A method for detecting arc faults with resistance to load fluctuation based on ratio-RMS

Abstract

The present invention discloses a method for detecting fault arcs against load fluctuations based on ratio-root mean square, and builds a fault arc detection platform; the collected current signal is calculated by the ratio-root mean square method to obtain a characteristic quantity Z that characterizes the difference between each current cycle and the adjacent current cycle, and if the value of Z exceeds the threshold, it is judged that a fault arc has occurred; for current cycles with larger Z values, its power and power change rate are calculated, and whether a fault arc has occurred is judged according to the number of current cycles with non-zero power change rates. The present invention only needs to sample and detect several points at the same position in different cycles to obtain quantitative data, and does not require continuous monitoring of the current waveform or conversion of the time domain and other domains, which greatly reduces the amount of data calculation, thereby achieving the purpose of greatly reducing the fault judgment time; the present invention has a simple structure, a small amount of calculation, and is easy to implement in hardware, and can achieve rapid identification of fault arcs while taking into account accuracy.

Description

Technical Field

[0001] The invention relates to a fault arc detection method, in particular to a ratio-root mean square-based load fluctuation-resistant fault arc detection method. Background Art

[0002] Fault arc is one of the causes of electrical fires. Since the occurrence of series fault arc is not accompanied by the characteristics of low voltage and high current, it is difficult to be identified by conventional protection devices. However, if allowed to develop, the continuously burning fault arc will generate high temperatures of 2000 to 4000°C, which will ignite surrounding objects and cause fire.

[0003] At present, there are three common methods for fault arc identification. The first is based on the physical characteristics of the fault arc, such as infrared, arc light, noise, etc., but the arc position is difficult to determine, and it is difficult to install sensors according to the arc position. At the same time, the physical characteristics of the arc are easily disturbed by the surrounding environment; the second is based on the characteristics of the electrical signal when the fault arc occurs, such as the rising rate and average value of the current signal, zero-break phenomenon and high-order harmonics, etc., and the threshold is set in advance to determine whether a fault arc occurs. However, due to the randomness of the arc, setting a hard threshold often does not achieve the ideal effect; the third is a fault arc identification method based on deep learning, but generally only in a single load experiment, the accuracy is high but the calculation is large. According to the requirements of UL1699, the detection of the fault arc needs to be completed within the specified period, which places high demands on the hardware. Comprehensively speaking, the fault arc detection method based on electrical signals is currently a more effective method for fault arc identification in multi-load circuits.

[0004] However, the traditional method of judging arc faults based on electrical signals often relies on hard thresholds or empirical thresholds. When the circuit environment changes or load switching is involved, the empirical threshold may change. At present, there is a lack of public and authoritative arc fault sample data sets in the field of fault detection, and the empirical threshold can only be determined based on the actual situation of the circuit load. Therefore, the traditional method of judging arc faults based on electrical signals also has certain limitations. Summary of the invention

[0005] In view of the problems existing in the above-mentioned prior art, the present invention provides a load fluctuation fault arc detection method based on ratio-root mean square, which only requires sampling in the time domain and performs calculation and analysis on the collected data, thereby reducing the fault judgment time and lowering the hardware requirements. The algorithm execution efficiency is high and the accuracy is high.

[0006] To achieve the above object, the present invention provides the following technical solution: a method for detecting arc faults against load fluctuation based on ratio-root mean square, comprising the following steps;

[0007] S1: Build a fault arc detection platform; the detection platform includes a power supply, a fault arc generating device, a load, a current transformer, a data processing system, a voltmeter and an oscilloscope. The power supply, the current transformer, the fault arc generating device and the load are connected in series in sequence, the voltmeter is connected in parallel with the fault arc generating device, and the current transformer, the voltmeter and the oscilloscope are connected to the data processing system respectively;

[0008] S2: For the collected current signal, the ratio-RMS method is used to calculate in the data processing system to obtain the characteristic quantity Z that characterizes the difference between each current cycle and the adjacent current cycle. If the value of Z exceeds the set threshold Z b , it is judged that a fault arc has occurred;

[0009] S3: For current cycles with larger Z values, the power and power change rate δ of the collected current and voltage signals are calculated in the data processing system. Whether a fault arc occurs is determined based on the number of current cycles in which the power change rate δ is not zero. If the power change rate δ for three consecutive cycles is not zero and is random, it can be determined that a fault arc exists. Conversely, if the power change rate δ within two cycles is not zero, it is determined to be caused by load fluctuation.

[0010] Furthermore, the fault arc generating device in S1 adopts the Cassie arc model.

[0011] Furthermore, the method for determining whether a fault arc is generated in step S2 is as follows:

[0012] Detecting the current amplitudes of points collected within three adjacent current cycles, and collecting three points in each current cycle;

[0013] The current amplitude values ​​of the three points collected in each current cycle are added together to obtain the algebraic sum S of the three points in each current cycle. n-1 , S n , S n+1 , and then calculate the ratio of the next current cycle to the previous current cycle to get D n-1 and D n ;

[0014] All calculated ratios are used with the root mean square formula to obtain the calculation formula for the characteristic quantity Z:

[0015] The value of Z exceeds the threshold Z b After that, it is determined that a series fault arc has occurred.

[0016] Furthermore, the power change rate δ in step S3 is calculated as follows:

[0017] According to the active power calculation formula Calculate the change of average power in each current cycle. For the collected discrete power data P1, P2, P3, corresponding to time t1, t2, t3, the discrete power data P i Corresponding to time t i , where i = 1, 2, 3, ..., the rate of change of power is:

[0018] Furthermore, there are multiple loads, which are connected in parallel, and each load is connected to the detection platform via a switch.

[0019] Furthermore, the electrodes of the fault arc generating device are copper carbon electrodes.

[0020] Compared with the prior art, the present invention only needs to sample and detect several points at the same position in different cycles to obtain quantitative data. There is no need to continuously monitor the current waveform or convert the time domain and other domains, which greatly reduces the amount of data calculation, thereby achieving the purpose of greatly reducing the fault judgment time, and also reduces the requirements for hardware, greatly reducing costs; the present invention has a simple structure, a small amount of calculation, and is easy to implement in hardware. The calculation in the time domain can promptly obtain the time when the fault occurs, and can achieve rapid identification of fault arcs while taking into account accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a circuit diagram of the arc fault detection platform of the present invention;

[0022] Figure 2 This is a flow chart of arc fault detection in the present invention;

[0023] Figure 3 The characteristic value Z of the present invention changes with the current cycle curve;

[0024] Figure 4 A power change rate diagram in the arc fault circuit of the present invention;

[0025] In the figure: 1. power supply, 2. fault arc generating device, 3. load, 4. current transformer, 5. data processing system, 6. voltmeter, 7. oscilloscope. DETAILED DESCRIPTION

[0026] The present invention will be further described below in conjunction with the accompanying drawings.

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0028] like Figure 2 As shown, the present invention provides a technical solution, including the following steps.

[0029] S1: Build a fault arc detection platform based on the actual household electricity consumption in residential areas; Figure 1 As shown, the detection platform includes a power supply 1, a fault arc generating device 2, a load 3, a current transformer 4, a data processing system 5, a voltmeter 6 and an oscilloscope 7; the power supply 1 adopts a single-phase electricity with a voltage level of 220V and a frequency of 50Hz; the electrode of the fault arc generating device 2 adopts a copper carbon electrode to simulate the arc formed by carbonization and ensure stable arcing; the data processing system 5 uses a computer installed with MATLAB; the power supply 1, the current transformer 4, the fault arc generating device 2 and the load 3 are connected in series in sequence to simulate the series fault arc, the voltmeter 6 is connected in parallel with the fault arc generating device 2, the current transformer 4, the voltmeter 6 and the oscilloscope 7 are respectively connected to the data processing system 5; then the change of the fault current in the line is monitored by the oscilloscope 7; there are multiple loads 3, multiple loads 3 are connected in parallel, and each load 3 is connected to the detection platform through a switch, and the access of the load 3 is realized by opening and closing the switch, and a single load 3 can be connected, and multiple loads 3 can also be connected.

[0030] The fault arc generating device 2 adopts the Cassie arc model, which assumes that the shape of the arc is approximately cylindrical, the middle channel of the arc is composed of gas, and there is a uniform temperature distribution on its cross section. The temperature of the arc is constant in time and space. From the energy balance principle, it can be obtained that:

[0031]

[0032] In formula (1), dq / dt is the change of energy stored in the arc column per unit length, e×i h is the input power per unit length of arc, i h is the instantaneous value of the arc current, e is the electric field strength in the arc column, P loss is the power loss per unit arc length. Since the arc resistance is relatively small, converting the arc model into the form of conductance can be analyzed more intuitively:

[0033]

[0034] In formula (2), g is the arc conductance. If And the arc length is L, the second formula in equation (2) can be rewritten as:

[0035]

[0036] Arc voltage u h=L×e, arc column power loss Substituting it into formula (3) we can get:

[0037]

[0038] Therefore, the dynamic arc equation of Cassie is as follows:

[0039]

[0040] In formula (5), T is the time constant in the Cassie arc equation, u c is the arc voltage constant. According to formula (5), the Cassie arc model is established in MATLAB in the data processing system (5) for simulation calculation.

[0041] S2: For the current data collected by the current transformer 4, the ratio-RMS method is used to calculate in MATLAB to obtain the characteristic quantity Z that characterizes the difference between each current cycle and the adjacent current cycle. According to whether the characteristic quantity Z is a value close to 0, the similarity between this current cycle and the adjacent current cycle is judged. The closer the value of the characteristic quantity Z is to 0, the higher the similarity. Conversely, the larger the value of the characteristic quantity Z is, that is, the closer it is to 0, the lower the similarity is, that is, the greater the difference is. If the value of the characteristic quantity Z exceeds the set threshold Z b , it is judged that a fault arc has occurred; the specific judgment method is:

[0042] For the nth current cycle to be detected, the current amplitudes of three random points in the nth current cycle are collected, and the algebraic sum representing the current characteristics of the nth current cycle is obtained by adding them up and recorded as S n For the current cycle n-1 before and the current cycle n+1 after the selected nth current cycle, three points at the same time position as the nth current cycle are selected respectively, and the algebraic sum of the current amplitude is calculated to obtain S n-1 and S n+1 , S n-1 and S n+1 Respectively represent the previous current cycle n-1 and the next current cycle n+1; then calculate the ratio of the next current cycle to the previous current cycle, and get All calculated ratios are converted into The calculation formula for the value of the characteristic quantity Z is obtained, that is, Among S, D and Z, the values ​​of D and Z play a decisive role, which has nothing to do with the value of S, and has nothing to do with the selected time within a current cycle. At the same time, S takes the sum of three current amplitudes, which can reduce the error and amplify the difference between fault and non-fault signals, which is conducive to the identification of fault arcs. The value of D is related to the relative value of the line itself, and has nothing to do with the absolute value of the line amplitude, that is, it only pays attention to the difference between different current cycles of the line, especially amplifying the difference between the normal current cycle and the fault current cycle. The calculation method of the value of Z has the effect of reducing errors and avoiding false operations caused by excessive hardware sensitivity. After multiple experimental verifications, when there is no fault in the line, the value of Z is maintained near a low value. When there is a series fault arc, the value of Z increases significantly. The current signals of nine current cycles are selected for verification.

[0043]

[0044] Table 1

[0045] As shown in Table 1, each Z value contains information of three current cycles. From the perspective of several consecutive current cycles, the sudden change of the circuit fault current value occurs within one current cycle, after which the value of S tends to be stable, but this change is mapped to the value of Z, which is manifested as a significant increase in the value of 2 Z. Multiple experiments have shown that when the value of Z exceeds 5, the probability of a fault arc exceeding 80%, and it can be noted that when no series fault arc occurs and the arc burns stably, the value of Z is close to 0. For the identification of fault arcs by relying on time domain methods, although the time of fault occurrence can be obtained, the influence of time domain load fluctuations cannot be eliminated. The accuracy of judgment can be improved by the power change rate.

[0046] S3: The fault arc detection method using time domain is susceptible to load disturbance during the detection process. For current cycles with large Z values, the power method is used in MATLAB to further determine whether a fault arc has occurred. For the collected current and voltage signals, the power and power change rate δ are calculated in MATLAB. The power change rate δ is calculated as follows: According to the active power calculation formula Calculate the change of average power in each current cycle. For the collected discrete power data P1, P2, P3, corresponding to time t1, t2, t3, the discrete power data P i Corresponding to time t i , where i = 1, 2, 3, ..., the rate of change of power is: Whether a fault arc occurs is determined by the number of current cycles in which the power change rate δ is not zero. If the power change rate δ for three consecutive cycles is not zero and has a large randomness, it can be determined that a fault arc exists. Conversely, if the power change rate δ within two cycles is not zero, it can be determined that it is caused by load fluctuation.

[0047] Figure 3 It is a smooth curve formed based on the data in Table 1 above. It can be seen intuitively that the value of Z jumps sharply from the 4th current cycle to the 5th current cycle. Judging the series fault arc based on the value of Z can achieve good results.

[0048] Figure 4 It is the power change rate of the series fault arc branch. It can be seen that due to the randomness of the fault arc, compared with the load mutation, the power of the fault arc branch will be non-zero for many times, that is, sudden changes, and each situation is not exactly the same. Under normal circumstances, the power change caused by load change usually only has a non-zero power change rate in one place or a short time span. According to this feature, fault arc identification with resistance to load disturbance in the time domain can be realized.

[0049] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

[0050] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any slight modification, equivalent substitution and improvement made to the above embodiment based on the technical essence of the present invention should be included in the protection scope of the technical solution of the present invention.

Claims

1. A method for detecting arc faults against load fluctuations based on ratio-root mean square, characterized in that: The steps include: S1: constructing a fault arc detection platform; the detection platform comprises a power supply (1), a fault arc generating device (2), a load (3), a current transformer (4), a data processing system (5), a voltmeter (6) and an oscilloscope (7); the power supply (1), the current transformer (4), the fault arc generating device (2) and the load (3) are connected in series in sequence, the voltmeter (6) is connected in parallel with the fault arc generating device (2), and the current transformer (4), the voltmeter (6) and the oscilloscope (7) are respectively connected to the data processing system (5); S2: The collected current signal is calculated in the data processing system (5) by the ratio-root mean square method to obtain a characteristic quantity Z that characterizes the difference between each current cycle and the adjacent current cycle. If the value of Z exceeds the set threshold Z b , it is judged that a fault arc has occurred; The method for determining whether a fault arc is generated in step S2 is as follows: Detecting the current amplitudes of points collected within three adjacent current cycles, and collecting three points in each current cycle; The current amplitude values ​​of the three points collected in each current cycle are added together to obtain the algebraic sum S of the three points in each current cycle. n-1 , S n , S n+1 , and then calculate the ratio of the next current cycle to the previous current cycle to get D n-1 and D n ; All calculated ratios are used with the root mean square formula to obtain the calculation formula for the characteristic quantity Z: The value of Z exceeds the threshold Z b After that, it is determined that a series fault arc has occurred; S3: For the current cycle with a larger Z value, the power and power change rate δ of the collected current and voltage signals are calculated in the data processing system (5). Whether a fault arc occurs is determined based on the number of current cycles in which the power change rate δ is not zero. If the power change rate δ for three consecutive cycles is not zero and is random, it is determined that a fault arc exists. Conversely, if the power change rate δ within two cycles is not zero, it is determined that it is caused by load fluctuation.

2. The method for detecting arc faults against load fluctuations based on ratio-RMS according to claim 1, characterized in that: The fault arc generating device (2) in S1 adopts the Cassie arc model.

3. The method for detecting arc faults against load fluctuations based on ratio-RMS according to claim 1, characterized in that: The power change rate δ calculation method in step S3 is: According to the active power calculation formula Calculate the change of average power in each current cycle. For the collected discrete power data P1, P2, P3, corresponding to time t1, t2, t3, the discrete power data P i Corresponding to time t i , where i = 1, 2, 3, ..., the rate of change of power is:

4. The method for detecting arc faults against load fluctuations based on ratio-RMS according to claim 1, characterized in that: There are multiple loads (3), the multiple loads (3) are connected in parallel, and each load (3) is connected to the detection platform via a switch.

5. The method for detecting arc faults against load fluctuations based on ratio-RMS according to claim 1, characterized in that: The electrodes of the fault arc generating device (2) are copper-carbon electrodes.

Images