Fault arc recognition method, system and device

Abstract

The present invention relates to the technical field of electrical fire monitoring. An embodiment of the present invention discloses a method, a system and a device for identifying a fault arc. The method for identifying a fault arc includes: obtaining the ADC raw data of an arc signal, where the ADC raw data includes current raw data; respectively processing the current raw data through FFT (Fast Fourier Transform) and extracting the second-order wavelet coefficients of the current to obtain a current comparison result; and judging whether a fault arc occurs according to the current comparison result. By using a multi-domain identification method based on FFT, the present invention extracts the characteristics of the distortion of the arc current from the perspective of a multi-dimensional domain, and infers whether a real fault arc is generated or it is due to the line conduction interference of a mixed load, thereby greatly improving the accuracy of arc detection.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical fire monitoring, and in particular, to a method, a system and a device for identifying a fault arc. Background Art

[0002] With the rapid development of China's economy and the continuous improvement of people's living standards, all kinds of electrical products and equipment are continuously spread throughout large cities and rural areas. As time goes by, these electrical products have gradually entered the "aging" stage, and there are potential electrical fire safety hazards. In recent years, fire accidents caused by electrical reasons have occurred frequently in China, and a considerable part of them are caused by arcs generated by electrical line failures. Generally, fault arcs can be divided into series-parallel carbonized path arcs, parallel metallic contact arcs and point contact arcs. Among them, the series-parallel carbonized path arc (hereinafter referred to as the serial arc) has extremely strong concealment. Because of its large impedance, the characteristics of virtual short and virtual open result in extremely small current generated when the arc occurs, which is very easy to be submerged by the current of other loads on the electrical line. The factors of superimposed mixed loads (such as inductive loads, inferior LED lighting rectifiers, etc.) make the arc on the electrical line not only have concealment, but also carry extremely high randomness, which in turn leads to inaccurate detection and misreporting and missed reporting of fault arcs. When the serial arc occurs continuously, the local temperature at the center of the arc area on the line can reach as high as 4000 degrees Celsius. The excessively high temperature for a long time can cause the insulation layer on the wire to gradually carbonize and fall off, and the tiny carbonized particles will float in the wire groove or the air. When the sparks on the arc splash a large number of tiny particles in the air and reach the ignition point, it will ignite the line and the surrounding flammable substances and even cause an explosion.

[0003] The identification of serial arcs is difficult and the possible hazards are also great. In view of the problems of low recognition rate and high misjudgment rate of the existing fault arc technology, it is necessary to further study the fault arc identification technology. Summary of the Invention

[0004] Embodiments of the present invention provide a method, a system and a device for identifying a fault arc to solve the problems of low recognition rate and high misjudgment rate of the existing fault arc technology.

[0005] On the one hand, the present invention provides a method for identifying a fault arc, including:

[0006] Obtaining the ADC raw data of the arc signal, where the ADC raw data includes the current raw data;

[0007] Processing the current raw data respectively through FFT (Fast Fourier Transform) and extracting the second-order wavelet coefficients of the current to obtain a current comparison result;

[0008] Judging whether a fault arc occurs according to the current comparison result.

[0009] Further, the current comparison result includes a first current comparison result and a second current comparison result. The steps of obtaining the comparison result by respectively processing the original current data through FFT (Fast Fourier Transform) and extracting the second-order wavelet coefficients of the current include:

[0010] Processing the original current data through FFT to obtain spectral data;

[0011] Analyzing the spectral data to obtain the first current comparison result;

[0012] Processing the original current data through extracting the second-order wavelet coefficients of the current to obtain first-period data and second-period data;

[0013] Comparing the first-period data and the second-period data to obtain the second current comparison result.

[0014] Further, the steps of processing the original current data through FFT to obtain spectral data include:

[0015] Processing the original current data through FFT to obtain frequency-domain data;

[0016] Filtering the frequencies below 2380 Hz in the frequency-domain data to obtain spectral data;

[0017] The steps of analyzing the spectral data to obtain the first current comparison result include:

[0018] Analyzing whether there is a high-frequency signal with less than or equal to 9 half-cycles or greater than or equal to 14 half-cycles within 100 half-cycles in 1 second in the spectral data. If so, the first current comparison result is negative.

[0019] Further, the steps of processing the original current data through extracting the second-order wavelet coefficients of the current to obtain first-period data and second-period data include:

[0020] Performing first-order wavelet filtering on the original current data to obtain a first-order wavelet coefficient table of the current;

[0021] Performing second-order wavelet filtering on the first-order wavelet coefficient table of the current and intercepting to obtain first-period data and second-period data. The first-period data and the second-period data are successively the data of the first two periods of the first-order wavelet coefficient table of the current, and the value range is (0 - 2π);

[0022] The steps of comparing the first-period data and the second-period data to obtain the second current comparison result include:

[0023] Subtract the data of the first cycle point by point from the data of the second cycle, and determine whether a jump occurs. If so, the second current comparison result is negative.

[0024] Further, the ADC raw data further includes voltage raw data. The step of determining whether a faulty arc occurs according to the current comparison result further includes:

[0025] When both the first current comparison result and the second current comparison result show that they are in the critical region, analyze the voltage zero-point data of the voltage raw data to obtain a voltage comparison result;

[0026] Determine whether a faulty arc occurs according to the current comparison result and the voltage comparison result.

[0027] Further, before the step of respectively processing the current raw data by FFT fast Fourier series transform and current second-order wavelet coefficient extraction to obtain a comparison result, it further includes:

[0028] Obtain a current value according to the current raw data, and obtain a voltage value according to the voltage raw data;

[0029] Judge whether the current value is greater than the rated current range, or the voltage value is in the faulty voltage range. If so, proceed to the next step.

[0030] Further, after the step of determining whether a faulty arc occurs according to the current comparison result and the voltage comparison result, it further includes:

[0031] Adjust the preset parameters according to the current comparison result and the voltage comparison result.

[0032] In order to solve the technical problems of low recognition rate and high false judgment rate in the existing faulty arc technology, the purpose of the present invention is to provide a faulty arc recognition method, including: obtaining the ADC raw data of the arc signal, where the ADC raw data includes current raw data; respectively processing the current raw data by FFT fast Fourier series transform and current second-order wavelet coefficient extraction to obtain a current comparison result; judging whether a faulty arc occurs according to the current comparison result.

[0033] Compared with the prior art, the present invention has the following advantages:

[0034] Through a multi-domain recognition method based on FFT fast Fourier series transform, extract the characteristics of arc current distortion from the perspective of a multi-dimensional domain, and infer whether a real faulty arc has occurred or it is due to line conduction interference of a mixed load, thereby greatly improving the accuracy of arc detection.

[0035] Another aspect of the present invention discloses a corresponding faulty arc recognition system, including:

[0036] An acquisition module, configured to acquire the ADC raw data of the arc signal, where the ADC raw data includes the current raw data;

[0037] A circuit comparison module, configured to process the current raw data respectively through FFT (Fast Fourier Transform) and extraction of the second-order wavelet coefficients of the current to obtain a current comparison result;

[0038] A judgment module, configured to judge whether a faulty arc occurs according to the current comparison result.

[0039] Further, it further includes:

[0040] A self-learning module, configured to adjust the preset parameters according to the current comparison result and the voltage comparison result.

[0041] Another aspect of the present invention discloses a corresponding arc recognition device, including:

[0042] A common-mode filtering anti-interference circuit, a current transformer, an operational amplifier circuit, a voltage sampling circuit, and the above-mentioned faulty arc recognition system. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required to be used in the description of the embodiments of the present invention. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to these drawings.

[0044] Figure 1 It is a flowchart of a fault recognition method according to an embodiment of the present invention.

[0045] Figure 2 It is a flowchart of a fault recognition method according to another embodiment of the present invention.

[0046] Figure 3 It is a true waveform diagram when a faulty arc occurs.

[0047] Figure 4 It is a schematic diagram of the test data waveform of simulating 50Hz superimposed with 2.38KHz.

[0048] Figure 5 It is a schematic diagram of the test data waveform of disassembling 50Hz and 2.38KHz through FFT time-domain to frequency-domain conversion.

[0049] Figure 6 It is a schematic diagram of a fault recognition system according to an embodiment of the present invention.

[0050] Figure 7 It is a schematic diagram of a fault recognition device according to an embodiment of the present invention. Specific Embodiments

[0051] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0052] Please refer to Figure 1 , a fault arc recognition method according to an embodiment of the present invention includes the following steps:

[0053] Step 101, obtain the ADC raw data of the arc signal, and the ADC raw data includes the current raw data.

[0054] In this embodiment, the product is installed at the forefront of the single load or hybrid load to be monitored to collect the ADC raw data of the arc signal. When a fault arc occurs, fault arc waveforms such as Figure 3 and Figure 4 will be generated. Taking the waveform of 50Hz superimposed with 2380Hz as an example, the fault current recognition system (which can be an MCU processing unit) collects the fault arc signal through a current transformer, and after passing through an operational amplifier circuit, a set of ADC raw data is obtained. Figure 4

[0055] Step 102, process the current raw data through FFT (Fast Fourier Transform) and extraction of the second-order wavelet coefficients of the current respectively to obtain a current comparison result.

[0056] Figure 5 In this step, on the one hand, substitute the ADC raw data into the FFT formula, and what is obtained is a set of data converted from the time domain to the frequency domain, such as Figure 5 the data contains information of 50Hz and 2380Hz. According to the national standard "GB14287.4 - 2014", the duration of the fault arc shall not exceed 0.42 milliseconds (i.e., it should be greater than 2.38KHz). We filter out the waveforms with frequencies below 2380Hz, that is, 50Hz is not used as the basis for identifying the fault arc. In addition, the national standard "GB14287.4 - 2014" also stipulates that if there are less than or equal to 9 half-cycles of fault arcs within one second, the detection is valid; if there are greater than or equal to 14 half-cycles of fault arcs within one second, the device issues an alarm. Then we need to calculate whether there are high-frequency signals with less than or equal to 9 half-cycles or greater than or equal to 14 half-cycles among 100 half-cycles within 1 second. If so, this is used as one of the main bases for identifying and determining the fault arc.

[0057] On the other hand, the above-mentioned ADC raw data is subjected to first-order wavelet filtering through the current first-order wavelet formula. The filtered data will become quite smooth, which helps to filter out most of the fundamental waves and highlight the high-frequency ramps. Then, second-order current wavelet filtering is performed. The data of two periods before and after the first-order wavelet is intercepted, and the value range is (0 to 2π). The data of the previous period (0 to π) is subtracted from the data of the latter period (π to 2π) point by point. We score according to half-cycle data as a record target. If the subtracted data result is greater than 0.05In√2, it is recorded as 1 point, otherwise it is recorded as 0 point; then there are 2 scoring periods within two periods (0 to 2π), and the value range of each scoring period is {00, 01, 10, 11}. If the scoring results of the previous and latter periods are inconsistent (a jump occurs), it proves the occurrence of a faulty arc. The more the number of jumps, the greater the possibility that the faulty arc persists. We use the scoring result characteristics of the second-order current wavelet coefficient as one of the factors for faulty arc determination.

[0058] Before this step, other current judgments can also be added, such as judging whether the current value is within the rated current range. S5. According to the national standard of "GB14287.4-2014", the current of the faulty arc should exceed 5% of the rated current value. The true current value within the same period is calculated using the ADC raw data, and then the faulty arc is restricted so that the determination system for faulty arc recognition can only be turned on when the detected current value is greater than 5% of the rated current, in order to avoid misjudgment and save computing power.

[0059] Step 103, determine whether a faulty arc occurs according to the current comparison result.

[0060] In the embodiment of the present invention, it is determined whether a faulty arc occurs according to the comparison result of the current. Among them, the recognition effect can be greatly improved only by analyzing the harmonic components of the FFT fast Fourier series transform and the characteristics of the second-order current wavelet coefficient. Of course, other judgment adjustments can also be added.

[0061] For example, as an implementation manner of an embodiment of the present invention, the three main conditions of the above current (the harmonic components of the FFT fast Fourier series transform, the characteristics of the second-order wavelet coefficients of the current, and the limitation of the rated current), plus the two additional conditions of the voltage (voltage zero-crossing tracking and the limitation of the voltage value range) can be used to screen and filter the information of the fault arc waveform layer by layer. The specific implementation process is as follows: First, convert the plotted points within one cycle into current and voltage values, and use the limitation of the rated current and the limitation of the voltage value range as necessary conditions. Then, calculate the weights of each condition. The weight of the current harmonic components based on the FFT fast Fourier series transform is 0.6, the weight of the characteristics of the second-order wavelet coefficients of the current is 0.3, and the weight of the voltage zero-crossing tracking is 0.1. If multiple harmonic frequencies are calculated through the FFT time-domain to frequency-domain conversion, plus the mutation of the second-order wavelet coefficients of the current, it can basically be determined that there is a fault arc; if the above two current weights reach the critical value, the weight of the voltage zero-crossing tracking plays a key role. Finally, the deep self-learning analysis system improves the recognition degree and comprehensive determination ability of the fault arc signal through preset parameters, self-learning microscopic control parameters, and weight ratios, so as to solve the problems of false alarms and missed alarms of the fault arc.

[0062] Among them, adding the judgment of two additional voltage signals will make the result more accurate. According to the national standard "GB14287.4-2014", the equipment should be able to work normally in an environment of AC187V to 242V. Therefore, the original ADC data is sampled through the voltage sampling circuit to calculate the real voltage value, and it is judged whether the voltage value is within AC187V to 242V. Because in most cases, once the load exceeds the working voltage range, it cannot work normally, and at this time, alarms of types such as undervoltage and overvoltage can be triggered. Normally, there are only two zero points (0, π) within a complete cycle. If there are more than two points, it can be considered whether there is a fault arc.

[0063] The fault arc recognition method of the embodiment of the present invention provides a more accurate judgment basis for identifying whether there is a fault arc, which helps to reduce the false alarm rate of the fault arc and improve the recognition rate of the arc.

[0064] Based on the improvement of the above embodiment of the fault arc recognition method, please refer to Figure 2 , the fault arc recognition method of another embodiment of the present invention includes:

[0065] Step 201, obtain the ADC raw data of the arc signal, and the ADC raw data includes current raw data and voltage raw data.

[0066] In the embodiment of the present invention, in order to further improve the accuracy of the determination system by judging the current characteristics, the judgment of the voltage characteristics can be added.

[0067] Step 202: Obtain the current value based on the original current data and obtain the voltage value based on the original voltage data.

[0068] Among them, the method for determining whether the current value is within the rated current range has been explained in the previous embodiment. The specific method for obtaining the current value from the original current data can be that, according to the national standard "GB14287.4 - 2014", the current of a fault arc should exceed 5% of the rated current value. First, retrieve the data point sequence of the original current data (the array sequence of current sampling) and substitute it into the following formula:

[0069]

[0070] where 2 q is the sampling length, and Xi is the data of the sampling point.

[0071] The specific method for obtaining the voltage value from the original voltage data can be that, according to the national standard "GB14287.4 - 2014", the equipment should be able to work normally in an environment of AC187V - 242V. The DC component of the AC voltage at n points is collected through a voltage sampling circuit and substituted into formula S16, and the obtained results are successively substituted into the following formulas:

[0072]

[0073] where Xn is the collected data point, n is the sampling length, and Wn is the weight value, which is taken as 1 here. After several cycles of conversion, the voltage value after weighted average filtering can be obtained, and the longer the cycle, the smoother the data.

[0074] Step 203: Determine whether the current value is greater than the rated current range or the voltage value is within the fault voltage range.

[0075] In this step, to determine whether the current value is greater than the rated current range, when I > 0.05In, we allow the fault arc determination system to be turned on; otherwise, the fault arc recognition process ends, so as to reduce misjudgment and save computing power. To determine whether the voltage value is within the rated voltage range, if the voltage value obtained in step 202 is within AC187V - 242V, the system will turn on the determination of the fault arc. When the current value is within the rated current range (such as I < 0.05In) and the voltage value is outside the fault voltage range, the recognition process can end.

[0076] Step 204: Process the original current data through FFT (Fast Fourier Transform) to obtain spectral data.

[0077] In the embodiment of the present invention, the current comparison result includes the first current comparison result and the second current comparison result.

[0078] Among them, step 204 specifically includes:

[0079] 1. Process the original current data through the FFT (Fast Fourier Transform) to obtain frequency-domain data.

[0080] In the embodiment of the present invention, harmonic components are obtained according to the FFT. Specifically, the current pulses in the original circuit data are sampled and plotted according to the sampling frequency and assembled into an original current data array sequence. The number of plotted points is N (i.e., the signal length. To achieve the optimal computing power of the system, generally N is taken as 2 to the power of q, where q≥8, and the larger the value of q, the higher the accuracy). Then, the FFT is performed on this array sequence, and the transformation formula is:

[0081]

[0082] where k takes values from 0, 1,..., N / 2 - 1.

[0083] Substitute the current array sequence into the transformation formula. Through the transformation from the time domain to the frequency domain, a set of frequency-domain data can be obtained, corresponding to the amplitude of the time-domain waveform frequency, so as to obtain the frequency magnitudes of various harmonics within a complete cycle.

[0084] 2. Filter the frequencies below 2380 Hz in the frequency-domain data to obtain spectral data.

[0085] Normally, the frequency-domain data will obtain the spectrum of 50 Hz. As Figure 3 shown, when a faulty arc occurs, it often contains a large number of multiple harmonics. Through time-frequency conversion, the frequencies of various sub-harmonics within a complete cycle can be obtained. According to the national standard "GB14287.4 - 2014", if there are less than or equal to 9 half-cycles of faulty arcs within one second, the detection is effective; if there are greater than or equal to 14 half-cycles of faulty arcs within one second, the device issues an alarm; and the duration of the above arcs shall not exceed 0.42 milliseconds (i.e., less than 2.38 KHz). Then, the multiple harmonics obtained in the frequency domain must be the 2nd and above, and the harmonics greater than 2.38 KHz are counted as the triggering conditions for identifying the arc. Therefore, the frequency-domain data of 2380 Hz and above should be taken as the spectral data for comparison.

[0086] Step 205: Analyze the spectral data to obtain the first current comparison result.

[0087] Among them, step 205 may specifically include:

[0088] Analyze whether there are high-frequency signals with less than or equal to 9 half-cycles or greater than or equal to 14 half-cycles within 100 half-cycles within one second in the spectral data. If so, the first current comparison result is negative.

[0089] Step 207: Extract the original current data through the second-order wavelet coefficients of the current to obtain the first-cycle data and the second-cycle data.

[0090] Among them, step 206 specifically includes:

[0091] 1. Perform first-order wavelet filtering on the original current data to obtain the first-order wavelet coefficient table of the current.

[0092] Still take the above-mentioned plotted data of the original current data (the array sequence of current sampling), and a set of data of the first-order wavelet coefficients of the current is obtained through the first-order wavelet calculation formula. The calculation formula is:

[0093]

[0094] Since x is the number of points sampled within one cycle, Ua represents the maximum sampling voltage, and Da is the maximum resolution of the ADC. According to the national standard "GB14287.4 - 2014", the current of the faulty arc should exceed 5% of the rated current value; therefore, we multiply 5% of the equipment rated current In by √2 and then divide by CT. CT is the current transformer ratio of 2000, and the DC component of the voltage at the peak value of the rated current is obtained and used as the quotient factor for each sampling point. By calculating the periodic set through the periodic function f(x), the first-order wavelet coefficient table of the current can be obtained.

[0095] 2. Perform second-order wavelet filtering on the first-order wavelet coefficient table of the current, and intercept to obtain the first-cycle data and the second-cycle data. The first-cycle data and the second-cycle data are the data of the first two cycles of the first-order wavelet coefficient table of the current in sequence, and the value range is (0 to 2π).

[0096] The coefficient table obtained after the first-order wavelet filtering of the current will be much smoother than the array sequence of the original current data. On this basis, the extraction of the second-order wavelet coefficients of the current is performed; within the first-order wavelet coefficient table of the current, the data of the first two cycles are intercepted, that is, the first-cycle data and the second-cycle data, and the value range is (0 to 2π).

[0097] Step 207: Compare the first-cycle data and the second-cycle data to obtain the second current comparison result.

[0098] Among them, step 207 can specifically include:

[0099] Subtract the data of the first-cycle data point by point from the data of the second-cycle data, and judge whether there is a jump. If so, the second current comparison result is negative.

[0100] Subtract the data of the first cycle, i.e., the data of the previous cycle (0 to π), point by point from the data of the second cycle, i.e., the data of the next cycle (π to 2π). Score based on half-cycle data as a recording target. If the subtracted data result is greater than 0.05In√2, record 1 point; otherwise, record 0 point. Then, within two cycles (0 to 2π), there are 2 scoring cycles in total, and the value range of each scoring cycle is {00, 01, 10, 11}. If the scoring results of the front and back two cycles are inconsistent (jumping), it proves the occurrence of a faulty arc. The more the number of jumps, the greater the possibility of the continuous occurrence of the faulty arc. At this time, the second current ratio result is negative. We use the scoring result feature of the second-order wavelet coefficient of this current as one of the factors for judging the faulty arc.

[0101] Step 208: When both the first current ratio result and the second current ratio result show that they are in the critical region, analyze the voltage zero-point data of the original voltage data to obtain the voltage ratio result.

[0102] In the embodiment of the present invention, when both the first current ratio result and the second current ratio result show that they are in the critical region, to more accurately judge the occurrence of the faulty arc and prevent the fire caused by the faulty arc, the voltage situation can be further analyzed to perform voltage zero-point tracking. Under normal circumstances, there are exactly two zero points within a complete cycle, which are point 0 and π respectively. Search for the zero point according to the voltage DC component of the AC voltage sampled by the voltage sampling circuit at n points. If there are more than two zero points within one cycle, we can consider whether there is a faulty arc.

[0103] Step 210: Judge whether a faulty arc occurs according to the current ratio result and the voltage ratio result.

[0104] In the embodiment of the present invention, through the above three main conditions of the current (the harmonic component of the FFT fast Fourier series transform, the feature of the second-order wavelet coefficient of the current, and the limitation of the rated current), plus the two additional conditions of the voltage (voltage zero-point tracking and the limitation of the voltage value range), the information of the faulty arc waveform can be screened and filtered layer by layer. The specific implementation process is as follows: First, convert the plotted points within one cycle into current and voltage values, and use the limitation of the rated current and the limitation of the voltage value range as necessary conditions. Then, calculate the weights of each condition. The weight of the current harmonic component based on the FFT fast Fourier series transform is 0.6, the weight of the feature of the second-order wavelet coefficient of the current is 0.3, and the weight of the voltage zero-point tracking is 0.1. If multiple harmonic frequencies are calculated through the FFT time-domain to frequency-domain conversion, plus the mutation of the second-order wavelet coefficient of the current, we can basically determine the existence of a faulty arc; if the weights of the above two currents reach the critical value, the weight of the voltage zero-point tracking plays a key role.

[0105] Step 211: Adjust the preset parameters according to the current comparison result and the voltage comparison result.

[0106] In this step, the recognition degree and comprehensive judgment ability of the fault arc signal can also be improved by preset parameters, self-learning microscopic control parameters and weight ratios, so as to solve the problems of false alarms and missed alarms of fault arcs.

[0107] Compared with the prior art, the present invention has the following advantages:

[0108] Through the multi-domain recognition method based on the FFT (Fast Fourier Transform) fast Fourier series transform, the characteristics of the arc current distortion are extracted from the perspective of the multi-dimensional domain, and it is inferred whether a real fault arc has occurred or it is due to the line conduction interference of the mixed load, thus greatly improving the accuracy of arc detection.

[0109] Please refer to Figure 3 , on the other hand, the present invention discloses a corresponding fault arc recognition system, which adopts the above-mentioned fault arc recognition method. The system includes:

[0110] An acquisition module, connected to the circuit comparison module, for acquiring the ADC raw data of the arc signal, and the ADC raw data includes the current raw data.

[0111] A circuit comparison module, connected to the judgment module, for respectively processing the current raw data through the FFT fast Fourier series transform and the extraction of the current second-order wavelet coefficients to obtain the current comparison result.

[0112] A judgment module, connected to the self-learning module, for judging whether a fault arc has occurred according to the current comparison result.

[0113] A self-learning module, for adjusting the preset parameters according to the current comparison result and the voltage comparison result.

[0114] Among them, the judgment module can also be used to analyze the voltage zero-point data of the voltage raw data to obtain the voltage comparison result when both the first current comparison result and the second current comparison result show that they are in the critical region, and judge whether a fault arc has occurred according to the current comparison result and the voltage comparison result.

[0115] Please refer to Figure 4 , on yet another aspect, the present invention also discloses a corresponding arc recognition device, including:

[0116] Common-mode filtering anti-interference circuit, current transformer, operational amplifier circuit, voltage sampling circuit, and the above-mentioned fault arc recognition system. Specifically, the arc recognition device provided by the present invention can be installed at the very front end of the electrical circuit of the single load or hybrid load to be measured. Among them, the device excludes the pre-stage interference signal through the common-mode filtering anti-interference circuit, filters out the conducted interference of the pre-stage to obtain the sampling data of the post-stage load; the live wire of the mains power passes through the high-frequency precision current transformer, and the tiny current signal induced by the secondary of the current transformer passes through the operational amplifier circuit, rectifies and filters the current signal into the voltage signal component of a unidirectional current pulse, and the amplitude of the voltage signal is controlled within the pin working voltage range of the fault recognition system; at the same time, a high-precision voltage-resistant resistor is connected in series between the live wire and the neutral wire, and the phase voltage is sampled in the form of series voltage division. Thus, the waveform sampling of signal such as voltage and current values and sine waves can be realized for the fault arc recognition system to call and be used as the basis for determining whether there is a real fault arc, and further adopt the aforementioned fault arc recognition method to improve the recognition rate of the fault arc and reduce its false judgment rate.

[0117] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in the embodiments of the present invention can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0118] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments and will not be repeated here.

[0119] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A method for identifying a faulty arc, characterized in that, Including: Obtaining the ADC raw data of the arc signal, where the ADC raw data includes current raw data; Processing the current raw data respectively through FFT (Fast Fourier Transform) and current second-order wavelet coefficient extraction to obtain a current comparison result; Judging whether a faulty arc occurs according to the current comparison result; The current comparison result includes a first current comparison result and a second current comparison result. The step of processing the current raw data respectively through FFT and current second-order wavelet coefficient extraction to obtain the comparison result includes: Processing the current raw data through FFT to obtain spectral data; Analyzing the spectral data to obtain the first current comparison result; Processing the current raw data through current second-order wavelet coefficient extraction to obtain first-period data and second-period data; Comparing the first-period data and the second-period data to obtain the second current comparison result; The step of processing the current raw data through current second-order wavelet coefficient extraction to obtain first-period data and second-period data includes: Performing first-order wavelet filtering on the current raw data to obtain a first-order wavelet coefficient table of current; Performing second-order wavelet filtering on the first-order wavelet coefficient table of current and intercepting to obtain the first-period data and the second-period data. The first-period data and the second-period data are the data of the first two periods of the first-order wavelet coefficient table of current in sequence, and the value range is (0 to 2π); The step of analyzing the spectral data to obtain the first current comparison result includes: Analyzing whether there are high-frequency signals with less than or equal to 9 half-cycles or greater than or equal to 14 half-cycles within 100 half-cycles in 1 second in the spectral data. If so, the first current comparison result is negative; The step of comparing the first-period data and the second-period data to obtain the second current comparison result includes: Subtracting the data of the first-period data point by point from the data of the second-period data and judging whether there is a jump. If so, the second current comparison result is negative.

2. The fault arc recognition method according to claim 1, wherein, The step of processing the current raw data through FFT to obtain spectral data includes: Processing the current raw data through FFT to obtain frequency-domain data; Filtering the frequencies below 2380 Hz in the frequency-domain data to obtain the spectral data.

3. The fault arc recognition method according to claim 1, wherein The ADC raw data further includes voltage raw data. The step of judging whether a faulty arc occurs according to the current comparison result further includes: When both the first current comparison result and the second current comparison result show being in the critical region, analyzing the voltage zero-point data of the voltage raw data to obtain a voltage comparison result; Judging whether a faulty arc occurs according to the current comparison result and the voltage comparison result.

4. The fault arc recognition method according to claim 3, wherein Before the step of processing the current raw data respectively through FFT and current second-order wavelet coefficient extraction to obtain the comparison result, it further includes: Obtain the current value according to the original current data, and obtain the voltage value according to the original voltage data; Judge whether the current value is greater than the rated current range, or the voltage value is within the fault voltage range. If so, proceed to the next step.

5. The fault arc identification method according to claim 3, wherein, After the step of judging whether a fault arc occurs according to the current comparison result and the voltage comparison result, it further includes: Adjust the preset parameters according to the current comparison result and the voltage comparison result.

6. A fault arc identification system, characterized in that, It includes: An acquisition module, configured to acquire the ADC original data of the arc signal, and the ADC original data includes the original current data; A circuit comparison module, configured to process the original current data through FFT (Fast Fourier Transform) and extraction of the second-order current wavelet coefficients respectively to obtain a current comparison result; A judgment module, configured to judge whether a fault arc occurs according to the current comparison result; The current comparison result includes a first current comparison result and a second current comparison result; The circuit comparison module includes: Process the original current data through FFT to obtain spectral data; Analyze the spectral data to obtain a first current comparison result; Process the original current data through extraction of the second-order current wavelet coefficients to obtain first-period data and second-period data; Compare the first-period data and the second-period data to obtain a second current comparison result; The analyzing the spectral data to obtain a first current comparison result includes: Analyze whether there is a high-frequency signal with less than or equal to 9 half-cycles or greater than or equal to 14 half-cycles within 100 half-cycles in 1 second in the spectral data. If so, the first current comparison result is negative; The comparing the first-period data and the second-period data to obtain a second current comparison result includes: Subtract the data of the first-period data point by point from the data of the second-period data, and judge whether there is a jump. If so, the second current comparison result is negative.

7. An arc recognition device, characterized in that, It includes: A common-mode filtering anti-interference circuit, a current transformer, an operational amplifier circuit, a voltage sampling circuit, and the fault arc identification system according to claim 6.

Images