Arc fault detection method, device and system based on DC fault feature identification, and medium

Abstract

The invention provides an arc fault detection method, device and system based on direct current fault feature identification and a medium. The method comprises the following steps: monitoring a voltage distortion coefficient of a phase voltage by taking a cyclic wave as a unit; judging whether the voltage distortion coefficient of a certain phase exceeds an out-of-limit threshold value or not; if the distortion coefficient of a certain phase voltage exceeds an out-of-limit threshold value, filtering the phase voltage to obtain the filtered phase voltage; selecting data for identification according to the filtered phase voltage; and performing parameter identification on the selected identification data to obtain an arcing voltage direct current value, judging whether arcing voltage exists or not according to the arcing voltage direct current value and a preset judgment threshold value, and further judging whether an arc fault exists or not. According to the invention, the problems of high difficulty in series arc fault detection and low detection accuracy in the current low-voltage alternating-current distribution line can be solved.

Description

Technical Field

[0001] The present invention relates to the field of electrical engineering measurement, and in particular to an arc fault detection method, device, system and medium based on DC fault feature identification. Background Art

[0002] Arc faults are common faults in low-voltage distribution lines, including parallel and series types. Among them, parallel arc faults are caused by damaged insulation between conductors. Their essence is arc short circuit, so they usually have a large fault current. This fault is relatively easy to detect; series arc faults are caused by loose terminals, poor contact of wires, or conductor damage caused by construction. Compared with parallel arc faults, series arc faults have a smaller fault current, so the speed of causing electrical fires is slower than that of parallel arc faults. However, such faults usually have the characteristics of continuous occurrence, and only a voltage of 20V is required to make them reignite periodically. Since the arc is essentially an ionization discharge phenomenon in which voltage breaks through the air, the temperature at the arc fault point is extremely high. The arc center temperature of 0.5A current can reach 2000~3000℃. Therefore, if such faults persist, the insulation layer of the wire will be decomposed by heat and eventually cause a fire. If there are flammable materials around the arc fault point, its harm is more urgent.

[0003] For series arc faults, the fault point and the load are in series in the line, and the fault current is similar to the load current. For general loads, the fault current is slightly smaller than the load current; for constant power loads, the fault current is slightly larger than the load current, so it is difficult to use the current amplitude characteristics to detect the fault. At present, such faults can only be detected by identifying the waveform characteristics of the fault current. According to the physical mechanism of the arc, the arc current usually shows the current waveform characteristics of increased high-frequency noise, the appearance of current zero rest area, the increase of current rise rate, the increase of current harmonic content, and the asymmetry of positive and negative half-cycles. At present, arc fault detection products on the market, such as AFCI or AFD, use one or more of the above characteristics for detection. A large number of research results at home and abroad also use mathematical tools to extract one or more of the above characteristics in the time domain or transform domain for detection. In recent years, with the popularization of machine learning technology in various fields, a large number of black box model detection methods using neural networks to integrate the above multiple characteristics have emerged. The above methods are all based on fault current characteristics for detection. The key disadvantage of such methods is that a large number of nonlinear load currents may also have the zero-break, high frequency, harmonic and other characteristics based on which the linear load fault current characteristics are based. Therefore, whether it is based on multi-feature criterion design or neural network-based feature fusion, its actual effect is difficult to apply to a variety of loads. The high detection accuracy mentioned in a large number of research literatures is only applicable to the selected training loads. For unknown loads, the generalization ability of the above algorithms still needs to be further studied.

[0004] Compared with the current characteristics, the arc voltage at the fault point has the unique characteristic of being less affected by the load type. Therefore, it is easier to establish a unified fault judgment criterion by identifying the existence of the fault arc voltage characteristics. Since the load impedance is much larger than the line impedance of the loop, the arc voltage information at the fault point is mainly reflected in the voltage of the monitoring point downstream of the fault point. Therefore, there are also related studies on identifying the arc voltage characteristics at the fault point to characterize the existence of arc faults. For example, the literature [Miao Xiren, Guo Yinting, Tang Jincheng, et al. Load-end arc fault voltage detection and morphological wavelet identification] extracts fault arc features by wavelet decomposition of load-end voltage, and the literature [Zhang Liping, Miao Xiren, Shi Dunyi. Research on low-voltage arc fault identification method based on EMD and ELM] uses empirical mode decomposition to extract load-end fault features. These two methods essentially use the waveform distortion caused by the arc voltage starting and breaking phase angle of the fault point to detect faults. In practice, the arc starting and breaking phase angle features will be affected by the load type and line parameters, and there is a problem of difficulty in setting feature thresholds, and this method cannot distinguish the load-end voltage distortion caused by the nonlinear voltage drop of the line. The literature [Gao Hongxin, Guo Fengyi, Tang Aixia, et al. Using load-end voltage to predict series fault arcs] uses wavelet packets to decompose the load-end voltage and uses the prediction variance of the extreme learning machine to extract the fault arc features. The actual detection effect of this method depends on the number of types of training loads. Reference [Zhao Yuan, Zhang Guanying, Wang Yao, et al. Series fault arc detection method based on load-end voltage analysis] proposed an upstream line fault detection method based on the differential mean of adjacent voltage waveforms at the load end. This method assumes that adjacent voltage waveforms at the load end have large differences when a fault occurs. In actual faults, the fault voltage cycles at the load end may also have high similarities. Therefore, the accuracy of this method is poor.

[0005] The essence of various current detection methods is to detect the zero-time distortion, high frequency, harmonics and other characteristics of the fault current. In terms of detection effect, whether using a single feature or a multi-feature fusion method, it is difficult to apply to a variety of loads, especially a large number of nonlinear loads. The load current is likely to have the same or similar current waveform characteristics as a linear load with a series arc fault; in addition, some current fault characteristics (such as harmonic content) will also be affected by the load power; some loads also have the current characteristics of a series fault arc when starting, which can cause misjudgment; some loads such as arc welding machines, arcs generated when brushed motors are working, and arcs generated when plugging and unplugging sockets have fault characteristics similar to series fault arcs, which increases the difficulty of detecting series fault arcs; for weak series arc faults with very small arc currents, the fault characteristic quantity in the current waveform is often not obvious, making this type of fault more difficult to detect.

[0006] The various detection methods using load-end voltage listed above all utilize the fault distortion point generated by the arc extinction-arcing process of the arc voltage at the fault point to detect. In actual systems, high-order harmonics can also generate distortion points on the load-end voltage, and their fault characteristics may be similar to the fault distortion points generated by the arc voltage, so it is easy to cause misjudgment; in addition, the voltage change formed by the arc extinction-arcing process of the arc voltage is related to the load type and line parameters, so that the fault characteristics of the fault distortion point can be in multiple frequency bands. Therefore, these algorithms have the problem of difficulty in selecting characteristic frequency bands in application. Summary of the invention

[0007] In order to solve the problems of difficulty in detecting series arc faults and low detection accuracy in the current low-voltage AC distribution lines, the present invention provides an arc fault detection method, device, system and medium based on DC fault feature identification. Different from the traditional detection method based on fault current characteristics, the present invention uses voltage information to identify and detect series arc faults, and can be implemented based on the power Internet of Things technology and with smart meters and smart electricity terminals as application carriers.

[0008] An arc fault detection method based on DC fault feature identification comprises the following steps:

[0009] Monitor the voltage distortion coefficient of phase voltage in cycles;

[0010] Determine whether the voltage distortion coefficient of a certain phase exceeds the over-limit threshold;

[0011] If the distortion coefficient of a phase voltage exceeds the over-limit threshold, the phase voltage is filtered to obtain a filtered phase voltage;

[0012] Selecting identification data according to the filtered phase voltage;

[0013] Parameter identification is performed on the selected identification data to obtain the magnitude of the arc voltage DC value, and whether the arc voltage exists is determined based on the magnitude of the arc voltage DC value and a preset determination threshold, thereby determining whether there is an arc fault.

[0014] Furthermore, the filtering process for the phase voltage includes: using a digital low-pass filter to filter out high-frequency noise above 1kHz in the phase voltage; using a digital comb filter to filter out odd harmonics within 150Hz to 1kHz in the phase voltage.

[0015] Furthermore, the selecting of identification data according to the filtered phase voltage includes:

[0016] The phase voltage u′ after filtering o (n) Perform FFT transformation and calculate u′ o (n) initial phase angle

[0017] In half-wave units, according to the initial phase angle and u′ o (n) The number of sampling points to find u′ o (n) positive and negative half-wave peak points P1 and P2;

[0018] The peak points P1 and P2 of the positive and negative half-waves are taken as the data selection centers of the positive and negative half-waves, and φ L The window width is used to select the data for identifying the positive and negative half-waves respectively, wherein the data identification window width selected for each half-wave does not exceed 90°.

[0019] Furthermore, the step of performing parameter identification on the selected identification data to obtain the magnitude of the arcing voltage DC value includes:

[0020] Construct an overdetermined equation AX=Y to perform parameter identification on the identification data within the selected identification window:

[0021]

[0022] In the formula,

[0023] In A, sin(1), sin(2)…sin(N), cos(1), cos(2),…cos(N) are standard sine and cosine data of N sampling points per cycle; A and Y are known numbers, and X is an unknown number; U arc is the DC value of the arc voltage, α, β and U arc All are quantities to be identified; is the collected voltage signal, n1, n2…n N It is the sampling point number in the selected data window.

[0024] An arc fault detection device based on DC fault feature identification, comprising:

[0025] Phase voltage monitoring module, used to monitor the voltage distortion coefficient of phase voltage in cycles;

[0026] The voltage distortion coefficient over-limit judgment module is used to judge whether the voltage distortion coefficient of a certain phase exceeds the over-limit threshold;

[0027] The phase voltage filtering module is used to filter the phase voltage if the distortion coefficient of a phase voltage exceeds the over-limit threshold to obtain a filtered phase voltage;

[0028] An identification data selection module, used for selecting identification data according to the filtered phase voltage;

[0029] The arc fault judgment module is used to perform parameter identification on the selected identification data to obtain the magnitude of the arc voltage DC value, and judge whether there is an arc voltage according to the magnitude of the arc voltage DC value and a preset judgment threshold, and then judge whether there is an arc fault.

[0030] Furthermore, the phase voltage filtering module performs filtering processing on the phase voltage, including: using a digital low-pass filter to filter out high-frequency noise above 1kHz in the phase voltage; using a digital comb filter to filter out odd harmonics within 150Hz to 1kHz in the phase voltage.

[0031] Furthermore, the identification data selection module selects identification data according to the filtered phase voltage, specifically including:

[0032] The phase voltage u′ after filtering o (n) Perform FFT transformation and calculate u′ o (n) initial phase angle

[0033] In half-wave units, according to the initial phase angle and u′ o (n) The number of sampling points to find u′ o (n) positive and negative half-wave peak points P1 and P2;

[0034] The peak points P1 and P2 of the positive and negative half-waves are taken as the data selection centers of the positive and negative half-waves, and φ L The window width is used to select the data for identifying the positive and negative half-waves respectively, wherein the data identification window width selected for each half-wave does not exceed 90°.

[0035] Furthermore, the arc fault judgment module performs parameter identification on the selected identification data to obtain the magnitude of the arcing voltage DC value, specifically including:

[0036] Construct an overdetermined equation AX=Y to perform parameter identification on the identification data within the selected identification window:

[0037]

[0038] In the formula,

[0039] In A, sin(1), sin(2)…sin(N), cos(1), cos(2),…cos(N) are standard sine and cosine data of N sampling points per cycle; A and Y are known numbers, and X is an unknown number; U arc is the DC value of the arc voltage, α, β and U arc All are quantities to be identified; is the collected voltage signal, n1, n2…n NIt is the sampling point number in the selected data window.

[0040] An arc fault detection system based on DC fault feature identification includes: a computer-readable storage medium and a processor;

[0041] The computer-readable storage medium is used to store executable instructions;

[0042] The processor is used to read the executable instructions stored in the computer-readable storage medium to execute the arc fault detection method based on DC fault feature identification.

[0043] A non-transitory computer-readable storage medium stores a computer program, which, when executed by a processor, implements the arc fault detection method based on DC fault feature identification.

[0044] The present invention aims to identify the DC arcing characteristics of the arc voltage at the fault point, and discloses a method for identifying the upstream line series arc fault by using the downstream monitoring point of the fault point. The method avoids the problems of the existing voltage detection method that normal harmonic distortion cannot be distinguished and the fault characteristic frequency band is difficult to determine. The fault characteristics are more direct, the physical meaning is clear, and the applicability to a wide range of load types. It is a new detection method suitable for smart meters and smart distribution terminals in the context of the power Internet of Things. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is the voltage and current waveform diagram of different types of load faults;

[0046] Figure 2 It is the equivalent circuit diagram of series arc fault;

[0047] Figure 3 It is the voltage waveform of the monitoring point under the series arc fault;

[0048] Figure 4 It is the waveform diagram of the monitoring point Uo when there is no arc fault and when there is an arc fault in the line;

[0049] Figure 5 It is a flow chart of an arc fault detection method based on DC fault feature identification according to an embodiment of the present invention. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] The arc voltage is composed of the cathode voltage drop, anode voltage drop and arc column voltage drop of the arc electrode. The arc gap of the low-voltage system is very short, mainly short arc, so the arc column voltage drop can be ignored. It is considered that the arc voltage is basically equal to the sum of the anode and cathode voltage drops near the fault point. For air media, the near-electrode voltage drop of copper and carbon materials is about 10V, so the actual arc voltage drop is generally about 20V, and it has a DC characteristic. Figure 1 As shown, although the fault current characteristics of different loads vary greatly, their fault voltage characteristics show obvious square wave characteristics, and their arcing voltages are all approximately DC characteristics and are around 20V.

[0052] like Figure 2 As shown in the figure, for the monitoring point downstream of the fault, the voltage at the monitoring point is:

[0053] u o (t) = u s (t)-u Z (t)-u arc (t) (1)

[0054] Among them, u Z (t) represents the line impedance voltage drop.

[0055] According to formula (1), when a fault occurs, the voltage u at the downstream monitoring point is o (t) contains arc voltage u arc (t) is the complete information. From the local characteristics, the arc voltage u arc (t) will make u o (t) produces distortion and an overall drop in voltage, wherein the distortion is caused by the rising / falling edge of the arc voltage, and the overall drop in voltage is caused by the arcing voltage drop of the arc.

[0056] In practice, the line impedance voltage drop will also cause u o (t) produces local distortion, so it is easy to make misjudgment when only using voltage distortion for fault detection; on the contrary, the voltage drop caused by arc burning at its downstream monitoring point can be used as a reliable criterion for the existence of arc fault.

[0057] like Figure 3As shown in the figure, when a fault occurs, the monitoring point downstream of the fault point contains the fault arc voltage information. If there is a fault arc voltage, the fault arc voltage will cause the voltage at the monitoring point to be distorted and produce voltage drop information. Figure 3 For the local waveform between the distortion points shown, the local signal It can be expressed as:

[0058]

[0059] In formula (2), u s (t) is the power supply (transformer) voltage, and its waveform is approximately a standard sine wave; u Z (t) is the line voltage drop, which includes fundamental and harmonic voltage drops (mainly odd harmonics); u arc (t) represents the arc voltage; w(t) represents the window selection of data.

[0060] For the local waveform described in equation (2), if the harmonic voltage drop u in equation (2) is ignored, Z (t), then

[0061]

[0062] In formula (3), Indicates u o (t), ω1 is the fundamental wave angular frequency, which is a known number. is the initial phase angle of the sampling waveform, which is an unknown number; U arc is the DC value of the arc voltage, which is an unknown number (in the low voltage system, this value is around 20V).

[0063] In formula (3), there are and U arc Three unknowns. is the collected voltage signal, sin(ω1t) and cos(ω1t) are the standard sine and cosine signal sampling values, both of which are known quantities.

[0064] Formula (3) is obtained by ignoring the harmonic voltage drop u from formula (2): Z (t) We can get that for formula (3), α, β and U arc That is, the target identification parameter. Theoretically, only The parameter identification can be completed with 3 sampling points (3 unknowns, 3 equations). However, in practice, The harmonic voltage drop u Z (t) is inevitable. At this time, the error of identification using the data of three sampling points is large. In order to improve the accuracy of identification, it is necessary to construct an overdetermined equation with more equations than the number of unknowns to solve α, β and U. arc Parameters are used to ensure the accuracy of parameter identification (such as using 50 equations to solve 3 unknowns).

[0065] Therefore, the identification principle of the present invention is as follows: taking half-wave as a unit, through a large number of The sampling values ​​and the corresponding number of standard sine and cosine data are used to complete the construction of the overdetermined equation, and the least squares error is minimized to identify α, β and U arc If U arc If it is greater than the set threshold, it is considered that there is an arcing voltage and this half-wave is considered to be a fault half-wave, otherwise it is considered that this half-wave is not a fault half-wave.

[0066] In order to further improve the accuracy of identification, before using equation (3) to construct the overdetermined equation for identification, Filtering is performed to eliminate high-frequency noise and odd harmonic voltage drops to improve the accuracy of identification.

[0067] According to the above principle analysis, the first aspect of the present invention provides an arc fault detection method based on DC fault feature identification, such as Figure 5 As shown, the method comprises the following steps:

[0068] Step (1): Monitor the phase voltage u in cycles o The voltage distortion coefficient of (n), the voltage waveform distortion coefficient ρ is defined as follows:

[0069]

[0070] Where U N is the true effective value of the load end voltage cycle signal; U1 is the effective value of its fundamental component.

[0071] Step (2): Determine whether the voltage distortion coefficient of a certain phase exceeds the threshold value (such as using 3% as the threshold value), then preliminarily determine that a series arc fault occurs in the phase line, and then proceed to step (3); otherwise, return to step (1).

[0072] Step (3): Filter the phase voltage to improve the identification accuracy. o (n) after filtering is recorded as u′ o (n), the filtering process consists of two parts:

[0073] ① Use a digital low-pass filter to filter out high-frequency noise above 1kHz;

[0074] ② Use a digital comb filter to filter out odd harmonics within 150Hz~1kHz.

[0075] Step (4): According to the filtered phase voltage u′ o (n) Select the identification data. The specific steps are as follows:

[0076] ① For the phase voltage u′ after filtering o (n) Perform FFT transformation and calculate u′ o (n) initial phase angle

[0077] ② Take half-wave as unit, according to the initial phase angle and u′ o (n) The number of sampling points to find u′ o (n) positive and negative half-wave peak points P1 and P2;

[0078] ③ Take the positive and negative half-wave peak points P1 and P2 as the data selection center of the positive and negative half-wave, and use φ L is the window width, and the selection of data for identification of positive and negative half-waves is completed respectively. To ensure the validity of the data, the data identification window width selected for each half-wave shall not exceed 90°;

[0079] Step (5): Identify the selected identification data to determine whether there is a fault. The specific steps are as follows:

[0080] ① According to formula (3), the overdetermined equation AX=Y is constructed to perform parameter identification on the identification data in the selected identification window, that is,

[0081]

[0082] In formula (5),

[0083] In the formula, sin(1), sin(2)…sin(N), cos(1), cos(2),…cos(N) in A are the standard sine and cosine data of N sampling points per cycle (when using, you only need to select a continuous section of data from the sine and cosine standard data, and the initial phase of the standard data has no effect on the algorithm. Taking 200 points per cycle as an example, the standard sine data is sin18°, sin36°, sin54°…sin360°, and the standard cosine data is cos18°, cos36°, cos54°…cos360°). In the formula, A and Y are known numbers, X is an unknown number; α, β and U arc All are quantities to be identified; n1, n2…n N It is the sampling point number in the selected data window.

[0084] ② According to the identification result of formula (5), if U arc If it is greater than the preset judgment threshold, it is considered that there is an arcing voltage and an arc fault exists in this half-wave; otherwise, it is considered that there is no arc fault in this half-wave.

[0085] Figure 4 It is the waveform of the monitoring point Uo when there is no arc fault and when there is an arc fault in the line. Figure 4After identifying the waveform in (a), U arc The identification result is 1.5V (actually it should be 0V). Figure 4 U in (b) arc The identification result is 19.2V (the actual arcing voltage at the fault point is 20V). Therefore, if 15V is set as the judgment threshold of the identification result, it is possible to determine whether there is an arc fault.

[0086] The present invention has the following advantages:

[0087] (1) Using voltage signals to identify the presence of series arc faults in upstream lines has the advantage of not being affected by load type.

[0088] (2) Using an overdetermined equation composed of multiple sampling points to identify the arc burning DC quantity can reduce the impact of harmonic voltage drop on identification and improve the identification accuracy.

[0089] (3) For voltage signals, power quality problems will not introduce DC voltage components into AC voltage. The DC voltage components introduced by the arc at the fault point are unique characteristics caused by the occurrence of arc phenomenon. Therefore, the present invention is theoretically unique in identifying arc-type grounding faults.

[0090] Another aspect of the present invention provides an arc fault detection device based on DC fault feature identification, comprising:

[0091] Phase voltage monitoring module, used to monitor the voltage distortion coefficient of phase voltage in cycles;

[0092] The voltage distortion coefficient over-limit judgment module is used to judge whether the voltage distortion coefficient of a certain phase exceeds the over-limit threshold;

[0093] The phase voltage filtering module is used to filter the phase voltage if the distortion coefficient of a phase voltage exceeds the over-limit threshold to obtain a filtered phase voltage;

[0094] An identification data selection module, used for selecting identification data according to the filtered phase voltage;

[0095] The arc fault judgment module is used to perform parameter identification on the selected identification data to obtain the magnitude of the arc voltage DC value, and judge whether there is an arc voltage according to the magnitude of the arc voltage DC value and a preset judgment threshold, and then judge whether there is an arc fault.

[0096] Another aspect of the present invention provides an arc fault detection system based on DC fault feature identification, comprising: a computer-readable storage medium and a processor;

[0097] The computer-readable storage medium is used to store executable instructions;

[0098] The processor is used to read the executable instructions stored in the computer-readable storage medium to execute the arc fault detection method based on DC fault feature identification described in the first aspect.

[0099] Another aspect of the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the arc fault detection method based on DC fault feature identification described in the first aspect is implemented.

[0100] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0101] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0102] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0103] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0104] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention is described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation of the present invention can still be modified or replaced by equivalents, and any modification or equivalent replacement that does not deviate from the spirit and scope of the present invention should be included in the scope of protection of the claims of the present invention.

Claims

1. An arc fault detection method based on DC fault feature identification, characterized in that: The steps include: Monitor the voltage distortion coefficient of phase voltage in cycles; Determine whether the voltage distortion coefficient of a certain phase exceeds the over-limit threshold; If the distortion coefficient of a phase voltage exceeds the over-limit threshold, the phase voltage is filtered to obtain a filtered phase voltage; Selecting identification data according to the filtered phase voltage; Parameter identification is performed on the selected identification data to obtain the magnitude of the arc voltage DC value, and whether the arc voltage exists is determined based on the magnitude of the arc voltage DC value and a preset determination threshold, thereby determining whether there is an arc fault.

2. The arc fault detection method based on DC fault feature identification according to claim 1, characterized in that: The filtering process for the phase voltage includes: using a digital low-pass filter to filter out high-frequency noise above 1kHz in the phase voltage; using a digital comb filter to filter out odd harmonics within 150Hz to 1kHz in the phase voltage.

3. The arc fault detection method based on DC fault feature identification according to claim 1, characterized in that: The step of selecting identification data according to the filtered phase voltage includes: The phase voltage u′ after filtering o (n) Perform FFT transformation and calculate u′ o (n) initial phase angle In half-wave units, according to the initial phase angle and u′ o (n) The number of sampling points to find u′ o (n) positive and negative half-wave peak points P1 and P2; The peak points P1 and P2 of the positive and negative half-waves are taken as the data selection centers of the positive and negative half-waves, and φ L The window width is used to select the data for identifying the positive and negative half-waves respectively, wherein the data identification window width selected for each half-wave does not exceed 90°.

4. The arc fault detection method based on DC fault feature identification according to claim 3, characterized in that: The step of performing parameter identification on the selected identification data to obtain the magnitude of the arcing voltage DC value includes: Construct an overdetermined equation AX=Y to perform parameter identification on the identification data within the selected identification window: In the formula, In A, sin(1), sin(2)…sin(N), cos(1), cos(2),…cos(N) are standard sine and cosine data of N sampling points per cycle; A and Y are known numbers, and X is an unknown number; U arc is the DC value of the arc voltage, α, β and U arc All are quantities to be identified; is the collected voltage signal, n1, n2…n N It is the sampling point number in the selected data window.

5. An arc fault detection device based on DC fault feature identification, characterized in that: include: Phase voltage monitoring module, used to monitor the voltage distortion coefficient of phase voltage in cycles; The voltage distortion coefficient over-limit judgment module is used to judge whether the voltage distortion coefficient of a certain phase exceeds the over-limit threshold; The phase voltage filtering module is used to filter the phase voltage if the distortion coefficient of a phase voltage exceeds the over-limit threshold to obtain a filtered phase voltage; An identification data selection module, used for selecting identification data according to the filtered phase voltage; The arc fault judgment module is used to perform parameter identification on the selected identification data to obtain the magnitude of the arc voltage DC value, and judge whether there is an arc voltage according to the magnitude of the arc voltage DC value and a preset judgment threshold, and then judge whether there is an arc fault.

6. The arc fault detection device based on DC fault feature identification according to claim 5, characterized in that: The phase voltage filtering module performs filtering processing on the phase voltage, including: using a digital low-pass filter to filter out high-frequency noise above 1kHz in the phase voltage; using a digital comb filter to filter out odd harmonics within 150Hz to 1kHz in the phase voltage.

7. The arc fault detection device based on DC fault feature identification according to claim 5, characterized in that: The identification data selection module selects identification data according to the filtered phase voltage, specifically including: The phase voltage u′ after filtering o (n) Perform FFT transformation and calculate u′ o (n) initial phase angle In half-wave units, according to the initial phase angle and u′ o (n) The number of sampling points to find u′ o (n) positive and negative half-wave peak points P1 and P2; The peak points P1 and P2 of the positive and negative half-waves are taken as the data selection centers of the positive and negative half-waves, and φ L The window width is used to select the data for identifying the positive and negative half-waves respectively, wherein the data identification window width selected for each half-wave does not exceed 90°.

8. The arc fault detection device based on DC fault feature identification according to claim 7, characterized in that: The arc fault judgment module performs parameter identification on the selected identification data to obtain the magnitude of the arcing voltage DC value, specifically including: Construct an overdetermined equation AX=Y to perform parameter identification on the identification data within the selected identification window: In the formula, In A, sin(1), sin(2)…sin(N), cos(1), cos(2),…cos(N) are standard sine and cosine data of N sampling points per cycle; A and Y are known numbers, and X is an unknown number; U arc is the DC value of the arc voltage, α, β and U arc All are quantities to be identified; is the collected voltage signal, n1, n2…n N It is the sampling point number in the selected data window.

9. An arc fault detection system based on DC fault feature identification, comprising: A computer readable storage medium and a processor; The computer-readable storage medium is used to store executable instructions; The processor is used to read the executable instructions stored in the computer-readable storage medium to execute the arc fault detection method based on DC fault feature identification according to any one of claims 1 to 4.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the arc fault detection method based on DC fault feature identification according to any one of claims 1 to 4 is implemented.

Images