A fault research and judgment method and system based on a transformer area intelligent terminal
By using the fault assessment method of the intelligent terminal in the transformer substation, and through feature extraction and decoupling processing of current and voltage data, a fault assessment index is generated, which solves the problem of detecting high-resistance grounding faults and achieves efficient and reliable fault location and early warning.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- LINFEN FENNENG POWER TECH TESTING CO LTD
- Filing Date
- 2026-02-27
- Publication Date
- 2026-06-05
AI Technical Summary
Existing technologies are insufficient to effectively detect and locate high-resistance grounding faults in distribution network areas, leading to long-term latency of the faults and potentially causing large-scale power outages or equipment damage.
The system acquires three-phase current and voltage data through intelligent terminals in the transformer substation, performs feature extraction and decoupling processing, constructs a reference signal and dynamic coupling coefficient, calculates fault residual energy and projection intensity index, generates a fault judgment index, and achieves accurate judgment of high-resistance grounding faults.
It improves the sensitivity and reliability of high-resistance grounding fault assessment, reduces false alarms and missed alarms, promptly detects potential hazards, and enhances the stability and safety of power supply.
Smart Images
Figure CN121741386B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of power grid management technology, and in particular to a fault diagnosis method and system based on intelligent terminals in distribution transformer areas. Background Technology
[0002] As a key link in the power system for distributing electrical energy to end users, the distribution network is directly related to the stability of power supply and the quality of power supply services. The distribution area is the basic unit for realizing power distribution and management in the distribution network. It usually refers to a power supply unit with the distribution transformer as the core, which is composed of the distribution transformer, high-voltage incoming line, low-voltage outgoing line, metering device, protection equipment and the low-voltage power supply area it covers. It is an important hub connecting the high-voltage side of the distribution network and end users.
[0003] The division and operation management of distribution substations are the core foundation for the refined operation and maintenance of the power distribution network. Their coverage is determined based on factors such as regional power load, geographical environment, and user distribution. They are responsible for the power transmission, voltage regulation, and load distribution functions for various users such as residents, industrial and commercial users within the region.
[0004] High-resistance grounding faults are a common and highly dangerous type of fault in the operation of power distribution networks. This fault usually refers to an abnormal electrical connection between the live parts of power distribution lines and electrical equipment and the ground through a high-resistance medium. The impedance value of the fault circuit is much higher than that of conventional metallic grounding faults, and the fault current is usually weak and has no obvious arc or short-circuit impact characteristics.
[0005] High-resistance grounding faults are difficult to detect and locate effectively using conventional protection devices such as overcurrent protection or zero-sequence protection. If the existence of the fault cannot be determined in time, the fault will remain dormant for a long time and gradually evolve into a serious fault, resulting in large-scale power outages or equipment damage in the transformer area. Therefore, it is necessary to assess whether a high-resistance grounding fault exists in the transformer area. Summary of the Invention
[0006] To more accurately determine whether a transformer substation has a high-resistance grounding fault, this application provides a fault assessment method and system based on a transformer substation smart terminal.
[0007] According to a first aspect of the embodiments of this application, a fault assessment method based on a smart terminal in a distribution transformer area is provided, comprising: acquiring three-phase current data of the low-voltage side of the distribution transformer through the smart terminal, and extracting features from the three-phase current data to obtain the fundamental zero-sequence current component, the fundamental negative-sequence current component, and the third harmonic current component; constructing a reference signal based on the fundamental negative-sequence current component and the third harmonic current component, and calculating the dynamic coupling coefficient between the fundamental zero-sequence current component and the reference signal; performing orthogonal decoupling processing on the fundamental zero-sequence current component based on the dynamic coupling coefficient to obtain the residual current component, and determining the fault residual energy index of the residual current component; acquiring three-phase voltage data of the low-voltage side of the distribution transformer through the smart terminal, determining the zero-sequence voltage component based on the three-phase voltage data, and calculating the projection intensity index of the zero-sequence voltage component and the residual current component; the projection intensity index is used to characterize the degree of in-phase relationship between the fault current and the zero-sequence voltage; generating a fault assessment index based on the fault residual energy index and the projection intensity index, and using the fault assessment index to assess whether a high-resistance grounding fault exists in the distribution transformer area.
[0008] This allows for a more accurate assessment of whether a high-resistance grounding fault exists in the transformer area.
[0009] Optionally, feature extraction is performed on the three-phase current data to obtain the fundamental zero-sequence current component, the fundamental negative-sequence current component, and the third harmonic current component, including: transforming the three-phase current data using the symmetrical component method to extract the fundamental negative-sequence current component and the fundamental zero-sequence current component; and performing spectral analysis on the three-phase current data using the fast Fourier transform to separate the third harmonic current component.
[0010] Optionally, a reference signal is constructed based on the fundamental negative sequence current component and the third harmonic current component, including: linearly superimposing the instantaneous value of the fundamental negative sequence current component with the weighted instantaneous value of the third harmonic current component to obtain the reference signal.
[0011] Thus, considering that the zero-sequence current in the low-voltage distribution area mainly originates from the asymmetrical and nonlinear loads of the three-phase load, a reference signal can be constructed by linear superposition to fit the background zero-sequence current pattern under normal operating conditions, providing a reference benchmark for subsequent decoupling.
[0012] Optionally, the dynamic coupling coefficient is determined by the following formula: ,in, The dynamic coupling coefficient is used to characterize the projection intensity of the fundamental zero-sequence current component onto the reference signal direction. The number of sampling points contained in the sliding calculation window. The sampling point number within the sliding calculation window. Let be the instantaneous value of the fundamental zero-sequence current component at the k-th sampling point. The instantaneous value of the reference signal at the k-th sampling point. This is used to prevent positive numbers with a denominator of zero.
[0013] Optionally, the fault residual energy index of the residual current component is determined as follows: the background zero-sequence current estimation component is obtained by multiplying the dynamic coupling coefficient by the reference signal; the background zero-sequence current estimation component represents the component of the fundamental zero-sequence current component that is related to the load imbalance characteristics; the difference between the fundamental zero-sequence current component and the background zero-sequence current estimation component is calculated to obtain the instantaneous value of the residual current component; the instantaneous value of the residual current component within the sliding calculation window is squared, and the arithmetic mean of the squares corresponding to different instantaneous values is used as the fault residual energy index; the fault residual energy index is used to characterize the independent abnormal energy in the signal that cannot be linearly characterized by the reference signal.
[0014] In this way, the original zero-sequence current is decomposed into a component related to the load characteristics and a residual component unrelated to the load characteristics by using the orthogonal projection principle. The residual component is the abrupt signal generated by the high-impedance grounding fault, thereby realizing the extraction of weak fault signals under strong background noise.
[0015] Optionally, the projection intensity indices of the zero-sequence voltage component and the residual current component are calculated, including: ,in, For projection intensity index; The rated phase voltage constant of the transformer substation; This represents the number of sampling points contained in the sliding calculation window. The sampling point number within the sliding calculation window; The instantaneous value of the zero-sequence voltage component at the k-th sampling point. This represents the instantaneous value of the fundamental zero-sequence current component at the kth sampling point; For dynamic coupling coefficients; The instantaneous value of the reference signal at the kth sampling point; This represents the instantaneous value of the residual current component at the k-th sampling point. To take the absolute value.
[0016] In this way, by utilizing the specific phase relationship between the zero-sequence voltage and the fault current when a high-resistance ground fault occurs, and by calculating the projected intensity, it is possible to further eliminate interference signals that are abnormal in amplitude but do not match the phase logic, thereby improving the robustness of the judgment.
[0017] Optionally, a fault assessment index is generated based on the fault residual energy index and the projection intensity index, including: multiplying the fault residual energy index by a preset balance coefficient to obtain a weighted energy component, and using the sum of the projection intensity index and the weighted energy component as the fault assessment index; the balance coefficient is used to adjust the weight of the fault residual energy index in the fault assessment index.
[0018] Optionally, before collecting the three-phase voltage and three-phase current data of the low-voltage side of the distribution transformer in real time via a smart terminal, the method further includes: monitoring the zero-crossing points of the three-phase voltage and three-phase current data in real time, calculating the time difference between the two zero-crossing points, determining the phase asynchrony deviation based on the time difference; and using a Lagrange interpolation algorithm to dynamically adjust the sampling index position of the three-phase current data according to the phase asynchrony deviation, so that the three-phase voltage and three-phase current data are aligned on the time axis.
[0019] This eliminates phase deviations in voltage and current data caused by differences in transformer transmission characteristics or varying delays in signal conditioning circuits, ensuring the accuracy of all subsequent phase-based projection calculations and avoiding misjudgments due to asynchronous sampling.
[0020] Optionally, the method further includes: performing trend analysis on the change data of the fault judgment index within a historical preset time period; if the trend analysis result indicates that the fault judgment index has not exceeded the preset action threshold within the historical preset time period but shows a monotonically increasing trend, outputting early warning information to indicate the deterioration of the insulation performance of the line.
[0021] This allows for the timely detection of potential hazards such as aging cable insulation or slow-growing trees affecting contact wires.
[0022] Optionally, the fault assessment index is used to assess whether a transformer substation has a high-resistance grounding fault, including: determining that a transformer substation has a high-resistance grounding fault when the fault assessment index is greater than a preset action threshold and the duration exceeds a preset duration; or determining that a transformer substation does not have a high-resistance grounding fault when the fault assessment index is less than or equal to the preset action threshold or the duration is less than or equal to the preset duration.
[0023] According to a second aspect of the embodiments of this application, a fault assessment system based on a smart terminal in a distribution area is provided, comprising: a processor and a memory, wherein the memory stores computer program instructions, and the computer program instructions, when executed by the processor, implement the steps of the fault assessment method based on a smart terminal in a distribution area provided in the first aspect of this application.
[0024] The technical solutions provided by the embodiments of this application may include the following beneficial effects: by constructing a reference signal containing negative sequence and third harmonic, the contribution of normal load imbalance to zero sequence current is quantified by using dynamic coupling coefficient, and then the background noise is stripped off by orthogonal decoupling to obtain a pure residual current component, eliminating the interference of three-phase imbalance and harmonics on fault detection. Furthermore, by introducing a projection intensity index of voltage dimension, the fault is located from two dimensions: energy amplitude and phase correlation, which improves the sensitivity and reliability of high-resistance grounding fault judgment.
[0025] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description
[0026] Figure 1 This is a flowchart illustrating a fault assessment method based on a smart terminal in a distribution area, according to an exemplary embodiment.
[0027] Figure 2 This is a trend graph of the dynamic coupling coefficient changing over time in the embodiments of this application;
[0028] Figure 3 This is a schematic diagram comparing the fault assessment index and the effective value of the zero-sequence current in the embodiments of this application;
[0029] Figure 4 This is a schematic diagram illustrating the structure of a fault assessment system based on a smart terminal in a transformer substation, according to an exemplary embodiment. Detailed Implementation
[0030] To more accurately determine whether a distribution transformer in a transformer substation has a high-resistance grounding fault, this application provides a fault assessment method and system based on a smart terminal in the transformer substation. Figure 1 This is a flowchart illustrating a fault assessment method based on a smart terminal in a distribution area, according to an exemplary embodiment. Figure 1 As shown, the method includes the following steps.
[0031] In step S101, the three-phase current data of the low-voltage side of the distribution transformer is obtained through the smart terminal, and the features of the three-phase current data are extracted to obtain the fundamental zero-sequence current component, the fundamental negative-sequence current component, and the third harmonic current component, respectively.
[0032] In one embodiment, before acquiring the three-phase current data of the low-voltage side of the distribution transformer through a smart terminal, a high-precision synchronous calibration of the data acquisition system is required. Specifically, the zero-crossing points of the three-phase voltage and current data are monitored in real time, and the time difference between their zero-crossing points is calculated. The phase asynchrony deviation is determined based on the time difference. Using a Lagrange interpolation algorithm, the sampling index position of the three-phase current data is dynamically adjusted according to the phase asynchrony deviation to align the three-phase voltage and current data on the time axis.
[0033] For the processing details of phase synchronization, the smart terminal can be configured with a sampling module. The sampling frequency can be set to 12.8kHz, that is, 256 points are sampled per cycle. Since the signal transmission paths of voltage transformers and current transformers are different, and considering the differences in filter capacitors, there is often a certain time difference when the physical signal arrives at the processor.
[0034] The zero-crossing points of voltage and current are captured by a hardware zero-crossing detection circuit, and the time difference between the two is calculated. Assuming the sampling interval is Then the decimal point index offset corresponding to the phase misalignment deviation is The fourth-order Lagrange interpolation formula can be used to reconstruct the time-aligned current sampling value using the original data of the current sampling point and the two sampling points before and after it.
[0035] In subsequent fault analysis, the phase relationship between voltage and current, i.e., the projection intensity index, is the key basis for determining whether a fault has actually occurred. If the phase delay caused by the physical channel is ignored, it will lead to a deviation in the calculation of the angle between the zero-sequence voltage and the residual current.
[0036] For example, a 30-degree phase angle deviation caused by physical channel delay may cause the resistive ground fault current, which is originally in phase, to exhibit capacitive characteristics, thus leading to missed detection by the algorithm. The Lagrange interpolation algorithm can achieve high-precision fractional delay compensation in the digital domain with moderate computational complexity, making it suitable for real-time processing in embedded terminals. Phase synchronization preprocessing eliminates the system error introduced by hardware channel differences, ensuring the accuracy of all subsequent feature indicators based on phasor analysis.
[0037] In one embodiment, feature extraction is performed on the three-phase current data to obtain the fundamental zero-sequence current component, the fundamental negative-sequence current component, and the third harmonic current component, respectively. This includes: transforming the three-phase current data using the symmetrical component method to extract the fundamental negative-sequence current component and the fundamental zero-sequence current component; and performing spectral analysis on the three-phase current data using the fast Fourier transform to separate the third harmonic current component.
[0038] At the detailed level of feature extraction, the time-domain sampled data that has been synchronously calibrated can first be windowed, for example, by using a Hanning window to reduce spectral leakage; then, a fast Fourier transform can be performed to convert the time-domain signal into a frequency-domain signal; in the frequency domain, the amplitude and phase of the fundamental component at a frequency of 50 Hz, as well as the amplitude and phase of the third harmonic component at a frequency of 150 Hz, can be directly extracted.
[0039] The fundamental component can be further decomposed into positive-sequence, negative-sequence, and zero-sequence components using a symmetrical component transformation matrix; among them, the fundamental zero-sequence current component is equal to the average of the positive-sequence, negative-sequence, and zero-sequence components.
[0040] The fundamental zero-sequence current is the main object of fault detection, but it also contains components generated by the imbalance of three-phase loads. The fundamental negative-sequence current mainly reflects the degree of asymmetry of the three-phase load, while the third harmonic current is mainly generated by single-phase rectifier loads such as variable frequency air conditioners. These third harmonics will be arithmetically superimposed on the neutral line in a three-phase four-wire system to form zero-sequence current interference.
[0041] In this way, through frequency domain analysis and symmetrical component transformation, the fundamental negative sequence current component can reflect the load characteristics, the third harmonic current component can reflect the nonlinear characteristics, and the fundamental zero sequence current component can reflect the zero sequence leakage level, enabling the intelligent terminal to grasp the load characteristics, nonlinear characteristics, and total zero sequence leakage level of the current distribution area.
[0042] In step S102, a reference signal is constructed based on the fundamental negative sequence current component and the third harmonic current component, and the dynamic coupling coefficient between the fundamental zero sequence current component and the reference signal is calculated. Based on the dynamic coupling coefficient, the fundamental zero sequence current component is orthogonally decoupled to obtain the residual current component, and the fault residual energy index of the residual current component is determined.
[0043] In one embodiment, constructing a reference signal based on the fundamental negative sequence current component and the third harmonic current component includes: linearly superimposing the instantaneous value of the fundamental negative sequence current component with the weighted instantaneous value of the third harmonic current component to obtain the reference signal.
[0044] In the specific operation of constructing the reference signal, the time-domain waveforms of the fundamental negative-sequence current component and the third harmonic current component can be taken. Considering that the superposition effect of the third harmonic on the neutral line is often stronger than the influence of the negative-sequence component, a preset empirical weighting coefficient can be introduced. and Using empirical weighting coefficients and The reference signal can be obtained by weighted summation of the time-domain waveforms of the fundamental negative sequence current component and the third harmonic current component. For example, the value is 1. For example, the value can be between 0.5 and 0.8, and the specific value can be determined according to the load type of the transformer area. This will not be elaborated further in the embodiments of this application.
[0045] Under normal operating conditions without grounding faults, the zero-sequence current on the neutral line mainly consists of two parts. One part is the unbalanced current caused by uneven power distribution of the three-phase load, which is strongly linearly correlated with the negative-sequence current. The other part is the odd-order harmonic current caused by nonlinear loads, with the third harmonic accounting for the largest proportion and superimposed on the neutral line.
[0046] The reference signal is a mathematical fit or prediction of the background zero-sequence current caused by load characteristics under normal operating conditions. If the actual detected zero-sequence current is highly similar to this reference signal, it means that the current zero-sequence current is mainly caused by normal load, and the distribution transformer in the area does not actually have a fault.
[0047] In this way, a dynamic background noise model is established by constructing a reference signal. The background noise model can be adjusted in real time as the load of the transformer area changes, such as the increase of negative sequence during the evening peak or the increase of harmonics, so that subsequent fault detection can adapt to environmental changes.
[0048] In one embodiment, the dynamic coupling coefficient is determined by the following formula: ,in, The dynamic coupling coefficient is used to characterize the projection intensity of the fundamental zero-sequence current component onto the reference signal direction. The number of sampling points contained in the sliding calculation window. The sampling point number within the sliding calculation window. Let be the instantaneous value of the fundamental zero-sequence current component at the k-th sampling point. The instantaneous value of the reference signal at the k-th sampling point. This is used to prevent positive numbers with a denominator of zero.
[0049] When calculating the dynamic coupling coefficient, for example, the number of sampling points within the sliding calculation window can be set. The value is 256, and real-time calculation is performed using an overlapping sliding method; Take a very small positive number, for example This is used to ensure the stability of numerical calculations.
[0050] By calculating the dynamic coupling coefficient, we can quantify what proportion of the current zero-sequence current can be explained by the reference signal. If the dynamic coupling coefficient is close to 1 or a certain stable constant, it means that the change of the fundamental zero-sequence current completely follows the change of the load, and no fault has occurred.
[0051] If a high-resistance ground fault occurs, the fault current will be superimposed on the fundamental zero-sequence current. Since the high-resistance fault current usually exhibits randomness and intermittency, the waveform characteristics of the high-resistance fault current are not related to the negative sequence or harmonic characteristics caused by the load. This will cause the calculated dynamic coupling coefficient to change abruptly or no longer fully explain the energy of the fundamental zero-sequence current.
[0052] In this way, the dynamic coupling coefficient can reflect the cause of zero-sequence current in real time, simplify the complex waveform relationship into a scalar coefficient, reduce the computational load of subsequent processing, and retain the most core correlation characteristics.
[0053] Figure 2This is a trend graph of the dynamic coupling coefficient changing over time in the embodiments of this application, such as... Figure 2 As shown, the dynamic coupling coefficient changes. During the normal phase, the value of the dynamic coupling coefficient remains stable at around 1, which indicates that the zero-sequence current energy at this time originates from the load imbalance characteristics.
[0054] like Figure 2 As shown, because the fault current after the fault occurs disrupts the original load characteristic balance, the measured current can no longer be linearly represented by the reference signal, causing the value of the dynamic coupling coefficient to fluctuate abruptly and rise to 1.1. The dynamic coupling coefficient can sensitively sense changes in waveform components.
[0055] In one embodiment, the fault residual energy index of the residual current component is determined by multiplying the dynamic coupling coefficient by the reference signal to obtain the background zero-sequence current estimation component; the background zero-sequence current estimation component characterizes the component of the fundamental zero-sequence current component related to the load imbalance characteristics; the difference between the fundamental zero-sequence current component and the background zero-sequence current estimation component is calculated to obtain the instantaneous value of the residual current component; the instantaneous value of the residual current component within the sliding calculation window is squared, and the arithmetic mean of the squares corresponding to different instantaneous values is used as the fault residual energy index; the fault residual energy index is used to characterize the independent abnormal energy in the signal that cannot be linearly characterized by the reference signal.
[0056] Background zero-sequence current estimation components equal residual current component equal Fault residual energy index equal .
[0057] The current of a high-resistance ground fault is often very small, for example, only 1A to 5A, while the normal zero-sequence current in the transformer area may be as high as tens of amperes. If we look directly at the change in the total amount, the fault signal will be completely submerged. By subtracting the relevant components, we can achieve soft measurement or adaptive filtering, which will filter out the zero-sequence current caused by load fluctuations that meets the expectations. The current obtained is highly likely to come from an external fault.
[0058] In this way, fault characteristics can be extracted. Even if the fault current is much smaller than the load unbalance current, as long as the waveform characteristics of the fault current are not completely consistent with the load characteristics, it can be significantly amplified in the residual energy index, thereby improving the sensitivity of detecting high-resistance grounding faults.
[0059] In step S103, the three-phase voltage data of the low-voltage side of the distribution transformer is obtained through the smart terminal, the zero-sequence voltage component is determined based on the three-phase voltage data, and the projection intensity index of the zero-sequence voltage component and the residual current component is calculated.
[0060] In one embodiment, calculating the projected intensity indices of the zero-sequence voltage component and the residual current component includes: ,in, For projection intensity index; The rated phase voltage constant of the transformer substation; This represents the number of sampling points contained in the sliding calculation window. The sampling point number within the sliding calculation window; The instantaneous value of the zero-sequence voltage component at the k-th sampling point. This represents the instantaneous value of the fundamental zero-sequence current component at the kth sampling point; For dynamic coupling coefficients; The instantaneous value of the reference signal at the kth sampling point; This represents the instantaneous value of the residual current component at the k-th sampling point. To take the absolute value.
[0061] For example, the rated phase voltage constant of the transformer area is 220V. In the process of calculating the projected intensity index, the zero-sequence voltage can be obtained first by three-phase voltage vector synthesis. When a single-phase ground fault occurs, the zero-sequence voltage will rise and the phase will be opposite to the fault phase voltage.
[0062] The projection intensity index is used to characterize the degree of in-phase relationship between the fault current and the zero-sequence voltage. There is a clear physical relationship between the zero-sequence voltage and the residual current generated by the fault. For resistive grounding faults, the two are basically in phase. For grounding through arc resistance, the two have slightly fluctuating phases but remain highly correlated. The summation process in the calculation formula of the projection intensity index can calculate the value of the cross-correlation function of the two signals at time zero, that is, calculate the effective power component.
[0063] While the energy index of residual current can detect anomalies, it may still be affected by certain special disturbances, such as the transient impact at the moment of starting a high-power motor. Although these disturbances produce residuals, they usually do not have a continuous and stable voltage-current phase relationship.
[0064] By introducing zero-sequence voltage as a reference vector and projecting the residual current onto the direction of zero-sequence voltage, it is possible to effectively distinguish between real grounding faults and spurious current disturbances. Only when the residual current has a significant projection onto the direction of zero-sequence voltage is it considered a valid fault signal. By introducing voltage dimension verification, the anti-interference capability and confidence level of fault judgment are improved.
[0065] In step S104, a fault assessment index is generated based on the fault residual energy index and the projection intensity index, and the fault assessment index is used to assess whether there is a high-resistance grounding fault in the transformer area.
[0066] In one embodiment, generating a fault assessment index based on a fault residual energy index and a projection intensity index includes: multiplying the fault residual energy index by a preset balance coefficient to obtain a weighted energy component, and using the sum of the projection intensity index and the weighted energy component as the fault assessment index; the balance coefficient is used to adjust the weight of the fault residual energy index in the fault assessment index.
[0067] The formula for calculating the fault assessment index can be expressed as follows: ,in, For projection intensity index, The preset balance coefficient, This refers to the residual energy index of the fault. The value can be preset according to the grounding method of the transformer area, which will not be described in detail in this embodiment.
[0068] The fault residual energy index is sensitive to amplitude, while the projection intensity index is sensitive to phase. Therefore, weighted fusion of the fault residual energy index and the projection intensity index can cover more types of fault scenarios.
[0069] For example, in the early stages of a fault, the current may be very small but the phase characteristics are obvious, and the projection intensity index plays a dominant role at this time. As the fault progresses, the current gradually increases and the waveform becomes distorted. At this time, the fault residual energy index rises rapidly, which can accelerate the tripping determination process.
[0070] In this way, by fusing multi-dimensional features, a highly robust fault criterion is constructed, which can ensure that high-resistance faults are not missed and transient disturbances are not misjudged.
[0071] In one embodiment, the determination of whether a transformer substation has a high-resistance grounding fault using a fault assessment index includes: determining that a transformer substation has a high-resistance grounding fault when the fault assessment index is greater than a preset action threshold and the duration exceeds a preset duration; or determining that a transformer substation does not have a high-resistance grounding fault when the fault assessment index is less than or equal to the preset action threshold, or the duration is less than or equal to the preset duration.
[0072] For example, the action threshold can be preset to 0.5 and the preset duration to 200ms. When the fault judgment index is determined to remain above 0.5 for 200ms, an alarm signal or trip command can be output.
[0073] There are a large number of transient processes in the power distribution network, such as lightning surges or load switching arcs. These transient processes may cause the fault assessment index to rise and then randomly drop to the normal level in a short period of time (e.g., 10ms-40ms). Without delayed confirmation, it is easy to cause frequent false trips. Moreover, the 200ms time window is sufficient to filter out the vast majority of transient interferences.
[0074] In this way, the dual constraints of threshold and time limit ensure the accuracy and selectivity of protection actions, avoid unnecessary power outages, and thus improve power supply reliability.
[0075] Figure 3 This is a schematic diagram comparing the fault assessment index and the effective value of the zero-sequence current in an embodiment of this application. Figure 3 The dashed line represents the monitoring indicator of the effective value of the zero-sequence current in existing technology. The effective value of the zero-sequence current can be determined using the root mean square of the effective value of the zero-sequence instantaneous component of the three-phase current; for example... Figure 3 As shown, since it is impossible to distinguish between load imbalance current and fault current, the effective value of zero-sequence current cannot accurately reflect the actual fault, which may lead to the risk of false alarm or missed alarm.
[0076] Figure 3 The solid line represents the fault assessment index calculated in this application, such as... Figure 3 As shown, the fault judgment index remains close to zero during normal operation and is unaffected by load fluctuations; it only exhibits a clear step action when a fault occurs. By comparison, it can be seen that the fault judgment index obtained by the embodiment of this application improves the signal-to-noise ratio, solves the problem that high-impedance grounding fault signals may be submerged under strong background noise, and effectively avoids possible false alarms or missed alarms.
[0077] In one embodiment, trend analysis can also be performed on the change data of the fault judgment index within a historical preset time period. If the trend analysis result indicates that the fault judgment index has not exceeded the preset action threshold within the historical preset time period but shows a monotonous upward trend, a warning message for indicating the deterioration of the insulation performance of the line can be output.
[0078] The smart terminal can store the average curve of the fault assessment index over the past 24 hours or week locally. The slope of the average curve can be calculated using a linear regression algorithm. If the instantaneous value of the fault assessment index is always lower than the action threshold, it means that no direct fault has occurred.
[0079] If the slope of the average curve of the fault assessment index is positive and greater than a certain warning slope threshold, or if the baseline value of the fault assessment index, i.e. the noise level, is gradually increasing, it means that the cable insulation layer is undergoing irreversible deterioration, such as water tree aging.
[0080] Traditional protection devices only activate after a fault occurs, but for power operation and maintenance, prevention is of greater value. By monitoring the slight upward trend of residual energy and projected intensity, early potential hazards that have not yet developed into complete breakdown faults can be detected. Operation and maintenance personnel can conduct inspections and insulation tests on specific line sections in advance based on the early warning information to ensure the power supply safety of the power supply lines.
[0081] Figure 4This is a schematic diagram illustrating the structure of a fault assessment system 1000 based on a smart terminal in a distribution area, according to an exemplary embodiment. (Refer to...) Figure 4 The fault assessment system 1000 based on the intelligent terminal of the distribution area includes a processor 1100 and a memory 1200. The memory 1200 stores computer program instructions. When the computer program instructions are executed by the processor 1100, they implement all or part of the steps of the fault assessment method based on the intelligent terminal of the distribution area in this application.
[0082] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only.
[0083] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.
Claims
1. A fault diagnosis method based on intelligent terminals in distribution transformer areas, characterized in that, include: The three-phase current data of the low-voltage side of the distribution transformer is obtained through a smart terminal, and the features of the three-phase current data are extracted to obtain the fundamental zero-sequence current component, the fundamental negative-sequence current component, and the third harmonic current component. A reference signal is constructed based on the fundamental negative-sequence current component and the third harmonic current component, and the dynamic coupling coefficient between the fundamental zero-sequence current component and the reference signal is calculated. , , Characterizes the projection intensity of the fundamental zero-sequence current component onto the reference signal direction. The number of sampling points contained in the sliding calculation window. The sampling point number within the sliding calculation window. The fundamental zero-sequence current component is in the first... The instantaneous value of each sampling point The reference signal at the 1st The instantaneous value of each sampling point. To prevent positive numbers with zero denominators, the fundamental zero-sequence current component is orthogonally decoupled based on the dynamic coupling coefficient to obtain the residual current component, and the fault residual energy index of the residual current component is determined. The three-phase voltage data of the low-voltage side of the distribution transformer is acquired through a smart terminal. The zero-sequence voltage component is determined based on the three-phase voltage data, and the projection intensity index of the zero-sequence voltage component and the residual current component is calculated. , ; The rated phase voltage constant of the transformer substation; The zero-sequence voltage component is in the first... The instantaneous value of each sampling point; For the residual current component in the first... The instantaneous value of each sampling point. To take the absolute value; the projected intensity index characterizes the degree of in-phase relationship between the fault current and the zero-sequence voltage; A fault assessment index is generated based on the fault residual energy index and the projected intensity index. The fault assessment index is then used to assess whether a high-resistance grounding fault exists in the transformer area.
2. The fault diagnosis method based on intelligent terminals in distribution areas according to claim 1, characterized in that, Feature extraction was performed on the three-phase current data to obtain the fundamental zero-sequence current component, the fundamental negative-sequence current component, and the third harmonic current component, including: The three-phase current data were transformed using the symmetrical component method to extract the fundamental negative sequence current component and the fundamental zero sequence current component; the three-phase current data were then subjected to spectrum analysis using the fast Fourier transform to separate the third harmonic current component.
3. The fault diagnosis method based on intelligent terminals in distribution areas according to claim 1, characterized in that, A reference signal is constructed based on the fundamental negative sequence current component and the third harmonic current component, including: The reference signal is obtained by linearly superimposing the instantaneous value of the fundamental negative sequence current component with the instantaneous value of the weighted third harmonic current component.
4. The fault diagnosis method based on intelligent terminals in distribution areas according to claim 1, characterized in that, The fault residual energy index of the residual current component is determined in the following way: The background zero-sequence current estimation component is obtained by multiplying the dynamic coupling coefficient by the reference signal; the background zero-sequence current estimation component characterizes the component of the fundamental zero-sequence current component that is related to the load imbalance characteristics. The difference between the fundamental zero-sequence current component and the estimated background zero-sequence current component is calculated to obtain the instantaneous value of the residual current component. The instantaneous values of the residual current components within the sliding calculation window are squared, and the arithmetic mean of the squares corresponding to different instantaneous values is used as the fault residual energy index. The fault residual energy index is used to characterize the independent anomalous energy in a signal that cannot be linearly characterized by a reference signal.
5. The fault diagnosis method based on a smart terminal in a distribution area according to claim 1, characterized in that, A fault assessment index is generated based on the fault residual energy index and the projected intensity index, including: The weighted energy component is obtained by multiplying the fault residual energy index by a preset balance coefficient. The sum of the projected intensity index and the weighted energy component is used as the fault judgment index. The balance coefficient is used to adjust the weight of the fault residual energy index in the fault judgment index.
6. The fault diagnosis method based on a smart terminal in a distribution area according to claim 1, characterized in that, Before collecting three-phase voltage and three-phase current data on the low-voltage side of the distribution transformer in real time via a smart terminal, the method further includes: The zero-crossing points of the three-phase voltage and three-phase current data are monitored in real time, and the time difference between the two zero-crossing points is calculated. The phase asynchrony deviation is determined based on the time difference. The sampling index position of the three-phase current data is dynamically adjusted according to the phase asynchrony deviation using the Lagrange interpolation algorithm to align the three-phase voltage and three-phase current data on the time axis.
7. The fault diagnosis method based on a smart terminal in a distribution area according to claim 1, characterized in that, The method further includes: Perform trend analysis on the changes in the fault assessment index within a preset historical time period. If the trend analysis results indicate that the fault assessment index has not exceeded the preset action threshold within the preset historical time period but shows a monotonous upward trend, output early warning information to indicate the deterioration of the insulation performance of the line.
8. A fault diagnosis system based on intelligent terminals in distribution substations, characterized in that, include: The processor and memory, wherein the memory stores computer program instructions, which, when executed by the processor, implement the fault assessment method based on a smart terminal in a distribution area according to any one of claims 1-7.