Three-phase inverter dc side leakage fault type discrimination method and electronic equipment

Abstract

The application provides a three-phase inverter DC side leakage fault type discrimination method and electronic equipment, the application carries out frequency spectrum time sequence characteristic analysis on the residual current of the three-phase inverter AC side, further analyzes the change of the residual current amplitude with time under different frequencies, and can understand the fault type of the inverter DC side leakage; through continuous energy monitoring of the residual current signal of the inverter, real-time current monitoring of the inverter is realized, and the sudden change of the leakage fault type or the instantaneous leakage can also be well monitored and discriminated. The application is aimed at the specific characteristics of the inverter leakage current harmonic component, the leakage current spectrum characteristics are different under different fault types, on the basis, the leakage fault type discrimination method based on the harmonic time-frequency feature extraction is proposed, the specific fault type of the inverter can be discriminated under various working conditions, and then the leakage fault position is located, and the problem that the leakage fault type discrimination and positioning are inaccurate in current engineering application is solved.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of leakage detection, and particularly relates to a three-phase inverter DC side leakage fault type discrimination method and electronic equipment. BACKGROUND

[0002] The three-phase grid-connected inverter is a key device for converting the direct current generated by the power generation system into alternating current. In the power generation system, direct current side leakage is a common fault condition. Leakage not only causes direct loss of energy and reduces the working efficiency of the inverter, but also may cause overheating, short circuit and even fire inside the device, posing a serious threat to personnel and property. In addition, the diversity of direct current side leakage fault types, including high resistance leakage, resistance-capacitance series leakage, resistance-capacitance parallel leakage and resonance leakage, each has its unique electrical characteristics, which makes the fault diagnosis and positioning extremely complex. This complexity requires more delicate and intelligent detection technology to ensure the safe operation of the system.

[0003] Traditional leakage discrimination methods often rely on periodic manual inspection or simple threshold alarm mechanisms. These methods are difficult to be effective in the face of the complexity and concealment of leakage. First, manual inspection is time-consuming and labor-intensive, and cannot achieve real-time monitoring. For intermittent or transient leakage, the moment of fault occurrence may be missed. Second, the alarm system based on fixed threshold can detect obvious leakage events, but lacks sensitivity for low-level leakage or changing leakage patterns, and is prone to false positives or false negatives. In addition, traditional methods also have deficiencies in fault positioning. Once an alarm is raised, maintenance personnel often need to spend a lot of time to locate the specific leakage fault position, which not only increases maintenance costs, but also may lead to the expansion of the fault range.

[0004] Patent CN107179492B discloses a leakage detection method for power distribution lines. This method simultaneously acquires the residual current signal and voltage signal of the power distribution line, separates and calculates the resistive component of the residual current according to the amplitude and phase relationship of the residual current and voltage at each frequency, obtains the resistive leakage current in the residual current, compares the resistive leakage current with a predefined threshold, and judges the leakage condition of the power distribution line. Although it can detect leakage events, it lacks sensitivity for low-level leakage or changing leakage patterns, is prone to false positives or false negatives, and cannot identify fault types and positions. SUMMARY

[0005] The purpose of the present application is to overcome the deficiencies of the prior art and provide a three-phase inverter DC side leakage fault type discrimination method and electronic equipment, which can identify the leakage fault type and improve the accuracy of leakage fault positioning.

[0006] To achieve the above purpose, the technical scheme adopted by the present application is:

[0007] A three-phase inverter DC side leakage fault type discrimination method, comprising the following implementation processes:

[0008] S1: Collecting grid-connected current signals of the grid-connected inverter AC side, adding three-phase grid-connected current signals to obtain a residual current signal;

[0009] S2: Wavelet packet decomposition and reconstruction are performed on the residual current signal to obtain amplitude-time curves and energy spectra of the residual current at different frequencies;

[0010] S3: According to the amplitude-time curves of the residual current at different frequencies, the fault type of leakage is judged; according to the energy spectra of the residual current at different frequencies, the energy change of the residual current is identified, and when the energy change is greater than or equal to a first threshold value, it is determined that the fault type of leakage changes or instantaneous leakage occurs.

[0011] The present application can understand the fault type of inverter DC side leakage by analyzing the frequency spectrum time sequence characteristics of the residual current of the three-phase inverter AC side, and analyzing the change of the residual current amplitude with time at different frequencies; through continuous energy monitoring of the residual current signal of the inverter, real-time current monitoring of the inverter is realized, and sudden change or instantaneous leakage of the fault type of leakage can also be well monitored and distinguished; the present application can identify the leakage fault type and improve the accuracy of leakage fault positioning.

[0012] The present application is aimed at the specific characteristics of inverter leakage current harmonic components, and the frequency spectrum characteristics of leakage current are different under different fault types, on the basis of which a leakage fault type discrimination method based on harmonic time-frequency feature extraction is proposed, which can distinguish the specific fault type of the inverter under various working conditions, and then locate the leakage fault position, solving the problem of inaccurate leakage fault type discrimination and positioning in current engineering applications.

[0013] Further, the implementation process of judging the fault type of leakage according to the amplitude-time curves of the residual current at different frequencies comprises:

[0014] After wavelet packet decomposition and reconstruction, the residual current has a DC component and an odd multiple switching frequency harmonic component, when the amplitudes of the DC component and the odd multiple switching frequency harmonic component are greater than or equal to a second threshold value and less than or equal to a third threshold value, it is determined that the fault type of leakage is parasitic resistance fault leakage;

[0015] When the amplitudes of the DC component and the odd multiple switching frequency harmonic component are greater than or equal to a fourth threshold value and less than or equal to a fifth threshold value, it is determined that the fault type of leakage is parasitic resistance and parasitic capacitance parallel fault leakage;

[0016] The remaining current after wavelet packet decomposition and reconstruction has only odd multiple switching frequency harmonic components, when the amplitude is greater than or equal to a second threshold value and less than or equal to a third threshold value, it is determined that the leakage fault type is a parasitic resistance and parasitic capacitance series fault leakage;

[0017] When the amplitude is greater than or equal to a fourth threshold value and less than or equal to a fifth threshold value, it is determined that the leakage fault type is a parasitic capacitance fault leakage.

[0018] Further, the fourth threshold value is less than or equal to 5 times the second threshold value, and the fifth threshold value is less than or equal to 5 times the third threshold value.

[0019] Based on the same inventive concept, the present application also provides an electronic device, comprising:

[0020] One or more processors;

[0021] A memory having one or more programs stored thereon, which, when executed by the one or more processors, cause the one or more processors to implement the steps of the three-phase inverter DC side leakage fault type discrimination method.

[0022] Based on the same inventive concept, the present application also provides a computer readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the three-phase inverter DC side leakage fault type discrimination method.

[0023] Compared with the prior art, the present application has the following advantages:

[0024] The present application analyzes the frequency spectrum time sequence characteristics of the residual current on the AC side of the three-phase inverter, and then analyzes the change of the residual current amplitude with time at different frequencies, so as to understand the fault type of the leakage on the DC side of the inverter; through continuous energy monitoring of the residual current signal of the inverter, real-time current monitoring of the inverter is realized, and sudden or instantaneous leakage of the leakage fault type can also be well monitored and distinguished; the present application can identify the leakage fault type and the leakage fault position, and improve the accuracy of leakage fault positioning.

[0025] The present application is aimed at the specific characteristics of the leakage current harmonic components of the inverter, and the frequency spectrum characteristics of the leakage current are different under different fault types, on the basis of which a leakage fault type discrimination method based on harmonic time-frequency feature extraction is proposed, which can distinguish the specific fault type of the inverter under various working conditions, and then locate the leakage fault position, solving the problem of inaccurate leakage fault type discrimination and positioning in current engineering applications.

[0026] The application can not only achieve real-time monitoring effect on the residual current change of the inverter, and find potential leakage problems in time, but also determine the fault type and position of the current time period by using the time sequence characteristics of the residual current, take corresponding repair measures for different fault types in time, capture even intermittent or instantaneous leakage, and accurate fault discrimination helps to reduce unnecessary maintenance work, ensures that the maintenance action is targeted, and improves the overall reliability and availability of the system. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1 A three-phase inverter DC side leakage fault type discrimination module of the application is shown in the figure.

[0028] Figure 2 A three-phase inverter DC side leakage fault type discrimination module of the application is shown in the figure.

[0029] Figure 3 A residual current amplitude and energy spectrum feature extraction flow chart of the application is shown in the figure.

[0030] Figure 4 A structure diagram of a non-isolated photovoltaic three-phase grid-connected inverter under different fault types is shown in the figure.

[0031] Figure 5 A residual current reconstruction node coefficient diagram under different frequencies is shown in the figure.

[0032] Figure 6 A residual current energy spectrum diagram of different nodes is shown in the figure.

[0033] Figure 7 A leakage fault type discrimination unit flow chart of the application is shown in the figure. DETAILED DESCRIPTION

[0034] The application will be described in detail below with reference to the embodiments. It should be noted that the embodiments in the application and the features in the embodiments can be combined with each other without conflict. For the convenience of description, if the words "up", "down", "left", "right" appear in the following text, they only mean the same direction as the up, down, left and right of the drawing itself, and do not limit the structure.

[0035] EMBODIMENT

[0036] As Figure 1 , Figure 2The three-phase inverter DC side leakage fault type discrimination system of the embodiment comprises a data acquisition unit, a time sequence analysis unit, a fault discrimination unit and an instantaneous leakage identification unit. Since the leakage current is generated on the loop formed by the power grid and the parasitic impedance and the ground, the data acquisition unit first acquires the amplitude, frequency and other information of the grid-connected current on the alternating current side by using the current sensor on the alternating current side. Since the residual current is the sum of the currents of the three phases A, B and C, the amplitude, frequency and other information of the residual current can be obtained by adding the grid-connected currents of the three phases. The residual current data is transmitted to the time sequence analysis unit, and the Wavelet Analyzer technology of the time sequence analysis unit is used to perform wavelet packet decomposition on the collected residual current signal to obtain the wavelet packet tree of the residual current data. The node reconstruction is performed on each node to obtain the amplitude-time curve change graph of the residual current under different frequencies and the time energy spectrum of each node under each frequency. Then, the fault discrimination unit discriminates and classifies the amplitude-time curve change graph of the residual current under different frequencies. Finally, the instantaneous leakage identification unit monitors the residual current energy of each frequency in unit time in real time. When an energy mutation or abnormality is found, it is judged that the leakage fault has occurred and intermittent or instantaneous leakage has occurred.

[0037] Figure 3 is a residual current time sequence fault feature extraction flowchart, and the machine interpretation of the fault feature extraction mechanism is as follows:

[0038] Due to the modulation effect of the three-phase inverter, a voltage component of the switching sub-harmonic is generated, and then a high-frequency zero-sequence leakage current of an integer odd multiple of the switching frequency is generated. Generally, the high-frequency zero-sequence leakage current component of one switching frequency is the most obvious. Now the time sequence characteristics of the residual current generated on the alternating current side of the three-phase inverter are analyzed. First, the wavelet packet decomposition is used to process the residual current signal data. The wavelet packet decomposition is widely used in the field of signal processing. It can decompose both low-frequency and high-frequency signals, and there is no redundancy or omission in the decomposition. Therefore, the signal containing a large amount of medium and high frequency information can be better analyzed in time and frequency localization. The wavelet packet decomposition and reconstruction principle is as follows:

[0039] Let x(t) be a time signal, denote the i-th wavelet packet on the j-th layer, and

[0040]

[0041] The wavelet packet decomposition is represented as:

[0042]

[0043] The wavelet packet reconstruction is represented as:

[0044]

[0045] Wherein, h(·) is a low-pass filter, g(·) is a high-pass filter, l, k are the decomposition layers, h, g are filter coefficients, h is related to the scaling function, and g is related to the wavelet function.

[0046] The original signal containing harmonics is decomposed into signals of different frequency bands, the results on the low frequency band are regarded as the fundamental component, and the high frequency band is each harmonic. The residual current signal is decomposed into each node by using wavelet packet decomposition, and each bottom layer node corresponds to a different frequency band. The node coefficient graph obtained by observing each node coefficient graph of the residual current data decomposed by the wavelet packet is the amplitude-time curve graph of the current decomposition in each frequency band. However, as the number of decomposition layers increases, the higher level wavelet packet decomposition allows the signal to be analyzed in a higher frequency range. Therefore, the frequency range corresponding to each node coefficient will be narrowed, that is, the higher frequency components are more finely decomposed and described, and as the frequency resolution increases, the time resolution will decrease accordingly. Therefore, as the number of layers increases, the time range of the node coefficient is no longer able to represent all the time domain as the original signal. Therefore, it is necessary to reconstruct each node of the bottom layer decomposition to obtain the reconstructed node coefficient graph, that is, the coefficient distribution graph used to reconstruct the original signal according to the node coefficient. At this time, the amplitude-time curve graph of the residual current at different frequencies can be obtained through the reconstructed coefficient distribution graph.

[0047] The wavelet packet decomposition can extract the energy features of the signal. The residual current signal is decomposed by 4 layers of wavelet packet, and each wavelet packet coefficient of the bottom layer is reconstructed to obtain the reconstructed signal X a,b , a∈(1,4), b∈(1,2 a )), wherein the subscript a represents the decomposition layer number of the wavelet packet coefficient used for reconstruction, and b represents the sorting position of the wavelet packet coefficient used for reconstruction in the a layer coefficient.

[0048] After obtaining the reconstructed signal X a,b , the energy E a,b is calculated:

[0049]

[0050] Wherein, n represents the sequence length, the energy is arranged in the original order, and the energy sequence Y a of the corresponding layer reconstruction signal is obtained. For example, the energy sequence of the 3rd layer is:

[0051] Y3=(E 3,1 ,E 3,2 ,E 3,3 ,E 3,4 ,E 3,5 ,E 3,6 ,E 3,7E 3,8 )

[0052] Figure 4 is the structure diagram of non-isolated photovoltaic three-phase grid-connected inverter under different fault types. According to the inverter model shown in Figure 4 The simulation parameters for collecting different components of leakage current are shown in Table 1.

[0053] Table 1 Simulation model parameters for collecting leakage current components under different fault types

[0054] Parameter Value Photovoltaic equivalent direct voltage U dc / V]]> 1000 Parasitic resistance R / Ω 500 C PV / μF 1 Filter inductance L / mH 5 Filter capacitor C f / μF 10 Grid side filter inductance L g / μH 100 Switching frequency f s / Hz 10000 Fundamental frequency f / Hz 50 Dead time delay time t d / μs 5

[0055] The simulation wavelet packet decomposition analysis shows that the leakage current under different fault types is shown in Table 2.

[0056] Table 2 Leakage current component table under different fault types

[0057]

[0058] According to Table 2, the DC component and the zero sequence high frequency component under the switching frequency harmonic in the leakage current are larger, which is the main reason, and the low frequency dead zone component is smaller and can be ignored. Therefore, the time sequence variation characteristics of the DC component and the high frequency component of the switching frequency integer times of the current signal are extracted by using wavelet packet decomposition. The original signal containing each harmonic is decomposed into signals under different frequency bands, the results on the low frequency band are regarded as the fundamental component, and the high frequency band is each harmonic. The current signal is decomposed into each node by using wavelet packet decomposition, and each bottom node corresponds to a different frequency band. The node coefficient graph of the residual current under different frequencies is observed, that is, the decomposition result of the original signal in each frequency band.

[0059] Figure 5 is the node coefficient graph of the residual current under different frequencies. The original signal is decomposed by 3-layer wavelet packet, the sampling frequency is 100000 Hz, and the simulation time is 0.8 s. The simulation time is divided into four time periods corresponding to different leakage fault types: 0-0.2 s for parasitic resistance fault; 0.2-0.4 s for parasitic resistance and parasitic capacitance series fault; 0.4-0.6 s for parasitic resistance and parasitic capacitance parallel fault; 0.6-0.8 s for parasitic capacitance fault, as shown in Table 3.

[0060] Table 3 Fault type time table

[0061] Fault type Time / s Parasitic resistance fault 0～0.2 Parasitic capacitance fault 0.2～0.4 Parasitic resistance and parasitic capacitance in series fault 0.4～0.6 Parasitic resistance and parasitic capacitance in parallel fault 0.6～0.8

[0062] After 3-layer decomposition of the residual current data, the frequency band is divided into 8 regions, and each sub-band width is 50000 / 8=6260 Hz. Considering the frequency band order disorder problem of wavelet packet decomposition, the wavelet packet decomposition nodes are (3, 0) to (3, 7), as shown in Table 4.

[0063] Table 4 Band division of wavelet packet decomposition

[0064] Decomposition node Frequency band range / Hz Harmonic component (3，0) 0～6250 Direct current component (3，1) 6250～12500 fs frequency harmonic component (3，2) 12500～18750 / (3，3) 18750～25000 2fs frequency sideband harmonic component (3，4) 25000～31250 3fs frequency harmonic component (3，5) 31250～37500 / (3，6) 37500～43750 4fs frequency sideband harmonic component (3，7) 43750～50000 /

[0065] The frequency band range of the (3, 0) node 7 is 0-6250 Hz, containing a direct current component, and the frequency band range of the (3, 1) node 8 is 6250-12500 Hz, containing a switching frequency harmonic component. The harmonic component of the node (3, 0) is mainly dominated by the direct current component, and it can be found that the leakage current caused by the parasitic resistance fault and the parasitic resistance and parasitic capacitance parallel fault in the loop will produce a relatively obvious direct current component. The node (3, 1) is mainly the switching frequency harmonic component, and it can be found that the leakage current caused by the parasitic capacitance fault and the parasitic resistance and parasitic capacitance parallel fault has a large value, which is the main reason for the leakage current component.

[0066] Figure 6 The residual current energy spectrum of different nodes is obtained by wavelet packet decomposition, and the image of the residual current signal energy sequence can be obtained, and it can be obviously seen that the distribution of the reconstructed signal energy of different nodes, that is, different frequency bands, has a significant difference. The frequency band range of the (3, 0) node 7 is 0-6250 Hz, containing a direct current component, and the frequency band range of the (3, 1) node 8 is 6250-12500 Hz, containing a switching frequency harmonic component. In the fixed frequency band interval, the energy of the current at each frequency is also distinguished, and due to the switching effect of the inverter, the energy is mainly distributed in the direct current side and the integer odd multiple switching frequency high frequency harmonic, that is, the node 7 and the node 8. And due to different fault types, the energy spectrum color and energy size are different, and when the energy spectrum color changes greatly or the energy size changes greatly, it can be judged that the fault type has changed.

[0067] Figure 7 The leakage fault type discrimination unit flow chart is shown in FIG. 7, and the machine explanation of the leakage fault type discrimination mechanism is as follows:

[0068] Due to the modulation effect of the three-phase inverter, a switching sub-harmonic voltage component is generated, and the Fourier series expression of the A-phase voltage waveform of the three-phase inverter is as follows:

[0069]

[0070] In the formula, ω = 2πf, f is the fundamental frequency, E d is the direct current power supply voltage, M is the modulation degree, N is the carrier ratio, m is the harmonic number relative to the carrier, n is the harmonic number relative to the modulation wave, is the initial phase of the modulation wave, J0, J n is the first kind of Bessel function.

[0071] The phase voltage waveforms of the three-phase inverter are mutually different by 120° in phase, and the Fourier series expression of the B-phase voltage waveform of the three-phase inverter is as follows:

[0072]

[0073] The Fourier series expression of the C-phase voltage waveform of the three-phase inverter is as follows:

[0074]

[0075] The first term in the above three expressions is a voltage fundamental wave, the second term is an m-th harmonic of a carrier wave, and the third term is an upper or lower sideband harmonic of the m-th harmonic of the carrier wave.

[0076] Note that the m-th harmonic of the carrier wave in the second term has the same phase of the voltage component of the integer multiple harmonic of the carrier wave generated by the A, B and C phases, and thus generates a certain amount of zero sequence voltage, and also generates a zero sequence current, the harmonic number of the current should be an integer multiple of the carrier wave, that is, an integer odd multiple of the switching harmonic. Since the switching frequency is generally a high frequency of kilohertz, the leakage current can be generated, and the form embodied is a high-frequency zero-sequence leakage current of an integer odd multiple of the switching frequency.

[0077] There is a parasitic impedance between the metal frame of the inverter and the ground, which is affected by the size of the equipment, the installation condition and the environment. Generally, the form of the parasitic impedance can be a parasitic resistance and a parasitic capacitance, and the leakage current caused by the fault conditions is also different. Through wavelet packet reconstruction analysis of the waveform, it can be concluded that if the fault type is a parasitic resistance leakage, then the direct current leakage component and the leakage component at the switching frequency odd integer multiple of the inverter will be detected, but the amplitude is small, the amplitude is greater than or equal to the second threshold value, and less than or equal to the third threshold value; if the fault type is a parasitic resistance and a parasitic capacitance in parallel leakage, then the direct current leakage component and the leakage component at the switching frequency odd integer multiple of the inverter will be detected, and the amplitude is very large, the amplitude is greater than or equal to the fourth threshold value, and less than or equal to the fifth threshold value. If the fault type is a parasitic resistance and a parasitic capacitance in series leakage, then only the leakage component at the switching frequency odd integer multiple of the inverter will be detected, but the amplitude is small, the amplitude is greater than or equal to the second threshold value, and less than or equal to the third threshold value; if the fault type is a parasitic capacitance leakage, then only the leakage component at the switching frequency odd integer multiple of the inverter will be detected, and the amplitude is very large, the amplitude is greater than or equal to the fourth threshold value, and less than or equal to the fifth threshold value. The fourth threshold value is less than or equal to 5 times the second threshold value, and the fifth threshold value is less than or equal to 5 times the third threshold value. According to this, it can be concluded that under different fault conditions, according to the frequency at which the leakage current is generated and the amplitude of the generated leakage current, it can be inferred that what kind of fault type causes the leakage, and a leakage fault type discrimination unit is established to judge the fault type of the inverter.

[0078] Preferably, the second threshold is 0A, the third threshold is 3A, the fourth threshold is 4A, and the fifth threshold is 10A.

[0079] The inverter DC side leakage fault type discrimination method is one of the key factors for safe and reliable operation of the inverter system. The traditional fault discrimination method is often limited by high cost of artificial detection, long discrimination time and difficulty in online monitoring. However, through the spectrum timing analysis of the inverter AC side leakage current, the specific situation of the DC side leakage fault type can be clearly understood in a more economical and efficient way, and real-time monitoring can be achieved. The time sequence characteristics such as waveform, amplitude and energy of the current can quickly respond to the change of the fault type, so as to timely and efficiently detect the intermittent or instantaneous leakage caused by the change of the fault type. According to the specific characteristics of the inverter leakage current harmonic component, it is clear that the frequency spectrum characteristics of the leakage current are different under different fault types. On this basis, a leakage fault type discrimination method based on harmonic time-frequency feature extraction is proposed, which can discriminate the specific fault type and position of the inverter under various working conditions, and solve the problem of inaccurate leakage fault type discrimination and positioning in current engineering applications.

[0080] The method effectively overcomes the limitations of traditional technology in inverter leakage current fault discrimination, such as long discrimination time, inaccurate fault type and positioning, etc. The method is simple to operate and can improve the detection accuracy. Not only can the real-time monitoring effect of the residual current change of the inverter be achieved, but also potential leakage problems can be found in time. At the same time, by using the time sequence characteristics of the residual current, the fault type and position of the current time period can be determined, and appropriate repair measures can be taken for different fault types. Even intermittent or instantaneous leakage can be captured. Accurate fault discrimination can help reduce unnecessary maintenance work, ensure targeted maintenance action, and improve the overall reliability and availability of the system.

[0081] Another embodiment of the present application provides an electronic device, comprising:

[0082] One or more processors;

[0083] A memory having one or more programs stored thereon, which, when executed by the one or more processors, cause the one or more processors to implement the steps of the three-phase inverter DC side leakage fault type discrimination method.

[0084] In some implementations, the memory can be a high-speed random access memory (RAM: Random Access Memory), and can also include a non-volatile memory, such as at least one disk memory.

[0085] In other implementations, the processor can be a central processing unit (CPU), a digital signal processor (DSP), or other types of general purpose processors, without limitation.

[0086] Another embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the three-phase inverter DC side leakage fault type discrimination method.

[0087] The above-mentioned embodiments should be understood as merely illustrative of the present application, and should not be used to limit the scope of the present application. After reading the present application, those skilled in the art can make various equivalent modifications to the present application, which fall within the scope of the appended claims.

Claims

1. A method for identifying the type of leakage current fault on the DC side of a three-phase inverter, characterized in that, The implementation process includes the following: S1: Collect the grid-connected current signal on the AC side of the grid-connected inverter, and add the grid-connected current signals of the three phases to obtain the residual current signal; S2: Perform wavelet packet decomposition and reconstruction on the residual current signal to obtain the amplitude-time variation curve and energy spectrum of the residual current at different frequencies; S3: Based on the amplitude variation curve of the residual current at different frequencies over time, determine the type of leakage fault; based on the energy spectrum of the residual current at different frequencies, identify the energy change of the residual current; when the energy change is greater than or equal to the first threshold, determine whether the leakage fault type is intermittent leakage or instantaneous leakage. The process of determining the type of leakage fault based on the amplitude-time variation curve of the residual current at different frequencies includes: The residual current after wavelet packet decomposition and reconstruction contains a DC component and a harmonic component that is an odd multiple of the switching frequency. When the amplitude of the DC component and the amplitude of the harmonic component that is an odd multiple of the switching frequency are greater than or equal to the second threshold and less than or equal to the third threshold, the leakage fault type is determined to be parasitic resistance fault leakage. When the amplitude of the DC component and the amplitude of the harmonic component at odd multiples of the switching frequency are greater than or equal to the fourth threshold and less than or equal to the fifth threshold, the leakage fault type is determined to be a parasitic resistance and parasitic capacitance parallel fault leakage. The residual current after wavelet packet decomposition and reconstruction contains only harmonic components that are odd multiples of the switching frequency. When the amplitude is greater than or equal to the second threshold and less than or equal to the third threshold, the leakage fault type is determined to be a series leakage fault caused by parasitic resistance and parasitic capacitance. When its amplitude is greater than or equal to the fourth threshold and less than or equal to the fifth threshold, the leakage fault type is determined to be parasitic capacitance leakage.

2. The method for determining the DC-side leakage fault type of a three-phase inverter according to claim 1, characterized in that, The fourth threshold is less than or equal to 5 times the second threshold, and the fifth threshold is less than or equal to 5 times the third threshold.

3. An electronic device, characterized in that, include: One or more processors; A memory having stored one or more programs that, when executed by one or more processors, cause the one or more processors to perform the steps of the method according to any one of claims 1-2.

4. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the steps of the method according to any one of claims 1-2.

Images