Intelligent sensing and identifying method for circuit fault of semiconductor hybrid solid-state circuit breaker

Abstract

The invention relates to the technical field of circuit monitoring, in particular to an intelligent sensing and identifying method for circuit faults of a semiconductor hybrid solid-state circuit breaker. Obtaining a fluctuation abnormal moment according to the amplitude characteristic and the continuous characteristic of the current fluctuation in the current time sequence curve; obtaining a suspected current anomaly moment according to the difference characteristic of the fluctuation anomaly degree of the fluctuation anomaly moment and the same historical moment; clustering is carried out according to the current characteristics and fluctuation characteristics of all suspected current abnormal moments in the current time sequence curve, and different clusters are obtained; obtaining the current anomaly degree of the clustering cluster according to the quantity characteristics of the suspected current anomaly moments in the clustering cluster and the distance characteristics in the clustering space; obtaining a current abnormity moment according to the current abnormity degree; a fault characteristic value is obtained according to the distribution characteristic of the current abnormity moment and the change characteristic of the current abnormity degree in the current time sequence curve, and circuit fault identification is carried out, so that the working stability of the circuit breaker is improved.

Description

Technical Field

[0001] The present invention relates to the field of circuit monitoring technology, and in particular to a method for intelligently sensing and identifying circuit faults in semiconductor hybrid solid-state circuit breakers. Background Art

[0002] A semiconductor hybrid solid-state circuit breaker is a power protection device that combines semiconductor devices with mechanical switches. It combines fast tripping capability with low conduction losses and is primarily used in medium- and high-voltage DC circuit systems. Current sensors typically monitor line current. When the current exceeds a safe value, it is identified as a line fault. The semiconductor hybrid solid-state circuit breaker instantly cuts off the current, quickly disconnecting and protecting the circuit. However, this method cannot provide early warning of potential faults; it simply disconnects the circuit when excessive current is detected. If the fault current is too high during power outage, the circuit breaker can overheat due to excessive load, triggering further circuit failures and increasing the risk of failure. Summary of the Invention

[0003] In order to solve the technical problem that when circuit monitoring is performed only based on current magnitude, a large fault current during power outage may easily lead to overheating of the circuit breaker due to excessive load, causing further circuit failure and increasing the risk of failure, the present invention aims to provide a method for intelligent perception and identification of semiconductor hybrid solid-state circuit breaker circuit faults. The technical solution adopted is as follows:

[0004] Obtaining a current timing curve of a circuit where the semiconductor hybrid solid-state circuit breaker is located;

[0005] Obtaining the degree of abnormal fluctuation at the fluctuation point based on the amplitude characteristics and duration characteristics of the current fluctuation in the current time series curve; obtaining the abnormal fluctuation moment based on the abnormal fluctuation degree; obtaining the degree of regularity of fluctuation based on the difference between the abnormal fluctuation moment and the degree of abnormal fluctuation at the same historical moment; obtaining the suspected abnormal current moment based on the abnormal fluctuation degree;

[0006] Clustering is performed based on the current characteristics and fluctuation characteristics of all suspected current abnormal moments in the current time series curve to obtain different clusters; obtaining the current abnormality degree of the cluster based on the number characteristics of the suspected current abnormal moments in the cluster and the distance characteristics in the cluster space; obtaining the current abnormal moment based on the current abnormality degree;

[0007] A fault characteristic value is obtained according to the distribution characteristics of the current abnormality moment and the change characteristics of the current abnormality degree in the current time series curve; and circuit fault identification is performed according to the fault characteristic value.

[0008] Furthermore, the step of obtaining the degree of abnormal fluctuation of the fluctuation point according to the amplitude characteristics and duration characteristics of the current fluctuation in the current time series curve includes:

[0009] The extreme point in the current timing curve is taken as the fluctuation point; the absolute value of the slope change rate of the data point in the current timing curve is calculated and normalized to obtain the change characteristic value of the data point; the data point whose change characteristic value exceeds the preset inflection point threshold is taken as the inflection point; the average value of the absolute value of the difference between the fluctuation point and the adjacent inflection points before and after in the current timing curve is calculated to obtain the fluctuation amplitude of the fluctuation point; the ratio of the fluctuation amplitude to the maximum value among all fluctuation amplitudes is calculated to obtain the amplitude characteristic value; the time interval between the adjacent inflection points before and after the fluctuation point in the current timing curve is calculated to obtain the fluctuation duration of the fluctuation point; the ratio of the fluctuation duration to the maximum value among all fluctuation durations is calculated and negatively correlated to obtain the duration characteristic value; the sum of the amplitude characteristic value and the duration characteristic value is calculated to obtain the degree of fluctuation abnormality of the fluctuation point.

[0010] Furthermore, the step of obtaining the abnormal fluctuation moment according to the abnormal fluctuation degree includes:

[0011] When the degree of fluctuation anomaly of the fluctuation point exceeds a preset fluctuation threshold, the moment at which the fluctuation point occurs is the fluctuation anomaly moment.

[0012] Furthermore, the step of obtaining the degree of fluctuation regularity based on the difference characteristics between the degree of fluctuation anomaly at the moment of fluctuation anomaly and the same moment in history includes:

[0013] The average of the absolute values ​​of the differences between the degree of abnormal fluctuations at the abnormal fluctuation moment and all the same historical moments is calculated and negatively correlated with each other to obtain the degree of regularity of fluctuations at the abnormal fluctuation moment.

[0014] Furthermore, the step of obtaining the suspected current abnormality time according to the fluctuation regularity includes:

[0015] When the degree of the fluctuation regularity at the abnormal fluctuation moment does not exceed a preset regularity threshold, the abnormal fluctuation moment is a suspected current abnormality moment.

[0016] Furthermore, the step of clustering the current characteristics and fluctuation characteristics of all suspected current anomaly moments in the current time series curve to obtain different clusters includes:

[0017] Clustering is performed according to the current values ​​corresponding to all suspected current anomaly moments, the fluctuation amplitudes and the fluctuation durations using a K-means clustering algorithm to obtain different clusters.

[0018] Furthermore, the step of obtaining the current anomaly degree of the cluster according to the quantity characteristics of the suspected current anomaly moments in the cluster and the distance characteristics in the cluster space includes:

[0019] In the cluster space, the average value of the Euclidean distances between all suspected current anomaly moments in the cluster and other suspected current anomaly moments of the nearest neighbors is calculated to obtain a distribution discrete value; the product of the inverse of the number of suspected current anomaly moments in the cluster and the distribution discrete value is calculated and normalized to obtain the current anomaly degree of the cluster.

[0020] Furthermore, the step of obtaining the current abnormality moment according to the current abnormality degree includes:

[0021] The suspected current abnormal moment in the cluster where the current abnormality exceeds a preset abnormality threshold is used as the current abnormal moment.

[0022] Furthermore, the step of obtaining a fault characteristic value according to the distribution characteristics of the current abnormality moment and the change characteristics of the current abnormality degree in the current time series curve includes:

[0023] The current timing curve is divided into a first half and a second half from the middle moment; the ratio of the number of current abnormal moments in the second half to that in the first half is calculated to obtain a first fault trend value; the ratio of the average values ​​of the current abnormality corresponding to the current abnormal moments in the second half to that in the first half is calculated to obtain a second fault trend value; and the product of the first fault trend value and the second fault trend value is calculated to obtain the fault characteristic value.

[0024] Furthermore, the step of identifying a circuit fault according to the fault characteristic value includes:

[0025] When the fault characteristic value exceeds a preset fault threshold, a circuit fault warning is issued.

[0026] The present invention has the following beneficial effects:

[0027] In the present invention, since current fluctuation characteristics can reflect whether a circuit current is abnormal, obtaining the degree of abnormal fluctuation can characterize the current fluctuation characteristics in the circuit, and further, based on the degree of abnormal fluctuation, the moment of abnormal current fluctuation can be preliminarily determined. Since abnormal fluctuation moments are not limited to circuit faults, but can also be caused by large load switching changes, obtaining the degree of regularity can determine whether the degree of abnormal fluctuation at the moment of abnormal fluctuation exhibits regular characteristics. This allows the distinction between abnormal fluctuation moments caused by load switching changes and those caused by actual circuit faults. Obtaining suspected abnormal current moments can identify current fluctuation moments caused by load switching changes at irregular times. Obtaining clusters can distinguish the current variation characteristics corresponding to different suspected abnormal current moments. Since current variation characteristics caused by circuit faults tend to have low similarity and low frequency, obtaining the degree of abnormal current can characterize the factors that caused the suspected abnormal current moments within the clusters. Obtaining the abnormal current moment can characterize the actual circuit fault moment. Obtaining the fault characteristic value can indicate whether the circuit fault has an increasing risk trend. Finally, circuit fault identification based on the fault characteristic value improves the accuracy of circuit fault identification and the stability of circuit breaker operation, preventing circuit faults from causing overheating and damage to the circuit breaker. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0029] Figure 1 A flow chart of a method for intelligently sensing and identifying circuit faults in a semiconductor hybrid solid-state circuit breaker provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0030] To further illustrate the technical means and effectiveness of the present invention in achieving its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effectiveness of a method for intelligently sensing and identifying circuit faults in a semiconductor hybrid solid-state circuit breaker according to the present invention. In the following description, references to different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0031] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0032] The specific scheme of the intelligent perception and identification method of semiconductor hybrid solid-state circuit breaker circuit fault provided by the present invention is described in detail below with reference to the accompanying drawings.

[0033] See also Figure 1 , which shows a flow chart of a method for intelligent perception and identification of semiconductor hybrid solid-state circuit breaker circuit faults provided by one embodiment of the present invention, the method comprising the following steps:

[0034] Step S1, obtaining a current timing curve of a circuit where a semiconductor hybrid solid-state circuit breaker is located.

[0035] Obtain a current timing curve of the circuit where the semiconductor hybrid solid-state circuit breaker is located. In an embodiment of the present invention, the current monitoring acquisition frequency is 10 Hz, and 10 seconds of current data are collected each time. Since the current will have slight fluctuations under normal conditions, in order to improve the accuracy of fault identification, the collected current data is smoothed to remove the slight fluctuation characteristics. The smoothed current data is projected into a two-dimensional rectangular coordinate system to construct a current timing curve, with the horizontal axis being time and the vertical axis being current. The current data within 10 seconds before the current moment is used as the latest current timing curve. The implementer can determine the acquisition frequency and timing curve length according to the implementation scenario.

[0036] Step S2, obtaining the degree of fluctuation anomaly of the fluctuation point according to the amplitude characteristics and duration characteristics of the current fluctuation in the current time series curve; obtaining the moment of fluctuation anomaly according to the degree of fluctuation anomaly; obtaining the degree of fluctuation regularity according to the difference characteristics between the degree of fluctuation anomaly at the moment of fluctuation anomaly and the same moment in history; obtaining the moment of suspected current anomaly according to the degree of fluctuation regularity.

[0037] First, the latest current timing curve is analyzed to determine whether there are any abnormalities in the real-time monitoring period. Abnormal current fluctuations indicate the risk of circuit failure. Under normal circumstances, the circuit current will have slight fluctuations, such as slow and slight fluctuations caused by power supply ripple or temperature. Abnormal fluctuations are often larger in amplitude than normal fluctuations, and the fluctuations appear and end faster. Therefore, the degree of fluctuation abnormality at the fluctuation point can be preliminarily obtained based on the amplitude characteristics and duration characteristics of the current fluctuations in the current timing curve.

[0038] Preferably, in an embodiment of the present invention, the step of obtaining the degree of fluctuation anomaly includes: taking the extreme points in the current time series curve as fluctuation points; it should be noted that the extreme points can be obtained using the existing AMPD peak search algorithm, and the extreme points are the peak points and trough points in the current time series curve where the fluctuation characteristics are more obvious. The absolute value of the slope change rate of the data points in the current time series curve is calculated and normalized to obtain the change characteristic value of the data points; the slope change rate is the slope difference between the data point and the previous data point. Under normal circumstances, the current data is relatively stable and the change characteristic value is small; when the current fluctuates, the inflection points at the beginning and end of the fluctuation show obvious slope changes and the change characteristic value is large. In an embodiment of the present invention, the normalization method is linear normalization. The change characteristic value of data points with consistent current change trends tends to 0, while the change characteristic value of data points where the current change trends differ tends to 1. Therefore, the preset inflection point threshold is set to 0.5, and the data points with change characteristic values ​​exceeding the preset inflection point threshold are regarded as inflection points. The inflection point represents the starting position of the current fluctuation. Calculate the average of the absolute values ​​of the differences between the fluctuation point and the adjacent inflection points before and after it in the current time series curve to obtain the fluctuation amplitude of that fluctuation point. A larger fluctuation amplitude indicates a more pronounced fluctuation and is more likely to be an abnormal fluctuation. Calculate the ratio of the fluctuation amplitude to the maximum value of all fluctuation amplitudes to obtain the amplitude characteristic value. The goal is to keep the fluctuation amplitude within the range of 0 to 1, eliminating the influence of dimension. A larger amplitude characteristic value indicates a greater degree of fluctuation and a more abnormal fluctuation. Calculate the time interval between the adjacent inflection points before and after the fluctuation point in the current time series curve to obtain the fluctuation duration of that fluctuation point. A shorter fluctuation duration indicates a faster fluctuation and a lower likelihood of slow fluctuation. Calculate the ratio of the fluctuation duration to the maximum value of all fluctuation durations and negatively correlate them to obtain the duration characteristic value. The goal is to keep the fluctuation duration within the range of 0 to 1, and the shorter the fluctuation duration, the closer the duration characteristic value is to 1, indicating a more likely abnormal fluctuation. Calculate the sum of the amplitude characteristic value and the duration characteristic value to obtain the degree of fluctuation anomaly at the fluctuation point; the greater the degree of fluctuation anomaly, the more likely the fluctuation is caused by circuit anomaly; the smaller the degree of fluctuation anomaly, the more likely the fluctuation is caused by normal circuit conditions. The formula for obtaining the degree of fluctuation anomaly includes:

[0039]

[0040] Where R i Indicates the abnormal degree of fluctuation of the ith fluctuation point, S i Indicates the fluctuation amplitude of the i-th fluctuation point, S max Indicates the maximum value of the fluctuation amplitude, represents the amplitude characteristic value, T i Indicates the fluctuation duration of the i-th fluctuation point, T max Indicates the maximum value of the fluctuation duration, exp() represents the exponential function with a natural constant as the base, Indicates the duration feature value.

[0041] Furthermore, the moment of abnormal fluctuation can be obtained based on the degree of abnormal fluctuation, specifically including: when the degree of abnormal fluctuation of the fluctuation point exceeds the preset fluctuation threshold, the moment of the fluctuation point is the abnormal fluctuation moment; in the embodiment of the present invention, the preset fluctuation threshold is 0.6. When the preset fluctuation threshold is exceeded, it means that the current fluctuation characteristics of this fluctuation are greatly different from those under normal circumstances, and the fluctuation characteristics are abnormal, and then the moment of the fluctuation point is taken as the abnormal fluctuation moment. The implementer can determine the preset fluctuation threshold according to the implementation scenario. After obtaining the abnormal fluctuation moment, it is necessary to further analyze whether it is caused by circuit abnormality; some large loads that change automatically in the circuit will perform load switching at a fixed time point every day. Such regular load change moments may be determined as abnormal fluctuation moments, but they are not caused by real circuit faults; therefore, it is necessary to obtain the degree of regularity of fluctuation based on the difference characteristics of the degree of abnormal fluctuation at the abnormal fluctuation moment and the same moment in history.

[0042] Preferably, in an embodiment of the present invention, the step of obtaining the degree of fluctuation regularity includes calculating the average of the absolute values ​​of the differences between the degree of fluctuation anomaly at the moment of abnormal fluctuation and all similar historical moments, and performing negative correlation mapping to obtain the degree of fluctuation regularity at the abnormal fluctuation moment. In this embodiment of the present invention, the negative correlation mapping is performed using the exp() function, and the range of the degree of fluctuation regularity is 0 to 1. When the degree of fluctuation anomaly at the same historical moment is similar to that at the abnormal fluctuation moment, it means that the current fluctuation at that moment is caused by a load switching change of a large load at a fixed time. Therefore, the greater the degree of fluctuation regularity, the more likely the abnormal fluctuation moment is caused by a normal load change; the smaller the degree of fluctuation regularity, the more likely the abnormal fluctuation moment is caused by a true circuit abnormal fluctuation. Furthermore, a suspected current abnormal moment can be obtained based on the degree of fluctuation regularity. Preferably, in this embodiment of the present invention, the step of obtaining the suspected current abnormal moment includes determining the abnormal fluctuation moment as a suspected current abnormal moment when the degree of fluctuation regularity at the abnormal fluctuation moment does not exceed a preset regularity threshold. Since the degree of fluctuation regularity corresponding to a normal load change tends to 1, while the degree of fluctuation regularity corresponding to a true circuit abnormality tends to 0, the preset regularity threshold in this embodiment of the present invention is 0.5. A suspected current abnormality moment means that the moment is highly likely to be caused by a circuit fault.

[0043] Step S3: clustering all the suspected current anomaly moments in the current time series curve according to the current characteristics and fluctuation characteristics to obtain different clusters; obtaining the current anomaly degree of the cluster according to the number characteristics of the suspected current anomaly moments in the cluster and the distance characteristics in the cluster space; and obtaining the current anomaly moment according to the current anomaly degree.

[0044] Because large loads controlled by manual switches occasionally exist in the circuit, the irregular switching of these loads may also cause suspected current anomalies. Therefore, further analysis is needed based on the differences between the current change characteristics of manually controlled load switches and those of circuit fault anomalies. Manually controlled load switches do not have regular characteristics, but the current change patterns caused by manually controlled load changes are similar. As the number of manual adjustments accumulates, current change characteristics with high similarity will reappear in the historical current timing curve. However, circuit faults such as overload, leakage, component aging, and arcing have different environments and fault severity at each fault, resulting in more random current change characteristics. The likelihood of similar current change patterns appearing in the historical current timing curve is low. Furthermore, different clusters can be obtained based on the current characteristics and fluctuation characteristics of all suspected current anomaly moments in the current time series curve. Preferably, in an embodiment of the present invention, the step of obtaining different clusters includes: clustering the current values, fluctuation amplitudes, and fluctuation durations corresponding to all suspected current anomaly moments using a K-means clustering algorithm to obtain different clusters. It should be noted that the K-means clustering algorithm belongs to the prior art, and the number of clusters is obtained using the elbow method. The specific clustering steps are not repeated here. Through clustering, suspected current anomaly moments with different fluctuation characteristics can be divided.

[0045] Furthermore, a greater number of suspected current anomaly moments within a cluster indicates a greater number of suspected current anomaly moments with similar current variation characteristics, making the suspected current anomaly moments in that cluster more likely to be caused by manual load changes. A smaller number of suspected current anomaly moments within a cluster indicates a smaller number of suspected current anomaly moments with similar current variation characteristics, making the suspected current anomaly moments in that cluster more likely to be caused by circuit faults. During the clustering process, the more similar the current variation characteristics between two suspected current anomaly moments, the shorter their Euclidean distance in the cluster space. Therefore, a shorter Euclidean distance between suspected current anomaly moments within a cluster indicates more similar current variation characteristics, making the suspected current anomaly moments in the cluster more likely to be caused by manual load changes. Conversely, a larger Euclidean distance indicates less similar current variation characteristics, making the suspected current anomaly moments in the cluster more likely to be caused by circuit faults. Therefore, the current anomaly degree of a cluster is determined based on the number of suspected current anomaly moments within the cluster and the distance characteristic in the cluster space.

[0046] Preferably, in an embodiment of the present invention, the step of obtaining the degree of current anomaly includes: calculating the average of the Euclidean distances between all suspected current anomaly moments in the cluster and other suspected current anomaly moments of the nearest neighbor in the cluster space to obtain a distribution dispersion value; a larger distribution dispersion value indicates a lower similarity in the current change characteristics at the suspected current anomaly moments in the cluster, and a higher likelihood of indicating a circuit fault. Calculating the product of the inverse of the number of suspected current anomaly moments in the cluster and the distribution dispersion value, and normalizing the result, to obtain the degree of current anomaly for the cluster; a larger current anomaly value indicates a higher likelihood that the suspected current anomaly moments in the cluster are caused by a circuit fault. Furthermore, the current anomaly moment can be obtained based on the current anomaly degree, specifically including: taking the suspected current anomaly moment in the cluster whose current anomaly degree exceeds the preset anomaly threshold as the current anomaly moment; since the value range of the current anomaly degree after linear normalization is 0 to 1, the current anomaly degree of the cluster corresponding to the manual control load change tends to 0, while the current anomaly degree of the cluster corresponding to the real current fault tends to 1, and the two are located at the two ends of the value range; therefore, in the embodiment of the present invention, the preset anomaly threshold is 0.5, and the implementer can determine it according to the implementation scenario.

[0047] Step S4, obtaining a fault characteristic value based on the distribution characteristics of the current abnormality moment and the change characteristics of the current abnormality degree in the current time series curve; and performing circuit fault identification based on the fault characteristic value.

[0048] Abnormal current moments represent moments of significant current fluctuation caused by non-load switching changes. As the number of abnormal current moments increases, the likelihood of a current fault increases. Therefore, a fault characteristic value is obtained based on the distribution characteristics of the abnormal current moments and the variation characteristics of the current abnormality degree in the current time series curve. Preferably, in an embodiment of the present invention, the step of obtaining the fault characteristic value includes: dividing the current time series curve from the middle moment into a first half and a second half; calculating the ratio of the number of abnormal current moments in the second half to the number in the first half to obtain a first fault trend value; when the second half has more abnormal current moments than the first half, the first fault trend value is greater than 1, indicating that the abnormal current moments are increasing and the circuit fault trend is significant. A second fault trend value is obtained by calculating the ratio of the average current abnormality degree corresponding to the abnormal current moments in the second half to the first half; when the average current abnormality degree is greater in the second half than in the first half, the second fault trend value is greater than 1, indicating that the risk of circuit fault is increasing. The fault characteristic value is obtained by multiplying the first fault trend value by the second fault trend value. The larger the fault characteristic value, the more obvious the fault trend of the circuit becomes, and the greater the risk of failure. The formula for obtaining the fault characteristic value includes:

[0049]

[0050] Where W represents the fault characteristic value, D1 represents the number of abnormal current moments in the first half, and D2 represents the number of abnormal current moments in the second half. represents the first fault trend value, H1 represents the average value of the current abnormality corresponding to the current abnormality moment in the first half, and H2 represents the average value of the current abnormality corresponding to the current abnormality moment in the second half. Indicates the second fault tendency value.

[0051] Furthermore, after obtaining the fault characteristic value, circuit fault identification can be performed based on the fault characteristic value, specifically including: when the fault characteristic value exceeds a preset fault threshold, a circuit fault warning is performed; in an embodiment of the present invention, the preset fault threshold is 1. When the fault characteristic value exceeds 1, it means that the current abnormality characteristics in the second half are more obvious than those in the first half, the trend of circuit failure increases, and the risk increases; in order to avoid the semiconductor hybrid solid circuit breaker directly breaking the circuit when the fault current is too large, causing the circuit breaker to overload, overheat, and be damaged, timely warning is required to improve the stability of the circuit breaker operation.

[0052] In summary, the embodiments of the present invention provide a method for intelligent perception and identification of circuit faults in semiconductor hybrid solid-state circuit breakers; obtain the fluctuation abnormality moment according to the amplitude characteristics and duration characteristics of the current fluctuation in the current timing curve; obtain the suspected current abnormality moment according to the difference characteristics of the fluctuation abnormality degree between the fluctuation abnormality moment and the same historical moment; cluster the current characteristics and fluctuation characteristics of all suspected current abnormality moments in the current timing curve to obtain different cluster clusters; obtain the current abnormality degree of the cluster cluster according to the number characteristics of the suspected current abnormality moments in the cluster cluster and the distance characteristics in the cluster space; obtain the current abnormality moment according to the current abnormality degree; obtain the fault characteristic value and perform circuit fault identification according to the distribution characteristics of the current abnormality moment in the current timing curve and the change characteristics of the current abnormality degree, thereby improving the stability of the circuit breaker operation.

[0053] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0054] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for intelligent perception and identification of circuit faults in semiconductor hybrid solid-state circuit breakers, characterized in that: The method comprises the following steps: Obtaining a current timing curve of a circuit where the semiconductor hybrid solid-state circuit breaker is located; Obtaining the degree of abnormal fluctuation at the fluctuation point based on the amplitude characteristics and duration characteristics of the current fluctuation in the current time series curve; obtaining the abnormal fluctuation moment based on the abnormal fluctuation degree; obtaining the degree of regularity of fluctuation based on the difference between the abnormal fluctuation moment and the degree of abnormal fluctuation at the same historical moment; obtaining the suspected abnormal current moment based on the abnormal fluctuation degree; Clustering is performed based on the current characteristics and fluctuation characteristics of all suspected current abnormal moments in the current time series curve to obtain different clusters; obtaining the current abnormality degree of the cluster based on the number characteristics of the suspected current abnormal moments in the cluster and the distance characteristics in the cluster space; obtaining the current abnormal moment based on the current abnormality degree; A fault characteristic value is obtained according to the distribution characteristics of the current abnormality moment and the change characteristics of the current abnormality degree in the current time series curve; and circuit fault identification is performed according to the fault characteristic value.

2. The intelligent perception and identification method for semiconductor hybrid solid-state circuit breaker circuit fault according to claim 1 is characterized in that: The step of obtaining the degree of abnormal fluctuation of the fluctuation point according to the amplitude characteristics and duration characteristics of the current fluctuation in the current time series curve comprises: The extreme point in the current timing curve is taken as the fluctuation point; the absolute value of the slope change rate of the data point in the current timing curve is calculated and normalized to obtain the change characteristic value of the data point; the data point whose change characteristic value exceeds the preset inflection point threshold is taken as the inflection point; the average value of the absolute value of the difference between the fluctuation point and the adjacent inflection points before and after in the current timing curve is calculated to obtain the fluctuation amplitude of the fluctuation point; the ratio of the fluctuation amplitude to the maximum value among all fluctuation amplitudes is calculated to obtain the amplitude characteristic value; the time interval between the adjacent inflection points before and after the fluctuation point in the current timing curve is calculated to obtain the fluctuation duration of the fluctuation point; the ratio of the fluctuation duration to the maximum value among all fluctuation durations is calculated and negatively correlated to obtain the duration characteristic value; the sum of the amplitude characteristic value and the duration characteristic value is calculated to obtain the degree of fluctuation abnormality of the fluctuation point.

3. The intelligent perception and identification method for semiconductor hybrid solid-state circuit breaker circuit fault according to claim 1 is characterized in that: The step of obtaining the abnormal fluctuation moment according to the abnormal fluctuation degree comprises: When the degree of fluctuation anomaly of the fluctuation point exceeds a preset fluctuation threshold, the moment at which the fluctuation point occurs is the fluctuation anomaly moment.

4. The intelligent perception and identification method for semiconductor hybrid solid-state circuit breaker circuit fault according to claim 1 is characterized in that: The step of obtaining the degree of fluctuation regularity according to the difference characteristics of the degree of fluctuation anomaly at the moment of fluctuation anomaly and the same moment in history comprises: The average of the absolute values ​​of the differences between the degree of abnormal fluctuations at the abnormal fluctuation moment and all the same historical moments is calculated and negatively correlated with each other to obtain the degree of regularity of fluctuations at the abnormal fluctuation moment.

5. The method for intelligent perception and identification of circuit faults in semiconductor hybrid solid-state circuit breakers according to claim 1, characterized in that: The step of obtaining the suspected current abnormality time according to the degree of the fluctuation regularity includes: When the degree of the fluctuation regularity at the abnormal fluctuation moment does not exceed a preset regularity threshold, the abnormal fluctuation moment is a suspected current abnormality moment.

6. The method for intelligent perception and identification of circuit faults in semiconductor hybrid solid-state circuit breakers according to claim 2, characterized in that: The step of clustering the current characteristics and fluctuation characteristics of all suspected current abnormality moments in the current time series curve to obtain different clusters includes: Clustering is performed according to the current values ​​corresponding to all suspected current anomaly moments, the fluctuation amplitudes and the fluctuation durations using a K-means clustering algorithm to obtain different clusters.

7. The method for intelligent perception and identification of circuit faults in semiconductor hybrid solid-state circuit breakers according to claim 1, characterized in that: The step of obtaining the current anomaly degree of the cluster according to the number characteristics of the suspected current anomaly moments in the cluster and the distance characteristics in the cluster space comprises: In the cluster space, the average value of the Euclidean distances between all suspected current anomaly moments in the cluster and other suspected current anomaly moments of the nearest neighbors is calculated to obtain a distribution discrete value; the product of the inverse of the number of suspected current anomaly moments in the cluster and the distribution discrete value is calculated and normalized to obtain the current anomaly degree of the cluster.

8. The method for intelligent perception and identification of circuit faults in semiconductor hybrid solid-state circuit breakers according to claim 1, characterized in that: The step of obtaining the current abnormality time according to the current abnormality degree comprises: The suspected current abnormal moment in the cluster where the current abnormality exceeds a preset abnormality threshold is used as the current abnormal moment.

9. The method for intelligently sensing and identifying circuit faults in semiconductor hybrid solid-state circuit breakers according to claim 1, characterized in that: The step of obtaining the fault characteristic value according to the distribution characteristics of the current abnormality moment and the change characteristics of the current abnormality degree in the current time series curve includes: The current timing curve is divided into a first half and a second half from the middle moment; the ratio of the number of current abnormal moments in the second half to that in the first half is calculated to obtain a first fault trend value; the ratio of the average values ​​of the current abnormality corresponding to the current abnormal moments in the second half to that in the first half is calculated to obtain a second fault trend value; and the product of the first fault trend value and the second fault trend value is calculated to obtain the fault characteristic value.

10. The method for intelligent perception and identification of circuit faults in semiconductor hybrid solid-state circuit breakers according to claim 1, characterized in that: The step of identifying a circuit fault according to the fault characteristic value comprises: When the fault characteristic value exceeds a preset fault threshold, a circuit fault warning is issued.

Images