Intelligent protection device and method for turn-to-turn short circuit of dry-type air-core reactor

Abstract

This application relates to the field of power electronic component manufacturing technology, specifically to an intelligent relay protection device and method for inter-turn short circuits in dry-type air-core reactors. This device, used for fault detection and relay protection of dry-type air-core reactors, comprises a data acquisition unit, a host unit, a communication unit, a relay protection unit, and a power supply unit. The host unit receives and processes power data acquired by the data acquisition unit. By analyzing the variation characteristics and noise interference characteristics of the power data in various intervals of the dry-type air-core reactor, it constructs an adaptive window size and a polynomial order to filter and denoise the power data. The denoised power data is then used to identify inter-turn short circuit faults and provide relay protection for the dry-type air-core reactor. This application aims to avoid over-smoothing of fault characteristics in the power data by SG filtering, thereby improving the usability of voltage and current data for subsequent inter-turn short circuit fault identification.

Description

Technical Field

[0001] This application relates to the field of power electronic component manufacturing technology, specifically to an intelligent relay protection device and method for inter-turn short circuit of a dry-type air-core reactor. Background Technology

[0002] In recent years, with the development of power systems towards high voltage and long-distance transmission, dry-type air-core reactors have become widely used in power systems due to their advantages of simple structure, good linearity, and ease of maintenance. They are typically constructed by connecting multiple reactors in parallel to compensate for reactive power in power lines. The most common fault in dry-type air-core reactors is local short circuits caused by insulation failure between adjacent turns of the winding. In the initial stage of an inter-turn short circuit fault, the current flowing through the reactor changes very little, making it difficult to detect these initial minute changes and preventing timely triggering of overcurrent protection. Protection is only triggered when the fault gradually develops into a single-phase short circuit or a phase-to-ground short circuit, leading to serious accidents and affecting the normal operation of the power system.

[0003] In the early stages of inter-turn short-circuit faults in dry-type air-core reactors, the changes in current data characteristics are not obvious. Existing fault detection methods often rely on relay protection based on a single threshold. However, due to voltage fluctuations and abnormal interference in the power system, the fault-detectable features in the current data are obscured by these fluctuations and interference characteristics. Traditional inter-turn short-circuit fault detection methods struggle to detect early fault symptoms, making it difficult to accurately set appropriate thresholds for relay protection. To improve the usability of current data, denoising is an effective method. Traditional SG filtering denoising methods, based on fixed filter window lengths and multiple orders, filter and denoise the voltage and current data of the reactor. This makes it difficult to identify early fault data characteristics and interference features, easily leading to the smooth filtering out of usable fault features, thus affecting subsequent identification of inter-turn short-circuit faults in the reactor. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides an intelligent relay protection device and method for inter-turn short circuit of dry-type air-core reactors. The specific technical solution adopted is as follows:

[0005] In a first aspect, embodiments of this application provide a method for intelligent relay protection of inter-turn short circuits in dry-type air-core reactors, the method comprising the following steps:

[0006] Step 1: Collect and preprocess the power data of the dry-type air-core reactor;

[0007] Step 2: Use variational mode decomposition to decompose each type of power data into multiple IMF components and divide the data into intervals;

[0008] Step 3: Utilize the difference between the absolute value of the mean amplitude of power data and the mean of the absolute values ​​of power data amplitude, as well as the difference in amplitude variation of power data between adjacent peaks, to sequentially analyze the symmetry and peak variability of each IMF component of the power data in each interval, and integrate them into the first feature index;

[0009] Step 4: Adaptively adjust the window size of the SG filter in each interval using the first feature index, and adaptively adjust the polynomial order of the SG filter in each interval using the spectral entropy characteristics of each IMF component of the power data in each interval; then use the improved SG filter to filter each IMF component of the power data in each interval, and reconstruct the power data by denoising each IMF component.

[0010] Step 5: Analyze the power factor change of the filtered and reconstructed power data to determine whether there is an inter-turn short circuit fault; if a fault is determined to have occurred, trigger the relay protection action.

[0011] Preferably, the power data includes voltage and current data of the dry-type air-core reactor.

[0012] Preferably, the preprocessing includes time series reconstruction and alignment preprocessing of the voltage and current data respectively; the reconstruction uses sinc interpolation to reconstruct the voltage and current data into a continuous time series, and the alignment uses a dynamic time planning algorithm to ensure the synchronization and consistency of the voltage and current data.

[0013] Preferably, the method for fusing the first feature index is as follows: ;in It is an exponential function with the natural constant e as the base. The first representing the current data The IMF component in the first Symmetry of each interval The first representing the current data The IMF component in the first Peak variability in each interval.

[0014] Preferably, the symmetry is calculated by the ratio of the absolute value of the mean current amplitude of each IMF component of the current data at all times in each interval to the mean of the absolute values ​​of the current amplitude at all times.

[0015] Preferably, the peak variability is determined by the mean of the absolute differences between all two adjacent local extrema in each interval for each IMF component of the current data.

[0016] Preferably, the window size of the SG filter is adaptively adjusted by linear interpolation between a preset maximum and minimum value based on a first feature index.

[0017] Preferably, the polynomial order is set hierarchically based on the normalized value of the spectral entropy characteristics of all intervals; the higher the spectral entropy, the lower the polynomial order.

[0018] Preferably, the method for obtaining the spectral entropy is as follows: using Hilbert transform, the instantaneous frequency of each IMF component of the power data at each sampling point in each interval is obtained; and the spectral entropy is calculated based on the instantaneous frequency distribution of all sampling points in each interval.

[0019] Secondly, this application embodiment also provides an intelligent relay protection device for inter-turn short circuit of dry-type air-core reactor. The device is used for fault detection and relay protection of dry-type air-core reactor and consists of a data acquisition unit, a host unit, a communication unit, a relay protection unit, and a power supply unit.

[0020] The data acquisition unit includes a voltage and current data acquisition module, which acquires the voltage and current data of the reactor through a voltage transformer and a current transformer.

[0021] The host unit uses an industrial-grade embedded processor to receive and process the data collected by the data acquisition unit. It uses the inter-turn short-circuit intelligent relay protection method for dry-type air-core reactors described above to perform real-time analysis of data changes in dry-type air-core reactors and output control commands for the dry-type air-core reactors.

[0022] The communication unit is responsible for uploading the collected data and test results to the power management center via a wireless network.

[0023] The relay protection unit provides relay protection for the dry-type air-core reactor through the control commands output by the main unit, so as to avoid major accidents caused by inter-turn short circuit faults.

[0024] The power supply unit is a 220V power supply that provides separate power to the relay protection device.

[0025] This application has at least the following beneficial effects:

[0026] This application addresses the problem that traditional SG filtering often smooths out slowly changing fault features and noise interference during denoising due to the minimal changes in power data caused by early inter-turn short-circuit faults in dry-type air-core reactors, making it difficult to retain useful fault features. This application constructs an adaptive filtering window and polynomial order by analyzing the waveform and frequency domain characteristics of power data in various intervals. This effectively eliminates noise interference in high-noise intervals, preventing noise interference from being identified as short-circuit fault features. Simultaneously, it performs a more refined fit within the actual short-circuit fault data interval, maximizing the retention of power data variation characteristics of inter-turn short-circuit faults. This avoids over-smoothing of fault features in the power data by SG filtering, improving the usability of voltage and current data for subsequent inter-turn short-circuit fault identification. Attached Figure Description

[0027] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 This is a schematic diagram of the built-in unit of the intelligent relay protection device for inter-turn short circuit of a dry-type air-core reactor according to this application.

[0029] Figure 2 This is a flowchart illustrating the steps of an intelligent relay protection method for inter-turn short circuits of a dry-type air-core reactor according to this application. Detailed Implementation

[0030] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an intelligent relay protection device and method for inter-turn short circuits of a dry-type air-core reactor proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

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

[0032] The following description, in conjunction with the accompanying drawings, details the specific scheme of the intelligent relay protection device and method for inter-turn short circuit of a dry-type air-core reactor provided in this application.

[0033] This application proposes an intelligent relay protection device for inter-turn short circuits in dry-type air-core reactors, hereinafter referred to as the relay protection device. The relay protection device is used for fault detection and relay protection of dry-type air-core reactors, and consists of a data acquisition unit, a host unit, a communication unit, a relay protection unit, and a power supply unit.

[0034] The data acquisition unit includes a voltage and current data acquisition module, which acquires the voltage and current data of the reactor through a voltage transformer and a current transformer.

[0035] The host unit uses an industrial-grade embedded processor to receive and process the data collected by the data acquisition unit. It can perform synchronous sampling, data calculation and analysis, and other operations. It uses the dry-type air-core reactor inter-turn short-circuit fault detection method designed in this application to analyze the data changes of the dry-type air-core reactor in real time and output control commands for the dry-type air-core reactor.

[0036] The communication unit is responsible for uploading the collected data and test results to the power management center via a wireless network.

[0037] The relay protection unit provides relay protection for the dry-type air-core reactor through the control commands output by the main unit, thereby preventing major accidents caused by inter-turn short circuit faults.

[0038] The power supply unit is a 220V power supply that provides separate power to the relay protection device.

[0039] The enclosure of the relay protection device shall be at least IP54 or IP65 rated and have electromagnetic interference immunity to prevent electromagnetic shocks in high-voltage power lines from damaging the normal operation of the relay protection device.

[0040] The device control process is as follows: power-on of the device → multi-channel data sampling → data preprocessing → data feature analysis → fault determination → relay protection of the device → data transmission, realizing full-process monitoring of the operation of the dry-type air-core reactor and inter-turn short-circuit fault identification.

[0041] See attached Figure 1 This embodiment provides a schematic diagram of the built-in unit of the relay protection device.

[0042] Another embodiment of this application provides a method for intelligent relay protection against inter-turn short circuits in dry-type air-core reactors. Please refer to [link to relevant documentation]. Figure 2 The method includes the following steps:

[0043] Step 1: Collect and preprocess the power data of the dry-type air-core reactor.

[0044] The power data in the dry-type air-core reactor is obtained through the data acquisition unit in the intelligent relay protection device for inter-turn short circuit of dry-type air-core reactor (hereinafter referred to as relay protection device). The power data includes the voltage and current data of the dry-type air-core reactor.

[0045] The voltage and current sampling frequency is 10kHz, and the data analysis time window is 5 minutes. The sampling frequency and analysis time window size can be set by the implementer according to the implementation scenario without special restrictions. However, to ensure the reliability of the acquired signals, the sampling frequency setting should satisfy the Nyquist sampling theorem.

[0046] Then, the voltage and current data are reconstructed into continuous time series signals using the sinc interpolation method. The sinc interpolation method is an existing technology, and its specific process will not be described in detail here.

[0047] The obtained time series signals are aligned using a dynamic time programming algorithm to ensure data synchronization and consistency. The dynamic time programming algorithm is an existing technology, and its specific process will not be elaborated further.

[0048] Step 2: By analyzing the variation characteristics and noise interference characteristics of the power data in the dry-type air-core reactor in various intervals, an adaptive window size and polynomial order are constructed to filter and denoise the power data.

[0049] In the detection of inter-turn short-circuit faults in dry-type air-core reactors, the early-stage power data changes are relatively small. Furthermore, noise interference exists in high-voltage power lines and reactors during operation. For example, the hollow structure of a dry-type air-core reactor results in significant magnetic flux leakage. This leakage causes random electromagnetic coupling interference to be induced in conductors or sensors around the reactor. This noise is superimposed on the current or voltage waveform, masking the subtle power data changes in the early inter-turn short-circuit fault. Traditional SG filtering uses a fixed window length and polynomial order to smooth the signal data. If the window length is too large or the polynomial order is too low, it will smooth out the current changes caused by the early inter-turn short-circuit fault, losing the accuracy of the data. Conversely, if the window length is too small or the polynomial order is too high, interference from random noise cannot be effectively filtered out.

[0050] To address the aforementioned issues, this application constructs an adaptive window size and polynomial order to filter and denoise the voltage and current data in various intervals by analyzing the variation characteristics and noise interference characteristics of voltage and current data in dry-type air-core reactors. This effectively eliminates noise interference in high-noise intervals, preventing noise interference from being identified as short-circuit fault features. At the same time, it preserves the power data variation characteristics of inter-turn short-circuit faults to the greatest extent in the actual short-circuit fault intervals, avoiding excessive smoothing of fault features by SG filtering, which would compromise its usability for subsequent inter-turn short-circuit fault identification.

[0051] First, the voltage and current data obtained from the reactor are expressed as follows: ,in Representing the first Voltage and current data at any given time. The time order is on the order of 1ms. This represents the total number of sampling points within the entire data analysis time window.

[0052] Considering the complex multi-frequency interference caused by impulse interference in power system lines and random interference from electromagnetic coupling in reactors, it is difficult to accurately analyze the voltage and current variation characteristics and noise interference characteristics throughout the entire power transmission signal. This application employs variational mode decomposition to decompose the voltage and current data separately, obtaining... One IMF component, The size can be set by the implementer according to the implementation scenario, without special restrictions. In this embodiment... The size is set to 6. Complex power data signals are decomposed into multiple IMF components at different frequency bands, facilitating the analysis of interference noise characteristics and power data variation characteristics at each component level.

[0053] Taking current data denoising as an example, The Each IMF component is represented as ,Will According to length Divided into equal-length intervals, a total of Each interval Its size can be set by the implementer according to the implementation scenario, without special restrictions. In this embodiment, The size is 2 seconds. The first IMF component In each interval The current data at time t is represented as .

[0054] In a normal reactor, the current flowing through it remains stable. When an inter-turn short-circuit fault occurs, the local insulation layer between the turns is broken down, leading to an increase in local circulating current. This causes a slight distortion in the current waveform and enhances the asymmetry of the current data. Electromagnetic interference and impacts in the power system lines can also cause random fluctuations in the current waveform. Therefore, the asymmetry of the current waveform can be used as one of the indicators for evaluating data characteristics.

[0055] Among them, the current data of the first The IMF component in the first Symmetry of each interval It can be represented as:

[0056]

[0057] in, The first representing the current data The IMF component in the first The absolute value of the mean of the current amplitude at all times within each interval. The first representing the current data The IMF component in the first The mean of the absolute values ​​of the current amplitude at all times within a given interval. The value indicates the degree of bias in the current data waveform. A larger value indicates that the waveform is biased to one side. The meaning is used to indicate the use of Normalization is performed.

[0058] The closer the value is to 0, the better the symmetry of the IMF component of the current data in that interval, and the less fault characteristics or noise interference characteristics there are in that interval. When the data is close to zero, the waveform remains a standard sine wave; however, when there is significant noise interference in this range, the waveform characteristics exhibit irregular changes, with the current IMF component waveform shifting towards one side. The larger the value between (0,1), the greater the proportion of bias energy caused by the change in data waveform due to inter-turn short circuit fault or the influence of noise, resulting in a worse symmetry of the current IMF component waveform.

[0059] Furthermore, since both inter-turn short-circuit faults and noise interference disrupt the symmetry of the waveform characteristics of the current IMF component data, it is difficult to accurately determine whether the abnormal data characteristics in this interval are noise interference based solely on waveform symmetry indicators, making it impossible to set an appropriate window size for filtering. Considering that inter-turn short-circuit faults generate harmonic interference in the power data, and that the current data gradually increases to the point of causing a major accident with overcurrent, this phenomenon causes the reactor current data to show a trend of increasing, while electromagnetic coupling interference and other noise effects present random noise impacts in the power data. Therefore, this application further determines whether there is noise interference in the abnormal waveform interval data based on the variation and fluctuation of the peak values ​​of each waveform in the current IMF component.

[0060] For any one Any sampling point in the interval If the current amplitude at the point satisfies Then record the sampling point. For a local extremum point, obtain The number of all local extreme points in the interval is denoted as . indivual.

[0061] The first current data The IMF component in the first Peak variability of each interval It can be represented as:

[0062]

[0063] in , These represent the first and second data points of the current data. The IMF component in the first In the interval, the first The, the The difference between the current amplitudes at local extreme points represents the magnitude of the change between two consecutive local extreme points of the current. Finally, the mean of the absolute differences between all adjacent local extreme points in the entire interval represents the variability of the current peak value in the interval of the current IMF component data.

[0064] The larger, the more it means The greater the variability of the peak current within the interval, the stronger the random noise impact from electromagnetic coupling, and vice versa. The smaller, the more... The lower the variability of the peak current in the interval, the more likely the change in the peak current is a slow trend caused by the inter-turn short-circuit fault.

[0065] In summary, the interval The first characteristic index can be expressed as:

[0066]

[0067] in It is an exponential function with the natural constant e as the base, and its purpose is to... Normalize to the order of [0,1] for easier summation. Perform calculations. This is how it's constructed. and and Proportional relationship, when The worse the waveform symmetry of the current IMF component in the interval and the greater the variability of the current peak value, The larger, the more it means The interval contains a large amount of noise interference, and vice versa. The interval represents normal current or inter-turn short-circuit fault characteristics.

[0068] exist The adaptive filtering window size of the SG filter over the interval can be expressed as:

[0069]

[0070] in, , are the maximum and minimum values ​​of the SG filter window, respectively, and their sizes are odd numbers within the range of values. These values ​​can be set by the implementer according to the implementation scenario without special restrictions. In this embodiment... , The sizes are set to 63 and 7 respectively.

[0071] when When there is a large amount of noise interference in an interval, in order to better smooth out and eliminate the random noise interference signal in that interval, The larger the value, the larger the final filter window size; conversely, when the value is smaller... When the interval mainly contains normal current data characteristics and inter-turn short-circuit fault characteristics, The smaller the value, the smaller the filter window will be, in order to avoid the SG filter from smoothly filtering out the slowly changing current characteristics and to retain as much of the current change characteristics caused by short-circuit faults as possible.

[0072] Furthermore, in the early inter-turn short-circuit faults of dry-type air-core reactors, the spectral entropy of the signal reflects the complexity of the spectral distribution of the current data. A range with a smaller spectral entropy indicates that the spectral capability of the signal is mostly concentrated on a few frequency components, and the change of the signal data is relatively smooth and stable. Conversely, when the spectral entropy is large, it indicates that the signal contains more random noise interference and the spectral energy distribution is discrete. If a higher polynomial order is used for fitting and filtering, the filtered signal will retain more details and it will be difficult to filter out noise signals.

[0073] The polynomial order in each interval is adaptively adjusted based on the spectral entropy characteristics of the current IMF component in each interval.

[0074] right Perform a Hilbert transform to obtain The instantaneous frequency of each sampling point in the interval is transformed by Hilbert transformation using existing technology; the specific process will not be elaborated here.

[0075] Furthermore, according to Calculate the spectral entropy of the instantaneous frequency distribution of all sampling points in the interval. Known The method for calculating the spectral entropy based on the instantaneous frequency of the sampling points is an existing technique, and its specific process will not be elaborated here. (Calculation...) Spectral entropy of all intervals .right Perform maximum and minimum value normalization to obtain the normalized value. Spectral entropy of the interval .

[0076] exist The order of the adaptive polynomial of the SG filter over the interval can be expressed as:

[0077]

[0078] in The initial value of the polynomial order of the preset SG filter can be set by the implementer according to the implementation scenario without special restrictions. In this embodiment, the value is 2.

[0079] This is how it is constructed. According to The spectral entropy within the interval is adaptively adjusted. When the spectral entropy is large, it indicates that the signal frequency distribution in that interval is more complex. To avoid the SG filter retaining too many detailed features during binomial fitting and failing to effectively filter out noise signals, a smaller polynomial order is used for smoothing. Conversely, when the spectral entropy is small, it indicates that the signal frequency distribution in that interval is more concentrated. To retain as many detailed current change features as possible, a larger polynomial order is used to fit the data in that interval more finely, avoiding the filtering out of inter-turn short-circuit fault features.

[0080] Finally, based on the adaptive filtering windows and adaptive polynomial orders obtained in the above steps, the IMF components of the current data are filtered and denoised in the SG filter. The denoised IMF components are then reconstructed to obtain the denoised current data. The IMF component reconstruction method is existing technology, and its specific process will not be described in detail.

[0081] Repeat the above steps to denoise the voltage data, and finally obtain the denoised voltage and current data.

[0082] Step 3: Based on the denoised power data, perform inter-turn short-circuit fault identification and relay protection for the dry-type air-core reactor.

[0083] The main unit of the relay protection device calculates the denoised voltage and current data to determine whether an inter-turn short circuit fault has occurred.

[0084] The method for determining inter-turn short-circuit faults refers to the relay protection operation method described in patent "CN109473949B Relay Protection Method and Device for Inter-turn Short-circuit Faults of Dry-type Air-core Parallel Reactor Groups". When the power factor change in the dry-type air-core reactor exceeds a set threshold, an inter-turn short-circuit fault is determined to have occurred in the dry-type air-core reactor. The set threshold is the minimum value of the power factor change. The specific calculation method for the power factor of the dry-type air-core reactor and the method for setting the minimum value of the power factor change are described in that patent, and the specific process will not be repeated here.

[0085] If an inter-turn short circuit fault occurs, the relay protection device controls the relay protection unit through the main unit to provide relay protection for the reactor.

[0086] The various embodiments in this application are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0087] It should be noted that, unless otherwise specified and limited, terms such as “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a circuit structure, article, or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such article or device. Without further limitations, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the article or device that includes said element. Furthermore, the term “and / or” as used herein includes any and all combinations of one or more of the associated listed items.

[0088] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention 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 invented in this application.

[0089] 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 method for intelligent relay protection of a dry-type air-core reactor against inter-turn short circuit, characterized in that, The method includes the following steps: Collect and preprocess the power data of the dry-type air-core reactor; Variational mode decomposition is used to decompose each type of power data into multiple IMF components and divide them into intervals; By utilizing the difference between the absolute value of the mean amplitude of power data and the mean of the absolute values ​​of power data amplitude, as well as the difference in amplitude variation of power data between adjacent peaks, the symmetry and peak variability of each IMF component of power data in each interval are analyzed sequentially and fused into the first feature index. The window size of the SG filter in each interval is adaptively adjusted using the first feature index, and the polynomial order of the SG filter in each interval is adaptively adjusted using the spectral entropy feature of each IMF component of the power data in each interval. Then, the improved SG filter is used to filter each IMF component of the power data in each interval, and the power data is reconstructed from the denoised IMF components. The power factor change is analyzed on the filtered and reconstructed power data to determine whether there is an inter-turn short circuit fault; if a fault is determined to have occurred, the relay protection is triggered. The fusion method for the first feature index is as follows: ;in It is an exponential function with the natural constant e as the base. The first representing the current data The IMF component in the first Symmetry of each interval The first representing the current data The IMF component in the first Peak variability in each interval.

2. The intelligent relay protection method for inter-turn short circuit of a dry-type air-core reactor as described in claim 1, characterized in that, The power data includes the voltage and current data of the dry-type air-core reactor.

3. A method for turn-to-turn short circuit intelligent protection of a dry-type air-core reactor as claimed in claim 2, characterized in that, The preprocessing includes time series reconstruction and alignment preprocessing of voltage and current data respectively; the reconstruction uses sinc interpolation to reconstruct voltage and current data into a continuous time series, and the alignment uses a dynamic time planning algorithm to ensure the synchronization and consistency of voltage and current data.

4. The method of claim 1, wherein the method is characterized by: The symmetry is calculated by the ratio of the absolute value of the mean current amplitude of each IMF component of the current data at all times in each interval to the mean of the absolute values ​​of the current amplitude at all times.

5. The method of claim 1, wherein the method further comprises: determining the fault type of the dry-type air-core reactor based on the fault current and the fault voltage. The peak variability is determined by the mean of the absolute differences between all two adjacent local extrema in each interval for each IMF component of the current data.

6. The method of claim 1, wherein the method further comprises: determining the fault type of the dry-type air-core reactor based on the comparison of the first and second voltage signals. The window size of the SG filter is adaptively adjusted by linear interpolation between a preset maximum and minimum value based on a first feature index.

7. The method of claim 1, wherein the method further comprises: determining the fault type of the dry-type air-core reactor based on the comparison of the measured current and the calculated current. The polynomial order is set hierarchically based on the normalized value of the spectral entropy characteristics of all intervals; the higher the spectral entropy, the lower the polynomial order.

8. A method for turn-to-turn short circuit intelligent protection of a dry-type air-core reactor as claimed in claim 7, characterized in that, The method for obtaining the spectral entropy is as follows: using Hilbert transform, the instantaneous frequency of each IMF component of the power data at each sampling point in each interval is obtained; based on the instantaneous frequency distribution of all sampling points in each interval, its spectral entropy is calculated.

9. An intelligent protection device for detecting a turn-to-turn short circuit in a dry-type air-core reactor, characterized in that, This device is used for fault detection and relay protection of dry-type air-core reactors, and consists of a data acquisition unit, a host unit, a communication unit, a relay protection unit, and a power supply unit. The data acquisition unit includes a voltage and current data acquisition module, which acquires the voltage and current data of the reactor through a voltage transformer and a current transformer. The host unit uses an industrial-grade embedded processor to receive and process the data collected by the data acquisition unit. It uses the inter-turn short-circuit intelligent relay protection method for dry-type air-core reactors as described in any one of claims 1-8 to perform real-time analysis of data changes in dry-type air-core reactors and output control commands for the dry-type air-core reactors. The communication unit is responsible for uploading the collected data and test results to the power management center via a wireless network. The relay protection unit provides relay protection for the dry-type air-core reactor through the control commands output by the main unit, so as to avoid major accidents caused by inter-turn short circuit faults. The power supply unit is a 220V power supply that provides separate power to the relay protection device.

Images