actual situation, generally They are equal and their sum is 1, for example, Both can be 0.5 or 0.5.

[0026] In one embodiment, the calculation steps for the high-frequency electric field hysteresis out-of-phase index are as follows: For each dummy action event, obtain the peak voltage time corresponding to each dummy action event, denoted as _____. At the peak voltage time Set the sampling interval around the center; In the formula, The sampling time window (e.g., 10 μs) is set to ensure complete coverage of the disturbance and recovery process; the sampling interval is... Total number of sampling points ; Indicates the total number of sampling points in the sampling interval; The voltage across the SPD sampled in the sampling interval is denoted as... , The voltage change rate is calculated using the differential method based on the voltage across the SPD collected in the sampling interval. ; ; Will Voltage change rate Remove units, in terms of voltage The corresponding value is the x-axis. electric field change rate The corresponding value is the ordinate. Construct a time series trajectory; plot the trajectory as a two-dimensional path to obtain the circular trajectory of the response during the disturbance process, and calculate the geometric center of the circular trajectory of the response. : , ; Calculate the geometric center of the trajectory of the response ring. Deviation from the start point of the disturbance distance The calculation formula is: In the formula, for The corresponding value of the disturbance initiation point voltage; for The corresponding value of the rate of change of voltage at the disturbance initiation point; The median voltage value in response to the circular trajectory As a dividing line, all points are divided into the left trajectory, i.e. The trajectory on the right, i.e. Calculate the area of ​​the trajectory on the left side respectively. and the area of ​​the trajectory on the right Area of ​​the trajectory on the left The calculation formula is: In the formula, This indicates the total number of points on the left trajectory. Indicates the first The x-coordinates of the trajectory points Indicates the first The ordinates of the trajectory points The left side shows the index of the points on the trajectory; the right side shows the area of ​​the trajectory. The calculation formula is: In the formula, This indicates the total number of points on the right-hand track. Indicates the first The x-coordinates of the trajectory points Indicates the first The ordinates of the trajectory points The index of the trajectory points on the right; The formula for calculating the area asymmetry index is as follows: In the formula, It is the area asymmetry index; The geometric center of the response ring trajectory Deviation from the start point of the disturbance distance Normalization is performed to map the values ​​to the 0-1 range. The normalized distance is then multiplied by the area asymmetry index to obtain the high-frequency electric field hysteresis out-of-phase index.

[0027] It should be noted that the sampling interval ( ) and time window width ( The selection of the sampling interval needs to balance the temporal resolution of the disturbance response characteristics with data integrity. The typical value is taken as 10ns to 100ns to ensure that high-frequency components and edge abrupt features can be captured; while The typical value range is 2μs to 10μs, which means extending 1μs to 5μs to the left and right of the disturbance center point. The purpose is to cover the entire disturbance rise phase, peak point, and recovery phase to ensure that the failed action process is completely recorded. The specific value should be flexibly adjusted in combination with the SPD response characteristics, system sampling rate, and target analysis accuracy.

[0028] It should be noted that the data involved in the above calculations mainly comes from the high-time-resolution voltage monitoring signals at both ends of the SPD (surge protector) during disturbances. These signals are typically acquired in real time by high-speed data acquisition devices (such as high-frequency oscilloscopes or digital sampling devices) installed at the input and output ends of the SPD. The sampling frequency needs to be high enough (e.g., ≥1MHz) to ensure that the complete high-frequency disturbance response process is captured. During monitoring operation, all voltage data are continuously recorded and buffered. When the voltage signal is detected to enter the dynamic action critical window (i.e., approaching the SPD conduction threshold), a voltage waveform segment of fixed duration before and after the disturbance peak is automatically extracted and used as the raw data input for a single "dummy action event". Subsequently, the voltage change rate is calculated using the differential method, thus forming a two-dimensional trajectory of disturbance voltage versus voltage change rate, which is used for the subsequent calculation of the high-frequency electric field hysteresis out-of-phase index.

[0029] It should be noted that the high-frequency electric field hysteresis out-of-phase index refers to the degree of nonlinear hysteresis and out-of-phase response exhibited by the internal electric field of a surge protector (SPD) under high-frequency disturbances. It measures the magnitude and asymmetry of the SPD's electric field response deviating from its normal linear and synchronous state under rapidly changing voltage stimuli. A larger high-frequency electric field hysteresis out-of-phase index indicates a more significant deviation of the SPD's internal electric field response loop from its normal trajectory, with more prominent nonlinear effects and asynchronous behavior. This suggests that the SPD's dielectric material may have problems such as local breakdown, dielectric fatigue, or charge trapping, causing it to fail to respond to voltage surges in a timely and accurate manner, thus reducing its protection effectiveness. This abnormal state means the SPD is in a sub-healthy stage, with an increased risk of performance degradation and a higher probability of future failure or failure modes. Therefore, timely replacement of the SPD is necessary to prevent damage to subsequent power grid equipment due to protection failure.

[0030] It should be noted that the high-frequency electric field hysteresis out-of-phase index calculated using the above method can extract the essential performance degradation information of the SPD under high-frequency disturbances from the geometric behavior characteristics of the dynamic response, avoiding misjudgments or omissions that may be caused by relying solely on a single instantaneous voltage or conduction characteristic value. Traditional methods often use single-point characteristics such as voltage amplitude, conduction time, and response speed as performance indicators, but these methods cannot effectively reflect the overall response state of the SPD under complex high-frequency disturbance conditions, especially lacking the ability to characterize dynamic details such as hysteresis, nonlinearity, and out-of-phase. The above method, by constructing a two-dimensional trajectory of "voltage-voltage change rate" and introducing two factors, namely trajectory geometric center offset and area asymmetry, is equivalent to capturing the changes in microscopic physical characteristics such as time-phase hysteresis and structural non-uniform response caused by aging or degradation of the SPD in the dynamic response. The "high-frequency electric field hysteresis out-of-phase index" constructed by multiplying the two is more sensitive to early sub-health states. Even if the SPD has not yet shown obvious failure, slight deviations in its trajectory structure can be sensitively captured, which helps in early warning and lifespan management. Meanwhile, it avoids dependence on specific preconditions such as frequency and waveform characteristics, thus possessing broader adaptability and practical engineering value. Therefore, compared to other methods, this calculation method based on the geometric characteristics of the response trajectory is more advantageous in complex perturbation contexts.

[0031] In one embodiment, the calculation steps for the transient coupling offset index are as follows: For each dummy action event, obtain the peak voltage time corresponding to each dummy action event, denoted as _____. ,by Set the sampling interval around the center; In the formula, The sampling time window is set to ensure complete coverage of the disturbance and recovery process; the sampling interval is... Total number of sampling points ; Indicates the total number of sampling points in the sampling interval; The voltages across the SPD corresponding to the sampling interval are extracted to obtain a voltage sequence, and the current corresponding to each voltage in the voltage sequence is extracted to obtain a current sequence. The voltage and current sequences are then normalized to the range of 0-1, and units are removed to obtain normalized voltage and current sequences. The first voltage in the normalized voltage sequence Each sample value, The first one in the normalized current sequence Each sample value; The normalized voltage and current are treated as a two-dimensional trajectory point. ,calculate The change in vector direction between each pair of adjacent trajectory points yields a sequence of trajectory angle deviations. Specifically... In the formula, To minimize the value and prevent the denominator from being zero, it is generally set to a value of [value missing]. ; Represents the first in the trajectory angle deviation sequence One trajectory angle deviation, ; The absolute difference between two adjacent trajectory angle deviations in the trajectory angle deviation sequence is calculated to obtain the trajectory angle jump. , For the first A trajectory angle jump degree, If the transient response is "decoupled", then the angle change is significant, that is, the trajectory direction changes drastically between consecutive points.

[0032] Calculate the mean of all trajectory angle jumps and normalize them, mapping them to the range of 0-1. Use the normalized mean trajectory angle jump as the transient coupling offset index; specifically, the normalization process involves dividing the mean trajectory angle jump by... .

[0033] It should be noted that commonly used methods for removing dimensions include Min-Max normalization and Z-Score normalization, which will not be elaborated upon here. Furthermore, during the calculation process, the voltage and current data involved are acquired in real time using high-speed data acquisition devices, typically connected to both ends of the SPD and its series circuit. Data acquisition relies on high-precision voltage probes and current sensors (such as Hall current sensors or current transformers), coupled with a data acquisition system with a high sampling rate. Each time a dummy event is detected, continuous sampling data within a certain time window is extracted, centered on the voltage peak time.

[0034] It should be noted that the transient coupling offset index reflects the stability and continuity of the coupling relationship between voltage and current in a surge protection device (SPD) when subjected to rapidly changing voltage disturbances. A large index indicates a significant "decoupling" phenomenon in the transient responses of voltage and current, meaning the correlation between them is weakened. This manifests as a sharp jump in the vector direction between normalized trajectory points, reflecting anomalies in charge migration, dielectric response, or connection status within the SPD. Such abnormal coupling offset is usually caused by the influence of complex electromagnetic interference such as high-frequency harmonics, PWM (pulse width modulation) edge disturbances, and ground potential fluctuations, leading to the current response failing to keep pace with voltage changes, resulting in phase shifts and amplitude mismatches. For example, when PWM modulation devices in the power grid frequently switch, the voltage signal changes rapidly. If the transient current within the SPD cannot change synchronously, the trajectory direction will jump. This jump reflects the instability and abnormality of the SPD response, indicating that the SPD may be in a sub-healthy state, such as material fatigue, structural aging, or loose connections. Therefore, the larger the transient coupling offset index, the more severe the interference the SPD is subjected to, the higher the risk of internal performance degradation, and the more significantly its protection effect and lifespan will be affected. By monitoring this index, potential SPD failures can be predicted in advance, preventing equipment failure at critical moments.

[0035] It should be noted that by calculating the transient coupling offset index based on the vector direction change and trajectory angle jump of the normalized two-dimensional voltage and current trajectories, the dynamic changes in the coupling between voltage and current in the SPD under transient disturbances can be accurately captured, especially reflecting subtle abnormal jumps in phase and amplitude between the two. This method avoids the nonlinear and abrupt characteristics that traditional indicators that rely solely on amplitude or phase difference may ignore, and more comprehensively represents the degree of "decoupling" in the transient response of the SPD, reflecting the influence of microscopic defects such as material aging, dielectric inhomogeneity, or loose connections within the SPD. Compared with other methods that may simply calculate correlation coefficients or peak differences, this method is more sensitive to complex disturbances with high frequency and rapid abrupt changes, thus more accurately reflecting the sub-health state and potential risks of the SPD.

[0036] In one embodiment, the steps of calculating the mean of the action failure index of the dummy action event sequence and outputting the SPD status monitoring result based on the mean of the action failure index are as follows: Compare the mean of the action failure index of the virtual action event sequence with the preset mean threshold. If the mean is not less than the preset mean threshold, it means that the SPD is in a sub-healthy state and its internal performance has deteriorated, and it needs to be replaced in time. If the mean is less than the preset mean threshold, it means that the SPD is currently performing well and has not deteriorated, so it does not need to be replaced for the time being.

[0037] It's important to note that calculating the mean of the failure index of the dummy operation event sequence and comparing it with a preset mean threshold is a crucial step in assessing the health of a surge protection device (SPD). Specifically, by averaging the failure index of all dummy operation events over a period of time, a comprehensive indicator of the SPD's overall performance can be obtained. If this mean reaches or exceeds the preset threshold, it indicates that the SPD has experienced numerous "failures," meaning it repeatedly approached protection activation but failed to fully operate. This usually signifies that its internal components may have begun to deteriorate, its function is nearing failure, it is in a sub-healthy state, and there is a significant risk of failure, requiring timely replacement to prevent equipment damage. Conversely, if the mean is significantly lower than the threshold, it indicates that the SPD's internal state is relatively stable, no abnormal accumulation has occurred, its performance remains good, and it can continue to operate safely without immediate replacement. For example, in a power system, if the failure index mean threshold is set at 0.6 and the measured mean is 0.7, maintenance personnel should pay attention and arrange for SPD replacement as soon as possible; if the mean is 0.3, it indicates that the equipment is in good condition and its service life can be extended. This method enables dynamic monitoring and early warning of SPD operating status, improving the security of equipment management.

[0038] Based on the same inventive concept, this invention also provides an SPD real-time status monitoring system. See [link to relevant documentation] Figure 2 , Figure 2 This invention provides a framework diagram of a real-time SPD status monitoring system, which includes: Virtual Action Module: Constructs a dynamic action critical window based on the SPD conduction threshold, and records all disturbances falling into the dynamic action critical window as virtual action events; Sequence module: Extracts virtual action events within a preset time window and constructs a sequence of virtual action events; Analysis module: Extracts action failure indication information after each virtual action in the virtual action event sequence, and constructs the action failure index for each virtual action event based on the action failure indication information; Monitoring module: Calculates the mean of the action failure index of the dummy action event sequence, and outputs the SPD status monitoring results based on the mean of the action failure index.

[0039] Based on the SPD real-time status monitoring system provided by the embodiments of the present invention, the above method can determine whether high-frequency harmonics, PWM edge disturbances and ground potential fluctuations have really affected the SPD performance according to the actual situation, and output the monitoring results in a timely manner, thereby achieving a balance between effective early warning and reducing false judgments and reducing the false judgment rate of SPD real-time status.

[0040] In one embodiment, the application of the analysis module includes: The information on attempted actions includes the high-frequency electric field hysteresis out-of-phase index and the transient coupling offset index. The high-frequency electric field hysteresis out-of-phase index and the transient coupling offset index are normalized and the units are removed. The normalized high-frequency electric field hysteresis out-of-phase index and the transient coupling offset index are then weighted and summed to obtain the attempted action index for each virtual action event.

[0041] In one embodiment, the analysis module further includes: Sampling interval module: For each dummy action event, obtain the peak voltage time corresponding to each dummy action event, and set the sampling interval with the peak voltage time as the center; Differential calculation module: Calculates the voltage change rate using the differential method based on the voltage across the SPD collected in the sampling interval; The response loop trajectory module removes units from the voltage and rate of change at both ends of the SPD, and uses the corresponding voltage values ​​as the horizontal axis. The value corresponding to the rate of change of the electric field is on the ordinate. Construct a time series trajectory; plot the trajectory as a two-dimensional path to obtain the response loop trajectory during the disturbance process, and calculate the geometric center of the response loop trajectory; Distance module: Calculates the distance between the geometric center of the response loop's trajectory and the starting point of the disturbance; Area calculation module: Calculates the median voltage value in response to the circular trajectory. As a dividing line, all points are divided into the left trajectory, i.e. The trajectory on the right, i.e. Calculate the area of ​​the trajectory on the left side respectively. and the area of ​​the trajectory on the right Area of ​​the trajectory on the left The calculation formula is: In the formula, This indicates the total number of points on the left trajectory. Indicates the first The x-coordinates of the trajectory points Indicates the first The ordinates of the trajectory points The left side shows the index of the points on the trajectory; the right side shows the area of ​​the trajectory. The calculation formula is: In the formula, This indicates the total number of points on the right-hand track. Indicates the first The x-coordinates of the trajectory points Indicates the first The ordinates of the trajectory points The index of the trajectory points on the right; Area asymmetry module: Calculates the area asymmetry index. The formula is as follows: In the formula, It is the area asymmetry index; High-frequency electric field hysteresis out-of-phase index module: The distance between the geometric center of the response ring trajectory and the starting point of the disturbance is normalized and mapped to the numerical range of 0-1. The normalized distance is then multiplied by the area asymmetry index to obtain the high-frequency electric field hysteresis out-of-phase index.

[0042] In one embodiment, the analysis module further includes: Sampling interval module: For each dummy action event, obtain the peak voltage time corresponding to each dummy action event, and set the sampling interval with the peak voltage time as the center; Normalization module: Extracts the voltage across the SPD corresponding to the sampling interval to obtain a voltage sequence, and extracts the current corresponding to each voltage in the voltage sequence to obtain a current sequence; normalizes the voltage sequence and current sequence to the range of 0-1 respectively to obtain normalized voltage and current sequences; where, The first voltage in the normalized voltage sequence Each sample value, The first one in the normalized current sequence Each sample value; Trajectory angle deviation module: Treats the normalized voltage and current as a two-dimensional trajectory point. ,calculate The change in vector direction between each pair of adjacent trajectory points yields a sequence of trajectory angle deviations. Specifically... In the formula, To minimize the value and prevent the denominator from being zero, it is generally set to a value of [value missing]. ; Represents the first in the trajectory angle deviation sequence One trajectory angle deviation, ; Indicates the total number of sampling points in the sampling interval; The trajectory angle jump module calculates the absolute difference between two adjacent trajectory angle deviations in the trajectory angle deviation sequence to obtain the trajectory angle jump. Transient Coupling Offset Index Module: Calculates the mean of all trajectory angle jumps, normalizes them, maps them to the range of 0-1, and uses the normalized mean of trajectory angle jumps as the transient coupling offset index.

[0043] In one embodiment, the monitoring module includes: The first comparison monitoring module compares the mean of the action failure index of the virtual action event sequence with the preset mean threshold. If the mean is not less than the preset mean threshold, it means that the SPD is in a sub-healthy state and its internal performance has deteriorated, requiring timely replacement. The second comparison monitoring module: If the mean value is less than the preset mean value threshold, it means that the SPD is currently performing well and has not deteriorated, and does not need to be replaced for the time being.

[0044] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A method for real-time monitoring of SPD status, characterized in that, Includes the following steps: A dynamic action critical window is constructed based on the SPD conduction threshold, and all disturbances falling into the dynamic action critical window are recorded as virtual action events. Extract virtual action events within a preset time window and construct a sequence of virtual action events; Extract the action failure indication information after each virtual action in the virtual action event sequence, and construct the action failure index for each virtual action event based on the action failure indication information; Calculate the mean of the action failure index of the dummy action event sequence, and output the SPD status monitoring results based on the mean of the action failure index.

2. The SPD status real-time monitoring method according to claim 1, characterized in that, The action failure indication information includes the high-frequency electric field hysteresis out-of-phase index and the transient coupling offset index. The steps for constructing the action failure index for each virtual action event based on the action failure indication information are as follows: The high-frequency electric field hysteresis out-of-phase index and transient coupling offset index are normalized and the units are removed. The normalized high-frequency electric field hysteresis out-of-phase index and transient coupling offset index are then weighted and summed to obtain the action failure index of each virtual action event.

3. The SPD status real-time monitoring method according to claim 2, characterized in that, The calculation steps for the high-frequency electric field hysteresis out-of-phase index are as follows: For each dummy action event, obtain the peak voltage time corresponding to each dummy action event, and set the sampling interval with the peak voltage time as the center; The voltage change rate is calculated using the differential method based on the voltage across the SPD collected in the sampling interval. Remove the units from the voltage and voltage change rate across the SPD, and plot the voltage values ​​on the x-axis. The value corresponding to the rate of change of the electric field is on the ordinate. Construct a time series trajectory; plot the trajectory as a two-dimensional path to obtain the response loop trajectory during the disturbance process, and calculate the geometric center of the response loop trajectory; Calculate the distance between the geometric center of the response loop's trajectory and the starting point of the disturbance; The median voltage value in response to the circular trajectory As a dividing line, all points are divided into the left trajectory, i.e. The trajectory on the right, i.e. Calculate the area of ​​the trajectory on the left side respectively. and the area of ​​the trajectory on the right Area of ​​the trajectory on the left The calculation formula is: In the formula, This indicates the total number of points on the left trajectory. Indicates the first The x-coordinates of the trajectory points Indicates the first The ordinates of the trajectory points The left side shows the index of the points on the trajectory; the right side shows the area of ​​the trajectory. The calculation formula is: In the formula, This indicates the total number of points on the right-hand track. Indicates the first The x-coordinates of the trajectory points Indicates the first The ordinates of the trajectory points The index of the trajectory points on the right; The formula for calculating the area asymmetry index is as follows: In the formula, It is the area asymmetry index; The distance between the geometric center of the response ring trajectory and the starting point of the disturbance is normalized and mapped to the numerical range of 0-1. The normalized distance is then multiplied by the area asymmetry index to obtain the high-frequency electric field hysteresis out-of-phase index.

4. The SPD status real-time monitoring method according to claim 2, characterized in that, The calculation steps for the transient coupling offset index are as follows: For each dummy action event, obtain the peak voltage time corresponding to each dummy action event, and set the sampling interval with the peak voltage time as the center; The voltages across the SPD corresponding to the sampling interval are extracted to obtain a voltage sequence, and the current corresponding to each voltage in the voltage sequence is extracted to obtain a current sequence. The voltage and current sequences are then normalized to the range of 0-1 to obtain normalized voltage and current sequences. The first voltage in the normalized voltage sequence Each sample value, The first one in the normalized current sequence Each sample value; The normalized voltage and current are treated as a two-dimensional trajectory point. ,calculate The change in vector direction between each pair of adjacent trajectory points yields a sequence of trajectory angle deviations. Specifically... In the formula, To minimize the value and prevent the denominator from being zero, it is generally set to a value of [value missing]. ; Represents the first in the trajectory angle deviation sequence One trajectory angle deviation, ; Indicates the total number of sampling points in the sampling interval; The absolute difference between two adjacent trajectory angle deviations in the trajectory angle deviation sequence is calculated to obtain the trajectory angle jump. Calculate the mean of all trajectory angle jumps and normalize them to map them to the range of 0-1. Use the normalized mean of trajectory angle jumps as the transient coupling offset index.

5. The SPD status real-time monitoring method according to claim 1, characterized in that, The steps for outputting SPD status monitoring results based on the mean of the action failure index are as follows: Compare the mean of the action failure index of the virtual action event sequence with the preset mean threshold. If the mean is not less than the preset mean threshold, it means that the SPD is in a sub-healthy state and its internal performance has deteriorated, and it needs to be replaced in time. If the mean is less than the preset mean threshold, it means that the SPD is currently performing well and has not deteriorated, so it does not need to be replaced for the time being.

6. A real-time monitoring system for SPD status, characterized in that, The system includes: Virtual Action Module: Constructs a dynamic action critical window based on the SPD conduction threshold, and records all disturbances falling into the dynamic action critical window as virtual action events; Sequence module: Extracts virtual action events within a preset time window and constructs a sequence of virtual action events; Analysis module: Extracts action failure indication information after each virtual action in the virtual action event sequence, and constructs the action failure index for each virtual action event based on the action failure indication information; Monitoring module: Calculates the mean of the action failure index of the dummy action event sequence, and outputs the SPD status monitoring results based on the mean of the action failure index.

7. The SPD real-time status monitoring system according to claim 6, characterized in that, The applications of the analysis module include: The information on failed actions includes the high-frequency electric field hysteresis out-of-phase index and the transient coupling offset index. The high-frequency electric field hysteresis out-of-phase index and the transient coupling offset index are normalized and the units are removed. The normalized high-frequency electric field hysteresis out-of-phase index and the transient coupling offset index are then weighted and summed to obtain the failed action index for each virtual action event.

8. The SPD status real-time monitoring system according to claim 7, characterized in that, The analysis module also includes: Sampling interval module: For each dummy action event, obtain the peak voltage time corresponding to each dummy action event, and set the sampling interval with the peak voltage time as the center; Differential calculation module: Calculates the voltage change rate using the differential method based on the voltage across the SPD collected in the sampling interval; The response loop trajectory module removes units from the voltage and voltage change rate across the SPD, and uses the corresponding voltage values ​​as the horizontal axis. The value corresponding to the rate of change of the electric field is on the ordinate. Construct a time series trajectory; plot the trajectory as a two-dimensional path to obtain the response loop trajectory during the disturbance process, and calculate the geometric center of the response loop trajectory; Distance module: Calculates the distance between the geometric center of the response loop's trajectory and the starting point of the disturbance; Area calculation module: Calculates the median voltage value in response to the circular trajectory. As a dividing line, all points are divided into the left trajectory, i.e. The trajectory on the right, i.e. Calculate the area of ​​the trajectory on the left side respectively. and the area of ​​the trajectory on the right Area of ​​the trajectory on the left The calculation formula is: In the formula, This indicates the total number of points on the left trajectory. Indicates the first The x-coordinates of the trajectory points Indicates the first The ordinates of the trajectory points The left side shows the index of the points on the trajectory; the right side shows the area of ​​the trajectory. The calculation formula is: In the formula, This indicates the total number of points on the right-hand track. Indicates the first The x-coordinates of the trajectory points Indicates the first The ordinates of the trajectory points The index of the trajectory points on the right; Area asymmetry module: Calculates the area asymmetry index. The formula is as follows: In the formula, It is the area asymmetry index; High-frequency electric field hysteresis out-of-phase index module: The distance between the geometric center of the response ring trajectory and the starting point of the disturbance is normalized and mapped to the numerical range of 0-1. The normalized distance is then multiplied by the area asymmetry index to obtain the high-frequency electric field hysteresis out-of-phase index.

9. The SPD status real-time monitoring system according to claim 7, characterized in that, The analysis module also includes: Sampling interval module: For each dummy action event, obtain the peak voltage time corresponding to each dummy action event, and set the sampling interval with the peak voltage time as the center; Normalization module: Extracts the voltage across the SPD corresponding to the sampling interval to obtain a voltage sequence, and extracts the current corresponding to each voltage in the voltage sequence to obtain a current sequence; normalizes the voltage sequence and current sequence to the range of 0-1 respectively to obtain normalized voltage and current sequences; where, The first voltage in the normalized voltage sequence Each sample value, The first one in the normalized current sequence Each sample value; Trajectory angle deviation module: Treats the normalized voltage and current as a two-dimensional trajectory point. ,calculate The change in vector direction between each pair of adjacent trajectory points yields a sequence of trajectory angle deviations. Specifically... In the formula, To minimize the value and prevent the denominator from being zero, it is generally set to a value of [value missing]. ; Represents the first in the trajectory angle deviation sequence One trajectory angle deviation, ; Indicates the total number of sampling points in the sampling interval; The trajectory angle jump module calculates the absolute difference between two adjacent trajectory angle deviations in the trajectory angle deviation sequence to obtain the trajectory angle jump. Transient Coupling Offset Index Module: Calculates the mean of all trajectory angle jumps, normalizes them, maps them to the range of 0-1, and uses the normalized mean of trajectory angle jumps as the transient coupling offset index.

10. A real-time SPD status monitoring system according to claim 6, characterized in that, The monitoring module includes: The first comparison monitoring module compares the mean of the action failure index of the virtual action event sequence with the preset mean threshold. If the mean is not less than the preset mean threshold, it means that the SPD is in a sub-healthy state and its internal performance has deteriorated, requiring timely replacement. The second comparison monitoring module: If the mean value is less than the preset mean value threshold, it means that the SPD is currently performing well and has not deteriorated, and does not need to be replaced for the time being.

Images