A fault analysis method and system based on an AC-DC hybrid power distribution network

Abstract

The present application belongs to the field of AC / DC hybrid power distribution network control, and particularly relates to a fault analysis method and system based on an AC / DC hybrid power distribution network, which comprises: establishing a transient analysis model of the AC / DC hybrid power distribution network; when a single-phase grounding fault occurs in a resonant grounding AC subnetwork, a low-frequency voltage signal is injected by using a voltage source converter under the condition that an arc suppression coil remains online, and fault section positioning is realized based on the amplitude distribution of the low-frequency zero sequence current; when a single-pole grounding fault occurs in a DC subnetwork, zero-mode current is extracted by modulus decomposition, and fault line identification is realized based on the directional difference of the zero-mode current at both ends of the line; the coupling response characteristics on the AC and DC sides of the voltage source converter are monitored, cross-verification with the main characteristics in the fault subnetwork is performed, and the fault type and position are determined. The present application can unify the analysis of AC / DC fault characteristics and overcome the defects of traditional methods in terms of positioning difficulty in weak fault, high-resistance fault and distributed power access scenarios.

Description

Technical Field

[0001] This invention belongs to the field of AC / DC hybrid distribution network control, and particularly relates to a fault analysis method and system based on AC / DC hybrid distribution networks. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] With the rapid promotion of distributed photovoltaic power and the development of flexible DC distribution networks, AC / DC hybrid distribution networks have become an important technological direction for improving distribution network operation efficiency and power supply reliability. However, under this new distribution mode, the transient steady-state characteristics of faults have changed significantly, posing a severe challenge to traditional fault location technologies.

[0004] To improve power supply reliability, medium-voltage AC distribution networks generally adopt non-effective grounding. During a single-phase ground fault, the fault current is extremely weak, making it difficult for traditional overcurrent protection to operate reliably. In addition, the multi-branch radial structure of the distribution network and the nonlinear response of distributed power sources further exacerbate the complexity of the fault transient process, making accurate fault location extremely difficult.

[0005] When a single-pole ground fault occurs in a DC distribution network, its fault characteristics exhibit a polarized pattern: low-resistance faults generate instantaneous and severe undervoltage and overcurrent surges, demanding millisecond-level speeds for protection operation; high-resistance faults exhibit extremely weak transient characteristics, while the high-frequency ripple and harmonics generated by power electronic equipment severely interfere with fault feature extraction. Furthermore, short DC feeders weaken the differences in transient characteristics between different lines, and the multi-terminal grid structure introduces complex multi-feeder effects, further exacerbating the difficulty of fault line identification.

[0006] Therefore, there is an urgent need to propose effective fault identification schemes for the fault characteristics under the new form of AC / DC hybrid distribution networks. Summary of the Invention

[0007] To overcome the shortcomings of the prior art, the present invention provides a fault analysis method and system based on AC / DC hybrid distribution networks.

[0008] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions: Firstly, a fault analysis method based on AC / DC hybrid distribution networks is disclosed, including: A transient analysis model of AC / DC hybrid distribution network is established. Based on the transient analysis model of AC / DC hybrid distribution network, fault characteristics analysis and coupling verification of AC / DC subnetwork are carried out. When a single-phase ground fault occurs in the resonant grounded AC subgrid, while the arc suppression coil remains online, a low-frequency voltage signal is injected using a voltage source converter, and the fault section is determined based on the amplitude distribution of the low-frequency zero-sequence current. When a single-pole grounding fault occurs in the DC subgrid, the zero-mode current is extracted by modulus decomposition, and the fault line is identified based on the difference in the direction of the zero-mode current at both ends of the line. Monitor the coupling response characteristics on both the AC and DC sides of the voltage source converter, cross-verify them with the main characteristics within the faulty subnet, and determine the fault type and location.

[0009] Furthermore, while the arc suppression coil remains online, a low-frequency voltage signal is injected using a voltage source converter, including: using the signal modulation capability of the voltage source converter to superimpose a three-phase symmetrical low-frequency voltage signal onto the AC system.

[0010] Furthermore, the ratio of the injection frequency of the low-frequency voltage signal to the power frequency is less than 1.

[0011] Furthermore, the location of the fault section is determined based on the amplitude distribution of the low-frequency zero-sequence current. Specifically, this includes: using the low-frequency full-network capacitor current as a benchmark, determining the section with the measured zero-sequence current less than the benchmark as a non-fault section or downstream of the fault point, and determining the section with the measured zero-sequence current between the upper and lower limits determined by the overcompensation degree of the arc suppression coil and the ratio of the injection frequency to the power frequency as the upstream section of the fault point.

[0012] Furthermore, the modulus decomposition uses a preset decoupling matrix to transform the positive and negative currents, and the zero-mode current is equal to the sum of the positive and negative currents multiplied by a fixed coefficient.

[0013] Furthermore, fault line identification is achieved based on the difference in direction of zero-mode current at both ends of the line. Specifically, this includes comparing the directions of the zero-mode current measured at both ends of the same line. If the directions at both ends are the same, the line is determined to be faulty; if the directions at both ends are opposite, the line is determined to be non-faulty.

[0014] Furthermore, it also includes, through transient analysis models, establishing a database of coupling response characteristics of key electrical quantities on both the AC and DC sides of the voltage source converter for fault scenarios with different fault locations, types, and transition resistances in advance; during online analysis, the measured coupling characteristics are matched and compared with the database.

[0015] Secondly, a fault analysis system based on an AC / DC hybrid distribution network is disclosed, including: The model building module is used to establish a transient analysis model of the AC / DC hybrid distribution network, and to perform fault characteristic analysis and coupling verification of the AC / DC subnetwork based on the transient analysis model of the AC / DC hybrid distribution network. The AC fault analysis module is used to determine the location of the fault section based on the amplitude distribution of the low-frequency zero-sequence current when a single-phase ground fault occurs in the resonant grounded AC subgrid, while the arc suppression coil remains online; The DC fault analysis module is used to extract the zero-mode current through modulus decomposition when a single-pole grounding fault occurs in the DC subgrid, and to identify the faulty line based on the directional difference of the zero-mode current at both ends of the line. The cross-validation module is used to monitor the coupling response characteristics of the AC and DC sides of the voltage source converter, and to cross-validate them with the main characteristics in the fault subnet to determine the fault type and location.

[0016] Thirdly, a computer device is disclosed, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to perform the steps of the method described above.

[0017] Fourthly, a computer-readable storage medium is disclosed having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0018] The above one or more technical solutions have the following beneficial effects: This invention establishes a refined transient analysis model covering converters, photovoltaics, and energy storage. With the arc suppression coil remaining online, a low-frequency voltage signal is injected through the converter to significantly amplify the difference in zero-sequence current distribution of single-phase grounding faults in a resonant grounding system. Adaptive segment location can be achieved without changing the system grounding method, solving the problem of difficulty in location due to weak fault current in traditional methods. It is suitable for complex distribution network environments with a high proportion of distributed power sources.

[0019] This invention combines a two-layer analysis architecture of power frequency and low frequency networks and introduces a zero-mode current direction discrimination mechanism based on modulus decomposition. This mechanism utilizes the unidirectionality of capacitor discharge current paths to make the difference in zero-mode current direction between faulty and non-faulty lines explicit without the need for complex signal processing, overcoming the shortcomings of weakened transient characteristic differences caused by multiple feeder effects and short feeders in DC distribution networks.

[0020] This invention overcomes the limitations of existing technologies that focus only on a single subnet, providing a cross-subnet fault collaborative sensing and cross-verification scheme based on VSC port coupling response. It utilizes the DC-side second harmonic ripple generated by AC faults and the AC-side transient power impulse generated by DC faults as auxiliary features, forming redundant verification with the main features of the faulty subnet. This not only maintains high robustness under extreme conditions such as high-impedance grounding and communication interruptions, but also requires no additional hardware investment, significantly improving the confidence level and engineering practicality of fault type identification.

[0021] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0022] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0023] Figure 1 This is a schematic diagram of the basic structure of the voltage source converter according to Embodiment 1 of the present invention; Figure 2 This is a diagram of a complete transient analysis simulation model of an AC / DC hybrid distribution network built on the PSCAD platform according to Embodiment 1 of the present invention. Figure 3 This is a flowchart of the fault analysis method based on AC / DC hybrid distribution network according to Embodiment 1 of the present invention; Figure 4 This is a simplified model diagram of an AC resonant grounding distribution network according to Embodiment 1 of the present invention. Figure 5 This is a low-frequency zero-sequence network diagram of a single-phase grounding fault according to Embodiment 1 of the present invention. Detailed Implementation

[0024] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0025] It should be noted that the terminology used herein is for the purpose of describing particular implementations only and is not intended to limit the exemplary implementations of the present invention.

[0026] Where there is no conflict, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0027] Example 1 This embodiment discloses a fault analysis method based on AC / DC hybrid distribution networks. (See attached document.) Figure 1 Figure 5 The fault analysis method based on AC / DC hybrid distribution network in this embodiment will be described in detail.

[0028] In the AC / DC hybrid distribution network involved in this invention, the AC subnetwork and the DC subnetwork are interconnected through a voltage source converter (VSC). The AC side can adopt various neutral point grounding methods such as resonant grounding, ungrounded, or low-resistance grounding, while the DC side is usually grounded with a low current. The AC distribution network with different grounding methods is coupled to the DC distribution network through the converter, forming a variety of fault scenarios.

[0029] Traditional fault analysis methods for single-type power grids are difficult to apply directly to this complex new power distribution configuration. Therefore, this invention proposes a fault characteristic analysis method for AC / DC hybrid power distribution networks, aiming to solve the following technical problems: Under non-effective grounding methods such as resonant grounding, the single-phase ground fault current is extremely weak, making it difficult for traditional overcurrent protection to operate reliably. Existing line selection methods based on symmetrical component method have poor adaptability to three-phase unbalanced lines and have large errors. Furthermore, the connection of distributed power sources further exacerbates the complexity of the transient process.

[0030] When grounding is low-resistance, the fault current and voltage change drastically, requiring protection to operate within milliseconds. When grounding is high-resistance, transient characteristics are significantly weakened, and the high-frequency ripple and harmonics generated by power electronic equipment severely interfere with fault feature extraction. In addition, the weakening of transient characteristic differences between lines under short DC feeders and the complex multi-feeder effect under multi-terminal grids further increase the difficulty of fault line identification.

[0031] In AC / DC hybrid distribution networks, AC faults can cause second harmonic ripple on the DC side voltage through the converter, while DC faults can cause transient power surges on the AC side. Traditional methods only focus on the electrical quantities of a single subnet, neglecting the effective utilization of cross-subnet coupling characteristics. This leads to insufficient robustness in fault type identification and localization under extreme conditions, such as communication interruptions, measurement anomalies, and high-resistance grounding.

[0032] This embodiment proposes a fault analysis method based on AC / DC hybrid distribution networks, such as... Figure 3 As shown, the specific steps include the following: Step 1: Establish a transient analysis model of the AC / DC hybrid distribution network including VSC, photovoltaic system and energy storage system. Based on the transient analysis model of the AC / DC hybrid distribution network, conduct fault characteristic analysis and coupling verification of AC / DC subnetwork. Figure 1 This is a schematic diagram of the basic structure of the voltage source converter involved in this embodiment, as shown below. Figure 1 As shown, the left side of VSC represents the DC network, and the right side represents the AC network. The AC output voltage of VSC is denoted as... The modulation voltage is denoted as , The phase difference between the modulation voltage and the AC side voltage. Equivalent impedance. Active power injected by the VSC into the AC grid. and reactive power satisfy: Based on the above basic structure and working principle, it can be seen that by controlling... and This allows for the independent control of active power exchange on the AC / DC side and reactive power on the AC side, while maintaining the stability of the DC side voltage.

[0033] In AC / DC hybrid distribution networks, the output characteristics of distributed photovoltaic (PV) systems directly affect the capacitor discharge process and transient current contribution during DC-side faults. The equivalent mathematical model of a PV array is as follows:

[0034] in, , These represent the output current and voltage of the photovoltaic array, respectively. For photocurrent, It is the reverse saturation current. The electron charge is 1.602 × 10⁻⁶. -19 C, This is the diode characteristic factor. is the Boltzmann constant, with a value of 1.381 × 10⁻⁶. -23 J / K, For temperature, , These are the equivalent series resistance and the parallel resistance, respectively. , These represent the number of photovoltaic cells connected in series and the number of cells connected in parallel, respectively.

[0035] This model can accurately reflect the current-voltage characteristics of photovoltaic arrays under different light and temperature conditions, providing a basis for calculating the charging and discharging process of photovoltaic side capacitors in fault analysis.

[0036] In AC / DC hybrid distribution networks, energy storage units play a crucial role in regulating the output performance of intermittent distributed power sources, ensuring the power demand of critical loads, and maintaining the stability of grid voltage and frequency.

[0037] When a unipolar ground fault occurs on the DC side, the energy storage system feeds current to the fault point through its DC / DC converter. Its response characteristics are influenced by the control strategy of the battery management system. This embodiment uses an equivalent circuit with a controlled voltage source and series internal resistance to model a single lithium battery. The single lithium battery model uses a controlled voltage source and a series resistor. Indicates terminal voltage The state of charge (SOC) satisfies:

[0038]

[0039] in, This refers to the internal potential of the battery. Polarization voltage, For battery capacity, , These are constants related to battery characteristics. This is the initial value of SOC. This refers to the charging and discharging current of the battery.

[0040] Based on the modeling of key components, a complete transient analysis and simulation model of an AC / DC hybrid distribution network was built using the PSCAD platform, such as... Figure 2 As shown, a simulation model of the distribution network system is used to study the steady-state operating characteristics of the AC / DC distribution network system.

[0041] This embodiment establishes a transient analysis model for AC / DC hybrid distribution networks, encompassing converters, photovoltaic systems, and energy storage systems. Under a unified framework, it conducts collaborative analysis of the fault transient characteristics of AC and DC subnetworks, providing a systematic analytical foundation for fault diagnosis of AC / DC hybrid distribution networks.

[0042] Step 2: When a single-phase ground fault occurs in the resonant grounded AC subgrid, while the arc suppression coil remains online, a low-frequency voltage signal is injected using a voltage source converter, and the fault location is determined based on the amplitude distribution of the low-frequency zero-sequence current.

[0043] The AC subgrid uses resonant grounding, meaning the neutral point is grounded via an arc suppression coil. For ease of analysis, the influence of distributed photovoltaic and energy storage is initially ignored, and a system is established as follows: Figure 4 The simplified model of the AC resonant grounding distribution network is shown.

[0044] According to Thevenin's theorem, the neutral point voltage to ground... This can be expressed as:

[0045] in: f is the system operating frequency, from which the inductive current generated by the arc suppression coil can be further calculated. for:

[0046] Capacitive current generated by line-to-ground capacitance for:

[0047] and The vector sum is the fault entry point current. :

[0048] To reduce grounding current and extinguish the arc while preventing series resonance overvoltage, the tap position of the arc suppression coil is determined based on the power frequency current, employing an overcompensation method with an overcompensation degree p of 5% to 10%. The corresponding calculation formula is:

[0049] Therefore, the inductance value L of the arc suppression coil can be determined:

[0050] As can be seen from the above, in a power frequency network, the inductive current generated by the arc suppression coil... With the total network capacitor current The amplitudes are very close, differing only in overcompensation. This means that the fault current is extremely weak, and traditional methods of locating faults based on steady-state quantities at power frequency are very likely to fail.

[0051] To overcome the weakness of fault characteristics in power frequency networks, this embodiment proposes to utilize the powerful signal modulation capability of the VSC converter to actively superimpose an additional three-phase symmetrical low-frequency voltage signal onto the AC system. .

[0052] In theory, a larger zero-sequence current amplitude is more beneficial for signal detection and fault location. However, excessive ground current can aggravate system interference and may affect the self-extinguishing of the arc at the grounding point. Therefore, a balance needs to be struck between detectability and arc extinguishing requirements.

[0053] When a high-resistance ground fault occurs, the low-frequency current flowing into the ground is primarily determined by the additional voltage and transition resistance, and is minimally affected by the injection frequency and the system capacitance current level. However, when a low-resistance ground fault occurs, the larger the distributed capacitance of the entire network and the lower the injection signal frequency, the larger the amplitude of the low-frequency current flowing into the ground, making it easier to detect and identify. But if the injection frequency is too low, it will lead to an excessively large current amplitude at the grounding point, which is not conducive to arc extinguishing.

[0054] Taking the above factors into consideration, this embodiment selects a frequency of 25Hz and an amplitude equal to the system's rated power frequency voltage, provided that the current detection technology has sufficiently high accuracy. 0.5% is used as an additional low-frequency voltage signal.

[0055] According to the principle of circuit superposition, the voltage or current response of any branch in the distribution network will be equal to the algebraic sum of the responses of each independent source acting alone. The system can be divided into two independent sub-networks, the power frequency network and the low frequency network, for analysis.

[0056] In low-frequency networks, the inductive reactance of the arc suppression coil decreases as the frequency decreases, while the line capacitive reactance increases, significantly amplifying the difference between the two. In extreme cases, when... When the value equals 0, the low-frequency network evolves into a direct grounding mode, and the fault characteristics are extremely obvious.

[0057] After a single-phase ground fault occurs, the low-frequency zero-sequence network of the resonant grounding system is as follows: Figure 5As shown, in low-frequency networks, the zero-sequence current flowing through the detection point of a healthy line, and the zero-sequence current flowing through the detection point downstream of the fault point on a faulty line, are both less than the capacitive current of the entire network at the corresponding frequency. :

[0058] remember Let the zero-sequence current be the detection point downstream of any healthy line or fault point, then:

[0059] Further analysis of the relative relationship between the total network capacitor current and inductor current reveals that the absolute magnitudes of the total network capacitor current and inductor current increase with the transition resistance R. f The value changes, and the relative magnitude of the two depends only on the power frequency overcompensation degree and the frequency:

[0060] Where k is the ratio of the power frequency to the injection frequency, k>1, meaning the ratio of the injection frequency to the power frequency of the low-frequency voltage signal is less than 1, and the injection frequency is a low-frequency signal less than 50Hz. Therefore, the zero-sequence current flowing through each detection point from the upstream of the fault point on the faulty line to the outlet of the main transformer can be obtained. With the total network capacitor current The inequality relationship is:

[0061] Therefore, the larger the value of k, the better. same The more obvious the difference, the more the low-frequency full network capacitor current is used as the benchmark. The section where the measured zero-sequence current is less than the benchmark is judged as the non-fault section or the downstream of the fault point. The section where the measured zero-sequence current is between the upper and lower limits determined by the overcompensation degree of the arc suppression coil and the ratio of the injection frequency to the power frequency is judged as the upstream section of the fault point.

[0062] At this point, a zero-sequence current loop with a large amplitude, mainly composed of inductive current, will be formed between the grounding point and the arc suppression coil. and The inherent relative amplitude relationship is used to measure the low-frequency zero-sequence current amplitude of each local unit, with the total network capacitive current as the basis. Based on this, an adaptive threshold value is generated online. By comparing the relationship between the measured current at each detection point and the threshold value, the section location of a single-phase grounding fault in the resonant grounding system can be adaptively achieved through mutual comparison.

[0063] In AC / DC hybrid distribution networks, the fault transient steady-state characteristics of distributed generation (DG) are closely related to the control strategies of its grid-connected converters. For example, low-voltage ride-through control, negative-sequence current suppression control, and current limiting mechanisms all complicate the fault transient process and exhibit multi-timescale evolution characteristics. Theoretically, such complex characteristics require more precise mathematical models to fully describe, posing challenges to traditional fault analysis.

[0064] However, due to engineering requirements such as power quality control, relay protection coordination, and overvoltage suppression, various distributed power sources are typically electrically connected to the distribution system through grid-connected transformers. According to the "Distributed Power Source Grid-Connected Operation Control Specification," the neutral point grounding method of the grid-connected transformer should be compatible with the grounding method of the connected power grid. Specifically, when a distributed power source is connected to a distribution network with a non-effectively grounded neutral point (including resonant grounding), its grid-connected transformer must adopt a system-side delta connection.

[0065] Under this wiring configuration, the zero-sequence current on the low-voltage side (DG side) of the transformer cannot cross to the high-voltage side (grid side), meaning the zero-sequence network path of the grid is naturally blocked by the transformer's delta winding. Therefore, regardless of whether distributed generation is connected, the amount of connected capacity, or the control strategy employed, the power frequency zero-sequence network and low-frequency zero-sequence network of the AC system remain unchanged in topology.

[0066] Based on the above analysis, the fault characteristic analysis and section location method based on low-frequency zero-sequence current amplitude distribution proposed in this embodiment relies solely on the zero-sequence network structure of the system as its core criterion, and is unaffected by the nonlinear control strategy of the DG grid-connected converter. Therefore, this method can effectively avoid the complex transient interference brought about by distributed energy access, and maintain good adaptability and reliability even in scenarios with a high proportion of distributed power access.

[0067] Step 3: When a single-pole grounding fault occurs in the DC subgrid, the zero-mode current is extracted by modulus decomposition, and the faulty line is identified based on the directional difference of the zero-mode current at both ends of the line.

[0068] In the DC subnetwork, this embodiment focuses on the characteristic analysis of single-pole grounding faults under low-current grounding conditions. Since the positive and negative lines of the DC distribution network are laid in parallel, there are electromagnetic and capacitive couplings between them. When a single-pole grounding fault occurs, an induced fault electrical component will be generated on the non-faulty pole line, resulting in an asymmetry of the positive and negative electrical quantities.

[0069] To eliminate the coupling effects between lines, this invention employs modulus decomposition to decouple the positive and negative current signals into independent one-mode and zero-mode components. Modulus decomposition uses a preset decoupling matrix to transform the positive and negative currents. The zero-mode current is equal to the sum of the positive and negative currents multiplied by a fixed coefficient. The decoupling matrix is ​​shown below:

[0070] in, , These correspond to the positive and negative electrical quantities of a DC line, respectively. , These correspond to the first modulus and the zero modulus, respectively. For decoupling matrix, This is the decoupling inverse matrix. Based on the above formula, the formula for calculating the zero-mode current of the line is:

[0071] In the formula, Zero-mode current, , These are the positive and negative currents.

[0072] When the distribution network is operating normally, the positive and negative currents are equal in magnitude and opposite in direction, therefore There is no zero-mode current in the circuit.

[0073] After a single-pole grounding fault occurs, the voltage of the faulty pole capacitor forms a fault loop with the line, the fault point, and the grounding resistance at the midpoint of the capacitor. The discharge current of the faulty pole capacitor and the discharge current of the line capacitor cause changes in the zero-mode current at both ends of the line.

[0074] Let the positive and negative currents before the fault be respectively , (satisfy After the fault, the changes in positive and negative currents were respectively , Then, the zero-mode current of the line after the fault can be expressed as:

[0075] in, , The currents at the positive and negative poles of the line before the fault occurred. This represents the change in current in the positive terminal circuit. This represents the change in current in the negative pole line.

[0076] Further simulation analysis reveals that after a single-pole grounding fault, the direction and magnitude of the steady-state zero-mode current at both ends of the line are determined by the capacitor discharge current. Since the direction of the capacitor discharge current is fixed, the zero-mode currents at both ends of the faulty line are in the same direction, while those at both ends of the non-faulty line are in opposite directions. Therefore, fault line identification is achieved based on the difference in the direction of the zero-mode currents at both ends of the line. Specifically, the directions of the zero-mode currents measured at both ends of the same line are compared; if the directions are the same, the line is identified as faulty; if the directions are opposite, the line is identified as non-faulty.

[0077] It should be noted that the amplitude of the capacitor discharge current decreases as the transition resistance increases. Under low-resistance fault conditions, the zero-mode current amplitude is large and its direction is clear, making the above criterion highly reliable. However, under high-resistance fault conditions, the discharge current is weak, resulting in a small zero-mode current amplitude at both ends of the line, and the direction determination may be affected by system noise and measurement errors.

[0078] In response to this situation, this embodiment does not rely on a single zero-mode current direction criterion, but instead uses a cross-validation mechanism of coupling response characteristics in subsequent steps for auxiliary judgment.

[0079] Step 4: Monitor the coupling response characteristics of the AC and DC sides of the voltage source converter, cross-verify them with the main characteristics in the faulty subnet, and determine the fault type and location.

[0080] After completing the fault characteristic analysis of the AC subnet and DC subnet respectively, this embodiment further utilizes the voltage source converter (VSC) as the information fusion point. By monitoring its coupling response characteristics on both the AC and DC sides during the fault, cross-validation is performed with the main features extracted from the fault subnet, thereby achieving accurate determination of the fault type and location.

[0081] When a single-phase ground fault occurs in the AC subnet, a zero-sequence component will appear in the AC side voltage. Although the VSC usually isolates the zero-sequence path through a transformer, the voltage asymmetry caused by the fault may still cause the negative-sequence current suppression circuit in the VSC controller to activate, or generate second harmonic ripple (100Hz) in the DC side voltage of the VSC.

[0082] The DC-side voltage signal of the VSC is sampled and first processed by a second-order low-pass filter with a cutoff frequency of 150Hz and a notch filter to filter out switching frequency ripple and high-frequency noise. Then, an adaptive bandpass filter with a center frequency locked at 100Hz is used to extract the 100Hz component (i.e., the second harmonic component) in the DC voltage ripple. Finally, the amplitude sequence of this component is output by Fourier transform or RMS calculation.

[0083] Therefore, monitoring the voltage ripple component at this specific frequency on the DC side can serve as supplementary evidence of an asymmetrical fault on the AC side. This coupling characteristic can provide important redundancy information, especially when communication is interrupted or AC-side measuring devices fail.

[0084] When a single-pole ground fault occurs on the DC side, the DC side voltage of the VSC will change abruptly. To maintain DC voltage stability, the VSC control system will rapidly adjust the active power output on its AC side. This rapid change in power will generate a transient impulse component in the AC side current, covering the inner loop control bandwidth of the VSC current.

[0085] Therefore, the three-phase currents at the VSC AC port are synchronously sampled to synthesize the magnitude of the space vector current to eliminate fundamental frequency component interference. A 5-level discrete wavelet decomposition of this magnitude is performed using the db4 wavelet basis function to extract the 3rd and 4th level detail coefficients corresponding to the VSC control bandwidth. The wavelet energy of these detail coefficients within the first power frequency cycle after the fault is calculated as a transient energy index. By quantifying the sudden increase in this transient energy relative to the steady-state state, the severity of the DC-side fault can be characterized.

[0086] When the DC fault transition resistance is small, this energy surge can be several times higher; as the transition resistance increases, the surge decreases accordingly, thus providing a DC fault redundancy identification method based on AC side response for AC-DC hybrid systems.

[0087] To achieve more intelligent cross-validation, this embodiment also pre-establishes a coupling response feature database. Using the transient analysis model constructed in step 1, simulations are performed on fault scenarios with different fault locations, types, and transition resistances. The coupling response features of key electrical quantities on both the AC and DC sides of the VSC are extracted, including but not limited to the amplitude of the DC side second harmonic ripple, the frequency band energy of the AC side transient power impulse, and the characteristic decay time constant, forming a standardized feature database. During online analysis, the measured coupling features are quickly matched and compared with the database, further improving the accuracy of fault type identification and section location.

[0088] The main characteristics within the fault subnet, such as zero-sequence voltage / current anomalies on the AC side and voltage mutations / zero-mode current non-zero on the DC side, coupled with the VSC port coupling characteristics, such as second harmonic ripple on the DC side and transient power surges on the AC side, constitute complementary fault information pairs.

[0089] When the primary feature and the coupled feature point to the same value, the judgment result of the fault type and location can be highly confident; when one type of feature is missing due to measurement or communication reasons, the other type of feature can still provide reliable redundant judgment basis. This cross-validation mechanism significantly improves the robustness and reliability of this method under complex operating conditions.

[0090] Example 2 Based on the method of Embodiment 1, the purpose of this embodiment is to provide a fault analysis system based on an AC / DC hybrid distribution network, including: The model building module is used to establish transient analysis models for AC / DC hybrid distribution networks. The AC fault analysis module is used to determine the location of the fault section based on the amplitude distribution of the low-frequency zero-sequence current when a single-phase ground fault occurs in the resonant grounded AC subgrid, while the arc suppression coil remains online; The DC fault analysis module is used to extract the zero-mode current through modulus decomposition when a single-pole grounding fault occurs in the DC subgrid, and to identify the faulty line based on the directional difference of the zero-mode current at both ends of the line. The cross-validation module is used to monitor the coupling response characteristics of the AC and DC sides of the voltage source converter, and to cross-validate them with the main characteristics in the fault subnet to determine the fault type and location.

[0091] Example 3 The purpose of this embodiment is to provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the above-described method.

[0092] Example 4 The purpose of this embodiment is to provide a computer-readable storage medium.

[0093] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the steps of the above-described method.

[0094] The steps and methods involved in the apparatus of the above embodiments correspond to those in Embodiment 1. For specific implementation details, please refer to the relevant description section of Embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood as including any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and enabling the processor to perform any of the methods in this invention.

[0095] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. The present invention is not limited to any particular combination of hardware and software.

[0096] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A fault analysis method based on AC / DC hybrid distribution networks, characterized in that, include: A transient analysis model of AC / DC hybrid distribution network is established. Based on the transient analysis model of AC / DC hybrid distribution network, fault characteristics analysis and coupling verification of AC / DC subnetwork are carried out. When a single-phase ground fault occurs in the resonant grounded AC subgrid, while the arc suppression coil remains online, a low-frequency voltage signal is injected using a voltage source converter, and the fault section is determined based on the amplitude distribution of the low-frequency zero-sequence current. When a single-pole grounding fault occurs in the DC subgrid, the zero-mode current is extracted by modulus decomposition, and the fault line is identified based on the difference in the direction of the zero-mode current at both ends of the line. Monitor the coupling response characteristics on both the AC and DC sides of the voltage source converter, cross-verify them with the main characteristics within the faulty subnet, and determine the fault type and location.

2. The fault analysis method based on AC / DC hybrid distribution network as described in claim 1, characterized in that, The method of injecting a low-frequency voltage signal using a voltage source converter while keeping the arc suppression coil online includes: using the signal modulation capability of the voltage source converter to superimpose a three-phase symmetrical low-frequency voltage signal onto the AC system.

3. The fault analysis method based on AC / DC hybrid distribution network as described in claim 1, characterized in that, The ratio of the injection frequency of the low-frequency voltage signal to the power frequency is less than 1.

4. The fault analysis method based on AC / DC hybrid distribution network as described in claim 1, characterized in that, The method of determining the fault section location based on the amplitude distribution of low-frequency zero-sequence current specifically includes: using the low-frequency full-network capacitor current as a benchmark, determining the section with the measured zero-sequence current less than the benchmark as a non-fault section or downstream of the fault point, and determining the section with the measured zero-sequence current between the upper and lower limits determined by the overcompensation degree of the arc suppression coil and the ratio of the injection frequency to the power frequency as the upstream section of the fault point.

5. The fault analysis method based on AC / DC hybrid distribution network as described in claim 1, characterized in that, The modulus decomposition uses a preset decoupling matrix to transform the positive and negative currents. The zero-mode current is equal to the sum of the positive and negative currents multiplied by a fixed coefficient.

6. The fault analysis method based on AC / DC hybrid distribution network as described in claim 1, characterized in that, The method of identifying faulty lines based on the difference in the direction of zero-mode current at both ends of the line specifically includes: comparing the directions of the zero-mode current measured at both ends of the same line; if the directions at both ends are the same, the line is determined to be faulty; if the directions at both ends are opposite, the line is determined to be non-faulty.

7. The fault analysis method based on AC / DC hybrid distribution network as described in claim 1, characterized in that, Also includes: Using the transient analysis model, a database of coupling response characteristics of key electrical quantities on both the AC and DC sides of the voltage source converter is pre-established for fault scenarios with different fault locations, types, and transition resistances. During online analysis, the measured coupling characteristics are matched and compared with the database.

8. A fault analysis system based on an AC / DC hybrid distribution network, characterized in that, include: The model building module is used to establish a transient analysis model of the AC / DC hybrid distribution network, and to perform fault characteristic analysis and coupling verification of the AC / DC subnetwork based on the transient analysis model of the AC / DC hybrid distribution network. The AC fault analysis module is used to determine the location of the fault section based on the amplitude distribution of the low-frequency zero-sequence current when a single-phase ground fault occurs in the resonant grounded AC subgrid, while the arc suppression coil remains online; The DC fault analysis module is used to extract the zero-mode current through modulus decomposition when a single-pole grounding fault occurs in the DC subgrid, and to identify the faulty line based on the directional difference of the zero-mode current at both ends of the line. The cross-validation module is used to monitor the coupling response characteristics of the AC and DC sides of the voltage source converter, and to cross-validate them with the main characteristics in the fault subnet to determine the fault type and location.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the fault analysis method based on AC / DC hybrid distribution network as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it performs the steps of the fault analysis method based on AC / DC hybrid distribution network as described in any one of claims 1-7.

Images