Abnormality identification method and device for current transformer

Abstract

The invention provides an abnormity identification method and device for a current transformer, and the method comprises the steps: obtaining a secondary current instantaneous value sequence and a secondary voltage instantaneous value sequence of a secondary side of the current transformer through starting the abnormity identification of the current transformer; extracting a voltage fundamental component from the secondary voltage instantaneous value sequence, and dividing the secondary current instantaneous value sequence into a first current segment and a second current segment based on a zero crossing point of the voltage fundamental component; respectively calculating a first distortion characteristic value and a second distortion characteristic value of current waveforms in the first current segment and the second current segment relative to an ideal sine wave; and constructing a differential evolution sequence of the first distortion characteristic value and the second distortion characteristic value on a time axis, and further judging whether the iron core of the current transformer has asymmetric excitation abnormity or not based on the differential evolution sequence. According to the technical problem provided by the invention, the asymmetric excitation abnormity of the iron core of the current transformer can be accurately identified.

Description

Method and device for anomaly identification of current transformers Technical Field

[0001] This application relates to the field of electrical performance anomaly identification technology, and more specifically, to a method and apparatus for anomaly identification of current transformers. Background Technology

[0002] As electrical equipment becomes increasingly complex and intelligent, accurate monitoring of its operating status has become a core element in ensuring the safe and stable operation of the system. Its performance anomaly identification technology aims to analyze the equipment operating data by collecting multi-dimensional characteristic parameters such as voltage, current, temperature and vibration in real time, and using signal processing and machine learning algorithms to extract abnormal patterns from massive amounts of data, distinguish between equipment aging, transient overload and hidden faults, thereby achieving early warning and accurate location of faults.

[0003] In existing electrical performance anomaly identification methods, the process begins by using sensors deployed at the equipment to collect real-time operating parameters such as voltage, current, and temperature. Then, signal processing techniques like Fourier transform and wavelet analysis are employed to convert the raw data to the frequency or time-frequency domain to extract key features reflecting the equipment's status. Next, the system compares these real-time features with a benchmark model representing the health status to determine if the current operating state deviates from the normal pattern. However, in the anomaly identification of current transformers, while the magnetization curve of the current transformer core should theoretically be symmetrical, when the core experiences manufacturing defects, long-term aging, DC intrusion, or local saturation, the excitation of its positive and negative half-cycles varies. The magnetic properties will exhibit asymmetry, which will be directly reflected in the difference in the positive and negative half-wave shapes of the secondary current waveform. However, the traditional distortion index based on full-wave Fourier analysis calculates the distortion energy of the positive and negative half-waves together, and cannot separate and compare the degree of nonlinear deformation of each half-wave. This leads to false alarms when there are symmetrical harmonic sources unrelated to the core anomaly. Conversely, when the distortion is small in the early stage of the core asymmetry anomaly, the distortion energy may be masked by the fundamental component and missed, thus causing the current transformer core asymmetry excitation anomaly to be misidentified. Therefore, how to accurately identify the core asymmetry excitation anomaly of the current transformer has become a difficult problem for the industry. Summary of the Invention

[0004] This application provides a method and apparatus for identifying anomalies in current transformers, which can accurately identify asymmetric excitation anomalies in the core of current transformers.

[0005] In a first aspect, this application provides a method for anomaly identification of a current transformer, comprising the following steps: initiating anomaly identification of the current transformer, acquiring a dataset of instantaneous electrical quantities on the secondary side of the current transformer, the dataset containing a sequence of instantaneous secondary current values ​​and a sequence of instantaneous secondary voltage values; extracting the fundamental voltage component from the sequence of instantaneous secondary voltage values, and dividing the sequence of instantaneous secondary current values ​​into a first current segment corresponding to the positive half-wave and a second current segment corresponding to the negative half-wave based on the zero-crossing point of the fundamental voltage component; calculating a first distortion feature value of the current waveform in the first current segment relative to an ideal sine wave, and a second distortion feature value of the current waveform in the second current segment relative to an ideal sine wave, the distortion feature value being used to characterize the degree of nonlinear deformation of the current waveform within the corresponding half-wave; constructing a differential evolution sequence of the first distortion feature value and the second distortion feature value on the time axis, and then determining whether there is an asymmetric excitation anomaly in the core of the current transformer based on the differential evolution sequence.

[0006] In some embodiments, extracting the fundamental voltage component from the secondary voltage instantaneous value sequence specifically includes: performing digital filtering on the secondary voltage instantaneous value sequence to obtain a preprocessed voltage sequence; performing zero-crossing detection on the preprocessed voltage sequence to determine the time interval between adjacent zero-crossings, and calculating the actual length of the current power frequency cycle based on the time interval; performing integer-cycle truncation on the preprocessed voltage sequence based on the actual length to obtain a voltage integer-cycle data window for calculation; performing Fourier transform on the discrete points within the voltage integer-cycle data window to calculate the real and imaginary coefficients of the fundamental voltage component, and reconstructing the fundamental voltage component based on the real and imaginary coefficients.

[0007] In some embodiments, dividing the instantaneous value sequence of the secondary current into a first current segment corresponding to the positive half-wave and a second current segment corresponding to the negative half-wave based on the zero-crossing point of the fundamental voltage component specifically includes: analyzing the fundamental voltage component to determine all zero-crossing moments of the fundamental voltage component on the time axis; using the zero-crossing moment as the dividing point, marking the time intervals corresponding to each sampling point in the instantaneous value sequence of the secondary current, marking the sampling points located between two adjacent zero-crossing moments where the voltage corresponding to the initial zero-crossing moment changes from negative to positive as positive half-wave belonging points, and marking the sampling points located between two adjacent zero-crossing moments where the voltage corresponding to the initial zero-crossing moment changes from positive to negative as negative half-wave belonging points; extracting the corresponding sampling points from the instantaneous value sequence of the secondary current according to the marking of the positive half-wave belonging points, and combining them in chronological order to form the first current segment; extracting the corresponding sampling points from the instantaneous value sequence of the secondary current according to the marking of the negative half-wave belonging points, and combining them in chronological order to form the second current segment.

[0008] In some embodiments, calculating the first distortion characteristic value of the current waveform in the first current segment relative to an ideal sine wave and the second distortion characteristic value of the current waveform in the second current segment relative to an ideal sine wave specifically includes: performing a discrete Fourier transform on the current waveform sequence in the first current segment to extract the amplitude coefficients of each harmonic component in the first current segment; calculating the total harmonic distortion rate of the first current segment as the first distortion characteristic value based on the amplitude coefficients of each harmonic component and the fundamental amplitude of the first current segment; performing a discrete Fourier transform on the current waveform sequence in the second current segment to extract the amplitude coefficients of each harmonic component in the second current segment; and calculating the total harmonic distortion rate of the second current segment as the second distortion characteristic value based on the amplitude coefficients of each harmonic component and the fundamental amplitude of the second current segment.

[0009] In some embodiments, constructing the differential evolution sequence of the first distortion feature value and the second distortion feature value on the time axis specifically includes: after the sampling period corresponding to the current power frequency cycle ends, reading the first distortion feature value and the second distortion feature value, and calculating the difference between the first distortion feature value and the second distortion feature value as the differential value of the current cycle; constructing a set of discrete points with the sampling period as the horizontal axis coordinate and the differential value as the vertical axis coordinate based on the differential value of the current cycle; calculating the quotient of the vertical axis difference divided by the horizontal axis interval for two adjacent differential values ​​in the set of discrete points to obtain the instantaneous change slope of the set of discrete points in each local interval; and combining the set of discrete points and the instantaneous change slope in each local interval into a differential evolution sequence.

[0010] In some embodiments, the instantaneous value sequence of the secondary current on the secondary side of the current transformer is obtained through the current transformer.

[0011] In some embodiments, a sequence of instantaneous secondary voltage values ​​on the secondary side of a current transformer is obtained through a voltage transformer.

[0012] Secondly, this application provides an anomaly identification device for a current transformer, used to execute an anomaly identification method for a current transformer. The device includes: an acquisition module, used to initiate anomaly identification of the current transformer and acquire a dataset of instantaneous electrical quantities on the secondary side of the current transformer, the dataset including a sequence of instantaneous secondary current values ​​and a sequence of instantaneous secondary voltage values; a processing module, used to extract the fundamental voltage component from the sequence of instantaneous secondary voltage values ​​and divide the sequence of instantaneous secondary current values ​​into a first current segment corresponding to the positive half-wave and a second current segment corresponding to the negative half-wave based on the zero-crossing point of the fundamental voltage component; the processing module is further used to calculate a first distortion feature value of the current waveform in the first current segment relative to an ideal sine wave and a second distortion feature value of the current waveform in the second current segment relative to an ideal sine wave, the distortion feature value being used to characterize the degree of nonlinear deformation of the current waveform within the corresponding half-wave; and an execution module, used to construct a differential evolution sequence of the first distortion feature value and the second distortion feature value on the time axis, and then determine whether there is an asymmetric excitation anomaly in the core of the current transformer based on the differential evolution sequence.

[0013] Thirdly, this application provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described abnormal identification method for current transformers.

[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for identifying anomalies in current transformers.

[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: In the anomaly identification method and apparatus for current transformers provided in this application, the anomaly identification of the current transformer is first initiated to obtain the instantaneous electrical quantity dataset of the secondary side of the current transformer. The instantaneous electrical quantity dataset includes a sequence of instantaneous secondary current values ​​and a sequence of instantaneous secondary voltage values. Secondly, the voltage fundamental component is extracted from the sequence of instantaneous secondary voltage values, and the sequence of instantaneous secondary current values ​​is divided into a first current segment corresponding to the positive half-wave and a second current segment corresponding to the negative half-wave based on the zero-crossing point of the voltage fundamental component. Then, a first distortion feature value of the current waveform in the first current segment relative to an ideal sine wave and a second distortion feature value of the current waveform in the second current segment relative to an ideal sine wave are calculated respectively. The distortion feature value is used to characterize the degree of nonlinear deformation of the current waveform in the corresponding half-wave. Finally, a differential evolution sequence of the first distortion feature value and the second distortion feature value on the time axis is constructed, and then the presence of an asymmetric excitation anomaly in the core of the current transformer is determined based on the differential evolution sequence.

[0016] Therefore, this application can accurately identify asymmetric excitation anomalies in the core of a current transformer. First, by acquiring the instantaneous values ​​of the secondary current and voltage on the secondary side of the current transformer, the application provides the original electrical input for the entire anomaly identification process, avoiding calculation deviations and misjudgments caused by missing or asynchronous data. Second, by accurately extracting the fundamental voltage component from the instantaneous voltage value sequence and dividing the instantaneous secondary current value sequence into positive and negative half-waves based on the zero-crossing point of the fundamental voltage component, the application achieves strict timing alignment between the current waveform and the excitation voltage, effectively separating the positive and negative half-wave currents coupled within the same cycle. This specifically highlights the half-wave differential distortion of the core under asymmetric conditions such as saturation, residual magnetism, and local hysteresis, solving the problem that traditional full-cycle analysis easily masks asymmetric faults and cannot distinguish between symmetrical disturbances and asymmetric faults. Then, the application calculates the first and second current segments relative to the ideal values. The distortion characteristic value of the sine wave enables accurate characterization of the distortion degree of the positive and negative half-wave waveforms, significantly improving the ability to perceive weak distortion and asymmetric distortion. This avoids the problem of false alarms when there are symmetrical harmonic sources unrelated to the core anomaly, as the nonlinear deformation degree of the positive and negative half-wave waveforms cannot be identified. Finally, a differential evolution sequence of the first and second distortion characteristic values ​​on the time axis is constructed, which can expand the static difference of a single period into time-series evolution information with amplitude characteristics, rate of change, and continuous trend, reflecting the magnitude of the half-wave distortion asymmetry and capturing the gradual process of anomaly. This effectively filters out false anomaly features caused by instantaneous interference and random noise. Subsequently, the presence of asymmetric excitation anomaly in the core is determined based on the differential evolution sequence, thereby realizing the identification of asymmetric faults in the current transformer core. In summary, the technical solution provided in this application can accurately identify asymmetric excitation anomalies in the core of a current transformer. Attached Figure Description

[0017] Figure 1 is an exemplary flowchart of an anomaly identification method for a current transformer according to some embodiments of the present application; Figure 2 is an exemplary flowchart of determining the fundamental voltage component according to some embodiments of the present application; Figure 3 is a structural schematic diagram of an anomaly identification device for a current transformer according to some embodiments of the present application; Figure 4 is a structural schematic diagram of a computer device for implementing the anomaly identification method for a current transformer according to some embodiments of the present application. Detailed Implementation

[0018] In the embodiments of this application, in order to clearly describe the technical solutions of the embodiments of this application, the terms "first" and "second" are used to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different. The technical features described by "first" and "second" have no sequential or size order.

[0019] In the embodiments of this application, the terms "in some embodiments" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "in some embodiments" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of terms such as "in some embodiments" or "for example" is intended to present the relevant concepts in a specific manner to facilitate understanding.

[0020] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0021] Referring to Figure 1, which is an exemplary flowchart of an anomaly identification method for a current transformer according to some embodiments of this application, the figure mainly includes the following steps: In step S101, anomaly identification of the current transformer is initiated, and the instantaneous electrical quantity dataset of the secondary side of the current transformer is obtained. The instantaneous electrical quantity dataset includes a sequence of instantaneous secondary current values ​​and a sequence of instantaneous secondary voltage values.

[0022] In specific implementation, anomaly identification of the current transformer is initiated, and the instantaneous electrical quantity dataset of the secondary side of the current transformer is obtained through the current transformer and the voltage transformer. Specifically, the instantaneous value sequence of the secondary current on the secondary side of the current transformer is obtained through the current transformer, and the instantaneous value sequence of the secondary voltage on the secondary side of the current transformer is obtained through the voltage transformer, so as to obtain the instantaneous electrical quantity dataset, which includes the instantaneous value sequence of the secondary current and the instantaneous value sequence of the secondary voltage.

[0023] It should be noted that, in this application, the instantaneous value sequence of secondary current refers to the set of discrete instantaneous current amplitudes on the secondary side of the current transformer, which records the transient change trajectory of the current waveform at each sampling moment; the instantaneous value sequence of secondary voltage refers to the set of discrete instantaneous voltage amplitudes on the secondary side of the current transformer, which can provide an accurate phase reference for subsequent analysis.

[0024] In step S102, the fundamental voltage component is extracted from the instantaneous secondary voltage value sequence, and the instantaneous secondary current value sequence is divided into a first current segment corresponding to the positive half-wave and a second current segment corresponding to the negative half-wave based on the zero-crossing point of the fundamental voltage component.

[0025] In some embodiments, referring to FIG2, which is an exemplary flowchart of determining the fundamental voltage component according to some embodiments of this application, the extraction of the fundamental voltage component from the secondary voltage instantaneous value sequence in this embodiment can be achieved by the following steps: In step S1021, the secondary voltage instantaneous value sequence is digitally filtered to obtain a preprocessed voltage sequence; in step S1022, the preprocessed voltage sequence is zero-crossing detected to determine the time interval between adjacent zero-crossings, and the actual length of the current power frequency cycle is calculated based on the time interval; in step S1023, the preprocessed voltage sequence is truncated to an integer period based on the actual length to obtain a voltage integer period data window for calculation; in step S1024, Fourier transform is performed on the discrete points within the voltage integer period data window to calculate the real and imaginary coefficients of the fundamental voltage component, and the fundamental voltage component is reconstructed based on the real and imaginary coefficients.

[0026] In specific implementation, firstly, a digital filtering algorithm is used to filter the instantaneous value sequence of the secondary voltage to obtain a preprocessed voltage sequence that reflects the main trend of voltage change. The digital filtering algorithm can be either a finite-length unit impulse response filter or an infinite-length unit impulse response filter from power signal processing; no specific limitation is made here. This is used to remove noise interference from the acquired signal. Secondly, zero-crossing detection is performed on the preprocessed voltage sequence. All points in the preprocessed voltage sequence where values ​​change from negative to positive or from positive to negative are traversed. Linear interpolation is used to calculate the signal change between two adjacent discrete sampling points, and the moment corresponding to zero amplitude is taken as the zero-crossing point. The time interval between two adjacent zero-crossing points in the same direction is determined. This time interval is the duration for the voltage signal to travel from one positive zero-crossing point to the next positive zero-crossing point or from one negative zero-crossing point to the next negative zero-crossing point. The time interval between these adjacent zero-crossing points in the same direction is determined as the actual length of the current power frequency cycle. This actual length is the voltage signal... The time required for the voltage signal to complete one full periodic change is determined by the small fluctuations in the grid frequency. Therefore, the actual period length is determined by the real-time detected time interval between adjacent zero-crossing points in the same direction, which avoids the phase shift and period truncation errors caused by a fixed 50Hz period. Then, based on this actual length, the preprocessed voltage sequence is truncated to an integer period to obtain a voltage integer period data window for subsequent analysis and calculation. This voltage integer period data window is a discrete voltage data set with a limited length and containing complete period information. Finally, the discrete Fourier transform algorithm is applied to all discrete sampling points within the voltage integer period data window to convert the time-domain voltage signal to the frequency domain for decomposition. The real and imaginary coefficients corresponding to the fundamental voltage component are calculated. These real and imaginary coefficients are frequency domain parameters characterizing the amplitude and phase information of the fundamental component. Then, the real and imaginary coefficients are synthesized into a standard sine wave with the same frequency and phase as the original signal using a trigonometric function reconstruction formula. Finally, a pure and phase-accurate voltage fundamental component is reconstructed.

[0027] It should be noted that the voltage fundamental component in this application is a voltage waveform that retains only the power frequency fundamental component after removing harmonics and interference, providing a precise time reference for subsequent current half-wave segmentation.

[0028] In some embodiments, dividing the instantaneous value sequence of the secondary current into a first current segment corresponding to the positive half-wave and a second current segment corresponding to the negative half-wave based on the zero-crossing points of the fundamental voltage component is achieved through the following steps: analyzing the fundamental voltage component to determine all zero-crossing moments of the fundamental voltage component on the time axis; using the zero-crossing moments as boundaries, marking intervals for the moments corresponding to each sampling point in the instantaneous value sequence of the secondary current; marking sampling points located between two adjacent zero-crossing moments where the voltage corresponding to the initial zero-crossing moment changes from negative to positive as positive half-wave belonging points, and marking sampling points located between two adjacent zero-crossing moments where the voltage corresponding to the initial zero-crossing moment changes from positive to negative as negative half-wave belonging points; extracting corresponding sampling points from the instantaneous value sequence of the secondary current based on the markings of the positive half-wave belonging points, and combining them in chronological order to form the first current segment; extracting corresponding sampling points from the instantaneous value sequence of the secondary current based on the markings of the negative half-wave belonging points, and combining them in chronological order to form the second current segment.

[0029] In specific implementation, firstly, all discrete sampling points of the fundamental voltage component are traversed, and the amplitude sign change of adjacent discrete sampling points is judged sequentially. When adjacent sampling points show a sign jump from negative to positive or from positive to negative, linear interpolation is used to calculate the signal change between the two adjacent discrete sampling points, and the time corresponding to the amplitude being zero is taken as the zero-crossing point. This determines all zero-crossing points of the fundamental voltage component on the time axis; these zero-crossing points are the time points corresponding to the amplitude of the fundamental voltage component changing from positive to negative or from negative to positive. Secondly, after completing all zero-crossing points... After time-crossing extraction, each zero-crossing moment is combined with its adjacent next zero-crossing moment to form a continuous time interval. The initial zero-crossing point of each time interval is marked with the voltage change direction. The time interval where the voltage changes from negative to positive at the initial zero-crossing point is marked as the positive half-wave interval, and the time interval where the voltage changes from positive to negative at the initial zero-crossing point is marked as the negative half-wave interval. For example, if a zero-crossing moment is when the voltage crosses from negative to positive, then the time interval from that moment to the next zero-crossing moment is the positive half-wave interval. Subsequently, the sampling time corresponding to each sampling point in the instantaneous value sequence of the secondary current is compared with... The positive and negative half-wave intervals are time-matched to determine the corresponding time interval in which the sampling time falls. Current sampling points falling into the positive half-wave interval are marked as positive half-wave origin points, meaning the sampling time falls within the positive half-wave interval of the voltage fundamental wave. Similarly, current sampling points falling into the negative half-wave interval are marked as negative half-wave origin points, meaning the sampling time falls within the negative half-wave interval of the voltage fundamental wave. Then, after marking all current sampling points within their respective intervals, the sampling times are ordered from the instantaneous secondary current... All sampling points marked as positive half-wave affiliation points are extracted from the value sequence and combined in an orderly manner to form a first current segment corresponding to the positive half-wave interval of the voltage fundamental wave. This first current segment is a set of secondary current data that only includes the positive half-wave time period of the voltage. Finally, all sampling points marked as negative half-wave affiliation points are extracted from the instantaneous value sequence of the secondary current according to the same timing rules as the first current segment and combined in an orderly manner to form a second current segment corresponding to the negative half-wave interval of the voltage fundamental wave. This second current segment is a set of secondary current data that only includes the negative half-wave time period of the voltage.

[0030] It should be noted that when a current transformer core experiences asymmetric excitation anomalies such as saturation, residual magnetism, or uneven local hysteresis loss, the excitation characteristics of the positive and negative half-cycles are usually inconsistent. This leads to asymmetric and differentiated waveform distortion characteristics of the secondary current in the positive and negative half-waves. Current methods using the full-cycle waveform for overall analysis easily mask the distortion differences between the half-waves and cannot accurately reflect the asymmetric fault characteristics of the core. By independently separating the positive and negative half-wave current waveforms that are originally coupled in the same cycle, independent and aligned analysis objects can be provided for subsequent quantification of the distortion degree of the two half-waves. This allows the distortion characteristics of the positive and negative half-waves to be extracted separately and compared differentially, thus providing a reliable data foundation and clear comparative basis for subsequent identification of asymmetric excitation anomalies, effectively improving the pertinence and sensitivity of anomaly identification.

[0031] In step S103, the first distortion characteristic value of the current waveform in the first current segment relative to the ideal sine wave and the second distortion characteristic value of the current waveform in the second current segment relative to the ideal sine wave are calculated respectively. The distortion characteristic value is used to characterize the degree of nonlinear deformation of the current waveform in the corresponding half-wave.

[0032] In some embodiments, the calculation of a first distortion characteristic value of the current waveform in the first current segment relative to an ideal sine wave and a second distortion characteristic value of the current waveform in the second current segment relative to an ideal sine wave are achieved by the following steps: performing a discrete Fourier transform on the current waveform sequence in the first current segment to extract the amplitude coefficients of each harmonic component in the first current segment; calculating the total harmonic distortion rate of the first current segment as the first distortion characteristic value based on the amplitude coefficients of each harmonic component and the fundamental amplitude of the first current segment; performing a discrete Fourier transform on the current waveform sequence in the second current segment to extract the amplitude coefficients of each harmonic component in the second current segment; calculating the total harmonic distortion rate of the second current segment as the second distortion characteristic value based on the amplitude coefficients of each harmonic component and the fundamental amplitude of the second current segment.

[0033] In specific implementation, firstly, a Discrete Fourier Transform (DFT) operation is performed on the current waveform sequence of the first current segment. The current waveform sequence is substituted point by point into the DFT formula to calculate the real and imaginary coefficients corresponding to different frequency components. These real and imaginary coefficients are the basic calculation results of the DFT in the frequency domain, representing the amplitude and phase of the corresponding frequency components. Then, the real and imaginary coefficients corresponding to each frequency component are squared and summed, and the square root of the summation is performed to obtain the amplitude coefficients corresponding to each frequency component. These amplitude coefficients are parameters representing the magnitude of the corresponding harmonic components. Secondly, the amplitude coefficient corresponding to the fundamental component obtained after the DFT of the first current segment is used as the fundamental amplitude of the first current segment. The amplitude coefficients of each harmonic component other than the fundamental component are squared and summed, and the square root of the summation is performed to obtain the total amplitude. The total harmonic effective value is calculated, and then the ratio of the total harmonic effective value to the fundamental amplitude is used to obtain the total harmonic distortion rate of the first current segment, which is used as the first distortion characteristic value. The first distortion characteristic value is a quantitative parameter characterizing the degree of nonlinear deformation of the positive half-wave current waveform. Then, using the same calculation process as the first current segment, a discrete Fourier transform is performed on the discrete time domain waveform sequence of the second current segment to obtain the amplitude coefficients of each harmonic component in the second current segment. Finally, the amplitude coefficients corresponding to the fundamental component obtained after the discrete Fourier transform of the second current segment are used as the fundamental amplitude of the second current segment. The amplitude coefficients of each harmonic component other than the fundamental component are squared and summed. The square root of the summation result is used to obtain the total harmonic effective value. Then, the ratio of the total harmonic effective value to the fundamental amplitude is used to obtain the total harmonic distortion rate of the second current segment, which is used as the second distortion characteristic value.

[0034] It should be noted that the second distortion characteristic value in this application is a quantitative parameter characterizing the degree of nonlinear deformation of the negative half-wave current waveform. In addition, when an asymmetric excitation anomaly occurs in the core of a current transformer, there are usually significant differences in the excitation saturation, hysteresis characteristics, and residual magnetism of the positive and negative half-waves, which directly leads to inconsistent waveform distortion degrees between the two half-waves. If only the overall distortion index of the whole wave is used, it cannot reflect this distortion difference between the half-waves, and it is difficult to distinguish between symmetrical disturbances and asymmetric faults. By using the total harmonic distortion rate to independently and quantitatively characterize the distortion of the positive and negative half-wave current waveforms, the nonlinear deformation of the waveform, which is originally difficult to judge intuitively, can be transformed into a numerically comparable characteristic quantity. This effectively highlights the half-wave distortion imbalance characteristics caused by asymmetric excitation and improves the sensitivity and reliability of identifying asymmetric faults such as core saturation, residual magnetism, and local hysteresis anomalies.

[0035] In step S104, a differential evolution sequence of the first distortion feature value and the second distortion feature value on the time axis is constructed, and then the presence of an asymmetric excitation anomaly in the core of the current transformer is determined based on the differential evolution sequence.

[0036] In some embodiments, the differential evolution sequence of the first distortion feature value and the second distortion feature value on the time axis is constructed by the following steps: after the sampling period corresponding to the current power frequency cycle ends, the first distortion feature value and the second distortion feature value are read, and the difference between the first distortion feature value and the second distortion feature value is calculated as the difference value of the current cycle; based on the difference value of the current cycle, a set of discrete points is constructed with the sampling period as the horizontal axis coordinate and the difference value as the vertical axis coordinate; for two adjacent difference values ​​in the set of discrete points, the quotient of the difference of their vertical coordinates divided by the horizontal coordinate interval is calculated sequentially to obtain the instantaneous change slope of the set of discrete points in each local interval; the set of discrete points and the instantaneous change slope in each local interval are combined to form a differential evolution sequence.

[0037] In specific implementation, firstly, the first and second distortion characteristic values ​​corresponding to the distortion levels of the positive and negative half-wave currents in the current cycle are read. The first and second distortion characteristic values ​​are subtracted to obtain their numerical difference within the same cycle, which is then used as the difference value for the current cycle. This difference value characterizes the asymmetry of the distortion levels of the positive and negative half-wave current waveforms within the same cycle. Secondly, using multiple consecutive sampling cycles as the horizontal axis of the time axis and the difference values ​​corresponding to each cycle as the vertical axis, the difference values ​​calculated cycle by cycle are arranged in chronological order to form a discrete set of points per cycle. This discrete set of points represents ordered numerical points reflecting the trend of the distortion difference over time. Then, for this discrete set... For each pair of adjacent discrete points in the point set, the known method of numerical differentiation is used to obtain the ordinate difference by subtracting the ordinate difference of the previous discrete point from the ordinate difference of the subsequent discrete point. Then, this ordinate difference is divided by the abscissa interval corresponding to the two adjacent periods to obtain the instantaneous change slope reflecting the rate of change of the ordinate value within the local interval. This instantaneous change slope is a numerical parameter used to describe the local change trend of the difference sequence. For example, when the ordinate difference between two adjacent periods increases, the instantaneous change slope is positive, indicating that the degree of asymmetry is increasing. Finally, the discrete point set arranged in chronological order is combined one by one with the instantaneous change slope of each corresponding local interval in time sequence to form a difference evolution sequence that can simultaneously reflect the magnitude of the distortion difference and its dynamic change trend.

[0038] It should be noted that the differential evolution sequence in this application refers to the characteristic sequence characterizing the development and change of asymmetric excitation anomaly of the current transformer core over time. When an asymmetric excitation anomaly occurs in the current transformer core, the difference between the positive and negative half-wave distortion characteristics is not fixed, but exhibits a continuous dynamic change pattern with the excitation state, load fluctuation, and residual magnetism accumulation. Relying solely on the distortion difference of a single cycle cannot reflect the development trend and persistence characteristics of the anomaly, and is easily susceptible to instantaneous interference, leading to misjudgment. By combining the distortion difference calculated cycle by cycle with the instantaneous change slope of adjacent cycles in a time sequence, the magnitude of the half-wave distortion asymmetry can be characterized from the amplitude dimension, and the speed and development trend of the asymmetric characteristics can be reflected from the slope dimension. Thus, the isolated distortion difference is transformed into continuous characteristic information with time correlation, providing a stable basis for subsequent judgment on whether there is an asymmetric excitation anomaly in the core.

[0039] In some embodiments, determining whether the core of the current transformer has an asymmetric excitation anomaly based on the differential evolution sequence is achieved through the following steps: determining a dynamic threshold discrimination interval for determining whether the core of the current transformer has an asymmetric excitation anomaly, the dynamic threshold discrimination interval including an adaptive upper threshold and an adaptive lower threshold; reading the latest differential value contained in the differential evolution sequence as the current differential value to be detected; comparing the current differential value to be detected with the adaptive upper threshold and the adaptive lower threshold respectively; if the current differential value to be detected is greater than the adaptive upper threshold or less than the adaptive lower threshold, then starting a duration timer to record the duration of the current differential value to be detected continuously exceeding the limit. Number of cycles; after the number of cycles recorded by the duration timer reaches a preset duration threshold, extract the local sequence segment corresponding to the time period recorded by the duration timer in the differential evolution sequence; calculate the difference between adjacent differential values ​​in the local sequence segment to obtain the instantaneous change slope of the local sequence segment in each local interval, and determine whether the positive and negative signs of all the instantaneous change slopes are consistent; if the positive and negative signs of all the instantaneous change slopes are consistent and the current differential value to be detected is always greater than the adaptive upper limit threshold or always less than the adaptive lower limit threshold within the duration threshold, then generate an alarm signal characterizing that the current transformer core has an asymmetric excitation abnormality caused by the accumulation of local weak magnetization.

[0040] In specific implementation, firstly, a dynamic threshold discrimination interval is determined for judging whether the core of the current transformer has an asymmetric excitation anomaly. This dynamic threshold discrimination interval includes an adaptive upper threshold and an adaptive lower threshold. Secondly, the latest time-series differential value is read from the differential evolution sequence as the current differential value to be detected. This current differential value represents the degree of asymmetry in the positive and negative half-wave distortion within the most recent cycle. The current differential value to be detected is compared with both the adaptive upper threshold and the adaptive lower threshold. If the current differential value to be detected is greater than the adaptive upper threshold or less than the adaptive lower threshold, then the current differential value is determined to be... When a value exceeds the limit, a duration timer is activated. This duration timer is a counting unit that uses the power frequency cycle as the counting unit to accumulate the number of consecutive exceedance cycles of the differential value. The count value is incremented after each power frequency cycle of discrimination, thus recording the number of consecutive exceedance cycles of the current differential value in real time. Furthermore, when the accumulated number of cycles recorded by the duration timer reaches a preset duration threshold—the minimum number of consecutive cycles used to distinguish between transient disturbances and persistent faults—a differential sequence corresponding to the consecutive exceedance time period recorded by the duration timer is extracted from the differential evolution sequence. This data is used as a local sequence segment for trend judgment. This local sequence segment is a continuous data set reflecting the changing pattern of difference values ​​during periods of continuous boundary crossing. Then, the difference between each pair of adjacent difference values ​​in the local sequence segment is calculated to obtain the change in adjacent difference values. This change is then divided by a fixed time interval between adjacent periods to obtain the instantaneous change slope of the local sequence segment within each adjacent period interval. This instantaneous change slope is a numerical parameter used to characterize the direction and speed of change of difference values ​​within the local time period. Finally, the signs of all instantaneous change slopes are compared one by one to determine the trend. The system checks whether all slope signs are positive or negative, i.e., whether they are completely consistent. If the signs of all instantaneous slope changes are consistent, and the current differential value under test remains greater than the adaptive upper threshold or less than the adaptive lower threshold throughout the entire duration threshold accumulated by the duration timer, i.e., there is no regression to the dynamic threshold discrimination interval, then it is determined that the current transformer core has an asymmetric excitation abnormality caused by the gradual accumulation of local weak magnetization, and an alarm signal is generated to identify this fault type. This alarm signal is an indication signal used to prompt maintenance personnel that the core has an asymmetric excitation abnormality.

[0041] The determination of the dynamic threshold discrimination interval for judging whether the core of the current transformer has an asymmetric excitation anomaly is achieved through the following steps: reading a historical data window of a preset length before the current moment in the differential evolution sequence, calculating the moving average and moving standard deviation of all differential values ​​within the historical data window; setting an adaptive upper threshold of the dynamic threshold discrimination interval based on the sum of the moving average and the moving standard deviation, and setting an adaptive lower threshold of the dynamic threshold discrimination interval based on the difference between the moving average and the moving standard deviation; and constructing the dynamic threshold discrimination interval for judging whether the core of the current transformer has an asymmetric excitation anomaly based on the adaptive upper threshold and the adaptive lower threshold.

[0042] In practice, firstly, according to a preset data length, historical difference value data that are sequentially continuous and of fixed length before the current discrimination time are selected from the difference evolution sequence to form a historical data window for real-time statistical analysis. This historical data window is a set of historical data that is continuously updated in a sliding manner to reflect the normal fluctuation level of recent difference values. Then, all difference values ​​within this historical data window are summed and divided by the total number of data points in the window to obtain the moving average, which represents the overall average level of the difference values. This moving average is a statistical parameter reflecting the central trend of recent difference values. Next, the squared difference between each difference value in the window and the moving average is calculated sequentially. The sum of all squared differences is divided by the total number of data points to obtain the variance. The square root of the variance is then taken to obtain the variance representing the degree of dispersion of the difference values. The moving standard deviation is a statistical parameter reflecting the recent fluctuation range of the difference value. Then, the calculated moving average and moving standard deviation are added together, and the sum is used as the adaptive upper limit threshold of the dynamic threshold discrimination interval. This adaptive upper limit threshold is a dynamic critical value used to judge positive abnormal deviations of the difference value. Simultaneously, the moving average and moving standard deviation are subtracted, and the difference is used as the adaptive lower limit threshold of the dynamic threshold discrimination interval. This adaptive lower limit threshold is a dynamic critical value used to judge negative abnormal deviations of the difference value. Finally, using the adaptive upper limit threshold as the upper bound of the interval and the adaptive lower limit threshold as the lower bound, a dynamic threshold discrimination interval that can adaptively adjust in real time according to operating conditions and is used to distinguish between normal fluctuations and abnormal states is constructed.

[0043] It should be noted that the dynamic threshold discrimination interval in this application refers to the adaptive numerical interval used for differential value out-of-bounds judgment. For example, when the operating conditions of the current transformer fluctuate slightly, the sliding average and sliding standard deviation will be updated synchronously, so that the adaptive upper limit threshold and the adaptive lower limit threshold are adjusted accordingly, thereby avoiding misjudgment or omission of fixed threshold when the operating conditions change.

[0044] In another aspect, in some embodiments, this application provides an anomaly identification device for a current transformer. Referring to FIG3, which is a structural schematic diagram of an anomaly identification device for a current transformer according to some embodiments of this application, the anomaly identification device for the current transformer includes: an acquisition module 201, a processing module 202, and an execution module 203, which are described as follows: The acquisition module 201 is mainly used to initiate anomaly identification of the current transformer and acquire the instantaneous electrical quantity dataset of the secondary side of the current transformer. The instantaneous electrical quantity dataset includes a sequence of instantaneous values ​​of the secondary current and a sequence of instantaneous values ​​of the secondary voltage. The processing module 202 is mainly used to extract the fundamental voltage component from the sequence of instantaneous values ​​of the secondary voltage. The processing module 202 is further configured to calculate, based on the zero-crossing point of the fundamental voltage component, the instantaneous value sequence of the secondary current into a first current segment corresponding to the positive half-wave and a second current segment corresponding to the negative half-wave; the processing module 202 is further configured to calculate, respectively, the first distortion characteristic value of the current waveform in the first current segment relative to the ideal sine wave, and the second distortion characteristic value of the current waveform in the second current segment relative to the ideal sine wave, the distortion characteristic value being used to characterize the degree of nonlinear deformation of the current waveform in the corresponding half-wave; the execution module 203, in this application, is mainly used to construct the differential evolution sequence of the first distortion characteristic value and the second distortion characteristic value on the time axis, and then determine whether there is an asymmetric excitation abnormality in the core of the current transformer based on the differential evolution sequence.

[0045] In addition, this application also provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described abnormal identification method for current transformers.

[0046] In some embodiments, referring to FIG4, this figure is a schematic structural diagram of a computer device for implementing an anomaly identification method for a current transformer according to some embodiments of this application. The anomaly identification method for a current transformer in the above embodiments can be implemented by the computer device shown in FIG4, which includes at least one processor 301, a communication bus 302, a memory 303, and at least one communication interface 304.

[0047] The processor 301 may be a general-purpose central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more devices used to control the execution of the anomaly identification method for the current transformer in this application.

[0048] The communication bus 302 can be used to transmit information between the aforementioned components.

[0049] The memory 303 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. The memory 303 may exist independently and be connected to the processor 301 via the communication bus 302. The memory 303 may also be integrated with the processor 301.

[0050] The memory 303 stores program code for executing the scheme of this application, and its execution is controlled by the processor 301. The processor 301 executes the program code stored in the memory 303. The program code may include one or more software modules. In the above embodiments, the determination of the anomaly identification method of the current transformer can be implemented by the processor 301 and one or more software modules in the program code in the memory 303.

[0051] Communication interface 304 uses any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0052] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0053] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0054] In addition, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for identifying anomalies in current transformers.

[0055] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0056] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for identifying anomalies in a current transformer, characterized in that, The process includes the following steps: initiating anomaly identification of the current transformer, acquiring the instantaneous electrical quantity dataset of the secondary side of the current transformer, the instantaneous electrical quantity dataset containing the sequence of instantaneous secondary current values ​​and the sequence of instantaneous secondary voltage values; extracting the fundamental voltage component from the sequence of instantaneous secondary voltage values, and dividing the sequence of instantaneous secondary current values ​​into a first current segment corresponding to the positive half-wave and a second current segment corresponding to the negative half-wave based on the zero-crossing point of the fundamental voltage component; calculating the first distortion characteristic value of the current waveform in the first current segment relative to an ideal sine wave, and the second distortion characteristic value of the current waveform in the second current segment relative to an ideal sine wave, the distortion characteristic value being used to characterize the degree of nonlinear deformation of the current waveform within the corresponding half-wave; A differential evolution sequence of the first distortion feature value and the second distortion feature value on the time axis is constructed, and then the presence of an asymmetric excitation anomaly in the core of the current transformer is determined based on the differential evolution sequence.

2. The method as described in claim 1, characterized in that, Extracting the fundamental voltage component from the instantaneous secondary voltage value sequence specifically includes: performing digital filtering on the instantaneous secondary voltage value sequence to obtain a preprocessed voltage sequence; performing zero-crossing detection on the preprocessed voltage sequence to determine the time interval between adjacent zero-crossings, and calculating the actual length of the current power frequency cycle based on the time interval; truncating the preprocessed voltage sequence into integer cycles based on the actual length to obtain a voltage integer cycle data window for calculation; performing Fourier transform on the discrete points within the voltage integer cycle data window to calculate the real and imaginary coefficients of the fundamental voltage component, and reconstructing the fundamental voltage component based on the real and imaginary coefficients.

3. The method as described in claim 1, characterized in that, Dividing the instantaneous value sequence of the secondary current into a first current segment corresponding to the positive half-wave and a second current segment corresponding to the negative half-wave based on the zero-crossing points of the fundamental voltage component specifically includes: analyzing the fundamental voltage component to determine all zero-crossing moments of the fundamental voltage component on the time axis; using the zero-crossing moments as boundaries, marking the time intervals corresponding to each sampling point in the instantaneous value sequence of the secondary current; marking the sampling points located between two adjacent zero-crossing moments where the voltage corresponding to the initial zero-crossing moment changes from negative to positive as positive half-wave belonging points, and marking the sampling points located between two adjacent zero-crossing moments where the voltage corresponding to the initial zero-crossing moment changes from positive to negative as negative half-wave belonging points; extracting the corresponding sampling points from the instantaneous value sequence of the secondary current based on the markings of the positive half-wave belonging points, and combining them in chronological order to form the first current segment; extracting the corresponding sampling points from the instantaneous value sequence of the secondary current based on the markings of the negative half-wave belonging points, and combining them in chronological order to form the second current segment.

4. The method as described in claim 1, characterized in that, Calculating the first distortion characteristic value of the current waveform in the first current segment relative to an ideal sine wave, and the second distortion characteristic value of the current waveform in the second current segment relative to an ideal sine wave, specifically includes: performing a Discrete Fourier Transform on the current waveform sequence in the first current segment to extract the amplitude coefficients of each harmonic component in the first current segment; calculating the total harmonic distortion rate of the first current segment as the first distortion characteristic value based on the amplitude coefficients of each harmonic component and the fundamental amplitude of the first current segment; performing a Discrete Fourier Transform on the current waveform sequence in the second current segment to extract the amplitude coefficients of each harmonic component in the second current segment; and calculating the total harmonic distortion rate of the second current segment as the second distortion characteristic value based on the amplitude coefficients of each harmonic component and the fundamental amplitude of the second current segment.

5. The method as described in claim 1, characterized in that, Constructing the differential evolution sequence of the first distortion feature value and the second distortion feature value on the time axis specifically includes: after the sampling period corresponding to the current power frequency cycle ends, reading the first distortion feature value and the second distortion feature value, and calculating the difference between the first distortion feature value and the second distortion feature value as the differential value of the current cycle; constructing a set of discrete points with the sampling period as the horizontal axis coordinate and the differential value as the vertical axis coordinate based on the differential value of the current cycle; calculating the quotient of the difference in the vertical coordinate divided by the horizontal coordinate interval for two adjacent differential values ​​in the set of discrete points to obtain the instantaneous change slope of the set of discrete points in each local interval; and combining the set of discrete points and the instantaneous change slope in each local interval into a differential evolution sequence.

6. The method as described in claim 1, characterized in that, The instantaneous value sequence of the secondary current on the secondary side of the current transformer is obtained through the current transformer.

7. The method as described in claim 1, characterized in that, The instantaneous value sequence of the secondary voltage on the secondary side of the current transformer is obtained by using a voltage transformer.

8. An anomaly identification device for a current transformer, used to execute the anomaly identification method for a current transformer as described in any one of claims 1 to 7, characterized in that, The device includes: an acquisition module for initiating anomaly identification of a current transformer and acquiring a dataset of instantaneous electrical quantities on the secondary side of the current transformer, the dataset containing a sequence of instantaneous secondary current values ​​and a sequence of instantaneous secondary voltage values; a processing module for extracting the fundamental voltage component from the sequence of instantaneous secondary voltage values ​​and dividing the sequence of instantaneous secondary current values ​​into a first current segment corresponding to the positive half-wave and a second current segment corresponding to the negative half-wave based on the zero-crossing point of the fundamental voltage component; the processing module is further configured to calculate a first distortion feature value of the current waveform in the first current segment relative to an ideal sine wave and a second distortion feature value of the current waveform in the second current segment relative to an ideal sine wave, the distortion feature value being used to characterize the degree of nonlinear deformation of the current waveform within the corresponding half-wave; and an execution module for constructing a differential evolution sequence of the first distortion feature value and the second distortion feature value on the time axis, and then determining whether there is an asymmetric excitation anomaly in the core of the current transformer based on the differential evolution sequence.

9. A computer device, characterized in that, The computer device includes a memory and a processor, the memory storing code, and the processor being configured to retrieve the code and execute the anomaly identification method for a current transformer as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the anomaly identification method for current transformers as described in any one of claims 1 to 7.

Images