Arc detection method, device and electronic equipment

Abstract

The present invention provides an arc detection method, device and electronic device. In the present invention, at least one arc characteristic value is used when performing arc detection, which can be analyzed from multiple characteristic angles to improve the accuracy of the arc detection method. In addition, the characteristic threshold used in the present invention takes into account the historical arc characteristic values ​​and environmental factors, and the obtained characteristic threshold has higher accuracy, thereby achieving higher accuracy in arc detection. In addition, in the present invention, the arc characteristic cumulative value of the current detection cycle corresponding to the arc characteristic vector and the characteristic threshold vector is determined, and the arc detection result is output when the sum of the historical arc characteristic cumulative value and the arc characteristic cumulative value meets the arc detection condition. That is, in the present invention, the situation where the arc characteristic value suddenly changes due to interference is taken into account, and the sum of the historical arc characteristic cumulative value and the arc characteristic cumulative value is used to analyze whether the arc detection condition is met, which can avoid pseudo-arc interference and thereby improve the accuracy of arc detection.

Description

Technical Field

[0001] The present invention relates to the field of arc detection, and more particularly to an arc detection method, device and electronic equipment. Background Art

[0002] With the increasing use of photovoltaic energy in power system scenarios, the frequency of fire accidents caused by DC arcs has also increased. Arcs are a gas discharge phenomenon. Due to aging of terminal blocks, poor contact, short circuits, etc., current flows through insulating media such as air to generate sparks, thus causing fire accidents. Therefore, in practical applications, the demand for configuring DC arc detection devices is gradually increasing.

[0003] In the existing DC arc detection technology, a characteristic value of the current is extracted, compared with a set threshold, and whether an arc is generated is determined based on the comparison result. This detection method has low accuracy. Summary of the invention

[0004] In view of this, the present invention provides an arc detection method, device and electronic equipment to solve the problem of low accuracy of the arc detection method.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] An arc detection method, comprising:

[0007] Determine at least one arc characteristic value of the current frequency domain signal detected in the current detection cycle, and form an arc characteristic vector with the at least one arc characteristic value;

[0008] Acquire historical arc characteristic values ​​of historical detection cycles, and determine an environmental coefficient based on an average value of the historical arc characteristic values;

[0009] Based on the average value of the historical arc characteristic values ​​and the environmental coefficient, a characteristic threshold of the arc characteristic value is obtained, and the characteristic threshold of the arc characteristic value is combined into a characteristic threshold vector;

[0010] Determine the arc feature cumulative value of the current detection cycle corresponding to the arc feature vector and the feature threshold vector;

[0011] When the sum of the historical arc feature accumulated value and the arc feature accumulated value meets the arc detection condition, an arc detection result is output.

[0012] Optionally, determining at least one arc characteristic value of the current frequency domain signal detected in the current detection cycle includes:

[0013] Acquire a current signal of the photovoltaic module in a current detection cycle, and perform Fourier analysis on a harmonic signal in the current signal to obtain a current frequency domain signal corresponding to the harmonic signal;

[0014] At least one arc characteristic value of the current frequency domain signal is extracted.

[0015] Optionally, the at least one arc characteristic value includes: an amplitude characteristic value, a linear correlation characteristic value and a discrete characteristic value; extracting at least one arc characteristic value of the current frequency domain signal includes:

[0016] Calculating the amplitude average value of each frequency point in the current frequency domain signal to obtain an amplitude characteristic value;

[0017] Determine an arc linear characteristic value of the current frequency domain signal, and calculate a linear correlation characteristic value corresponding to the arc linear characteristic value;

[0018] The sum of the amplitude differences of adjacent frequency points in the current frequency domain signal is calculated to obtain a discrete eigenvalue.

[0019] Optionally, determining the environmental coefficient based on the average value of the historical arc characteristic value includes:

[0020] Calculating an average value of the historical arc characteristic values, and determining an environmental coefficient calculation formula according to a magnitude relationship between the average value of the historical arc characteristic values ​​and a preset amplitude threshold;

[0021] Based on the environmental coefficient calculation formula, the environmental coefficient of the photovoltaic module is calculated.

[0022] Optionally, obtaining a characteristic threshold of the arc characteristic value based on an average value of the historical arc characteristic value and the environmental coefficient includes:

[0023] The product of the average value of the historical arc characteristic value and the environmental coefficient is used as the characteristic threshold of the arc characteristic value.

[0024] Optionally, determining the arc feature cumulative value of the current detection cycle corresponding to the arc feature vector and the feature threshold vector includes:

[0025] The difference between the arc feature vector and the feature threshold vector is calculated to obtain a difference vector, and the arc feature cumulative value of the current detection cycle corresponding to the difference vector is determined.

[0026] Optionally, calculating the difference between the arc feature vector and the feature threshold vector to obtain the difference vector includes:

[0027] Calculating the difference between the arc characteristic vector and the corresponding component in the characteristic threshold vector;

[0028] Determining the component value of the component based on a comparison relationship between the difference and a preset threshold;

[0029] The component values ​​of the components are combined to obtain a difference vector.

[0030] Optionally, determining the arc feature cumulative value of the current detection cycle corresponding to the difference vector includes:

[0031] Based on the accumulated positive and negative values ​​corresponding to the target quadrant where the difference vector is located and the vector modulus of the difference vector, the accumulated value of the arc characteristics of the current detection cycle is obtained.

[0032] Optionally, based on the accumulated positive and negative values ​​corresponding to the target quadrant where the difference vector is located and the vector modulus of the difference vector, the accumulated value of the arc feature of the current detection cycle is obtained, including:

[0033] Based on the component values ​​of each component of the difference vector, determine the target hexagram where the difference vector is located, and obtain the cumulative positive and negative values ​​corresponding to the target hexagram;

[0034] The vector modulus of the difference vector is calculated, and the accumulated positive and negative values ​​are used as the signs of the vector modulus of the difference vector to obtain the accumulated value of the arc characteristics of the current detection cycle.

[0035] Optionally, when the sum of the historical arc feature cumulative value and the arc feature cumulative value meets the arc detection condition, outputting the arc detection result includes:

[0036] When the sum of the historical arc feature cumulative value and the arc feature cumulative value is not less than a preset cumulative threshold, an arc detection result is output.

[0037] An arc detection device, comprising:

[0038] A feature determination module, used to determine at least one arc feature value of the current frequency domain signal detected in the current detection cycle, and to form an arc feature vector from the at least one arc feature value;

[0039] A coefficient determination module, used for obtaining historical arc characteristic values ​​of historical detection cycles, and determining an environmental coefficient based on an average value of the historical arc characteristic values;

[0040] A vector determination module, used for obtaining a characteristic threshold value of the arc characteristic value based on an average value of the historical arc characteristic value and the environmental coefficient, and forming a characteristic threshold value vector with the characteristic threshold value of the arc characteristic value;

[0041] A cumulative value determination module, used to determine the arc feature cumulative value of the current detection cycle corresponding to the arc feature vector and the feature threshold vector;

[0042] The detection result determination module is used to output the arc detection result when the sum of the historical arc feature accumulated value and the arc feature accumulated value meets the arc detection condition.

[0043] An arc detection device, comprising: a memory and a processor;

[0044] Wherein, the memory is used to store programs;

[0045] The processor calls the program and is used to execute the above arc detection method.

[0046] An inverter comprises the above arc detection device.

[0047] Optionally, the method further includes: a control device, wherein when the arc detection result output by the arc detection device indicates that an arc fault is detected, the control device controls the inverter to shut down.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] The present invention provides an arc detection method, device and electronic device. In the present invention, at least one arc characteristic value is used when performing arc detection. Compared with the method of using only one characteristic value, it can be analyzed from multiple characteristic angles to improve the accuracy of the arc detection method. In addition, the characteristic threshold used in the present invention takes into account the historical arc characteristic value and environmental factors, and the accuracy of the characteristic threshold obtained is higher, and then the accuracy of arc detection using the characteristic threshold vector composed of the characteristic threshold is higher. In addition, in the present invention, the arc characteristic cumulative value of the current detection cycle corresponding to the arc characteristic vector and the characteristic threshold vector is determined, and the arc detection result is output when the sum of the historical arc characteristic cumulative value and the arc characteristic cumulative value meets the arc detection condition. That is, in the present invention, the situation that the arc characteristic value suddenly changes due to interference is taken into account, and the sum of the historical arc characteristic cumulative value and the arc characteristic cumulative value is used to analyze whether the arc detection condition is met, which can avoid pseudo arc interference and improve the accuracy of arc detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0051] Figure 1 A method flow chart of an arc detection method provided by an embodiment of the present invention;

[0052] Figure 2 An arc comparison diagram provided by an embodiment of the present invention;

[0053] Figure 3 A method flow chart of another arc detection method provided by an embodiment of the present invention;

[0054] Figure 4 Another arc comparison diagram provided by an embodiment of the present invention;

[0055] Figure 5 Another arc comparison diagram provided by an embodiment of the present invention;

[0056] Figure 6 Another arc comparison diagram provided by an embodiment of the present invention;

[0057] Figure 7 A method flow chart of another arc detection method provided by an embodiment of the present invention;

[0058] Figure 8 A schematic diagram of a hexagram provided in an embodiment of the present invention;

[0059] Fig. 9 A schematic structural diagram of an arc detection device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0060] 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.

[0061] With the increasing use of photovoltaic energy in power systems, the frequency of fire accidents caused by DC arcs has also increased. Arcing is a gas discharge phenomenon. Due to aging of terminal blocks, poor contact, short circuits, etc., current flows through insulating media such as air to generate sparks, thereby causing fire accidents. In addition, due to the power generation mechanism of photovoltaic modules, electricity is continuously generated under sunlight. If the inverter does not have an arc detection function, once an arc occurs, serious fire accidents often occur. Therefore, in photovoltaic power generation systems, the demand for configuring DC arc detection devices is becoming increasingly strong.

[0062] In existing DC arc detection technology, time-frequency feature analysis has become the main basis for detecting arc generation. It is usually necessary to compare an extracted feature value with a set threshold to determine whether an arc has occurred.

[0063] Since DC arc is a complex physical phenomenon, the characteristic quantities exhibited in different environments or operating conditions are not the same. Therefore, selecting only one characteristic value to determine whether an arc has occurred is not widely accepted by the market. In addition, the fixed threshold setting is easily disturbed by environmental factors such as the dynamic changes in temperature and light intensity in the photovoltaic environment and the large amount of switching noise. Inappropriate threshold design will produce high-frequency false alarms or reduce the effectiveness of arc detection in certain field environments.

[0064] Therefore, there is an urgent need for an arc detection method that can improve the accuracy of DC arc detection, greatly reduce the number of false arc alarms, enhance the safety and reliability of photovoltaic power generation systems, and improve power generation efficiency.

[0065] After research, the inventors found that in order to avoid the problem of insufficient accuracy of detection by a single eigenvalue and the problem of false arc interference affecting the normal operation of the photovoltaic power generation system, an arc detection method, device and electronic equipment were proposed to improve the accuracy of DC arc detection and greatly reduce the number of false arc alarms, thereby enhancing the safety and reliability of the photovoltaic power generation system and improving the power generation efficiency.

[0066] Specifically, in the present invention, at least one arc characteristic value is used when performing arc detection. Compared with the method of using only one characteristic value, it can be analyzed from multiple characteristic angles to improve the accuracy of the arc detection method. In addition, the characteristic threshold used in the present invention takes into account the historical arc characteristic values ​​and environmental factors, and the accuracy of the characteristic threshold obtained is higher, and then the accuracy of arc detection using the characteristic threshold vector composed of the characteristic threshold is higher. In addition, in the present invention, the arc characteristic cumulative value of the current detection cycle is obtained by the cumulative positive and negative values ​​corresponding to the target hexagram where the difference vector is located and the vector modulus of the difference vector, and the arc detection result is output when the sum of the historical arc characteristic cumulative value and the arc characteristic cumulative value meets the arc detection condition. That is, in the present invention, the situation where the arc characteristic value suddenly changes due to interference is taken into account, and the sum of the historical arc characteristic cumulative value and the arc characteristic cumulative value is used to analyze whether the arc detection condition is met, which can avoid pseudo-arc interference and improve the accuracy of arc detection.

[0067] Based on the above content, an embodiment of the present invention provides an arc detection method, and the execution subject may be an arc detection device. Figure 1 , which may include:

[0068] S11. Determine at least one arc characteristic value of the current frequency domain signal detected in the current detection cycle, and form an arc characteristic vector with the at least one arc characteristic value.

[0069] In this embodiment, arc detection is performed according to a detection cycle, and each detection cycle may be 3 ms.

[0070] In practical applications, each photovoltaic string corresponds to a current transformer of an arc detection device, and its input cable passes through the magnetic ring of the corresponding current transformer to extract the AC signal of the photovoltaic string current, and obtain the current signal of the photovoltaic module in the current detection cycle. The current signal is a time domain current signal. In the embodiment of the present application, if an arc occurs in a photovoltaic string, the corresponding fault string can be accurately operated.

[0071] Then, Fourier analysis is performed on the harmonic signal in the time domain current signal to obtain a current frequency domain signal corresponding to the harmonic signal. Then, at least one arc characteristic value of the current frequency domain signal is extracted.

[0072] In another implementation of the present invention, at least one arc characteristic value includes: an amplitude characteristic value, a linear correlation characteristic value and a discrete characteristic value. Generally speaking, it is preferred to simultaneously determine the amplitude characteristic value, the linear correlation characteristic value and the discrete characteristic value. In addition, it can also be at least two or at least one of the amplitude characteristic value, the linear correlation characteristic value and the discrete characteristic value.

[0073] In the actual calculation and analysis process, refer to Figure 2 , obtain the current frequency domain signal when the machine is operating normally and arcing occurs in a certain frequency band. Figure 2 The frequency domain signals during normal operation and arcing are given, where the frequency band (the horizontal axis represents frequency) is selected as 40-55kHz.

[0074] It can be observed that in normal operation, the frequency domain signal strength at 48kHz, the multiple of the machine switching frequency, is relatively large. In shorter frequency bands, such as 40-44kHz and 51-55kHz, the signals at each frequency point are relatively weak and change linearly with almost no fluctuation. In contrast, when an arc occurs, the signal strength at each frequency point in 40-44kHz and 51-55kHz increases significantly, and the linearity of each shorter frequency band is destroyed, and the fluctuation between discrete frequency points is also aggravated.

[0075] Therefore, by comparing Figure 2 The changes before and after the arc occurs can obtain the corresponding arc characteristic values ​​and generate the corresponding characteristic vectors.

[0076] Among them, the arc eigenvector includes amplitude eigenvalue, linear eigenvalue and discrete eigenvalue. Compared with the characteristics of variance, skewness, kurtosis, etc., the amplitude eigenvalue, linear eigenvalue and discrete eigenvalue have stronger resistance to environmental interference and reflect the characteristics of the real arc more accurately.

[0077] Reference Figure 3 , extracting at least one arc characteristic value of the current frequency domain signal may include:

[0078] S21. Calculate the amplitude average value of each frequency point in the current frequency domain signal to obtain an amplitude characteristic value.

[0079] Specifically, the arc amplitude characteristic value refers to the average amplitude value of each frequency point in the selected frequency band, which is represented by a:

[0080]

[0081] Where N represents the number of frequency points in the selected frequency band, x i Represents the current frequency domain signal intensity at the i-th frequency point.

[0082] In practical applications, 200 arc amplitude characteristic values ​​are obtained when the machine is operating normally and when an arc occurs for comparison. The comparison results are as follows: Figure 4 shown.

[0083] Among them, the frequency band is selected as 51-55kHz. Figure 4 The arc characteristic value in is obtained from Figure 2 The arc characteristic value of the frequency domain signal of 51-55kHz is cut out from 40-55kHz.

[0084] Specifically, sampling is performed within one cycle and the frequency domain analysis is performed to obtain Figure 2 The frequency point signals in the range of 51-55kHz are selected to calculate the arc eigenvalues. The three eigenvalues ​​correspond to the even inverted amplitude eigenvalue, linear eigenvalue and discrete eigenvalue. Figure 4 , Figure 5 and Figure 6 After that, the next cycle continues to extract the amplitude eigenvalue, linear eigenvalue, and discrete eigenvalue, and obtains Figure 4 , Figure 5 and Figure 6 The second point of , and so on, until the 200th point is obtained.

[0085] Of course, in actual applications and specific detection needs, it can be adjusted to other frequency bands. This application does not specifically limit the method of selecting the frequency band, and all of them belong to the protection scope of this application.

[0086] It should be noted that when the inverter is operating normally, the amplitude characteristic value of each string current is relatively low, basically between 60-270, and when an arc occurs, a large number of high-frequency harmonic signals appear in the string current, and the arc amplitude characteristic value in the selected frequency band also increases accordingly, stepping to between 180-900, and a large amount of data is above 300. Therefore, the amplitude characteristic value provided in the embodiment of the present application can effectively distinguish whether an arc occurs in the string.

[0087] S22. Determine an arc linear characteristic value of the current frequency domain signal, and calculate a linear correlation characteristic value corresponding to the arc linear characteristic value.

[0088] Specifically, the arc linear eigenvalue refers to the transformed form of the linear correlation coefficient of the frequency points in the selected frequency band, represented by r. Among them, the arc linear eigenvalue, also known as the conventional linear correlation coefficient, r can be expressed as:

[0089]

[0090] Where N represents the number of frequency points in the selected frequency band, x i represents the current frequency domain signal intensity at the i-th frequency point, and They represent the average current frequency domain signal strength and the average frequency point of N frequency points respectively.

[0091] The conventional linear correlation coefficient r is a value of [-1,1]. The closer r is to 0, the weaker the linear correlation between the frequency points. The closer r is to -1 or 1, the stronger the linear correlation between the frequency points.

[0092] The linear correlation characteristic value is transformed from the conventional linear correlation coefficient r, which is more suitable for arc detection judgment. Specifically, in order to facilitate the subsequent threshold calculation and simplify the integration process, the conventional linear correlation coefficient r is transformed as follows:

[0093] c=100-100*|r|

[0094] That is, the linear correlation eigenvalue c∈[0,100], the larger the c, the weaker the linear correlation.

[0095] In actual application, 200 arc linear characteristic values ​​are obtained for comparison when the machine is operating normally and when arcing occurs. The comparison results are as follows: Figure 5 shown.

[0096] Among them, the frequency band is also selected as 51-55kHz.

[0097] It should be noted that when the inverter is operating normally, the current of each string is linearly distributed between the frequency points in the selected shorter frequency band, and the arc linear characteristic value is small, basically between 0-20. Occasionally, a small number of data points greater than 20 will also be eliminated in the integration method of the embodiment of the present application to prevent false alarms of false arcing. When a real arc occurs, the linearity of the high-frequency signal of each string current in the selected frequency band is destroyed, and the arc linear characteristic value suddenly changes, jumping to between 20-100. Therefore, the arc linear characteristic value provided in the embodiment of the present application and the linear correlation characteristic value calculated based on the arc linear characteristic value can effectively distinguish whether an arc occurs in the string.

[0098] S23, calculating the sum of amplitude differences of adjacent frequency points in the current frequency domain signal to obtain a discrete eigenvalue.

[0099] Specifically, the discrete eigenvalue refers to the discrete index between frequency points in the selected frequency band, represented by d:

[0100]

[0101] Where N represents the number of frequency points in the selected frequency band, x i Represents the current frequency domain signal intensity at the i-th frequency point.

[0102] In practical applications, 200 arc discrete characteristic values ​​are obtained when the machine is operating normally and when arcing occurs for comparison. The comparison results are shown in the figure below. Figure 6 shown.

[0103] Among them, the frequency band is also selected as 51-55kHz.

[0104] It should be noted that when the inverter is operating normally, the current of each string fluctuates slightly between the frequency points in the selected frequency band, and the arc discrete characteristic value is small, basically between 0 and 40. When an arc occurs, the arc discrete characteristic value in the selected frequency band also rises sharply, jumping to between 50 and 300. Therefore, the discrete characteristic value provided in the embodiment of the present application can also effectively distinguish whether an arc occurs in the string.

[0105] After obtaining the three arc characteristic values, the corresponding arc characteristic vector can be generated. The arc characteristic vector can be expressed as:

[0106] A=(a,c,d)

[0107] S12. Obtain historical arc characteristic values ​​of historical detection cycles, and determine an environmental coefficient based on an average value of the historical arc characteristic values.

[0108] In this embodiment, the historical arc characteristic values ​​of each historical detection cycle of the arc detection device since its installation may be obtained. In addition, the historical arc characteristic values ​​of a certain stage or a certain number of historical detection cycles may also be obtained.

[0109] Specifically, N1 and N2 historical arc characteristic values ​​may be obtained respectively, where N1 is 200 and N2 is 4000. Then, the average value of the N1 historical arc characteristic values ​​is calculated. To reduce the calculation burden and storage pressure, the average value of the N1 historical arc characteristic values ​​may be used as the average value of the N2 historical arc characteristic values.

[0110] In addition, the average value of N2 historical arc characteristic values ​​can be directly obtained and calculated without considering the calculation burden and storage pressure.

[0111] It should be noted that the historical arc characteristic values ​​need to be updated in real time to adapt to different site environments and different operating conditions.

[0112] Then, an environmental coefficient is determined based on the average value of the historical arc characteristic values.

[0113] Specifically, each arc characteristic value corresponds to an environmental coefficient, and the environmental coefficients corresponding to the amplitude characteristic value a, the linear correlation characteristic value c, and the discrete characteristic value d are respectively called ta, tc, and td.

[0114] Among them, the values ​​of ta, tc, and td are related to the actual operating status of each string of the photovoltaic power station, and are adaptively adjusted to meet different on-site environments.

[0115] Due to the differences in the on-site environment of different power stations, affected by the dynamic changes of light intensity and temperature, noise, inconsistent aging of different terminal blocks and other interferences, the operating status of each string group can be divided into three levels: excellent, good and poor according to the average value of N2 historical arc characteristic values.

[0116] The higher the average value of N2 historical arc characteristic values, the closer the on-site environment is to the severe level, and the smaller the values ​​of ta, tc, and td are; the lower the average value of N2 historical arc characteristic values, the closer the on-site environment is to the excellent level, and the larger the values ​​of ta, tc, and td are.

[0117] The values ​​of ta, tc, and td are related to the actual operating conditions of the photovoltaic strings. Since each photovoltaic string corresponds to a magnetic ring for sampling, each photovoltaic string can be adaptively adjusted according to its own operating conditions.

[0118] The operating status of each photovoltaic string can be divided into three levels: excellent, good, and poor according to the average value y and quartile of the historical arc characteristic value, which can be specifically expressed as:

[0119]

[0120] Among them, Q1 is the first quartile of N2 historical data, that is, 25% of the historical data are less than or equal to Q1; Q3 is the third quartile of N2 historical data, that is, 75% of the historical data are less than or equal to Q3; Q is the quarterback difference of N2 historical data, Q=Q3-Q1, reflecting the degree of dispersion of historical data; y is the average of N1 historical data; E is the environmental coefficient when the operating state is excellent; S is the environmental coefficient when the operating state is poor.

[0121] When y≤Q1, the operating condition of the string group is excellent, and ta, tc, and td are calculated using the first formula in the above formula; when y≥Q3, the operating condition of the string group is poor, and ta, tc, and td are calculated using the third formula in the above formula; when Q1<y<Q3, the operating condition of the string group is good, and ta, tc, and td are calculated using the second formula in the above formula.

[0122] In the above three formulas for calculating the environmental coefficient, the setting of the upper limit value E of ta, tc, and td can improve the detection rate when the average value y is small; the setting of the lower limit value S of ta, tc, and td can reduce the false arc alarm when the average value y is high. Among them, E can be selected as 4, and S can be selected as 1.5.

[0123] By using the above formula, ta, tc, and td are calculated respectively. That is to say, based on the average value of the historical arc characteristic value, the environmental coefficient is determined, including:

[0124] Calculate the average value of the historical arc characteristic values, and determine the environmental coefficient calculation formula based on the relationship between the average value of the historical arc characteristic values ​​and the preset amplitude threshold, that is, select the first formula, the second formula or the third formula mentioned above, and calculate the environmental coefficient of the photovoltaic module, that is, ta, tc, and td, based on the environmental coefficient calculation formula.

[0125] S13. Based on the average value of the historical arc characteristic values ​​and the environmental coefficient, obtain a characteristic threshold of the arc characteristic value, and form a characteristic threshold vector with the characteristic threshold of the arc characteristic value.

[0126] In this embodiment, the product of the average value of N1 historical arc characteristic values ​​and the environmental coefficient can be used as the characteristic threshold of the arc characteristic value. In this embodiment, directly judging the difference in environment with historical data can not only distinguish factors such as weather and light, but also more directly reflect and process other interferences that affect the historical arc characteristic data, such as grid quality, differences in photovoltaic components, etc.

[0127] It should be noted that the feature threshold vector in the present invention is proposed for multiple feature values, and has a more comprehensive advantage than a single threshold.

[0128] S14. Determine the arc feature cumulative value of the current detection cycle corresponding to the arc feature vector and the feature threshold vector.

[0129] Specifically, the difference between the arc feature vector and the feature threshold vector may be calculated to obtain a difference vector, and the arc feature cumulative value of the current detection cycle corresponding to the difference vector may be determined.

[0130] Among them, refer to Figure 7, calculating the difference between the arc feature vector and the feature threshold vector to obtain the difference vector may include:

[0131] S31. Calculate the difference between the arc feature vector and the corresponding component in the feature threshold vector.

[0132] In this embodiment, the arc characteristic vector is A=(a, c, d), and the characteristic threshold vector is T=(ta, tc, td).

[0133] When the two are subtracted, the corresponding components in the vectors are subtracted, that is, the values ​​of a-ta, c-tc, and d-td are calculated.

[0134] S32: Determine the component value of the component based on a comparison relationship between the difference and a preset threshold.

[0135] In this embodiment, the preset threshold is 0, that is, the numerical values ​​of a-ta, c-tc, and d-td are compared with 0. If they are greater than 0, the component value of the component is 1; if they are equal to 0, the component value of the component is 0; if they are less than 0, the component value of the component is -1.

[0136] S33. Combine the component values ​​of the components to obtain a difference vector.

[0137] Specifically, the components obtained in step S32 are combined according to the corresponding component positions to obtain a difference vector. For example, the arc feature vector is (3, 4, 5), the feature threshold vector is (0, 4, 9), and the difference values ​​of the corresponding components are 3, 0, and -4, respectively. The component values ​​of the components are 1, 0, and -1, respectively, and the final difference vector is (1, 0, -1).

[0138] After determining the difference vector, determine the accumulated value of the arc feature of the current detection cycle corresponding to the difference vector. Specifically, based on the accumulated positive and negative values ​​corresponding to the target quadrant where the difference vector is located and the vector modulus of the difference vector, obtain the accumulated value of the arc feature of the current detection cycle. Specifically, including:

[0139] 1) Based on the component values ​​of each component of the difference vector, determine the target hexagram where the difference vector is located, and obtain the cumulative positive and negative values ​​corresponding to the target hexagram.

[0140] After the difference vector is determined, it is necessary to determine the hexagram in which the difference vector is located to determine the direction of the difference vector modulus accumulation. The specific direction refers to the positive or negative sign of the difference vector modulus.

[0141] The relationship between the hexagram and the space vector symbol is as follows Figure 8As shown in the figure, when the photovoltaic power generation system is operating normally, due to the magnitude relationship between the arc characteristic vector and the characteristic threshold vector, the difference vector is basically at the position of (-1,-1,-1) in the space coordinate system, that is, when operating normally, the difference vector is basically at the VIIth hexagram. However, when an arc occurs in a group of strings in the photovoltaic power generation system, the arc characteristic values ​​of the current of the group of strings rise sharply, and the arc characteristic vector is compared with the characteristic threshold vector. The difference vector obtained is basically at the position of (1,1,1) in the space coordinate system, that is, when an arc occurs, the arc integral vector is basically at the Ith hexagram.

[0142] In practical applications, the normal operation of the machine will be subject to various interferences, causing a sudden change in a certain arc characteristic value. For example, when operating under the same working conditions, only the string current suddenly changes, and the harmonic content of each frequency band of the string current is likely to increase accordingly, but the sudden change in the string current will not affect the linear characteristics and discrete characteristics of a certain frequency band. Therefore, the sudden change in current may cause the arc amplitude characteristic value to increase, causing the corresponding arc integral component to be greater than 0. This application defines this interference that causes a sudden change in a certain arc characteristic value as pseudo-arc interference.

[0143] The position of the vector form in the hexagram can intuitively determine the possibility of arc occurrence in this frequency band (I has the greatest possibility, II / IV / V has a high possibility, III / VI / VIII has a low possibility, and VII has the least possibility). The size of the possibility is realized by the vector modulus, avoiding the weight calculation that needs to be considered for multiple eigenvalues ​​in other schemes.

[0144] In order to prevent the influence of false arcs, when two or more of the three components of the difference vector satisfy that they are greater than 0, that is, the difference vector is in the I, II, IV, and V limits, the difference vector modulus accumulates positively, that is, the difference vector modulus takes a positive sign, that is; otherwise, the difference vector modulus accumulates negatively, that is, the difference vector modulus takes a positive sign.

[0145] In practical applications, based on the component values ​​(specifically the positive and negative signs) of each component of the difference vector, the target quadrant (the above-mentioned 8 quadrants I, II, IV, V, etc.) where the difference vector is located is determined, and the cumulative positive and negative values ​​corresponding to the target quadrant, that is, the positive and negative signs, are obtained.

[0146] 2) Calculate the vector modulus of the difference vector, and use the accumulated positive and negative values ​​as the signs of the vector modulus of the difference vector to obtain the accumulated value of the arc characteristics of the current detection cycle.

[0147] Specifically, after the vector modulus of the difference vector is calculated, the accumulated positive and negative values ​​are used as the positive and negative signs of the vector modulus of the difference vector, so as to obtain the accumulated value of the arc feature of the current detection cycle.

[0148] For example, if the difference vector is (1,0,-1), then the arc characteristic cumulative value is

[0149] S15. When the sum of the historical arc feature accumulated value and the arc feature accumulated value meets the arc detection condition, output the arc detection result.

[0150] In this embodiment, when the sum of the historical arc feature cumulative value and the arc feature cumulative value is not less than a preset cumulative threshold, the arc detection result is output.

[0151] Specifically, all historical arc feature cumulative values ​​are obtained, and when the sum of the historical arc feature cumulative values ​​and the arc feature cumulative value is greater than or equal to a preset cumulative threshold (specifically an upper limit value), it is determined that an arc fault is detected.

[0152] It should be noted that the upper limit value in this application example is 1 max Indicates that I max The value of needs to balance the arc detection duration and prevent false arc alarms, such as I max =50.

[0153] Among them, I max When the value is large, the arc detection cycle is prolonged, and the difficulty of arc detection will increase due to the real-time update mechanism of the characteristic threshold vector; I max When the value is small, the accumulated value of the integral arc characteristic caused by the pseudo arc may trigger the upper limit value, thereby causing a false alarm of the pseudo arc and threatening the normal operation of the photovoltaic power generation system.

[0154] In addition, in the embodiment of the present invention, when the sum of the historical arc feature cumulative value and the arc feature cumulative value is less than or equal to the lower limit value, the sum of the cumulative values ​​is set to the lower limit value.

[0155] In this application example, the lower limit is denoted by I min Indicates that I min =0, because the arc characteristic cumulative value is often at I min Therefore, I min It also affects the duration of the arc detection cycle and the ability to prevent false arc alarms. When the sum of the accumulated values ​​is less than or equal to the lower limit, the sum of the accumulated values ​​is set to the lower limit.

[0156] In this embodiment, when performing arc detection, at least one arc characteristic value is used. Compared with the method of using only one characteristic value, it can be analyzed from multiple characteristic angles to improve the accuracy of the arc detection method. In addition, the characteristic threshold used in the present invention takes into account the historical arc characteristic values ​​and environmental factors, and the accuracy of the characteristic threshold obtained is higher, and then the accuracy of arc detection using the characteristic threshold vector composed of the characteristic threshold is higher. In addition, in the present invention, the arc characteristic cumulative value of the current detection cycle is obtained by the cumulative positive and negative values ​​corresponding to the target hexagram where the difference vector is located and the vector modulus of the difference vector, and the arc detection result is output when the sum of the historical arc characteristic cumulative value and the arc characteristic cumulative value meets the arc detection condition. That is, in the present invention, the situation where the arc characteristic value suddenly changes due to interference is taken into account, and the sum of the historical arc characteristic cumulative value and the arc characteristic cumulative value is used to analyze whether the arc detection condition is met, which can avoid pseudo-arc interference and improve the accuracy of arc detection.

[0157] Optionally, based on the above-mentioned embodiment of the arc detection method, another embodiment of the present invention provides an arc detection device, referring to Fig. 9 ,include:

[0158] A feature determination module 11 is used to determine at least one arc feature value of the current frequency domain signal detected in the current detection cycle, and to form an arc feature vector from the at least one arc feature value;

[0159] A coefficient determination module 12, used to obtain historical arc characteristic values ​​of historical detection cycles, and determine the environmental coefficient based on the average value of the historical arc characteristic values;

[0160] A vector determination module 13, configured to obtain a characteristic threshold of the arc characteristic value based on an average value of the historical arc characteristic value and the environmental coefficient, and to form a characteristic threshold vector from the characteristic threshold of the arc characteristic value;

[0161] The cumulative value determination module 14 is used to determine the arc feature cumulative value of the current detection cycle corresponding to the arc feature vector and the feature threshold vector;

[0162] The detection result determination module 15 is used to output an arc detection result when the sum of the historical arc feature cumulative value and the arc feature cumulative value meets the arc detection condition.

[0163] Further, the feature determination module 11 includes:

[0164] A signal processing submodule, used to obtain the current signal of the photovoltaic module in the current detection cycle, and perform Fourier analysis on the harmonic signal in the current signal to obtain the current frequency domain signal corresponding to the harmonic signal;

[0165] The characteristic value extraction submodule is used to extract at least one arc characteristic value of the current frequency domain signal.

[0166] Furthermore, the at least one arc characteristic value includes: an amplitude characteristic value, a linear correlation characteristic value and a discrete characteristic value; and the characteristic value extraction submodule includes:

[0167] A first calculation unit, used to calculate the amplitude average value of each frequency point in the current frequency domain signal to obtain an amplitude characteristic value;

[0168] A second calculation unit, used to determine the arc linear characteristic value of the current frequency domain signal, and calculate the linear correlation characteristic value corresponding to the arc linear characteristic value;

[0169] The third calculation unit is used to calculate the sum of amplitude differences of adjacent frequency points in the current frequency domain signal to obtain a discrete eigenvalue.

[0170] Further, the coefficient determination module 12 is used to determine the environmental coefficient based on the average value of the historical arc characteristic value, specifically for:

[0171] The average value of the historical arc characteristic values ​​is calculated, and according to the relationship between the average value of the historical arc characteristic values ​​and a preset amplitude threshold, an environmental coefficient calculation formula is determined, and based on the environmental coefficient calculation formula, the environmental coefficient of the photovoltaic module is calculated.

[0172] Further, the vector determination module 13 is used to obtain the characteristic threshold of the arc characteristic value based on the average value of the historical arc characteristic value and the environmental coefficient, specifically for:

[0173] The product of the average value of the historical arc characteristic value and the environmental coefficient is used as the characteristic threshold of the arc characteristic value.

[0174] Further, the cumulative value determination module 14 includes:

[0175] The cumulative value determination submodule is used to calculate the difference between the arc feature vector and the feature threshold vector to obtain a difference vector, and determine the arc feature cumulative value of the current detection cycle corresponding to the difference vector.

[0176] Furthermore, the cumulative value determination submodule includes:

[0177] A difference calculation unit, used for calculating the difference between the arc characteristic vector and the corresponding component in the characteristic threshold vector;

[0178] A component value determination unit, configured to determine a component value of the component based on a comparison relationship between the difference and a preset threshold;

[0179] A combining unit is used to combine the component values ​​of the components to obtain a difference vector.

[0180] Furthermore, the cumulative value determination submodule includes:

[0181] The cumulative value determination unit is used to obtain the arc feature cumulative value of the current detection cycle based on the cumulative positive and negative values ​​corresponding to the target quadrant where the difference vector is located and the vector modulus of the difference vector.

[0182] Further, the cumulative value determination unit is specifically used for:

[0183] Based on the component values ​​of each component of the difference vector, the target quadrant where the difference vector is located is determined, and the cumulative positive and negative values ​​corresponding to the target quadrant are obtained, the vector modulus of the difference vector is calculated, and the cumulative positive and negative values ​​are used as the sign of the vector modulus of the difference vector to obtain the cumulative value of the arc characteristic of the current detection cycle.

[0184] Further, the detection result determination module 15 is specifically used for:

[0185] When the sum of the historical arc feature cumulative value and the arc feature cumulative value is not less than a preset cumulative threshold, an arc detection result is output.

[0186] In this embodiment, when performing arc detection, at least one arc characteristic value is used. Compared with the method of using only one characteristic value, it can be analyzed from multiple characteristic angles to improve the accuracy of the arc detection method. In addition, the characteristic threshold used in the present invention takes into account the historical arc characteristic values ​​and environmental factors, and the accuracy of the characteristic threshold obtained is higher, and then the accuracy of arc detection using the characteristic threshold vector composed of the characteristic threshold is higher. In addition, in the present invention, the arc characteristic cumulative value of the current detection cycle corresponding to the arc characteristic vector and the characteristic threshold vector is determined, and the arc detection result is output when the sum of the historical arc characteristic cumulative value and the arc characteristic cumulative value meets the arc detection condition. That is, in the present invention, the situation where the arc characteristic value suddenly changes due to interference is taken into account, and the sum of the historical arc characteristic cumulative value and the arc characteristic cumulative value is used to analyze whether the arc detection condition is met, which can avoid pseudo-arc interference and improve the accuracy of arc detection.

[0187] It should be noted that, for the working process of each module, sub-module and unit in this embodiment, please refer to the corresponding description in the above embodiment, which will not be repeated here.

[0188] Optionally, based on the above-mentioned embodiments of the arc detection method and device, another embodiment of the present invention provides an arc detection device, including: a memory and a processor;

[0189] Wherein, the memory is used to store programs;

[0190] The processor calls the program and is used to execute the above arc detection method.

[0191] Optionally, based on the above embodiment of the arc detection device, another embodiment of the present invention provides an inverter, comprising the above arc detection device.

[0192] Furthermore, the inverter further includes: a control device, wherein when the arc detection result output by the arc detection device indicates that an arc fault is detected, the control device controls the inverter to shut down.

[0193] That is, in the present invention, an arc detection device is used to detect whether an arc fault exists, and when an arc fault exists, the arc fault information is sent to the control device, and the control device controls the inverter to shut down. The control device may be a DSP (Digital Signal Process).

[0194] In this embodiment, when performing arc detection, at least one arc characteristic value is used. Compared with the method of using only one characteristic value, it can be analyzed from multiple characteristic angles to improve the accuracy of the arc detection method. In addition, the characteristic threshold used in the present invention takes into account the historical arc characteristic values ​​and environmental factors, and the accuracy of the characteristic threshold obtained is higher, and then the accuracy of arc detection using the characteristic threshold vector composed of the characteristic threshold is higher. In addition, in the present invention, the arc characteristic cumulative value of the current detection cycle corresponding to the arc characteristic vector and the characteristic threshold vector is determined, and the arc detection result is output when the sum of the historical arc characteristic cumulative value and the arc characteristic cumulative value meets the arc detection condition. That is, in the present invention, the situation where the arc characteristic value suddenly changes due to interference is taken into account, and the sum of the historical arc characteristic cumulative value and the arc characteristic cumulative value is used to analyze whether the arc detection condition is met, which can avoid pseudo-arc interference and improve the accuracy of arc detection.

[0195] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An arc detection method, characterized in that: include: Determine at least one arc characteristic value of the current frequency domain signal detected in the current detection cycle, and form an arc characteristic vector with the at least one arc characteristic value; Obtaining historical arc characteristic values ​​of historical detection cycles, and determining an environmental coefficient corresponding to a string operation condition based on an average value of the historical arc characteristic values; Based on the average value of the historical arc characteristic values ​​and the environmental coefficient, a characteristic threshold value of the arc characteristic value is obtained, and the characteristic threshold value of the arc characteristic value is combined into a characteristic threshold vector; Determine the arc feature cumulative value of the current detection cycle corresponding to the difference between the arc feature vector and the feature threshold vector; When the sum of the historical arc feature accumulated value and the arc feature accumulated value meets the arc detection condition, an arc detection result is output.

2. The arc detection method according to claim 1, characterized in that: Determining at least one arc characteristic value of the current frequency domain signal detected in the current detection cycle, including: Acquire a current signal of the photovoltaic module in a current detection cycle, and perform Fourier analysis on a harmonic signal in the current signal to obtain a current frequency domain signal corresponding to the harmonic signal; At least one arc characteristic value of the current frequency domain signal is extracted.

3. The arc detection method according to claim 2, characterized in that: The at least one arc characteristic value includes: an amplitude characteristic value, a linear correlation characteristic value, and a discrete characteristic value; extracting at least one arc characteristic value of the current frequency domain signal includes: Calculating the amplitude average value of each frequency point in the current frequency domain signal to obtain an amplitude characteristic value; Determine an arc linear characteristic value of the current frequency domain signal, and calculate a linear correlation characteristic value corresponding to the arc linear characteristic value; The sum of the amplitude differences of adjacent frequency points in the current frequency domain signal is calculated to obtain a discrete eigenvalue.

4. The arc detection method according to claim 1, characterized in that: Based on the average value of the historical arc characteristic values, the environmental coefficient corresponding to the string operation condition is determined, including: Calculating an average value of the historical arc characteristic values, and determining an environmental coefficient calculation formula according to a magnitude relationship between the average value of the historical arc characteristic values ​​and a preset amplitude threshold; Based on the environmental coefficient calculation formula, the environmental coefficient corresponding to the string operation condition of the photovoltaic components is calculated.

5. The arc detection method according to claim 1, characterized in that: Based on the average value of the historical arc characteristic value and the environmental coefficient, a characteristic threshold of the arc characteristic value is obtained, including: The product of the average value of the historical arc characteristic value and the environmental coefficient is used as the characteristic threshold of the arc characteristic value.

6. The arc detection method according to claim 1, characterized in that: Determining the arc feature cumulative value of the current detection cycle corresponding to the difference between the arc feature vector and the feature threshold vector includes: The difference between the arc feature vector and the feature threshold vector is calculated to obtain a difference vector, and the arc feature cumulative value of the current detection cycle corresponding to the difference vector is determined.

7. The arc detection method according to claim 6, characterized in that: Calculating the difference between the arc feature vector and the feature threshold vector to obtain a difference vector includes: Calculating the difference between the arc characteristic vector and the corresponding component in the characteristic threshold vector; Determining the component value of the component based on a comparison relationship between the difference and a preset threshold; The component values ​​of the components are combined to obtain a difference vector.

8. The arc detection method according to claim 6, characterized in that: Determining the arc feature cumulative value of the current detection cycle corresponding to the difference vector includes: Based on the accumulated positive and negative values ​​corresponding to the target quadrant where the difference vector is located and the vector modulus of the difference vector, the accumulated value of the arc characteristics of the current detection cycle is obtained.

9. The arc detection method according to claim 8, characterized in that: Based on the accumulated positive and negative values ​​corresponding to the target quadrant where the difference vector is located and the vector modulus of the difference vector, the accumulated value of the arc feature of the current detection cycle is obtained, including: Based on the component values ​​of each component of the difference vector, determine the target hexagram where the difference vector is located, and obtain the cumulative positive and negative values ​​corresponding to the target hexagram; The vector modulus of the difference vector is calculated, and the accumulated positive and negative values ​​are used as the signs of the vector modulus of the difference vector to obtain the accumulated value of the arc characteristics of the current detection cycle.

10. The arc detection method according to claim 1, characterized in that: When the sum of the historical arc feature accumulated value and the arc feature accumulated value meets the arc detection condition, outputting the arc detection result includes: When the sum of the historical arc feature cumulative value and the arc feature cumulative value is not less than a preset cumulative threshold, an arc detection result is output.

11. An arc detection device, characterized in that: include: A feature determination module, used to determine at least one arc feature value of the current frequency domain signal detected in the current detection cycle, and to form an arc feature vector from the at least one arc feature value; A coefficient determination module, used to obtain historical arc characteristic values ​​of historical detection cycles, and determine the environmental coefficient corresponding to the string operation condition based on the average value of the historical arc characteristic values; A vector determination module, used for obtaining a characteristic threshold value of the arc characteristic value based on an average value of the historical arc characteristic value and the environmental coefficient, and forming a characteristic threshold value vector with the characteristic threshold value of the arc characteristic value; A cumulative value determination module, used to determine the arc feature cumulative value of the current detection cycle corresponding to the difference between the arc feature vector and the feature threshold vector; The detection result determination module is used to output the arc detection result when the sum of the historical arc feature accumulated value and the arc feature accumulated value meets the arc detection condition.

12. An arc detection device, characterized in that: include: Memory and processor; Wherein, the memory is used to store programs; The processor calls the program and is used to execute the arc detection method as described in claims 1-10.

13. An inverter, characterized in that: Comprising the arc detection device as claimed in claim 12.

14. The inverter according to claim 13, characterized in that: Also includes: A control device is provided for controlling the inverter to shut down when the arc detection result output by the arc detection device indicates that an arc fault is detected.

Images