AC / DC power supply monitoring system

Abstract

The invention discloses an AC / DC power supply monitoring system, and relates to the technical field of AC / DC power supply monitoring, and the system comprises a feature collection module, a time sequence coupling module and a damage monitoring module. The feature collection module is used for collecting an amplitude change curve and a time domain duration window for low-amplitude transient overvoltage which does not exceed a safety threshold; dividing a monitoring time domain section and calculating an overvoltage generation density; the time sequence coupling module executes time sequence adjacency correction on adjacent overvoltage, extracts a damage effective action section and generates a superposition action curve; and the damage monitoring module calculates a damage superposition coefficient based on the superposition effect curve and divides early warning risk levels. According to the invention, accurate monitoring of low-amplitude transient overvoltage accumulated damage is realized, equipment aging and sudden faults are avoided, and the operation stability of an AC / DC power supply system is improved.

Description

Technical Field

[0001] This invention relates to the field of AC / DC power supply monitoring technology, specifically to an AC / DC power supply monitoring system. Background Technology

[0002] AC / DC power supply systems are widely used in industrial production, energy storage and other fields. Their operational stability directly affects the safety and reliability of related equipment. Among them, transient overvoltage is a key factor threatening the safety of AC / DC power supply systems. Most existing monitoring technologies focus on overvoltages that exceed the safety threshold, but ignore the potential hazards of low-amplitude transient overvoltages that do not exceed the safety threshold. Although a single application of such low-amplitude transient overvoltage may not cause obvious damage, repeated timing effects can have a cumulative effect. Long-term effects can lead to accelerated aging of the equipment insulation layer, shortened lifespan of power devices, and ultimately cause sudden failures. For such low-amplitude transient overvoltages, existing technologies do not analyze their temporal correlation and lack a temporal correction mechanism for adjacent overvoltages within the same monitoring period, thus failing to accurately capture their damage superposition patterns. In addition, the existing solutions do not provide a method for extracting the effective damage segment of low-amplitude transient overvoltages, and the monitoring relies only on a single overvoltage event for judgment. They have not built a damage superposition coefficient calculation system based on the entire monitoring cycle, and cannot realize graded early warning of cumulative equipment damage. Therefore, there is an urgent need for an AC / DC power supply monitoring system. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides an AC / DC power supply monitoring system that solves the problems of insufficient monitoring and lack of timing correlation analysis for accumulated damage from low-amplitude transient overvoltages.

[0004] To achieve the above objectives, the present invention provides the following technical solution: an AC / DC power supply monitoring system, comprising: The feature acquisition module acquires the amplitude change curve and time domain duration window of each low amplitude transient overvoltage in the AC / DC power supply system that does not exceed the safety threshold. It divides the time axis into equal-length continuous monitoring time domain segments based on the preset continuous monitoring duration and calculates the overvoltage occurrence density of each monitoring time domain segment. The time domain duration window is the time period from the occurrence start time of the low amplitude transient overvoltage to the decay end time. The timing coupling module performs timing adjacency correction on adjacent low-amplitude transient overvoltages within the same monitoring time domain segment based on the time domain duration window of the low-amplitude transient overvoltage and the corresponding monitoring time domain segment. It also extracts the effective damage segment from the amplitude change curve of the corrected low-amplitude transient overvoltage and generates the superimposed effect curve corresponding to each monitoring time domain segment. The damage monitoring module calculates the damage superposition coefficient based on the superposition curve corresponding to each monitoring time segment and classifies the early warning risk level.

[0005] As a further aspect of the present invention, the specific steps for calculating the overvoltage occurrence density for each continuous monitoring time segment are as follows: Calculate the time domain overlap duration between the time domain duration window of each low-amplitude transient overvoltage and the corresponding monitoring time domain interval, that is, the intersection duration between the time domain duration window of the overvoltage and the monitoring time domain interval; From the amplitude variation curve of this low-amplitude transient overvoltage, the amplitude interval corresponding to the time-domain overlap duration is selected, the mean value α of the amplitude corresponding to the amplitude interval is calculated, and the damage equivalence coefficient is calculated based on the mean value of the amplitude. Multiply the time-domain overlap duration by the damage equivalence coefficient to obtain the damage equivalent overlap duration, and divide the damage equivalent overlap duration by the duration of the monitoring time-domain interval corresponding to the low-amplitude transient overvoltage to obtain the damage correlation overlap ratio between the low-amplitude transient overvoltage and the corresponding monitoring time-domain interval. For each monitoring time segment, the damage correlation overlap ratio of all low-amplitude transient overvoltages in that time segment is accumulated to obtain the equivalent total number of overvoltage damages in that time segment. Then, this value is divided by the duration of the corresponding monitoring time segment to obtain the overvoltage occurrence density of low-amplitude transient overvoltages in that monitoring time segment.

[0006] As a further aspect of the present invention, according to the formula Calculate the damage equivalence coefficient, where Us is the safety threshold for low-amplitude transient overvoltage, Up is the peak amplitude of a single low-amplitude transient overvoltage, and q is the influence factor.

[0007] As a further aspect of the present invention, the specific operation for performing timing adjacency correction is as follows: Extract overvoltage occurrence density during the monitoring time domain Simultaneously, the time-domain duration window T of all low-amplitude transient overvoltages within the monitoring time domain is correlated, and a time difference identification threshold is constructed based on the two. Where mean() is the mean function; For all low-amplitude transient overvoltages within this monitoring time period, they are sorted according to the chronological order of their occurrence start times. The temporal correlation between adjacent low-amplitude transient overvoltages is compared one by one, specifically the time difference Δt between the decay termination time of the preceding low-amplitude transient overvoltage and the occurrence start time of the following low-amplitude transient overvoltage. This is combined with the aforementioned determined... Identify adjacency types and correction strategies: like If the two are found to be heavily overlapping adjacencies, then depth-time correction is performed. like If the condition is met, it is determined to be a slightly overlapping adjacency, and a simplified timing correction is performed. like If it is, it is determined to be a close-spaced adjacency, and no timing correction is required, but its adjacency type is marked; like If the timing is loosely spaced, it is determined to be a loosely spaced adjacency, and no timing correction is required, but the timing interval Δt needs to be recorded.

[0008] As a further aspect of the present invention, the execution of deep timing correction specifically includes: extracting the amplitude variation curves of two low-amplitude transient overvoltages, and locating the peak moment of the amplitude variation curve of the previous low-amplitude transient overvoltage. The starting time of the subsequent low-amplitude transient overvoltage. Revised to ; The simplified timing correction specifically includes: only adjusting the decay termination time of the previous low-amplitude transient overvoltage. Revised to .

[0009] As a further aspect of the present invention, the specific steps for extracting the effective damage segment are as follows: Based on the amplitude variation curve of the corrected low-amplitude transient overvoltage, which contains several continuous sampling points Where i∈[1,n], the horizontal axis represents time t, and the vertical axis represents the amplitude U. The start time of occurrence, This is the time when the decay ends; For each continuous sampling point in the amplitude variation curve , According to the formula Calculate the rate of change of amplitude within this sampling interval. ; From the start time of the sampling point Traverse all backwards Find the first boundary sampling point where the rate of change of amplitude changes from increasing to constant and stable. , and the time at that dividing point Defined as the inflection point of amplitude evolution However, it is necessary to ensure ; by Using this as the unique dividing point, the time interval is... The corresponding amplitude change curve is marked as the amplitude rising segment, time interval The corresponding amplitude change curve is marked as the amplitude decay segment, and the amplitude decay segment is regarded as the effective damage segment of this low amplitude transient overvoltage. Simultaneously record key data of the effective target segment of the injury, including the start time. Termination time Time-domain duration window T, sampling point amplitude and the rate of change of amplitude at adjacent sampling points .

[0010] As a further aspect of the present invention, the judgment rule for the increasing amplitude change rate is: three or more consecutive adjacent amplitude change rates satisfy monotonically increasing and are all positive values; the judgment rule for the uniform and stable amplitude change rate is: three or more consecutive adjacent amplitude change rates satisfy the absolute value of the difference between two adjacent amplitude change rates ≤ ε, and the amplitude change rates are all negative values, where ε is a preset minimum error amount.

[0011] As a further aspect of the present invention, the specific steps for generating the superposition curve corresponding to each monitoring time segment are as follows: For a single monitoring time segment, core data of the effective damage action period corresponding to all low-amplitude transient overvoltages within that time segment are extracted, and a unified reference time axis for that monitoring time segment is determined, i.e., the start time of that time segment is taken as... The termination time is The time axis range is obtained as follows The time intervals of all effective damage action segments within the time domain are mapped one by one onto the unified reference time axis according to their actual occurrence order, while retaining the amplitude and amplitude change rate of each effective damage action segment. For all effective damage segments after timing alignment, perform in-phase amplitude direct superposition operation, that is, take the sampling time of the unified reference time axis as the anchor point, sum the amplitudes of all effective damage segments at the same sampling time, obtain the total superposition amplitude at that sampling time, and form the superposition curve; For the amplitude abrupt change points that appear in the superposition curve, the amplitude change rate of each effective damage segment before and after the abrupt change point is extracted, and the amplitude abrupt change points are corrected.

[0012] As a further aspect of the present invention, the specific rule for determining the amplitude abrupt change point is: if the amplitude difference at the junction of two adjacent effective damage segments is less than 1 / 3, the amplitude difference between the two segments is less than 1 / 3. If the amplitude change is not observed, the transition point is determined to be a point of abrupt change in amplitude; otherwise, it is determined to be a normal amplitude transition. Here, Δtr is the sampling time interval of the unified reference time axis. This represents the rate of change of amplitude at the end of the previous effective damage period. This is the amplitude at the end of the effective damage segment of the previous injury. This is the amplitude at the beginning of the next effective damage segment; The specific operation for correcting amplitude abrupt change points is as follows: extract the amplitude change rate at the beginning of the next effective action segment. ; According to the formula Calculate the correction value for the amplitude abrupt change point .

[0013] As a further aspect of the present invention, the specific steps for calculating the damage superposition coefficient are as follows: Using the entire monitoring cycle as a unified reference time axis, the superposition curves corresponding to each monitoring time segment are mapped one by one to the unified reference time axis according to the order of their corresponding monitoring time segments. Feature extraction is performed on each curve in the set of damage effect curves, including the duration of non-zero amplitude of the curve T_i, the peak amplitude of the curve U_pi, and the uniform decay rate of the curve v_i; The duration of the non-zero amplitude of the curve refers to the sum of the durations of all continuous non-zero amplitude segments after removing the zero amplitude points of the damage curve; The peak amplitude of the curve refers to the set of non-zero amplitude points obtained from the damage curve, and the maximum value among them is taken; The uniform decay rate of the curve refers to the calculation of the slope of each continuous sampling point pair from the peak value of the curve to the last point of the set of non-zero amplitude points, starting from the peak value of the curve, and taking the arithmetic mean of all the single-segment decay rates. According to the formula Calculate the damage superposition factor Grade, where β is the inherent thermal relaxation recovery factor of the insulation layer and power devices of the DC power supply equipment.

[0014] This invention provides an AC / DC power supply monitoring system, which has the following advantages compared with the prior art: (1) This invention collects the amplitude change curve and time-domain duration window of low-amplitude transient overvoltage, and calculates the overvoltage occurrence density by combining the time-domain overlap duration and damage equivalence coefficient, accurately quantifying the damage contribution of a single overvoltage, and providing reliable data support for subsequent time-series correlation analysis and damage superposition calculation; (2) This invention uses adaptive timing adjacency correction to standardize the timing relationship of adjacent overvoltages in the same monitoring time domain segment, avoid damage misjudgment caused by timing distortion, extract the effective damage action segment to obtain the corresponding superimposed action curve, and correct the amplitude change point of the superimposed action curve to ensure that the curve truly reflects the damage superposition law. (3) The present invention calculates the damage superposition coefficient based on the superposition effect curve of the whole monitoring cycle, and corrects it in combination with the thermal relaxation recovery characteristics of the equipment, so as to truly reflect the overall cumulative damage level, and then classifies the risk level through a graded early warning mechanism. Attached Figure Description

[0015] Figure 1 This is the system principle block diagram of the present invention; Figure 2 This is a flowchart illustrating the steps involved in calculating the overvoltage occurrence density according to the present invention. Detailed Implementation

[0016] 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.

[0017] like Figure 1 This invention provides an AC / DC power supply monitoring system; As an embodiment of this application, the specific steps include the following: The feature acquisition module acquires the amplitude change curve and time domain duration window of each low amplitude transient overvoltage in the AC / DC power supply system that does not exceed the safety threshold. It divides the time axis into equal-length continuous monitoring time domain segments based on the preset continuous monitoring duration and calculates the overvoltage occurrence density of each monitoring time domain segment. The time domain duration window is the time period from the occurrence start time of the low amplitude transient overvoltage to the decay end time. The timing coupling module performs timing adjacency correction on adjacent low-amplitude transient overvoltages within the same monitoring time domain segment based on the time domain duration window of the low-amplitude transient overvoltage and the corresponding monitoring time domain segment. It also extracts the effective damage segment from the amplitude change curve of the corrected low-amplitude transient overvoltage and generates the superimposed effect curve corresponding to each monitoring time domain segment. The damage monitoring module calculates the damage superposition coefficient based on the superposition curve corresponding to each monitoring time segment and classifies the early warning risk level.

[0018] As a second embodiment of this application, it is implemented based on the first embodiment, except that this embodiment includes: The feature acquisition module acquires the amplitude change curve and time-domain duration window of each low-amplitude transient overvoltage in the AC / DC power supply system that does not exceed the safety threshold. Existing technologies only focus on transient overvoltages that exceed the safety threshold, while completely ignoring low-amplitude transient overvoltages that do not exceed the safety threshold: such overvoltages do not cause obvious visible damage to equipment insulation or power devices when they act once, but multiple time-series superpositions (such as dense occurrences in a short period of time) will produce a cumulative effect, and long-term effects will lead to accelerated equipment aging and sudden failures. The time-domain duration window is the time period from the start of the low-amplitude transient overvoltage to the end of its decay. The starting point of the occurrence refers to the moment when the amplitude of the low-amplitude transient overvoltage first exceeds the background voltage fluctuation range of the AC / DC power supply system during normal operation, and begins to cause damage to the equipment insulation layer and power devices. The attenuation termination time refers to the moment when the amplitude of the low-amplitude transient overvoltage first falls back to the background voltage fluctuation range of the AC / DC power supply system after natural attenuation, and no longer causes any damage to the equipment. Based on the preset continuous monitoring duration, the time axis is divided into equal-length continuous monitoring time domain segments, and the overvoltage occurrence density of each continuous monitoring time domain segment is calculated. Dividing the monitoring time domain into equal-length continuous segments allows discrete overvoltage events to be categorized according to the time dimension, enabling more accurate capture of the temporal correlation and damage superposition patterns at different time periods. Specifically, the following operations can be performed: First, a fixed continuous monitoring duration is preset (i.e., the standard duration of a single time domain segment). Taking the start-up time of the AC / DC power supply monitoring system as the origin of the time axis, several continuous, non-overlapping, and gapless time intervals are sequentially divided on the time axis according to the preset continuous monitoring duration. Each time interval is a continuous monitoring time domain segment of equal length. At the same time, each time domain segment corresponds to a unique time range, and the duration of all time domain segments is completely consistent.

[0019] The timing coupling module performs timing adjacency correction on adjacent low-amplitude transient overvoltages within the same monitoring time segment, based on the time-domain duration window of low-amplitude transient overvoltages and the corresponding monitoring time segment. The specific operation of performing the timing adjacency correction is as follows: Extract overvoltage occurrence density during the monitoring time domain Simultaneously, the time-domain duration window T of all low-amplitude transient overvoltages within the monitoring time domain is correlated, and a time difference identification threshold is constructed based on the two. Where mean() is the mean function. The range of values ​​varies Adaptive adjustment to changes; This directly reflects the temporal density of overvoltages within the monitoring time domain, during periods of high overvoltage (i.e., Within a large range, even minute-level sequential intervals between adjacent overvoltages can lead to damage aggregation because the previous overvoltage has not yet fully decayed before the next one occurs, requiring a smaller interval to be set. This is used to identify such minute time differences and avoid misclassification as interval adjacency, while in sparse periods (i.e., Within a small time frame, a slightly larger time interval between adjacent overvoltages is within the normal time sequence distribution and does not require strict identification. In this case, a larger interval can be set. It can balance computational efficiency and accuracy; For all low-amplitude transient overvoltages within this monitoring time period, they are sorted according to the chronological order of their onset times. The temporal correlation between adjacent overvoltages is compared one by one, specifically the time difference Δt between the decay termination time of the preceding overvoltage and the onset time of the following overvoltage. This is combined with the previously determined... Accurately identify adjacency types and correction strategies: like If the temporal overlap exceeds the critical value, it is determined to be a heavily overlapping adjacency, and depth temporal correction is performed to avoid misjudgment of damage superposition. The execution depth timing correction specifically includes: extracting two overvoltage amplitude change curves and locating the peak moment of the previous overvoltage amplitude change curve. The starting time of the next overvoltage event Revised to Ensure that the corrected timing difference between the two overvoltage cycles is To avoid timing distortion caused by over-correction; like If the temporal overlap does not exceed the critical value, it is determined to be a slight overlap and adjacency, and a simplified temporal correction is performed to match the actual damage superposition. The simplified timing correction specifically includes: only adjusting the decay termination time of the previous overvoltage. Revised to This reduces the temporal overlap range, making it more consistent with the actual damage accumulation pattern. like This indicates that if the timing interval is less than the critical value, it is determined to be a close-interval adjacency, and no timing correction is required, but its adjacency type is marked. like If the timing interval exceeds the critical value, it is determined to be a loosely adjacent interval, and no timing correction is required, but the timing interval Δt needs to be recorded. Then, the effective damage segment is extracted from the amplitude variation curve of the corrected low-amplitude transient overvoltage. The specific steps are as follows: Based on the amplitude variation curve of the corrected overvoltage, which contains several continuous sampling points Where i∈[1,n], the horizontal axis represents time t, and the vertical axis represents the amplitude U. The start time of occurrence, This is the time when the decay ends; For each continuous sampling point in the amplitude variation curve , According to the formula Calculate the rate of change of amplitude within this sampling interval. ,when When the value is positive, it indicates an increase in amplitude; conversely, when the value is negative, it indicates a decrease in amplitude. The larger the absolute value of this value, the faster the amplitude changes. From the start time of the sampling point Traverse all backwards Find the first boundary sampling point where the rate of change of amplitude changes from increasing to constant and stable. , and the time at that dividing point Defined as the inflection point of amplitude evolution At the same time, Apply position constraints, i.e. This indicates that it is within the first half of the time domain duration window T of a single overvoltage event; The rule for determining the increasing rate of amplitude change is as follows: Three or more consecutive adjacent amplitude change rates satisfy monotonically increasing and all are positive values; The rule for determining whether the rate of change of amplitude is uniform and stable is as follows: Three or more consecutive adjacent amplitude change rates satisfy the condition that the absolute value of the difference between two adjacent amplitude change rates is ≤ ε, and all amplitude change rates are negative, where ε is a preset minimum error amount; For example, the time-domain duration window T of a single overvoltage (equal to...) The sampling points of the amplitude variation curve are: to Calculate point by point to ,in, , , All are positive and ,later , , All are negative and , Then the boundary sampling point pair is , ,but ,and If the positional constraints are met, then it is located as the inflection point of amplitude evolution; by Using this as the unique dividing point, the time interval is... The corresponding amplitude change curve is marked as the amplitude rising segment, time interval The corresponding amplitude change curve is marked as the amplitude decay segment, and the amplitude decay segment is regarded as the effective damage segment of this low amplitude transient overvoltage. Simultaneously record key data of the effective target segment of the injury, including the start time. Termination time Time-domain duration window T, sampling point amplitude and the rate of change of amplitude at adjacent sampling points ; Generate the superposition curve corresponding to each monitoring time segment. The specific steps are as follows: For a single monitoring time segment, core data of the effective damage action period corresponding to all low-amplitude transient overvoltages within that time segment are extracted, and a unified reference time axis for that monitoring time segment is determined, i.e., the start time of that time segment is taken as... The termination time is The time axis range is obtained as follows Finally, the time intervals of all effective damage action segments within the time domain are mapped one by one to the unified reference time axis according to their actual occurrence order, completing the time alignment of all effective damage action segments, while preserving the amplitude and amplitude change rate of each effective damage action segment. The time series of each effective damage segment extracted above are their own relative time series, and they are not assigned to a unified time dimension of the monitoring time domain segment. If they are directly superimposed, the superimposition result will be distorted due to the inconsistent time series benchmark, and it will not be able to reflect the actual time series distribution of damage superimposition within the time domain segment. For all effective damage segments after timing alignment, perform in-phase amplitude direct superposition operation, that is, take the sampling time of the unified reference time axis as the anchor point, sum the amplitudes of all effective damage segments at the same sampling time, obtain the total superposition amplitude at that sampling time, and form the superposition curve; For the amplitude abrupt change points that appear in the superposition curve, the amplitude change rate before and after the abrupt change point of each effective damage segment is extracted, and the amplitude abrupt change points are corrected accordingly. The amplitude abrupt change point refers to the temporal transition point between two adjacent effective damage segments, i.e., the termination time of the previous effective damage segment. The start time of the next effective damage segment ,and Because the amplitude at the end of the effective damage segment of the previous injury Amplitude at the start of the next effective damage segment There is a jump, which causes the amplitude of the superposition effect curve to rise or fall sharply at that moment. This jump is an objective result of the timing connection, not a sudden change in the actual damage amplitude of the overvoltage. The specific rules for determining abrupt changes in amplitude are as follows: If the amplitude difference at the junction of two adjacent effective damage segments is If the amplitude change is not observed, the transition point is determined to be a point of abrupt change in amplitude; otherwise, it is determined to be a normal amplitude transition. Here, Δtr is the sampling time interval of the unified reference time axis. The rate of change of amplitude at the moment when the effective damage segment of the previous injury ends; The specific operation for correcting the amplitude abrupt change point is as follows: Extract the amplitude at the termination time of the previous effective damage segment from the recorded effective damage segments. Amplitude change rate Extract the amplitude at the start time of the next effective action segment. Amplitude change rate ; According to the formula Calculate the correction value at the amplitude abrupt change point to ensure that the corrected amplitude is consistent with the rate evolution of the previous effective action segment, and at the same time connects the amplitude of the next effective action segment.

[0020] The damage monitoring module calculates the damage superposition coefficient based on the superposition curve corresponding to each monitoring time segment. The specific operation is as follows: Using the entire monitoring cycle as a unified reference time axis, the superimposed effect curves corresponding to each monitoring time segment are mapped one by one to the unified reference time axis according to the order of their corresponding monitoring time segments, ensuring that all curves are sequentially continuous, non-overlapping, and gapless, forming a set of damage effect curves covering the entire monitoring cycle. Feature extraction is performed on each curve in the set of damage effect curves, including the duration of non-zero amplitude of the curve T_i, the peak amplitude of the curve U_pi, and the uniform decay rate of the curve v_i; The duration of the non-zero amplitude of the curve refers to the sum of the durations of all continuous non-zero amplitude segments after removing the zero amplitude points of the damage curve; The peak amplitude of the curve refers to the set of non-zero amplitude points obtained from the damage curve, and the maximum value among them is taken; The uniform decay rate of the curve refers to the calculation of the slope of each continuous sampling point pair from the peak value of the curve to the last point of the set of non-zero amplitude points, starting from the peak value of the curve, and taking the arithmetic mean of all the single-segment decay rates. According to the formula Calculate the damage superposition coefficient Grade, where β is the inherent thermal relaxation recovery coefficient of the insulation layer and power devices of the DC power supply equipment. This coefficient is a factory-preset inherent parameter of the equipment, and its value range is [0,1]. The larger β is, the stronger the recovery capability of the equipment is, and the larger the corresponding correction range is, to ensure that Grade truly reflects the overall cumulative damage degree of the equipment. The warning risk level is divided according to the damage superposition coefficient Grade, specifically including: like If it is classified as low risk, then it is classified as low risk. If it is classified as medium risk, then it is classified as medium risk. Those that are classified as high-risk, among which, , These are the lower and upper limits of the damage superposition coefficient, respectively, and both need to be adjusted according to the actual environment.

[0021] As a third embodiment of this application, this embodiment further discloses a method for calculating the overvoltage occurrence density during a monitoring time domain period, based on embodiments one and two. Figure 2 As shown, the specific content includes: Calculate the time domain overlap duration between the time domain duration window of each low-amplitude transient overvoltage and the corresponding continuous monitoring time domain interval, that is, the intersection duration between the time domain duration window of the overvoltage and the monitoring time domain interval. From the amplitude variation curve of this low-amplitude transient overvoltage, the amplitude interval corresponding to the time-domain overlap duration is selected, and the mean value α of the amplitude corresponding to this amplitude interval is calculated. Based on the mean value of the amplitude, the damage equivalence coefficient is calculated, and the specific expression is as follows: ; Where k is the damage equivalence coefficient, with a value range of (0,1], which is used to quantify the contribution of overvoltage to the actual damage of the equipment within the time-domain overlap period. The larger k is, the greater the damage contribution, and vice versa. It is directly related to the actual damage characteristics of overvoltage. Us is the safety threshold for low-amplitude transient overvoltage. This value is preset based on the specifications of AC / DC power supply equipment and the withstand characteristics of insulation materials, and serves as a reference for Up. Up represents the peak value of a single low-amplitude transient overvoltage, which can be directly extracted from the amplitude change curve of that low-amplitude transient overvoltage. It reflects the amplitude intensity of that overvoltage. The closer Up is to Us, the higher the potential damage risk of that overvoltage, and the more important it is to strengthen the influence of α on k. Conversely, the lower the potential damage risk, the less important it is to weaken the influence of α on k. In the above formula To determine the correlation coefficient of the peak overvoltage amplitude, a trigonometric function is constructed to control the range of the coefficient within (0,1]. At the same time, the nonlinear growth characteristics of the sin() function are utilized to more accurately reflect the influence of Up on the correlation coefficient, making the subsequent calculation of k more consistent with the actual damage scenario. q is an influencing factor, ranging from (1,2]. This value acts as a non-linear factor of the power function, further amplifying the impact of changes in α on k. When α is small, its small changes can be mitigated by... Achieve a significant change in k; Multiply the aforementioned time-domain overlap duration by the damage equivalence coefficient to obtain the damage equivalent overlap duration, and divide the damage equivalent overlap duration by the duration of the monitoring time-domain interval corresponding to the low-amplitude transient overvoltage to obtain the damage correlation overlap ratio between the low-amplitude transient overvoltage and the corresponding monitoring time-domain interval. The damage contribution of low-amplitude transient overvoltages is not determined solely by the overlap duration: under the same overlap duration, overvoltages with slow amplitude decay (i.e., high average amplitude during the overlap period) contribute far more to the damage to the equipment than overvoltages with fast amplitude decay (i.e., low average amplitude during the overlap period). For each monitoring time segment, the damage correlation overlap ratio of all low-amplitude transient overvoltages in that time segment is accumulated to obtain the equivalent total number of overvoltage damages in that time segment. Then, this value is divided by the duration of the corresponding monitoring time segment to obtain the overvoltage occurrence density of low-amplitude transient overvoltages in that monitoring time segment.

[0022] Some of the data in the above formulas are numerical calculations with dimensions removed, and the contents not described in detail in this specification are all prior art known to those skilled in the art.

[0023] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. An AC / DC power supply monitoring system, characterized in that, include: The feature acquisition module acquires the amplitude change curve and time domain duration window of each low amplitude transient overvoltage in the AC / DC power supply system that does not exceed the safety threshold. It divides the time axis into equal-length continuous monitoring time domain segments based on the preset continuous monitoring duration and calculates the overvoltage occurrence density of each monitoring time domain segment. The time domain duration window is the time period from the occurrence start time of the low amplitude transient overvoltage to the decay end time. The timing coupling module performs timing adjacency correction on adjacent low-amplitude transient overvoltages within the same monitoring time domain segment based on the time domain duration window of the low-amplitude transient overvoltage and the corresponding monitoring time domain segment. It also extracts the effective damage segment from the amplitude change curve of the corrected low-amplitude transient overvoltage and generates the superimposed effect curve corresponding to each monitoring time domain segment. The damage monitoring module calculates the damage superposition coefficient based on the superposition curve corresponding to each monitoring time segment and classifies the early warning risk level.

2. The AC / DC power supply monitoring system according to claim 1, characterized in that, The specific steps for calculating the overvoltage occurrence density for each continuous monitoring time segment are as follows: Calculate the time domain overlap duration between the time domain duration window of each low-amplitude transient overvoltage and the corresponding monitoring time domain interval, that is, the intersection duration between the time domain duration window of the overvoltage and the monitoring time domain interval; From the amplitude variation curve of this low-amplitude transient overvoltage, the amplitude interval corresponding to the time-domain overlap duration is selected, the mean value α of the amplitude corresponding to the amplitude interval is calculated, and the damage equivalence coefficient is calculated based on the mean value of the amplitude. Multiply the time-domain overlap duration by the damage equivalence coefficient to obtain the damage equivalent overlap duration, and divide the damage equivalent overlap duration by the duration of the monitoring time-domain interval corresponding to the low-amplitude transient overvoltage to obtain the damage correlation overlap ratio between the low-amplitude transient overvoltage and the corresponding monitoring time-domain interval. For each monitoring time segment, the damage correlation overlap ratio of all low-amplitude transient overvoltages in that time segment is accumulated to obtain the equivalent total number of overvoltage damages in that time segment. Then, this value is divided by the duration of the corresponding monitoring time segment to obtain the overvoltage occurrence density of low-amplitude transient overvoltages in that monitoring time segment.

3. The AC / DC power supply monitoring system according to claim 2, characterized in that, According to the formula Calculate the damage equivalence coefficient, where Us is the safety threshold for low-amplitude transient overvoltage, Up is the peak amplitude of a single low-amplitude transient overvoltage, and q is the influence factor.

4. The AC / DC power supply monitoring system according to claim 1, characterized in that, The specific operation for performing timing adjacency correction is as follows: Extract overvoltage occurrence density during the monitoring time domain Simultaneously, the time-domain duration window T of all low-amplitude transient overvoltages within the monitoring time domain is correlated, and a time difference identification threshold is constructed based on the two. Where mean() is the mean function; For all low-amplitude transient overvoltages within this monitoring time period, they are sorted according to the chronological order of their occurrence start times. The temporal correlation between adjacent low-amplitude transient overvoltages is compared one by one, specifically the time difference Δt between the decay termination time of the preceding low-amplitude transient overvoltage and the occurrence start time of the following low-amplitude transient overvoltage. This is combined with the aforementioned determined... Identify adjacency types and correction strategies: like If the two are found to be heavily overlapping adjacencies, then depth-time correction is performed. like If the condition is met, it is determined to be a slightly overlapping adjacency, and a simplified timing correction is performed. like If it is, it is determined to be a close-spaced adjacency, and no timing correction is required, but its adjacency type is marked; like If the timing is loosely spaced, it is determined to be a loosely spaced adjacency, and no timing correction is required, but the timing interval Δt needs to be recorded.

5. The AC / DC power supply monitoring system according to claim 4, characterized in that, The execution depth timing correction specifically includes: extracting the amplitude variation curves of two low-amplitude transient overvoltages, and locating the peak moment of the amplitude variation curve of the previous low-amplitude transient overvoltage. The starting time of the subsequent low-amplitude transient overvoltage. Revised to ; The simplified timing correction specifically includes: only adjusting the decay termination time of the previous low-amplitude transient overvoltage. Revised to .

6. The AC / DC power supply monitoring system according to claim 1, characterized in that, The specific steps for extracting the effective damage segment are as follows: Based on the amplitude variation curve of the corrected low-amplitude transient overvoltage, which contains several continuous sampling points Where i∈[1,n], the horizontal axis represents time t, and the vertical axis represents the amplitude U. The start time of occurrence, This is the time when the decay ends; For each continuous sampling point in the amplitude variation curve , According to the formula Calculate the rate of change of amplitude within this sampling interval. ; From the start time of the sampling point Traverse all backwards Find the first boundary sampling point where the rate of change of amplitude changes from increasing to constant and stable. , and the time at that dividing point Defined as the inflection point of amplitude evolution However, it is necessary to ensure ; by Using this as the unique dividing point, the time interval is... The corresponding amplitude change curve is marked as the amplitude rising segment, time interval The corresponding amplitude change curve is marked as the amplitude decay segment, and the amplitude decay segment is regarded as the effective damage segment of this low amplitude transient overvoltage. Simultaneously record key data of the effective target segment of the injury, including the start time. Termination time Time-domain duration window T, sampling point amplitude and the rate of change of amplitude at adjacent sampling points .

7. The AC / DC power supply monitoring system according to claim 6, characterized in that, The judgment rule for increasing amplitude change rate is: three or more consecutive adjacent amplitude change rates satisfy monotonically increasing and are all positive values; the judgment rule for uniform and stable amplitude change rate is: three or more consecutive adjacent amplitude change rates satisfy the absolute value of the difference between two adjacent amplitude change rates ≤ ε, and the amplitude change rates are all negative values, where ε is a preset minimum error amount.

8. The AC / DC power supply monitoring system according to claim 1, characterized in that, The specific steps for generating the superposition curve corresponding to each monitoring time segment are as follows: For a single monitoring time segment, core data of the effective damage action period corresponding to all low-amplitude transient overvoltages within that time segment are extracted, and a unified reference time axis for that monitoring time segment is determined, i.e., the start time of that time segment is taken as... The termination time is The time axis range is obtained as follows The time intervals of all effective damage action segments within the time domain are mapped one by one onto the unified reference time axis according to their actual occurrence order, while retaining the amplitude and amplitude change rate of each effective damage action segment. For all effective damage segments after timing alignment, perform in-phase amplitude direct superposition operation, that is, take the sampling time of the unified reference time axis as the anchor point, sum the amplitudes of all effective damage segments at the same sampling time, obtain the total superposition amplitude at that sampling time, and form the superposition curve; For the amplitude abrupt change points that appear in the superposition curve, the amplitude change rate of each effective damage segment before and after the abrupt change point is extracted, and the amplitude abrupt change points are corrected.

9. The AC / DC power supply monitoring system according to claim 8, characterized in that, The specific rule for determining the amplitude abrupt change point is as follows: if the amplitude difference at the junction of two adjacent effective damage segments is greater than the threshold value... If the amplitude change is not observed, the transition point is determined to be a point of abrupt change in amplitude; otherwise, it is determined to be a normal amplitude transition. Here, Δtr is the sampling time interval of the unified reference time axis. This represents the rate of change of amplitude at the end of the previous effective damage period. This is the amplitude at the end of the effective damage segment of the previous injury. This is the amplitude at the beginning of the next effective damage segment; The specific operation for correcting amplitude abrupt change points is as follows: extract the amplitude change rate at the beginning of the next effective action segment. ; According to the formula Calculate the correction value for the amplitude abrupt change point .

10. The AC / DC power supply monitoring system according to claim 1, characterized in that, The specific steps for calculating the damage superposition coefficient are as follows: Using the entire monitoring cycle as a unified reference time axis, the superposition curves corresponding to each monitoring time segment are mapped one by one to the unified reference time axis according to the order of their corresponding monitoring time segments. Feature extraction is performed on each curve in the set of damage effect curves, including the duration of non-zero amplitude of the curve T_i, the peak amplitude of the curve U_pi, and the uniform decay rate of the curve v_i; The duration of the non-zero amplitude of the curve refers to the sum of the durations of all continuous non-zero amplitude segments after removing the zero amplitude points of the damage curve; The peak amplitude of the curve refers to the set of non-zero amplitude points obtained from the damage curve, and the maximum value among them is taken; The uniform decay rate of the curve refers to the calculation of the slope of each continuous sampling point pair from the peak value of the curve to the last point of the set of non-zero amplitude points, starting from the peak value of the curve, and taking the arithmetic mean of all the single-segment decay rates. According to the formula Calculate the damage superposition factor Grade, where β is the inherent thermal relaxation recovery factor of the insulation layer and power devices of the DC power supply equipment.

Images