Fault feature analysis method for connecting large-scale distributed power supply to rural power distribution network

Abstract

The invention provides a large-scale distributed power supply access rural power distribution network fault feature analysis method, which comprises the following steps of 1, analyzing rural power distribution network fault feature types, including short-circuit fault features, small-current grounding fault features and arc light high-resistance fault features; 2, establishing a fault equivalent model of the distributed power supply; and step 3, analyzing the fault characteristics of the rural power distribution network containing distributed power supply access. According to the scheme, the influence of large-scale distributed power supply access on the short-circuit fault characteristics of the power distribution network is analyzed, and the influence of various influence factors such as the type, the position and the fault impedance of the distributed power supply on the amplitude, the phase and the like of fault current is systematically analyzed and summarized; the limitation of a traditional short-circuit fault differential protection algorithm applied to a distributed power supply access scene is discussed, and a foundation is laid for follow-up related research of a rural power distribution network.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network fault analysis, in particular to a method for analyzing fault characteristics of a large-scale distributed power supply connected to a rural distribution network. Background Art

[0002] my country's vast rural and mountainous distribution networks feature long power lines, wide distribution areas, poor network carrying capacity, and low automation levels. These networks severely restrict the integration of large-scale distributed renewable energy and pose a threat to system power supply reliability. Furthermore, these weak distribution networks, characterized by weak network topology, weak distribution lines, and poor communication conditions, make line maintenance difficult, failure rates high, and repairs challenging, making fault location and rapid resolution challenging. Furthermore, asymmetric delays and random information loss in power carrier / 4G / 5G communications hinder the use of globally synchronized measurement information for fault diagnosis. With the high penetration of renewable energy, weak distribution networks in rural and other areas face diverse fault scenarios and varying fault characteristics, making it difficult to obtain comprehensive global fault information. This severely limits the emergency response and power supply protection capabilities of these networks and threatens regional power supply reliability. Therefore, analyzing the changing trends in fault characteristics associated with the integration of large-scale distributed power sources into weak distribution networks in rural and mountainous areas is crucial to provide a foundation for subsequent research. With the influx of distributed power sources, primarily wind and photovoltaic power, the power flow distribution, fault characteristics, and signal components of the distribution network will undergo significant changes, posing significant challenges to network operation control and fault diagnosis. First, the integration of distributed power sources (DGs) has transformed the traditional single-source radial structure of distribution networks into a more complex dual-source or even multi-source structure. Fault currents are multi-source, bidirectional, and low-amplitude, limiting the coordination of traditional fault protection schemes and the application of feeder automation. Second, power electronics devices cause signals to exhibit strong harmonic and interharmonic characteristics, making fault signature extraction difficult and reducing the reliability of fault diagnosis. Furthermore, active distribution networks are noisy, and power electronics operate nonlinearly, making weak fault signatures easily drowned out by noise. All of these factors hinder the analysis of fault signatures in rural distribution networks. Summary of the Invention

[0003] In view of this, the purpose of the present invention is to provide a method for analyzing the fault characteristics of large-scale distributed power supply access to rural distribution networks, analyze the impact of large-scale distributed power supply access on the short-circuit fault characteristics of the distribution network, systematically analyze and summarize the impact of various influencing factors such as distributed power supply type, location, fault impedance, etc. on the fault current amplitude, phase, etc., and discuss the limitations of the traditional short-circuit fault differential protection algorithm used in the distributed power supply access scenario, laying the foundation for subsequent related research on rural distribution networks.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: a method for analyzing the fault characteristics of large-scale distributed power sources connected to a rural distribution network, comprising the following steps:

[0005] Step 1: Analyze the fault characteristics of rural distribution networks, including short-circuit fault characteristics, low-current grounding fault characteristics, and arc high-resistance fault characteristics;

[0006] Step 2: Establish a fault equivalent model for distributed power sources, including an inverter-type distributed power source fault equivalent model and a small hydropower fault equivalent model that takes into account the excitation control response;

[0007] Step 3: Analyze the fault characteristics of the rural distribution network with distributed generation access, including the analysis of short-circuit fault characteristics after distributed generation access, the analysis of small current grounding fault characteristics after distributed generation access, and the impact assessment of large-scale distributed generation access on traditional fault protection methods.

[0008] In a preferred embodiment, in step 1, the short circuit fault characteristics include three-phase short circuit, two-phase short circuit, single-phase ground short circuit, and two-phase ground short circuit;

[0009] In a three-phase short circuit, for the steady-state current characteristics, the approximate calculation formula for the effective value of the three-phase short-circuit current of the rural distribution line is:

[0010]

[0011] Where U N is the system rated voltage; c is the voltage coefficient; cU N is the system equivalent voltage source voltage; Z s1 is the system positive sequence impedance after the medium voltage busbar of the substation; Z L1 R is the positive sequence impedance of the circuit formed by the line between the substation busbar and the fault point and the ground; k is the fault resistance;

[0012] Record the rated short-circuit capacity S at the busbar k With rated line voltage U LP , the system positive sequence impedance is calculated as:

[0013]

[0014] The three-phase short-circuit current consists of a steady-state component and a transient component. The non-periodic component in the transient short-circuit current makes the effective value of the transient three-phase short-circuit current greater than the steady-state effective value. The expression of the transient three-phase short-circuit current is:

[0015]

[0016] Where: I m is the current amplitude when the system is operating normally; I pm is the current amplitude of the short-circuit cycle; α is the phase angle of the power supply voltage; It is the angle between the current and the circuit voltage when the system is operating normally; is the phase angle between the short-circuit current and the loop voltage; T a is the decay time constant of the non-periodic component current; the expression of the short-circuit current periodic component is:

[0017]

[0018] The expression of the non-periodic component of short-circuit current is:

[0019]

[0020] The three-phase short-circuit impulse current is related to the short-circuit phase angle and the grid time constant. The smaller the short-circuit phase angle, the larger the time constant and the higher the impulse current amplitude.

[0021] In a two-phase short circuit, the positive sequence impedance and negative sequence impedance of the three-phase symmetrical line are equal; the medium voltage distribution network is far away from the system power supply, and the approximate calculation formula for the effective value of the two-phase short circuit current is:

[0022]

[0023] If the transition resistance is zero, then

[0024]

[0025] In a single-phase ground short circuit, the effective value of the short-circuit current is simplified to

[0026]

[0027] R n The neutral point grounding resistance of the main transformer;

[0028] In a two-phase ground short circuit, the short-circuit currents of the two fault phases are equal, and the effective value calculation formula is:

[0029]

[0030] Where: Z1=Z S1 +Z L1 , Z0=Z S0 +Z L0 , a=e 120j is the operation factor;

[0031] When two phases are short-circuited to ground, the effective value of the grounding current at the fault point is

[0032]

[0033] In a low-resistance grounded distribution network, Z0>>Z1, the above two equations are further simplified to

[0034]

[0035] This shows that when a two-phase ground short circuit occurs in a low-resistance grounded distribution network, the short-circuit current of the fault phase is basically the same as that when a two-phase short circuit occurs.

[0036] In a preferred embodiment, in step 1, the small current grounding fault characteristics include the steady-state characteristics of the single-phase grounding fault in the neutral point ungrounded system, the steady-state characteristics of the single-phase grounding fault in the neutral point resonant grounded distribution network, and the transient characteristics of the small current grounding fault.

[0037] The steady-state characteristics of a single-phase grounding fault in a neutral point ungrounded system include:

[0038] 1) When single-phase grounding occurs, the voltage relative to the fault drops to zero, and the voltage relative to the non-fault increases to the original value. times, which is the line voltage, and zero-sequence voltage appears in the entire system; 2) The zero-sequence current of the non-fault line is equal to the sum of the three-phase capacitance current to ground, and the direction of capacitive reactive power flows from the busbar to the line; 3) The zero-sequence current of the fault line is the sum of the capacitance current to ground of all non-fault lines, and the direction of capacitive reactive power flows from the line to the busbar; 4) The zero-sequence current of the non-fault line leads the zero-sequence voltage by 90°, and the zero-sequence current of the fault line lags the zero-sequence voltage by 90°;

[0039] The steady-state characteristics of a single-phase ground fault in a neutral-grounded resonant distribution network include: The condition of the non-fault line is the same as that of a neutral-ungrounded system; the zero-sequence current is equal to the sum of the three-phase capacitive currents to ground during normal operation, and the capacitive reactive power flows from the busbar to the line; the zero-sequence current of the fault line is no longer equal to the sum of the zero-sequence currents of the non-fault lines, and its capacitive reactive power flows from the busbar to the line, just like in the non-fault line.

[0040] The transient characteristics of a low-current grounding fault include: the initial fault phase angle, the grounding point transition resistance, and the feeder parameters. These three conditions affect the transient characteristics of the zero-mode current of a low-current grounding fault, and their effects on the fault transient current are coupled. The transient current of a single-phase grounding fault is divided into capacitive and inductive components. At the onset of the fault, the phase determines the ratio of the capacitive transient to the inductive transient, which in turn affects the amplitude of the transient current. When the fault occurs at the instant when the phase voltage approaches its maximum value, within the first transient half-wave, the transient zero-mode voltage has the opposite polarity to the zero-mode current of the fault line, but the same polarity as the transient zero-mode current of the non-fault line. In a neutral-ungrounded distribution network, the transient zero-mode current of the fault line flows from the line to the busbar, while the transient zero-mode current of the non-fault line flows from the busbar to the line. The derivatives of the transient zero-mode current and the zero-mode voltage of the fault line are always opposite in polarity, while the derivatives of the transient zero-mode current and the zero-mode voltage of the non-fault line are always the same in polarity.

[0041] In a preferred embodiment, in step 1, the arc high-resistance fault characteristics include steady-state characteristics—phase current, power frequency amplitude and phase of voltage, steady-state characteristics—power frequency zero-sequence voltage, power frequency amplitude of zero-sequence current, steady-state characteristics—low-order harmonic amplitude and phase, steady-state characteristics—randomness, transient characteristics—transient high-frequency signal, and transient characteristics—waveform morphology.

[0042] In a preferred embodiment, in step 2, the inverter-type distributed power supply fault equivalent model includes a distributed power supply fault analysis model considering constant power control and a distributed power supply fault analysis model considering low voltage ride-through;

[0043] In the distributed power supply fault analysis model considering constant power control, PQ control is used when the distributed power supply inverter is connected to the distribution network and operated in grid-connected mode. PQ control controls the active power and reactive power output of the inverter. Under normal operating conditions, it maintains a constant active power output while also having a certain reactive power regulation capability. Under PQ control, the output voltage and frequency are determined by the grid.

[0044] During grid-connected operation, the AC side of the inverter obtains the grid current and voltage, performs Park transformation, and decouples the current and voltage into active and reactive power to obtain the instantaneous active and reactive power output by the distributed power supply.

[0045]

[0046] Since the d-axis coincides with the grid-connected point voltage vector in Park transformation, we have

[0047]

[0048] Deduced:

[0049]

[0050] i dref and i qref is the given active current and reactive current, i d and i q It is the active component and reactive component of the inductor current. After the feedback current is compared with the given current, the SPWM modulation wave is obtained through PI regulation and inverse DQ transformation. The SPWM output controls the switch off to form a closed-loop control system, thereby controlling the inverter output current, and thus controlling the output active power and reactive power. The phase-locked voltage is the grid voltage u a 、u b 、u c , the phase-locked loop provides angle reference for DQ conversion and inverse DQ conversion;

[0051] In the analysis model of distributed generation faults with low voltage ride-through, from the time the fault occurs until the voltage recovers to 0.9 pu, the reactive power output by the DG should track the voltage change at the grid connection point and meet the following requirements:

[0052]

[0053] Among them, u g.f is the per-unit voltage value of the distributed power access point after a fault occurs, I q.f is the reactive current value output by the distributed power supply after the fault, I N is the rated current value; in actual engineering, in order to provide sufficient reactive power support, u g.f When it is in the interval [0.2, 0.9], the coefficient is 1.5; therefore, the direct grid-connected distributed power supply considering low voltage ride-through is equivalent to a voltage-controlled current-controlled source;

[0054] Considering the inverter's overcurrent capability and low voltage ride-through control, the output active current setting is:

[0055]

[0056] Among them, I d.f is the active power output when the system fails, P m is the active power output at the fault point, I max is the maximum current allowed to flow through the inverter. Therefore, considering the LVRT control strategy and inverter current limiting constraints during distribution network faults, as well as the provision of greater reactive power support when the voltage drops significantly, the relationship between the inverter output fault current and the grid connection point voltage is as follows:

[0057]

[0058] The active component of the fault current is superimposed with the grid connection point voltage vector, and δ represents the phase angle of the positive sequence voltage at the grid connection point;

[0059]

[0060] From the following formula, we can see that there is a boundary voltage u x Make the inverter output current just reach the output current amplitude limit; that is:

[0061]

[0062] Let P m =1,I max =2I N , then u x =0.537, the inverter reaches the current limiting constraint boundary; at this time, the phase angle of the fault current is:

[0063]

[0064] When u g.f When it is in the interval [0.537, 0.9], the inverter can adjust the reactive power output while ensuring that the active power output remains unchanged, and is in the grid-connected control stage; when u g.f When the inverter is in the interval [0.2, 0.537], it is constrained by the control strategy and the inverter current limiting condition, and cannot guarantee the active power output remains unchanged, and is in the low voltage ride-through control stage;

[0065] When an asymmetric fault occurs, the active component of the fault output current is oriented toward the positive sequence voltage component of the grid connection point, that is, the inverter grid-connected distributed power supply is equivalent to a current source controlled by the positive sequence fault voltage of the grid connection point, that is:

[0066]

[0067] In a preferred embodiment, in step 2, the small hydropower fault equivalent model taking into account the excitation control response includes a simplified excitation system model and a small hydropower VCCS fault equivalent model;

[0068] In the simplified model of the excitation system, U ref is the set terminal reference voltage; Ut is the terminal voltage; ΔU is the terminal voltage deviation value; E f is the stator excitation potential output to the generator; T A 、T B1 、T B2 、T C1 and T C2 K is the lead / lag time constant of each link; R and K A is the gain multiple of each link; V Rmax is the peak value of the generator excitation potential;

[0069] Since the self-shunt static excitation system has a high response speed, the dynamic process of the excitation system can be ignored when calculating the fault current, and its excitation characteristic equation is simplified to:

[0070] E f =min(K v (U ref -U t ),V Rmax )

[0071] Where: min(·,·) is the minimum function; Kv is the gain coefficient of the excitation system, and the gain coefficient of the small hydropower unit is between 30 and 150;

[0072] In the small hydropower VCCS fault equivalent model, the steady-state d-axis and q-axis stator voltage and current relationship after the small hydropower unit fault is:

[0073]

[0074] Where: ud, uq and id, iq are the d-axis and q-axis components of the stator voltage and stator current respectively, Xd and Xq are the d-axis and q-axis components of the stator steady-state synchronous reactance respectively, Eq is the steady-state open-circuit potential, ra is the stator resistance, and the above variables are all per-unit values;

[0075] Since the stator reactance of the synchronous generator is much larger than the stator resistance, ra≈0 is assumed in the analysis. In the steady-state process after a fault, there is no current in the damping windings D and Q, and only the excitation winding potential exists on the stator q-axis. Therefore, the stator steady-state open-circuit potential is equal to the stator excitation potential Ef, that is, Eq=Ef. After making the above assumptions, it is simplified to:

[0076]

[0077] As can be seen from the above formula, the fault output current of small hydropower is related to the terminal voltage and the excitation potential. Since small hydropower is usually directly connected to the distribution network in a "T" connection mode and the connection line is very short, it is approximately assumed that the terminal voltage of the small hydropower unit is equal to the voltage UPCC at the common connection point PCC. The small hydropower unit under fault ride-through is equivalent to a VCCS model in which the fault output current IG is controlled by the voltage at the PCC.

[0078] In a preferred embodiment, in step 3, the short-circuit fault characteristic analysis after the distributed generation is connected specifically includes: 1) when a three-phase short-circuit fault occurs upstream of the feeder where the DG is located, the DG is decoupled from the system power supply side circuit, and the DG does not affect the short-circuit current supplied by the system; the connection of the DG does not affect the short-circuit current characteristics of the distribution network;

[0079] 2) When the fault is not severe and the DG is outputting active current, the grid connection point voltage is affected by the DG output power, first increasing and then decreasing as the DG output power increases. When the fault is severe and the DG is only outputting reactive current, the grid connection point voltage is affected by the DG rated power, first increasing and then decreasing as the DG rated power increases. The change pattern of the grid connection point voltage is independent of the fault type and fault location.

[0080] 3) When the fault is not severe and the DG is outputting active current, the short-circuit current flowing through the upstream feeder where the DG is located first decreases and then increases with the increase of the DG access capacity; the short-circuit positive sequence current flowing through the downstream feeder where the DG is located and the adjacent feeders first increases and then decreases with the increase of the DG access capacity; when the fault is severe and the DG is only outputting reactive current, the current flowing through all protections of the distribution network is only related to the DG rated power and has nothing to do with the DG output power;

[0081] 4) Although the common bus voltage and the current of the fault branch adjacent to the DG first increase and then decrease with the increase of the DG access capacity, the impact of the access of the adjacent line DG on the common bus voltage and the fault branch fault current is very small due to the support of the system on the common bus voltage;

[0082] 5) At the DC input end, the inverter generally adopts constant power control mode, and the output characteristics of the inverter have the characteristics of a current source; the pulse width modulation signal frequency of the inverter is several thousand hertz, and the response speed is only a few milliseconds. Therefore, ignoring the transient process of the inverter itself, the maximum output current is generally 1.2-1.5 times the rated current;

[0083] 6) When a distribution network fault occurs, the IIDG will maintain power supply to the distribution network through the inverter before it is disconnected from the grid due to short-circuit protection. The output short-circuit current is related to the specific control strategy adopted during the fault stage. Taking the more commonly used constant power control as an example, when a distribution network fault occurs, the inverter maintains its pre-fault active power and reactive power output. When an asymmetric fault occurs, the inverter first calculates the positive sequence voltage at the grid connection point, and then calculates the short-circuit current that the inverter needs to output. If the grid connection point voltage is low, the calculated target output current exceeds its maximum allowable output current. The inverter will maintain the output current amplitude at the maximum current level, and the active and reactive power output of the inverter will also be proportionally reduced. If the grid connection point voltage is lower than the set threshold value, the inverter will stop outputting current.

[0084] In a preferred embodiment, in step 3, the analysis of the impact of steady-state fault characteristics specifically includes analysis of the impact of steady-state fault characteristics and analysis of the impact of transient characteristics.

[0085] In a preferred embodiment, in step 3, the impact assessment of large-scale distributed power access on traditional fault protection methods includes phase sequence current amplitude differential protection analysis and traditional phase sequence current phase difference protection analysis;

[0086] Following the traditional scalar product braking differential protection expression, the braking equation that only reflects the current amplitude information is given as:

[0087]

[0088] In the formula, the independent variables are the fault current amplitudes on both sides; the coefficient k determines the braking characteristics and is assigned by the difference in current amplitude before and after the fault. The specific assignment method is:

[0089]

[0090] Where θ is defined as the equivalent phase angle, expressed as:

[0091]

[0092] Where, the current with subscript f corresponds to the fault current at the protection installation; the current with subscript n corresponds to the load current during normal operation;

[0093] Positive sequence current phase angle mutation and the direction of the total current phase mutation

[0094]

[0095] Where i pre is the steady-state current phasor value before the fault.

[0096] Compared with the prior art, the present invention has the following beneficial effects: The present invention analyzes in detail the fault characteristics of four typical fault types for traditional radial rural / mountainous distribution networks. For typical small current grounding faults in rural / mountainous distribution networks, the steady-state and transient characteristics of single-phase grounding faults under two modes, namely, neutral point ungrounded and grounded through arc suppression coils, are analyzed respectively. The impact of distributed power access on the fault characteristics of the distribution network is related to the type of distributed power supply, fault location, distributed power supply neutral point wiring method, distributed power supply control method, distributed power supply location, distributed power supply capacity, etc. The impact of large-scale distributed power supply access on the short-circuit fault characteristics of the distribution network is analyzed, and the influence of various influencing factors such as distributed power supply type, location, fault impedance, etc. on the fault current amplitude, phase, etc. is systematically analyzed and summarized. The limitations of the traditional short-circuit fault differential protection algorithm applied in the distributed power supply access scenario are discussed. In general, an increase in the penetration rate of motor-type DGs (especially above 50%) will reduce the reliability of traditional amplitude criteria, leading to their failure and even loss of the basis for their calibration. Phase criteria are superior to amplitude criteria, but without a fully configured measurement device, it is difficult to comprehensively consider the effects of DG switching, time-varying DG output, changes in transition resistance, and load fluctuations within the zone, and to distinguish these from fluctuations in the fault section characteristics. The DG grounding method in my country's low-current grounding system is ungrounded. The structure of the positive- and negative-sequence networks will change due to the addition of distributed generation (DGs), which will cause diagnostic criteria designed for phase components to lose their original reliability. However, the zero-sequence network is not affected by DG addition, so diagnostic criteria designed based on zero-sequence quantities will, in principle, maintain their original reliability. However, it is necessary to consider the reflection of injected harmonics in the zero-sequence network after DG addition. BRIEF DESCRIPTION OF THE DRAWINGS

[0097] Figure 1 A schematic diagram of calculating fault loop impedance according to a preferred embodiment of the present invention;

[0098] Figure 2 This is a fault composite sequence network diagram of a two-phase short circuit in a preferred embodiment of the present invention;

[0099] Figure 3 A composite sequence network diagram of a single-phase grounding fault according to a preferred embodiment of the present invention;

[0100] Figure 4 This is a composite sequence network diagram of a two-phase ground short circuit in a preferred embodiment of the present invention;

[0101] Figure 5 Schematic diagram and phasor diagram of a single-phase grounding fault in a distribution network with an ungrounded neutral point according to a preferred embodiment of the present invention, wherein (a) is a schematic diagram of a single-phase grounding fault, and (b) is a phasor diagram of three-phase voltage and current;

[0102] Figure 6 Schematic diagram of single-phase grounding current distribution in a neutral point ungrounded system according to a preferred embodiment of the present invention;

[0103] Figure 7 This is a schematic diagram of a zero-sequence equivalent network of a single-phase grounded distribution network with an ungrounded neutral point according to a preferred embodiment of the present invention;

[0104] Figure 8 The range of the fault section is determined in the preferred embodiment of the present invention.

[0105] Figure 9 Schematic diagram and phasor diagram of a single-phase grounding fault in a neutral point resonant grounded distribution network according to a preferred embodiment of the present invention, wherein (a) is a schematic diagram of a single-phase grounding fault, and (b) is a phasor diagram of three-phase voltage and current;

[0106] Figure 10 This is a schematic diagram of a single-phase grounding fault in a resonant grounding distribution network according to a preferred embodiment of the present invention;

[0107] Figure 11 A simplified composite model network diagram of a single-phase grounding fault in a resonant grounding distribution network according to a preferred embodiment of the present invention;

[0108] Figure 12 Schematic diagram of an equivalent circuit for transient analysis of a single-phase fault in a resonant grounded power distribution network according to a preferred embodiment of the present invention;

[0109] Figure 13 This is a voltage waveform diagram of an arc grounding fault in a 10kV distribution network according to a preferred embodiment of the present invention;

[0110] Figure 14 A simplified zero-sequence equivalent circuit diagram of a high-resistance grounding fault in different grounding systems according to a preferred embodiment of the present invention;

[0111] Figure 15 Figure 2 shows the low-order harmonic content of different grounding medium high-resistance faults in a preferred embodiment of the present invention, where (a) is wet soil, (b) is dry cement road, (c) is wet cement road, (d) is asphalt concrete, and (e) is the frequency spectrum distribution of different faults.

[0112] Figure 16 The measured zero-sequence current of a 10 kV system arc high-resistance fault in a preferred embodiment of the present invention, where (a) is a cement pole, (b) is dry land, (c) is dry cement, (d) is wet land, (e) is a dry asphalt road surface, and (f) is a wet cement ground.

[0113] Figure 17 Figure 1 shows the polarity relationship between the transient zero-mode voltage and the transient zero-mode current of the fault line in a preferred embodiment of the present invention, wherein (a) is a schematic diagram of the original waveforms of the transient voltage and current, and (b) is a diagram showing the relationship between the transient voltage derivative and the transient current;

[0114] Figure 18 This is an equivalent zero-sequence network diagram of a typical distribution network with n outgoing lines according to a preferred embodiment of the present invention;

[0115] Figure 19 A diagram showing the relationship between the sinusoidal component and the distortion component in the fault current of a resonant grounding system according to a preferred embodiment of the present invention;

[0116] Figure 20 10kV resonant grounding system arc high-resistance fault waveform diagram of different lines; (a) is wet soil, (b) is wet cement;

[0117] Figure 21 This is a block diagram of the inverter grid-connected outer loop control operation according to a preferred embodiment of the present invention;

[0118] Figure 22 This is a block diagram of the inverter grid-connected inner loop control operation according to a preferred embodiment of the present invention;

[0119] Figure 23 This is an IEEE standard model diagram of the self-shunt static excitation system of the preferred embodiment of the present invention;

[0120] Figure 24 This is a diagram of a fault equivalent model of a voltage-controlled current source of a small hydropower unit according to a preferred embodiment of the present invention;

[0121] Figure 25 This is a topological diagram of a 10kV rural distribution network including small hydropower in a preferred embodiment of the present invention;

[0122] Figure 26 This is a PSCAD small and medium hydropower grid-connected model diagram of a preferred embodiment of the present invention;

[0123] Figure 27 1 is a graph showing the characteristic changes of a small hydropower fault in a preferred embodiment of the present invention, where (a) is the excitation voltage, (b) is the fault output current, and (c) is the PCC point voltage.

[0124] Figure 28This is a diagram of an analysis model of an equivalent two-terminal power supply system including a distributed power supply in a preferred embodiment of the present invention;

[0125] Figure 29 A schematic diagram of the short-circuit current of a distributed power supply according to a preferred embodiment of the present invention;

[0126] Figure 30 An active power distribution network with an ungrounded neutral point and its exploded diagram according to a preferred embodiment of the present invention;

[0127] Figure 31 This is a fault line analysis diagram and exploded diagram of a preferred embodiment of the present invention;

[0128] Figure 32 An active power distribution network with a neutral point grounded via an arc suppression coil and its exploded diagram according to a preferred embodiment of the present invention;

[0129] Figure 33 This is a fault line analysis diagram and exploded diagram of a preferred embodiment of the present invention;

[0130] Figure 34 This is a diagram of a rotating DG fault model according to a preferred embodiment of the present invention;

[0131] Figure 35 The inverter DG fault model of the preferred embodiment of the present invention is

[0132] Figure 36 The impact of a single DG access on the line-mode network according to a preferred embodiment of the present invention, wherein (a) shows the impact of DG access at different locations on the line-mode network, and (b) shows the line-mode equivalent circuit of the DG and its grid-connected transformer;

[0133] Figure 37 Schematic diagram of a DG grid-connected transformer adopting a Yn / △ connection mode according to a preferred embodiment of the present invention;

[0134] Figure 38 This is the system zero-mode equivalent network when the DG grid-connected transformer adopts the Yn / △ connection mode in the preferred embodiment of the present invention. Among them, (a) is the equivalent network, and (b) is the equivalent circuit;

[0135] Figure 39 A typical power distribution network topology diagram according to a preferred embodiment of the present invention;

[0136] Figure 40 This is a diagram showing the relationship between the positive sequence current phase angle changes in criterion 1 of a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0137] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0138] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present application belongs.

[0139] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application; as used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form, and it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or their combinations.

[0140] Analysis of fault characteristics of large-scale distributed power generation connected to rural distribution networks, reference questions 1-40;

[0141] Under the goal of "carbon peak and carbon neutrality", it is proposed to accelerate the construction of a new power system with new energy as the main body, and the distribution network will also become a key link in accepting new energy. With the large-scale access of DGs mainly composed of wind power and photovoltaics, the power flow distribution, fault characteristics and signal components of the distribution network will change significantly, which brings severe challenges to the operation control and fault diagnosis of the distribution network. First, the access of DG has changed the traditional single-source radial structure of the distribution network, making it a more complex dual-source or even multi-source structure. The fault current has multi-source, bidirectional and low-amplitude characteristics, which limits the coordination of traditional fault protection schemes and feeder automation applications. Secondly, power electronic equipment causes the signal to have strong harmonic and interharmonic characteristics, making it difficult to extract fault characteristics and reducing the reliability of fault diagnosis. In addition, the active distribution network has strong noise, and the power electronic equipment has nonlinear characteristics when working. Weak fault characteristics are easily submerged by noise.

[0142] refer to Figure 1-40 This paper focuses on analyzing the types and output characteristics of distributed power sources, establishing a small hydropower fault model that takes into account the influence of excitation control methods, and discussing the steady-state and transient characteristics of short-circuit faults and small current grounding faults in rural / mountainous distribution networks with DG access. It also analyzes the impact of distributed power access on traditional protection and fault diagnosis and location.

[0143] 1 Analysis of Fault Characteristics of Rural / Mountainous Distribution Networks

[0144] 1.1 Short circuit fault characteristics

[0145] Assuming that the three-phase parameters are symmetrical and ignoring the influence of load, line distributed capacitance and its parallel compensation capacitor, the characteristics of different types of faults in a typical radial rural / mountainous topology distribution network are as follows.

[0146] (1) Three-phase short circuit

[0147] For steady-state current characteristics, the approximate calculation formula for the effective value of the three-phase short-circuit current of the rural / mountainous distribution line is:

[0148]

[0149] Where U N is the system rated voltage; c is the voltage coefficient; cU N is the system equivalent voltage source voltage; Z s1 is the system positive sequence impedance after the medium voltage busbar of the substation; Z L1 is the positive sequence impedance of the fault circuit (the circuit formed by the line between the substation busbar and the fault point and the ground); R k is the fault resistance.

[0150] The positive sequence impedance of the fault circuit is equal to the sum of the positive sequence impedances of the line sections through which the short-circuit current flows. Figure 1 Taking the radial line in as an example, when a fault occurs in k1, the positive sequence impedance of the fault loop is the sum of the positive sequence impedances between SA, AD and the line segment from node D to the fault point k1; when a fault occurs in k2, the positive sequence impedance of the fault loop is the sum of the positive sequence impedances between line segments SA, AB, BC and the line segment from node C to the fault point k2.

[0151] If the rated short-circuit capacity at the busbar is S k With rated line voltage U LP , the system positive sequence impedance can be calculated as:

[0152]

[0153] In my country, the rated short-circuit capacity of a 10kV distribution network is approximately 100 to 500MVA, the maximum effective value of the three-phase short-circuit current is approximately 6 to 30kA, and the system positive sequence impedance is approximately 1.0 to 0.2Ω. According to the Chinese national standard GB / T 15544.1-2013 "Calculation of Short-Circuit Current in Three-Phase AC Systems - Part 1: Current Calculation", the value of the voltage coefficient c when calculating the maximum and minimum short-circuit currents is: max with c min They are 1.0 / 0.95 (220 / 380V) and 1.1 / 1.0 (3~35kV) respectively.

[0154] The minimum short-circuit current occurs when a two-phase short circuit occurs, and the short-circuit current is 0.86 times the three-phase short-circuit current. When calculating the maximum and minimum short-circuit currents, the voltage coefficient c is taken as 1.10 and 1.00 respectively. Therefore, in the medium-voltage distribution network, for the same operating mode, the minimum short-circuit current should be 0.78 times the maximum short-circuit current.

[0155] The three-phase short-circuit current consists of the above-mentioned steady-state component and transient component. The non-periodic component in the transient short-circuit current makes the effective value of the transient three-phase short-circuit current greater than the steady-state effective value. The expression of the transient three-phase short-circuit current is:

[0156]

[0157] Where: I m is the current amplitude when the system is operating normally; I pm is the current amplitude of the short-circuit cycle; α is the phase angle of the power supply voltage; It is the angle between the current and the circuit voltage when the system is operating normally; is the phase angle between the short-circuit current and the loop voltage; T a is the decay time constant of the non-periodic component current. The expression of the short-circuit current periodic component is:

[0158]

[0159] The expression of the non-periodic component of short-circuit current is:

[0160]

[0161] The maximum instantaneous value of a three-phase short-circuit current occurs approximately half a cycle after the short circuit. This value is related not only to the amplitude of the periodic component but also to the starting value of the non-periodic component. In the most severe short-circuit scenario, the maximum instantaneous value of the three-phase short-circuit current is called the inrush current. The inrush current of a three-phase short-circuit is related to the short-circuit phase angle and the grid time constant. The smaller the short-circuit phase angle, the larger the time constant, and the higher the inrush current amplitude, reaching a maximum of 2.8 times the effective value of the steady-state short-circuit current.

[0162] (2) Two-phase short circuit

[0163] According to the symmetrical component method, the composite sequence network when two phases are short-circuited is as follows: Figure 2 As shown, where U p is the phase voltage of the system equivalent voltage source; Z s1 、Z s2 are the positive sequence impedance and negative sequence impedance of the system behind the medium voltage busbar of the substation; Z L1 、Z L2 are the positive sequence impedance and negative sequence impedance of the fault circuit respectively; R k is the short-circuit point transition resistance.

[0164] The positive-sequence impedance and negative-sequence impedance of the three-phase symmetrical line are equal; the medium-voltage distribution network is far away from the system power supply, and the difference between the negative-sequence impedance and the positive-sequence impedance of the system can be ignored. The approximate calculation formula for the effective value of the two-phase short-circuit current is:

[0165]

[0166] If the transition resistance is zero, then

[0167]

[0168] That is, the effective value of the two-phase metallic short-circuit current is 0.87 times the single-phase metallic short-circuit current.

[0169] (3) Single-phase grounding

[0170] The composite sequence network of a single-phase ground short circuit in a high current grounded distribution network is as follows: Figure 3 As shown, where Z s is the zero-sequence impedance of the system behind the medium-voltage busbar of the substation; Z L0 is the zero-sequence impedance of the fault line. The other parameters are the same as those of the circuit shown in the figure above.

[0171] In a neutral point directly grounded distribution network, Z S0 Equal to the main transformer zero-sequence impedance Z t0 In a low resistance grounded distribution network, there is Z S0 =3R n +Z t0 , where R n The grounding resistance of the neutral point of the main transformer is R n Relatively large, generally above 100Ω, much larger than Z t0 、Z S1 、Z L1 、Z L0 , so the effective value of the short-circuit current can be simplified to

[0172]

[0173] It shows that the single-phase short-circuit current of the low-resistance grounded distribution network mainly depends on the neutral point grounding resistance and transition resistance.

[0174] (4) Two-phase ground short circuit

[0175] The composite sequence network when two phases are short-circuited to ground is as follows Figure 4 shown.

[0176] When two phases are short-circuited to ground, the short-circuit currents of the two fault phases are equal, and the effective value calculation formula is:

[0177]

[0178] Where: Z1=Z S1 +Z L1 , Z0=Z S0 +Z L0 , a=e 120j is the operation factor.

[0179] When two phases are short-circuited to ground, the effective value of the grounding current at the fault point is

[0180]

[0181] In a low-resistance grounded distribution network, Z0>>Z1, the above two equations are further simplified to

[0182]

[0183] This shows that when a two-phase ground short circuit occurs in a low-resistance grounded distribution network, the fault phase short-circuit current is basically equal to that when a two-phase short circuit occurs, and the grounding current at the fault point is about 0.5 times the single-phase ground short-circuit current.

[0184] In general, the characteristics of short-circuit faults in general radial rural / mountainous distribution networks can be summarized as follows:

[0185] Three-phase short-circuit steady-state current increases. For example, in a 10kV rural / mountainous distribution network with a rated short-circuit capacity of approximately 100-500MVA, the maximum effective value of the three-phase short-circuit current is approximately 6-30kA. The maximum instantaneous transient current occurs approximately half a cycle after the short circuit occurs, depending on the short-circuit phase angle and the grid time constant, and can reach up to 2.8 times the effective value of the steady-state short-circuit current. The three-phase voltage decreases, and zero-sequence current and voltage disappear. In a two-phase ground fault, the current in both phases increases, the voltage decreases, and zero-sequence current and voltage appear. In a two-phase short-circuit fault, the current in both phases increases, the voltage decreases, and zero-sequence current and voltage appear. The currents in the two faulted phases are essentially in opposite phases, and the effective value of the metallic short-circuit current in both phases is 0.87 times the metallic short-circuit current in a single-phase fault. In a single-phase ground fault, the current in one phase increases, the voltage decreases, and zero-sequence current and voltage appear. The zero-sequence current is in phase with the faulted phase current, while the zero-sequence voltage is in opposite phase with the faulted phase voltage.

[0186] 1.2 Characteristics of small current grounding fault

[0187] According to the different neutral point grounding methods, the low-current grounding system can be divided into a neutral point ungrounded system and a neutral point resonant grounding system. The fault characteristics of the above two systems are analyzed respectively.

[0188] (1) Steady-state characteristics of single-phase grounding fault in neutral point ungrounded system

[0189] 1) Three-phase voltage and current characteristics

[0190] Figure 5The diagram shows a single-phase ground fault diagram and voltage-current phasor relationship diagram for a distribution network system with an ungrounded neutral point. When the system is not faulty, the three-phase lines have the same capacitance to ground, C0. Each phase flows through a capacitive current that leads the phase voltage by 90°, and the sum of the three-phase capacitive currents is zero. When a phase A ground fault occurs, the grounding capacitor of the faulty phase is short-circuited, the current flowing through the faulty phase is zero, the voltage to ground is zero, and the voltage to ground of the non-faulty phases B and C increases. times, Figure 5 (b) shows its phasor relationship.

[0191] After phase A is grounded, the three-phase voltage to ground is:

[0192]

[0193] The zero-sequence voltage at the ground fault point is the sum of the three phase-to-ground voltages:

[0194]

[0195] like Figure 6 The figure shows the fault analysis of a distribution network with an ungrounded neutral point and n outgoing lines. Assume that line L n A phase ground short circuit fault occurs, C in the figure 01 、C 0n are the capacitance of each line to ground.

[0196] Analyzing the non-fault line L1, the phase-to-ground capacitance currents flowing through L1 are:

[0197]

[0198] Similarly, the current flowing through the non-fault line L i The capacitance currents of each phase relative to ground are:

[0199]

[0200] The current flowing through the fault point is the sum of all non-fault phase-to-ground capacitance currents in the distribution network:

[0201]

[0202] Where C 0∑ It is the sum of the capacitance current per phase to ground in the distribution system.

[0203] For the fault line Ln, the capacitance current flowing through phases B and C is and The current flowing back from phase A is the sum of the capacitance currents of phases B and C of the entire system, i.e. the grounding point current. The capacitance currents of each phase to ground of the fault line Ln are:

[0204]

[0205] From the above formula, we can get that the non-fault line L i The zero-sequence current at the head end is the sum of its three-phase capacitance currents to ground, and its direction flows from the busbar to the line, with a phase leading the system zero-sequence voltage by 90°.

[0206] Fault line L n The zero-sequence current flowing through the first end is:

[0207]

[0208] It can be seen that the zero-sequence current of the fault line is equal to the sum of the ground capacitance currents of all non-fault components (excluding the fault line itself). Its direction is from the line to the bus, which is opposite to the non-fault line, and its phase lags behind the zero-sequence voltage by 90°.

[0209] According to the above analysis, the zero-sequence equivalent network in case of single-phase grounding fault is as follows: Figure 7 As shown, It is the zero-sequence virtual voltage source voltage of the grounding point, which is approximately equal to the voltage before the fault at the fault point, but opposite in direction. The zero-sequence capacitance of the power grid is equal to the capacitance relative to the ground. The series zero-sequence impedance of the line can be ignored relative to the capacitance relative to the ground.

[0210] From the above analysis, we can get the fault characteristics of a single-phase grounding fault in a distribution network with an ungrounded neutral point:

[0211] 1) When single-phase grounding occurs, the voltage relative to the fault drops to zero, and the voltage relative to the non-fault increases to the original value. times, which is the line voltage, and at the same time, zero sequence voltage appears in the whole system;

[0212] 2) The zero-sequence current of the non-fault line is equal to the sum of the three-phase capacitance currents, and the direction of capacitive reactive power is from the busbar to the line;

[0213] 3) The zero-sequence current of the fault line is the sum of the capacitance currents of all non-fault lines to ground, and the direction of capacitive reactive power is from the line to the busbar;

[0214] 4) The zero-sequence current of the non-fault line leads the zero-sequence voltage by 90°, and the zero-sequence current of the fault line lags the zero-sequence voltage by 90°.

[0215] (3) Other steady-state characteristics

[0216] The zero-sequence admittance of each section is the difference between the zero-sequence admittance at the beginning and end of the section. The zero-sequence admittance angles of the fault section and the healthy section are quite different. Figure 8 As shown in FIG, the section that satisfies the phase angle relationship of the shaded part is the fault section.

[0217] In addition, the amplitude of the zero-sequence 5th harmonic current upstream of the fault point is large, and it flows from the fault point to the busbar; the amplitude of the zero-sequence 5th harmonic current downstream of the fault point is small, and it flows from the fault point to the end of the line.

[0218] (2) Steady-state characteristics of single-phase grounding fault in neutral point resonant grounding distribution network

[0219] 1) Three-phase voltage and current characteristics

[0220] The neutral point resonant grounding method refers to connecting an inductive arc suppression coil between the neutral point and the earth. When a single-phase grounding fault occurs in the system, the inductive current in the arc suppression coil is used to compensate for the grounding capacitance current, so that the grounding point current is reduced and the arc is extinguished by itself. Figure 9 The figure shows a schematic diagram and phasor diagram of a single-phase grounding fault in a resonant grounding distribution network. During normal operation, since the neutral point voltage to ground is zero, there is no current in the arc suppression coil, which is equivalent to being disconnected. After a single-phase grounding fault, a short-circuit current forms between the grounding point and the arc suppression coil grounding point. The neutral point voltage rises to a phase voltage, which acts on the arc suppression coil and generates an inductive current. At the grounding fault point, this inductive current offsets the capacitive current at the grounding fault point, thereby reducing the current at the grounding point. The phasor diagram is shown as follows: Figure 9 As shown in (b).

[0221] Figure 10 The figure shows the fault analysis of the distribution network resonant grounding system with n outgoing lines. Assume that line L n A phase A ground short circuit fault occurs, and L is the inductance of the arc suppression coil.

[0222] For non-fault line L i For analysis, flow through L i The capacitance currents of each phase relative to ground are:

[0223]

[0224] The inductive current of the arc suppression coil passes through the fault point and returns along the fault phase. Therefore, the current flowing through the fault point is increased by an inductive component.

[0225]

[0226] The sum of the system's ground capacitance current The phases are opposite, so the current at the fault point is reduced by the current of the arc suppression coil. The compensation degree of the arc suppression coil is expressed by the compensation degree P:

[0227]

[0228] 1) Full compensation (P = 0). P = 0, that is, I L -IC∑ , the current resonant circuit operates exactly at the resonance point, and the capacitive current and the inductive current completely cancel each other out. However, operating the grid in this mode will generate resonant overvoltage, causing the neutral point voltage to increase. Therefore, full compensation is not adopted in actual applications.

[0229] 2) Undercompensation (P<0). L C∑ , the compensated ground current is still capacitive. When a component in the system is removed, full compensation may occur, so it is generally not adopted.

[0230] 3) Overcompensation (P>0). That is, I L >I C∑ When a single-phase ground fault occurs, the ground current is inductive. Because it does not cause resonant overvoltage, it is widely used. In actual practice, P is generally taken as 5%-10%. Based on this, the inductance and resistance of the arc suppression coil can be calculated as:

[0231]

[0232] r L =10%×X L =10%×2×π×f (23)

[0233] For the fault line Ln, the capacitance currents relative to ground can be obtained by the same logic:

[0234]

[0235] 2) Zero-sequence current characteristics

[0236] When a single-phase grounding fault occurs in a resonant grounding system, the zero-sequence current flowing through the head end of the non-fault line Li is the same as when the neutral point is not grounded:

[0237]

[0238] For the fault line, the current flowing through the Ln line is the current after compensation by the arc suppression coil, that is, the zero-sequence current flowing through the head end of the fault line Ln is:

[0239]

[0240] When overcompensation is used, the inductive current flowing through the fault point is greater than the system's capacitive current to ground, and the residual current after compensation becomes inductive. In this case, the zero-sequence current at the head of the fault line becomes inductive, flowing from the busbar to the line, the same direction as for non-faulty lines. Therefore, it is difficult to identify the faulty line based on the magnitude and direction of the zero-sequence current.

[0241] From this, we can conclude the similarities and differences between a resonant system with a single-phase grounding fault and a neutral point ungrounded system:​

[0242] 1) The situation of the non-fault line is the same as that of the neutral point ungrounded system. The zero-sequence current is equal to the sum of the three-phase capacitance currents to ground during normal operation. The direction of capacitive reactive power flows from the busbar to the line.

[0243] 2) The zero-sequence current of the fault line is no longer equal to the sum of the zero-sequence currents of the non-fault lines, and the direction of its capacitive reactive power becomes from the bus to the line, just like the non-fault line.

[0244] (3) Transient characteristics of small current ground fault

[0245] 1) Fault transient characteristic analysis network

[0246] The composite mode network for transient analysis of small current ground fault is similar to the composite sequence network for asymmetric component analysis, where e(t) is the power supply voltage, R k is the grounding resistance value. For line-mode networks, the system impedance is much smaller than the distributed capacitance reactance. Therefore, in the 1-mode and 2-mode loops, the influence of the distributed capacitance can be ignored. The zero-mode power supply loop is connected in series with a large arc suppression coil inductance. Therefore, the influence of the grounding transformer inductance and resistance can be ignored. Thus, the simplified composite mode network is obtained as follows: Figure 11 shown.

[0247] According to the simplified composite network, the transient analysis equivalent circuit is obtained as follows: Figure 12 As shown, R is the sum of 2 times the line mode loop resistance, the zero mode resistance from the fault point to the busbar, and 3 times the grounding resistance; L is the sum of 2 times the line mode loop inductance and the zero mode inductance from the fault point to the busbar, that is, R = 2(R s1 +R L1 )+R L0 +3R k , L=2(L s1 +L L1 )+L L0 .

[0248] For the neutral point ungrounded distribution network, the power supply end of the zero-mode network is open circuit. Therefore, by deleting the arc suppression coil, the corresponding single-phase grounding fault simplified zero-mode equivalent network can be obtained. There are three energy storage elements, namely zero-mode capacitor, line mode inductor and arc suppression coil inductor (the arc suppression coil inductance value of the neutral point ungrounded system is 0), 3L p >>L, the zero-mode capacitor charges faster, and the influence of the arc suppression coil can be ignored when analyzing the capacitor current. It is assumed that the zero-mode capacitor and the line-mode inductor are exchanging energy quickly. According to circuit analysis theory, the virtual ideal voltage source It is equivalent to the opposite value of the phase voltage at the fault point before the fault occurs, that is, U φm is the amplitude of the system zero-mode voltage, ω is the power frequency angular frequency, is the initial phase angle of the fault.

[0249] When a fault occurs in the power grid, the voltage balance equation of the zero-mode capacitor circuit is:

[0250]

[0251] Where u L is the voltage across the inductor L, u R is the voltage across the resistor, u C is the voltage across the zero-mode capacitor C0, u C The initial value of is the voltage of the equivalent zero-mode capacitor before the fault, i C is the capacitive current flowing through the fault point, and its initial value is 0. Perform Laplace transform and calculate it according to u C and i C The initial value of , we get:

[0252]

[0253] Where i C From the steady-state power frequency component i C.p and the transient component i of the oscillation decay C.t It consists of two parts. is the amplitude of the power frequency grounding capacitance current.

[0254] During the transient fault process, the iron core of the arc suppression coil is not saturated, and the arc suppression coil circuit equation is:

[0255]

[0256] Where u Lp Arc suppression coil 3L p The voltage across the terminals, u R is the voltage across the resistor, n is the number of turns of the arc suppression coil, is the current flowing through the arc suppression coil. Before the fault occurs, the potential difference between the two ends of the arc suppression coil is 0, and no current flows, so φ Lp The initial value of is 0. Using this condition, we can get:

[0257]

[0258] Where, is the time constant of the arc suppression coil circuit, we can get:

[0259]

[0260] According to the analysis of capacitive current and inductive current, the total current flowing through the grounding point is:

[0261]

[0262] Where if.p is the steady-state component of the fault current, i f.t is the transient component of the fault current.

[0263] 2) Fault transient characteristics

[0264] When the phase voltage of the feeder exceeds the peak value, that is, A single-phase ground fault occurs in the power grid. According to the expression of the fault transient current component, the capacitive current component has a maximum value i C.t.max , the inductive current component reaches its minimum value of 0. Therefore, the fault transient current only contains the attenuated oscillating capacitive current. At this time, the transient fault current component i f.t.1 for:

[0265]

[0266] When the phase voltage of the feeder passes through zero, that is, A single-phase ground fault occurs in the power grid. According to the expression of the fault transient current component, the inductive current component has a maximum value i L.t.max , the capacitive current component reaches its minimum value i C.t.min The ground fault that occurs when the phase voltage crosses zero is greatly affected by the arc suppression coil. At this time, the transient fault current component i f.t.2 for:

[0267]

[0268] In summary, the frequency range of inductive transients in the current component is 0Hz-50Hz, while the frequency range of capacitive transients in the current component is 300Hz-3000Hz. The frequency difference between the two is so great that they cannot compensate each other.

[0269] Based on the above analysis, the transient fault characteristics of a single-phase grounding fault in a small current grounding distribution network can be obtained:

[0270] 1) The three conditions of fault initial phase angle, grounding point transition resistance and feeder parameters are the fault conditions that affect the transient characteristics of zero-mode current of small current grounding fault. The influence of these conditions on fault transient current is coupled;

[0271] 2) The transient current of a single-phase ground fault is divided into capacitive and inductive components. At the beginning of the fault, the phase determines the ratio of capacitive transient to inductive transient, which in turn affects the amplitude of the transient current.

[0272] 3) When the fault occurs at the moment when the phase voltage is close to the maximum value, in the first transient half-wave, the transient zero-mode voltage has the opposite polarity to the zero-mode current of the fault line, but has the same polarity as the transient zero-mode current of the non-fault line;

[0273] 4) In a distribution network with an ungrounded neutral point, the transient zero-mode current of the fault line flows from the line to the busbar, while the transient zero-mode current of the non-fault line flows from the busbar to the line;

[0274] 5) The derivatives of the transient zero-mode current and zero-mode voltage of the fault line are always of opposite polarity, while the derivatives of the transient zero-mode current and zero-mode voltage of the non-fault line are always of the same polarity.

[0275] 1.3 Arc high resistance fault characteristics

[0276] High-resistance faults are a significant type of fault in distribution networks, primarily caused by line breakage, sagging, or contact with trees, which results in contact between the line conductor and the grounding medium. Therefore, they typically manifest as single-phase ground faults in overhead lines, and are less likely to occur in cable lines. High-resistance grounding media, such as cement, sand, soil, asphalt, and trees, have uneven surfaces, making it difficult for line conductors to establish stable and reliable contact. Therefore, at the higher voltage levels of the distribution network, the air gap between the conductor and the grounding medium is prone to ionization, forming an arc. Statistics show that, with the exception of lines falling into water, the vast majority of high-resistance grounding faults form an electrical connection with the grounding medium via arcing, which is the primary reason why high-resistance faults are often referred to as arc-type high-resistance faults.

[0277] When a high-resistance arc fault occurs, the arc first undergoes a period of transient arcing (also known as an unstable arcing phase), lasting from a few cycles to tens of seconds. It may then enter a stable arcing phase, or the arc may be extinguished. Due to the arc's poor stability and susceptibility to environmental interference, unstable and stable arcing phases typically alternate. The arc waveform exhibits nonlinear characteristics due to its inherent discharge mechanism. During the unstable arcing phase, due to intermittent arcing and extinction, the waveform exhibits irregular waveform distortion and random fluctuations.

[0278] The characteristics of arc high resistance fault are analyzed as follows:

[0279] (1) Steady-state characteristics—power frequency amplitude and phase of phase current and voltage

[0280] The difficulty in detecting high-resistance faults lies primarily in the subtle nature of the fault signature. Algorithms that use traditional steady-state parameters such as power frequency amplitude and phase to detect high-resistance faults are generally incapable of reliable operation. As early as the 1990s, Professor Russell's team at Texas A&M University conducted systematic artificial fault experiments on high-resistance faults. They conducted over 200 high-resistance fault tests (with the line broken and grounded via a high-impedance dielectric) on 10 12.5kV feeders in five substations. Only 35 of these faults were cleared by traditional overcurrent protection devices. Low current levels and the presence of arcing are the two primary characteristics of these high-resistance faults. Most fault currents range from 0 to 50A, and the waveforms exhibit varying degrees of nonlinearity. Because the current amplitudes of most high-resistance faults are too small, phase calculation errors are significant, and fault identification based solely on phase is prone to misjudgment. Regarding voltage, once the transition resistance reaches a certain value, the phase voltage in a high-resistance fault maintains essentially the same amplitude and phase as before the fault, without any nonlinear distortion. Therefore, using power frequency amplitude and phase to detect high-resistance faults is basically ineffective.

[0281] (2) Steady-state characteristics—power frequency amplitude of power frequency zero-sequence voltage and zero-sequence current

[0282] At present, since high-resistance faults are mainly single-phase grounding faults and low-current grounding systems generally have a higher system zero-sequence impedance, high-resistance faults in low-current grounding systems can still be detected by the criterion that the zero-sequence voltage reaches 15% of the phase voltage. Figure 14 is the simplified zero-sequence equivalent circuit of a high-resistance ground fault, where u f (t) is the virtual power supply at the fault point, which has the same voltage amplitude and opposite phase to the voltage of the fault phase before the fault; R is the positive sequence resistance, negative sequence resistance, zero sequence resistance of the line from the fault point to the busbar and three times the grounding point transition resistance R f In the case of high-resistance grounding, since the fault grounding resistance is relatively large, it can be approximately considered that R is equal to 3R f And ignore the influence of line impedance; Z is the zero-sequence impedance of the system to ground. In an ungrounded system, C0 is the zero-sequence capacitance reactance of the system to the ground; in the arc suppression coil grounding system, is the parallel impedance of LP and C0; in a low resistance grounding system, C0 and R n The parallel impedance.

[0283] Therefore, the zero-sequence voltage amplitude U0 of the system can be calculated by the phase voltage amplitude U before the fault. p express:

[0284]

[0285] The system zero-sequence impedance can be considered unchanged under different fault transition resistances, so Z can be calculated based on the zero-sequence current during a metallic grounding fault. Assuming that the fault current (i.e., the sum of the zero-sequence capacitive currents in the system) during a metallic single-phase grounding fault in a 10kV neutral point ungrounded system is I f ≈3I0=10A, the zero-sequence impedance of the system can be calculated according to the following formula:

[0286]

[0287] If the zero-sequence voltage starting value for detecting high-resistance fault is set to U0>15%U p , then substitute the above formula into (36) to get R f ≈3.8 kΩ , which is to set the high-resistance fault transition resistance value that can be detected by this criterion.

[0288] For resonant grounding system, due to the compensation capacitance current of arc suppression coil, the fault current will be reduced when a metallic single-phase grounding fault occurs in 10kV system. If it is 5A, then the zero-sequence impedance of the system can be calculated to be about 3.464kΩ. The zero-sequence voltage starting value U0 is set to be greater than 15% U p The ability to withstand transition resistance reaches approximately 7.6 kΩ, which basically meets the requirements of most high-resistance fault detection. Therefore, in a resonant grounding system, when the arc suppression coil operates within a good compensation range, the high-resistance fault detection method based on zero-sequence voltage can meet the requirements in most cases.

[0289] For a low-resistance grounding system, since the zero-sequence admittance of the system to the ground is much larger than the low resistance of the neutral point, the zero-sequence impedance Z is approximately 3R N , where R N is the resistance value of the neutral point connection (set to 10Ω). When the high resistance fault detection criterion is set to U0>15%U p When the transient resistance is less than 60Ω, this criterion is clearly invalid. Currently, high-resistance fault detection in low-resistance grounding systems is still achieved through zero-sequence current protection devices. By configuring a definite-time or inverse-time high-sensitivity zero-sequence current overcurrent protection device (current setting of approximately 3A), low-resistance grounding systems can generally detect high-resistance faults of 1 to 2 kΩ.

[0290] However, while zero-sequence voltage can theoretically detect high-resistance faults in low-current grounding systems, actual systems often have a certain damping rate (due to the system line resistance and the inherent damping of the arc suppression coil). Furthermore, some systems incorporate series damping in the arc suppression coil to reduce neutral point voltage offset during normal operation. This results in many actual field operations failing to detect high-resistance faults. Therefore, relying solely on zero-sequence voltage amplitude for detection requires consideration of the balanced configuration of the arc suppression coil series damping and detuning.

[0291] (3) Steady-state characteristics—low-order harmonic amplitude and phase

[0292] Arc nonlinearity stems from the energy exchange and transmission process of the arc gap plasma electron flow as the external excitation voltage and current change. Simply put, when the arc current is about to cross zero, the arc temperature drops rapidly and the arc column input power approaches zero. The arc gap resistance exhibits a "high resistance state" before and after the current crosses zero, resulting in a "zero-off period" in the arc current. Odd harmonic methods, represented by the third harmonic method, detect faults by detecting the increase in odd harmonic content caused by the "zero-off period" and the resulting unique phase relationship with the fault phase voltage (e.g., the third harmonic phase and the fault phase voltage are opposite near the zero-crossing point). Methods based on even harmonics and interharmonics are primarily used to detect harmonic content anomalies caused by intermittent arcing, but such anomalies generally do not have unique phase relationship characteristics.

[0293] With the neutral point not directly grounded and three-phase power supply, the zero-sequence current only contains the component caused by network imbalance (i.e., imbalance of grounding admittance). Faults in cable lines generally occur at the joints, and the probability of high-resistance faults is not high. Therefore, the fault characteristics of most high-resistance faults are relatively limited by the interference of unbalanced operation. However, at this time, due to the small amplitude of the zero-sequence current, the method based on the amplitude and phase of low-order harmonics is easily affected by measurement errors or background noise, and because the fault current will produce different degrees of nonlinearity under different fault grounding media, the harmonic amplitude content generated by the fault is also different. For example Figure 15 The low-order harmonic content of several measured high-resistance fault currents in 10kV systems is shown. When the fault grounding medium is relatively moist and the material has strong insulation, the arc nonlinearity is weaker and the harmonics generated are fewer. Especially under noise interference, it is difficult to reliably distinguish between the fault state and the normal state.

[0294] (4) Steady-state characteristics—randomness

[0295] In arc-type high-resistance faults, the arc is easily extinguished at the zero-crossing point due to the small fault current. However, as the voltage increases after the zero-crossing point, the arc gap is easily re-breakdown at the high voltage point, that is, the arc reignites. The intermittent burning and extinguishing of the arc is also affected by the external environment, such as the jumping process when the wire falls to the ground (branch hitting the tree), the ablation of the grounding medium, changes in the natural environment, etc. Therefore, different scenarios will cause the characteristics of high-resistance faults to vary greatly. Based on the analysis of 10kV field measurement data, arc-type high-resistance faults can be roughly divided into the following two categories:

[0296] 1): After the fault, the unstable arcing process lasts for a long time, accompanied by the rapid ablation of the grounding medium (the fundamental frequency current amplitude continues to increase with time), and at the same time exhibits an intermittent arcing process (a transient process with instantaneous changes in current amplitude).

[0297] 2): After the fault occurs, the arc quickly enters the stable arcing stage after a short transient process, without significant random fluctuations and intermittent arcing and extinction phenomena, such as Figure 16 (d)-(f).

[0298] (5) Transient characteristics—transient high-frequency signals

[0299] The intermittent arcing and extinguishing process in the unstable arcing stage of high-resistance fault is actually a process of continuous connection and disconnection between the power grid and the grounding medium. Whether it is the moment of fault occurrence (first arcing) or the subsequent arc extinction and arc reignition, they will be accompanied by transient processes caused by changes in the system operating state (such as Figure 17 ), this transient process is generally reflected as a rapidly decaying high-frequency signal. During the stable arcing phase, the nonlinear distortion of the current also generates a large number of high-frequency harmonics. Therefore, methods based on transient high-frequency signals can theoretically detect both unstable and stable arcing processes in high-resistance faults.

[0300] In a low-current grounding system, the derivatives of the transient zero-mode current and voltage on the fault line always have opposite polarities; the derivatives of the transient zero-mode current and voltage on the non-fault line always have the same polarity. This allows for fault line selection. Because the parameters and frequency characteristics of the lines on either side of the fault differ significantly, the transient processes on both sides of the fault can be considered independent. Specifically, the upstream transient process is generated jointly by the fault-to-busbar section and all healthy lines, while the downstream transient process is generated only by the lines and loads in the terminal section from the fault point.

[0301] A ground fault transient process exhibits multiple distinct transient components. The component with the lowest frequency and closest to the power frequency generally has the largest amplitude. This component, known as the transient main resonant component, is generated by the series resonance between the inductive line-mode network and the capacitive zero-mode network. Its transient current distribution exhibits a clear pattern. Therefore, the main resonant component best represents the fault transient characteristics. The following analysis of fault transients refers to the transient main resonant component.

[0302] A systematic analysis of the transient process of a single-phase high-resistance ground fault in a resonant grounding system was conducted. During a high-resistance fault, the main resonant frequency decreases. In this case, the inductive reactance of the arc suppression coil cannot be assumed to be significantly greater than the capacitive reactance of the ground capacitance, so the arc suppression coil inductance cannot be ignored. Because the line-mode and zero-mode impedances of the line are relatively small, the transient process during a high-resistance ground fault is primarily a parallel resonance between the system capacitance and the arc suppression coil. Therefore, the line inductance can be ignored. In the underdamped phase, the fault point current is a power frequency current with an attenuated DC component. This DC component is directly related to the transition resistance, and the effect of the distance from the fault point to the busbar on the DC component is negligible. The power frequency current amplitude decreases with increasing transition resistance, and the attenuation factor of the transient DC component decreases with increasing transition resistance. The overdamped phase is a damped oscillation process. The oscillation frequency increases with increasing transition resistance. When this oscillation frequency is approximately equal to the power frequency, it is combined with the virtual power supply at the fault point, resulting in a significant beat frequency phenomenon.

[0303] In summary, when the transition resistance reaches a certain value, the transient process of a high-resistance fault has no high-frequency components, and the resonant frequency is close to the power frequency. This conclusion is consistent with the phenomena shown in the measured data. The transient resistance tolerance of the currently used transient measurement method is approximately 1 to 2 kΩ, which is affected by line parameters and environment.

[0304] In addition to transient high-frequency components generated by system operating state changes, domestic and international scholars have also extensively studied high-frequency harmonic components caused by current nonlinearity. Considering that this nonlinearity is temporally distributed near the current zero-crossing point, a series of time-frequency domain feature analysis methods have emerged, such as wavelet transform, Hilbert-Huang transform, Twi-Williams distribution, empirical mode decomposition, and variational mode decomposition. For some high-resistance faults, the "zero-break" distortion of the current near the zero-crossing point causes the fault waveform to form a peaked wave. High-frequency harmonic components can be extracted near the peak, which is much higher than at other times. This sudden increase in high-frequency harmonic content or the temporal distribution of the high-frequency signal (near the peak) can effectively detect a certain number of high-resistance faults. However, the degree of nonlinearity in high-resistance faults is affected by the fault scenario. In humid environments or where the grounding dielectric is highly insulating, the ionization process of the arc or dielectric breakdown does not vary significantly with changes in current amplitude. The fault current waveform is relatively smooth at the distorted location, with no peaks, and thus does not produce a significant increase in high-frequency harmonics compared to before the fault.

[0305] In addition, a large number of studies are currently focusing on how to effectively extract high-frequency harmonic components. Judging the occurrence of high-resistance faults based solely on the increase in high-frequency harmonic content may actually lead to misjudgment of the operation of nonlinear loads in the distribution network, such as zero-sequence CT excitation inrush current, saturation, and arc furnaces.

[0306] In summary, relying solely on transient high-frequency signals of high-resistance faults cannot reliably detect high-resistance faults with high transition resistance (1 kΩ), nor can it reliably detect some high-resistance faults with weaker nonlinearity. Detecting high-resistance faults solely based on increased high-frequency harmonic content carries a high risk of misjudgment.

[0307] (6) Transient characteristics—waveform shape

[0308] Due to economic and technical constraints, or incomplete fault feature analysis, existing fault diagnosis algorithms, using various signal processing techniques to extract certain features, fail to capture the majority of typical arc high-resistance fault characteristics. In fact, with the exception of the steady-state power frequency amplitude and phase, and transient characteristics at the moment of system state change, the typical features utilized by existing algorithms listed above are all due to the nonlinearity of arc high-resistance fault current. Therefore, directly addressing the nonlinearity of the current waveform can better grasp the fault characteristics. Waveform morphology features employ this approach. By utilizing waveform morphology features and properly designing their description methods, they can mitigate the influence of nonlinear loads and avoid misdiagnosis.

[0309] Taking the resonant grounding system as an example, for a typical n-line distribution network, its zero-sequence network is shown in (38). 0i represents the equivalent zero-sequence capacitance to ground of feeder i; C 0L is the equivalent zero-sequence capacitance of the substation line to ground; L is the zero-sequence inductance of the arc suppression coil (numerically equal to three times the inductance of the arc suppression coil); R HIF is the zero-sequence transition resistance of high-resistance fault (equal to three times the transition resistance in terms of value); u f is a virtual voltage source, which has the same amplitude and opposite phase to the voltage at the fault point before the fault occurs; u 0b is the busbar zero-sequence voltage; i 0f is the fault zero sequence current; i 0i is the zero-sequence current of feeder i; and are the zero-sequence capacitance current to ground of feeder i and transformer line respectively; i 0L is the zero-sequence current flowing through the arc suppression coil.

[0310] Assuming that a high-resistance fault occurs on feeder n, the zero-sequence capacitance current i of the feeder to ground is 0n for:

[0311]

[0312] Relative to the zero-sequence current of the arc suppression coil, the zero-sequence capacitance current of the transformer line is Very small and can be ignored.

[0313] For the stable arcing process of a high-resistance fault, the zero-sequence current can be expressed as the superposition of an ideal sinusoidal component and a distortion component caused by a reaction arc (or breakdown of a solid dielectric), and the distortion component does not contain a sinusoidal component:

[0314]

[0315] Where, The phase of the zero-sequence capacitance current of the healthy line is represented by Similarly, for the arc suppression coil zero sequence current i 0L , zero sequence current of each healthy line And the zero-sequence current i of the fault line 0n can be decomposed into independent sinusoidal components and superposition components, which can be expressed as as well as Therefore, formula (38) can be further expressed as:

[0316]

[0317] Where, and I ML They are and The effective value of the two sinusoidal components; ω represents the power frequency angular velocity.

[0318] For the zero-sequence voltage u 0b , which can also be decomposed into sinusoidal components and distortion components

[0319]

[0320] Where, It can be expressed as:

[0321]

[0322] The figure below shows the fault current and its sinusoidal and distorted components in When the mathematical expression or characteristics of the distortion component are unknown, it is difficult to theoretically discuss the difference between the zero-sequence current distortion characteristics of the fault line and the non-fault line, or at different positions of the line. Therefore, for the convenience of discussion, we will denote the distortion component of the fault zero-sequence current as Simplified to a periodic piecewise function exist The mathematical expression is shown in the following formula and Figure 19 :

[0323]

[0324] Where -1<τ<0 and It can be seen from the formula, The maximum value is equal to

[0325] When the distortion offset is ignored, about Axis symmetry, and meets the following conditions, reflecting Symmetry within a period:

[0326]

[0327] The above expression can be written as the following second-order nonhomogeneous linear equation:

[0328]

[0329] Will exist Substitute the mathematical expression of into the solution and we get:

[0330]

[0331] Similarly, you can exist and Substitute the mathematical expressions of the three intervals into the solution. We can get three groups of and The expressions in these three intervals, taking into account and By establishing the boundary condition equation at the continuity of the interval boundary, we can obtain the coefficients a0 and a1 equal to 0. Finally, the zero-sequence current of each outgoing line can be expressed as:

[0332] i) Transformer line zero sequence current i 0L :

[0333]

[0334] Distortion component of zero-sequence current in transformer lines The distortion component of the fault zero sequence current In a resonant grounding system, the detuning degree v reflects the ability of the arc suppression coil to compensate for the system capacitive current, where v = 1-1 / ω 2 LC 0∑ In most cases, v∈[-0.1,0), at which point ω 2 LC 0∑ ∈[0.9091,1). Therefore, ω 2 LC 0∑ is much larger than 1, so in equations (46) and (47), A L >0,

[0335] Since the arc resistance of an arc high-resistance fault is spike-shaped, and the fault transition resistance generally shows the series connection of the arc resistance and the earth impedance, the distortion component of the fault current (fault zero-sequence current) reduces the amplitude of the original sinusoidal component. That is, the distortion component and the sinusoidal component have opposite signs. We call this the "reverse superposition" of the distortion component on the sinusoidal component. According to Equation (46), the transformer line zero-sequence current and the fault current have the same phase, but the distortion component of the transformer line zero-sequence current and the fault current distortion component have opposite signs. Therefore, the "superposition" effect of the distortion component is also opposite, that is, the distortion component of the transformer line zero-sequence current is "positively superposed" on its sinusoidal component.

[0336] ii) Healthy line zero sequence current i 0i(i≠n) :

[0337]

[0338] Similar to the analysis in i), the distortion component of the zero-sequence current of the healthy line is The distortion component of the fault zero-sequence current The signs are the same, but the phases of the sound line zero-sequence current and the fault zero-sequence current are opposite. Therefore, the distortion component of the sound line zero-sequence current is also "positively superimposed" on its sinusoidal component.

[0339] iii) Fault line zero sequence current i 0n :

[0340]

[0341] Similar to the analysis of i) and ii), the zero-sequence current of the fault line is the superposition of the reverse fault current and the fault line-to-ground capacitance current. The superposition effect of its distortion components depends on the magnitude of the zero-sequence capacitance current of the line. According to the above, when ω 2 LC 0∑ ∈[0.9091,1), 1-4ω 2 LC 0∑ <0. When C 0n / C 0∑ <1-1 / 4ω 2 LC 0∑ When 1-4ω 2 L(C 0∑ -C 0n )<0, at this time, the distortion component of the zero-sequence current of the fault line is in the same phase and has the same sign as the distortion component of the fault zero-sequence current, thus having a "reverse superposition" on the sinusoidal component of the fault line.

[0342] The superposition effect of the distortion components ultimately reflects the shape of the waveform distortion, such as Figure 20 These are the zero-sequence current waveforms measured on different lines for two sets of arc-flash high-resistance faults in a measured resonant grounding system. The differences in distortion caused by conductors contacting different grounding media also manifest differently in the "superposition effect" on different lines. When the distortion is large, the fault zero-sequence current and the fault line zero-sequence current exhibit distinct "zero-rest" characteristics, while the amplitude on the non-fault line increases abnormally near the original "zero-crossing" point of the current. When the distortion is small, the "zero-rest" distortion characteristics of the fault zero-sequence current and the fault line zero-sequence current are weakened but still exist. While the amplitude increase near the zero-crossing point of the non-fault line current exists, it does not cause significant distortion and remains primarily sinusoidal.

[0343] In summary, from the perspective of waveform distortion morphology, it is possible to detect, select, and even locate arc high-resistance faults with different distortion characteristics. The above deduction is limited to the perspective of resonant grounding systems. Using the same approach, it is easy to deduce the waveform distortion differences between different lines in arc high-resistance faults in ungrounded and low-resistance grounding systems. However, in reality, there are simpler methods for line selection or location in these two grounding systems, such as the zero-sequence current amplitude and phase method. Therefore, the fault waveform distortion characteristics of these two neutral-grounded systems will not be discussed in detail here.

[0344] 2. Fault equivalent model of distributed power supply

[0345] Distributed generation (DG) can be divided into two types according to the interface type: rotary type and inverter type. This paper focuses on analyzing the fault equivalent model of photovoltaic and small hydropower stations, which are widely used in rural / mountainous areas.

[0346] 2.1 Fault equivalent model of inverter-type distributed power supply

[0347] Typically, photovoltaic power generation systems, fuel cells, and energy storage devices are connected to the distribution grid through inverters, known as inverter-interfaced distributed generation (IIDG). Grid-connected via the inverter interface, their output exhibits strong nonlinearity, with output characteristics largely determined by the inverter control strategy. Therefore, accounting for the influence of control strategies is essential for studying the output characteristics of distributed generation (DGs).

[0348] (1) Distributed power supply fault analysis model considering constant power control

[0349] Currently, common inverter control methods include power control, constant voltage and frequency control, and droop control. The most common control strategy for grid-connected inverters is constant power control based on grid voltage vector orientation. Constant voltage and frequency control and droop control are primarily used when connected to a microgrid.

[0350] Distributed power inverters typically use PQ control when connected to the distribution network. PQ ​​control regulates the active and reactive power output of the inverter. Under normal operating conditions, it maintains a constant active power output while also providing a certain degree of reactive power regulation. Under PQ control, the output voltage and frequency are determined by the grid.

[0351] Figure 22 The figure shows the block diagram of the inverter grid-connected outer loop operation. During grid-connected operation, the AC side of the inverter obtains the grid current and voltage, performs Park transformation, and decouples the current and voltage into active and reactive power to obtain the instantaneous active and reactive power output by the distributed power supply.

[0352]

[0353] Since the d-axis coincides with the grid-connected point voltage vector in Park transformation, we can get

[0354]

[0355] It can be deduced that:

[0356]

[0357] i dref and i qref is the given active current and reactive current, i d and i q It is the active component and reactive component of the inductor current. After the feedback current is compared with the given current, the SPWM modulation wave is obtained through PI regulation and inverse DQ transformation. The SPWM output controls the switch off, forming a closed-loop control system, thereby controlling the inverter output current, and thus controlling the output active power and reactive power. The phase-locked voltage is the grid voltage ua 、u b 、u c , the phase-locked loop provides an angle reference for DQ conversion and inverse DQ conversion.

[0358] It can be seen that when the fault current does not reach the overcurrent protection limit of the power electronic device, the distributed power supply with constant power control can be equivalent to a controlled voltage source with a series reactance, thereby ensuring the constant output power of the distributed power supply. Due to nonlinear elements such as power electronics, the equivalent reactance of the inverter-connected distributed power supply may be nonlinear.

[0359] However, constant output active power can only be guaranteed under normal operating conditions or when the fault point is far enough from the DG access point that the system's equivalent impedance changes minimally. When a fault occurs near the DG access point, the fault current generated by the DG reaches the overcurrent protection limit, making the DG a constant current source. Once the low voltage ride-through protection limit is exceeded, the DG ceases operation.

[0360] (2) Analysis model of distributed power supply failure considering low voltage ride-through

[0361] With the rapid penetration of distributed generation (DG), to prevent large-scale power outages caused by DG disconnection during faults, the State Grid Corporation of China requires DG to remain connected for a certain period of time during grid faults and to support grid voltage stability. This is achieved by adopting low voltage ride-through (LVRT) control. Specifically, from the time a fault occurs until the voltage recovers to 0.9 pu, the reactive power output of the DG must track the voltage changes at the grid connection point and meet the following requirements:

[0362]

[0363] Among them, u g.f is the per-unit voltage value of the distributed power access point after a fault occurs, I q.f is the reactive current value output by the distributed power supply after the fault, I N is the rated current value. In actual engineering, in order to provide sufficient reactive power support, u g.f When the coefficient is in the range [0.2, 0.9], it is usually set to 1.5. Therefore, the direct grid-connected distributed power supply considering low voltage ride-through can be equivalent to a voltage-controlled current-controlled source.

[0364] Considering the inverter's overcurrent capability and low voltage ride-through control, the output active current setting is:

[0365]

[0366] Among them, I d.f is the active power output when the system fails, Pm is the active power output at the fault point, I max The maximum current allowed to flow through the inverter. Therefore, considering the LVRT control strategy and inverter current limiting constraints during distribution network faults, as well as the provision of greater reactive power support when the voltage drops significantly, the relationship between the inverter output fault current and the grid connection point voltage is as follows:

[0367]

[0368] The active component of the fault current is overlapped with the grid connection point voltage vector, and δ represents the phase angle of the positive sequence voltage at the grid connection point. Then the fault current can be expressed as Equation (58). Substituting Equation (57) into Equation (58) yields Equation (59).

[0369]

[0370] From formula (60), we can see that there is a boundary voltage u x Make the inverter output current just reach the output current amplitude limit. That is:

[0371]

[0372] Let P m =1,I max =2I N , then u x =0.537, the inverter reaches the current limiting constraint boundary. At this time, the phase angle of the fault current is:

[0373]

[0374] When u g.f When it is in the interval [0.537, 0.9], the inverter can adjust the reactive power output while ensuring that the active power output remains unchanged, and is in the grid-connected control stage; when u g.f When the inverter is in the interval [0.2, 0.537], it is constrained by the control strategy and the inverter current limiting condition, and cannot guarantee the active power output remains unchanged, and is in the low voltage ride-through control stage.

[0375] The above analysis is based on the premise that a symmetrical fault occurs in the system. When an asymmetrical fault occurs, the active component of the fault output current is oriented towards the positive sequence voltage component of the grid connection point. That is, the inverter grid-connected distributed power supply is equivalent to a current source controlled by the positive sequence fault voltage at the grid connection point, that is:

[0376]

[0377] 2.2 Equivalent model of small hydropower fault considering excitation control response

[0378] As a renewable, clean, distributed power source, small hydropower is poised for widespread adoption in regions rich in hydropower resources. Small hydropower units are typically connected to distribution networks using a T-connection, significantly increasing the number of branch lines. The rapid response of the excitation system significantly impacts the output current of small hydropower faults. By equating small hydropower units to voltage-controlled current sources (VCCSs) controlled by the voltage at the point of common connection (PCC), a ground fault analysis model for distribution networks containing small hydropower clusters is constructed based on multi-port network theory and ideal transformers.

[0379] (1) Simplified model of excitation system

[0380] With the rapid development of power semiconductor technology, almost all salient-pole synchronous generators, including small hydropower plants, have begun to adopt self-shunt static excitation systems. This type of excitation system uses a thyristor rectifier as the excitation power element. The rectifier conduction angle is changed according to the deviation between the terminal voltage and the reference voltage, thereby controlling the potential across the excitation winding and maintaining the stability of the terminal voltage. IEEE standard model of self-shunt static excitation system. ref is the set terminal reference voltage; Ut is the terminal voltage; ΔU is the terminal voltage deviation value; E f is the stator excitation potential output to the generator; T A 、T B1 、T B2 、T C1 and T C2 K is the lead / lag time constant of each link; R and K A is the gain multiple of each link; V Rmax is the peak value of the generator excitation potential.

[0381] Since the self-shunt static excitation system has a high response speed (up to tens of milliseconds), the dynamic process of the excitation system can be ignored when calculating the fault current, and its excitation characteristic equation can be simplified to:

[0382] E f =min(K v (U ref -U t ),V Rmax )(61)

[0383] Where: min(·,·) is the minimum function; Kv is the gain coefficient of the excitation system. The gain coefficient of small hydropower units is generally between 30 and 150.

[0384] (2) Small hydropower VCCS fault equivalent model

[0385] After obtaining a simplified model of the excitation system, in order to analyze the impact of the excitation potential Ef on the electrical quantities on the stator side of the small hydropower station, it is necessary to establish the small hydropower stator voltage and current equations that take Ef into account. Since the main purpose of studying the fault characteristics of small hydropower units is to provide a reference for the setting of adaptive protection for distribution networks containing small hydropower clusters, and in the adaptive protection of distribution networks, the post-fault steady-state value is generally used as the setting reference value. Therefore, the steady-state output current of the small hydropower generator after a fault can be directly analyzed. The relationship between the steady-state d-axis and q-axis stator voltages and currents after a small hydropower unit fault is:

[0386]

[0387] Where ud, uq and id, iq are the d-axis and q-axis components of the stator voltage and stator current, respectively; Xd and Xq are the d-axis and q-axis components of the stator steady-state synchronous reactance, respectively; Eq is the steady-state open-circuit potential; ra is the stator resistance; and all the above variables are per-unit values.

[0388] Since the stator reactance of a synchronous generator is much larger than the stator resistance, ra can be assumed to be 0 in the analysis. During the post-fault steady-state process, there is no current in the damping windings D and Q, and only the excitation winding potential exists on the stator q-axis. Therefore, the stator steady-state open-circuit potential is equal to the stator excitation potential Ef, that is, Eq = Ef. With the above assumptions, Eq (62) can be simplified to:

[0389]

[0390] From formula (62), it can be seen that the fault output current of small hydropower is related to the terminal voltage and the excitation potential. Since small hydropower is usually directly connected to the distribution network in a "T" connection mode and the connection line is very short, it can be approximately considered that the terminal voltage of the small hydropower generator is equal to the voltage at the common connection point (PCC) UPCC. The comprehensive formula (62) can be used to make the small hydropower unit under fault ride-through equivalent to a Figure 24 The VCCS model shown in Figure 1 shows that the fault output current IG is controlled by the voltage at PCC. Figure 24 Where: f(UPCC,d) is a function including the d-axis component UPCC,d of the voltage at the PCC; g(Ef,UPCC,q) is a function including Ef and the q-axis component UPCC,q of the voltage at the PCC.

[0391] (3) Simulation verification

[0392] In order to verify the effectiveness and accuracy of the established small hydropower VCCS fault equivalent model, the following Figure 25 The 10kV three-outlet rural distribution network model with distributed small hydropower is shown. Except for line 1 which is an overhead line, the other lines are mixed lines containing cables and overhead lines. Figure 26This is the grid-connected small hydropower model in PSCAD. The total system capacitive current is 34A, the line end load is a constant power load, and the power factor is set to 0.85. The transformer ratio is 110kV / 10.5kV, and the rated capacity is 50MVA. The neutral point is resonantly grounded, and the arc suppression coil is connected to the system neutral point via a grounding transformer. The arc suppression coil compensation is set to 10% overcompensation and the damping rate is 15%. The rated power of small hydropower station DG1 is 0.5MW, and the q-axis synchronous reactance X is 0. q1 Synchronous reactance X with d-axis d1 They are 0.614pu and 1.533pu respectively, and the excitation gain coefficient K v1 =80; the rated power of small hydropower DG2 is 1MW, and the q-axis synchronous reactance is X q2 Synchronous reactance X with d-axis d2 They are 0.6pu and 1.05pu respectively, and the excitation gain coefficient K v2 =30; small hydropower DG1 and DG2 excitation peak voltage V Rmax The per-unit value is set to 4.5pu according to the typical value of hydropower units.

[0393] Assume that a single-phase grounding fault occurs at k2 during the simulation 2s, and record the changes in the excitation voltage, fault output current, and PCC voltage of the small hydropower DG1. Figure 27 It can be seen that when the fault occurs at t = 2s, the small hydropower excitation system detects the voltage drop at the PCC point and immediately increases the excitation voltage E f , thereby increasing the output current of the small hydropower station. Simultaneously, the voltage at the PCC point also recovered somewhat under the action of the excitation control system. Therefore, when analyzing faults in distribution networks containing small hydropower stations, it is necessary to consider the impact of the excitation control system on the output characteristics of small hydropower stations.

[0394] 3. Analysis of Fault Characteristics of Rural / Mountainous Distribution Networks with Distributed Generation Access

[0395] 3.1 Analysis of short-circuit fault characteristics after distributed generation is connected

[0396] In the case of distributed power access, it can be equivalent to Figure 28 The dual power system analysis model shown in the figure has the system voltages on the M side and the N side being E M and E N , the equivalent impedances are Z M and Z N , where E M For the main grid power supply, E N For distributed power supply.

[0397] When a single-phase grounding fault occurs at point F through the transition resistor, the measured voltage at node M is used as the reference value, and the phase angle of the fault voltage is ΦMF It can be decomposed into the sum of the phase angle between the measured voltage and the measured current and the phase angle between the measured current and the fault voltage, as shown in the following equation.

[0398]

[0399] Wherein, the phase angle between the measured voltage and the measured current is known, and the phase angle between the measured current and the fault voltage is the phase angle between the measured current and the fault current, as shown in the following formula.

[0400]

[0401] Because the current at the fault point is unmeasurable, the phase angle between the measured current and the fault current cannot be directly calculated. However, when a fault occurs in the system, the fault current and the fault sequence current are in phase, and the sequence current at the measurement installation and the fault sequence current can be approximately assumed to be in phase. That is, the measured sequence current at node M is in phase with the fault sequence current. Therefore, the phase angle between the measured current and the fault voltage is approximately equal to the phase angle between the measured current and the sequence current at node M, as shown in the following equation.

[0402]

[0403] Where C is the negative sequence current distribution coefficient, and I2M is the negative sequence current on the M side. The phase angle φ can be obtained MF As shown below.

[0404]

[0405] The fault phase angle will affect the transient characteristics of the fault voltage phase angle. In terms of duration, the transient process of the fault phase is longer, which can reach 1 / 3 cycle, while the transient process of other non-fault phases is about 1 / 5 cycle.

[0406] In general, the impact of DG on the short-circuit fault characteristics of the distribution network can be summarized as follows:

[0407] 1) When a three-phase short-circuit fault occurs upstream of the feeder where the DG is located, the DG is decoupled from the system power supply circuit and does not affect the short-circuit current supplied by the system. Therefore, the connection of the DG does not affect the short-circuit current characteristics of the distribution network.

[0408] 2) When the fault is not severe and the DG is outputting active current, the grid connection point voltage is affected by the DG output power, initially increasing and then decreasing as the DG output power increases. When the fault is severe and the DG is only outputting reactive current, the grid connection point voltage is affected by the DG rated power, initially increasing and then decreasing as the DG rated power increases. The change pattern of the grid connection point voltage is independent of the fault type and location.

[0409] 3) When the fault is not severe and the DG is outputting active current, the short-circuit current flowing through the feeder upstream of the DG first decreases and then increases with the increase of the DG access capacity. The short-circuit positive-sequence current flowing through the feeder downstream of the DG and adjacent feeders first increases and then decreases with the increase of the DG access capacity. When the fault is severe and the DG is only outputting reactive current, the current flowing through all protection systems in the distribution network is only related to the DG rated power and is not related to the DG output power.

[0410] 4) Although the common bus voltage and the current of the fault branch adjacent to the DG first increase and then decrease with the increase of the DG access capacity, due to the support of the system on the common bus voltage, the access of the adjacent line DG has very little impact on the common bus voltage and the fault branch fault current.

[0411] 5) On the DC input side, the inverter generally uses constant power control, and the inverter's output characteristics are similar to those of a current source. The inverter's pulse width modulation signal frequency is several thousand hertz, and the response speed is only a few milliseconds. Therefore, the inverter's own transient processes can be ignored, and the maximum output current is generally 1.2-1.5 times the rated current.

[0412] 6) During a distribution network fault, the IIDG will maintain power to the distribution network through the inverter until it disconnects due to short-circuit protection. The output short-circuit current depends on the specific control strategy used during the fault phase. Taking the more commonly used constant power control as an example, when a distribution network fault occurs, the inverter maintains its pre-fault active and reactive power output. When an asymmetric fault occurs, the inverter first calculates the positive sequence voltage at the grid connection point, and then determines the short-circuit current the inverter needs to output. If the grid connection point voltage is low and the calculated target output current exceeds its maximum allowable output current, the inverter maintains the output current amplitude at the maximum current level, and the active and reactive power output of the inverter is proportionally reduced. If the grid connection point voltage falls below a set threshold (e.g., 20% of the rated voltage), the inverter stops outputting current. Therefore, under normal fault conditions, the fault current is 1-1.5 times the rated current.

[0413] In summary, different types of distributed generation (DGs) exhibit varying short-circuit current characteristics during fault conditions, with different current characteristics during the initial and steady-state phases. The figure below shows the short-circuit current variation over time for four different types of DGs under direct coupling. The horizontal axis represents time, and the vertical axis represents the current multiple relative to the rated current.

[0414] 3.2 Analysis of small current grounding fault characteristics after distributed power supply access

[0415] (1) Analysis of the impact on steady-state fault characteristics

[0416] The study of a small current ground fault in an active distribution network with an ungrounded neutral point is conducted, and the fault phase is phase A. Taking a dual-outlet system as an example, according to the superposition theorem, the original system can be regarded as the superposition of three single power supply systems, such as Figure 30 As shown in Figure 1, the arrows represent the direction of capacitive current flow. (a) is the original system, and (b), (c), and (d) are the corresponding three single-power systems.

[0417] in, and are the A-phase, B-phase and C-phase currents at the head end of the original system line n respectively; and are the A-phase, B-phase and C-phase currents at the head end of line n of single power supply system 1 respectively; and They are the A-phase, B-phase and C-phase currents at the head end of line n of the single power supply system 2; and They are the A-phase, B-phase and C-phase currents at the head end of line n of a single power supply system, and n=1,2.

[0418] The current of each phase at the head end of a healthy line is:

[0419]

[0420] It can be seen that the amplitude of the current in phase A of the healthy line will change, but the direction will not change, flowing from the bus to the line; the current amplitude of phases B and C of the healthy line will change. When the distributed power supply of the healthy line reverses power to the network, the phase current of the healthy line will reverse.

[0421] The current of each phase at the head end of the fault line is:

[0422]

[0423] It can be seen that the amplitude of the current in phase A of the fault line will change, but the direction will not change, flowing from the line to the bus; the current amplitudes of phases B and C of the fault line will change. When the distributed power supply of the fault line reverses the power supply to the network, the phase current of the healthy line will reverse.

[0424] Study the fault line and analyze the zero sequence current upstream and downstream of the fault point, such as Figure 31 The arrows represent the direction of capacitive current flow. (a) is the original system, and (b) and (c) are the corresponding two single-power systems.

[0425] in, and are the phase A, phase B and phase C currents upstream of the original line fault point respectively; and are the phase A, phase B, and phase C currents downstream of the original line fault point respectively; and Decompose the A-phase, B-phase and C-phase currents upstream of the fault point of line 1 respectively; and Decompose the A-phase, B-phase and C-phase currents downstream of the fault point of line 1 respectively; and Decompose the A-phase, B-phase and C-phase currents upstream of the fault point of line 2 respectively; and The A-phase, B-phase, and C-phase currents downstream of the fault point on line 2 are decomposed respectively.

[0426] The current of each phase upstream of the fault point is:

[0427]

[0428] The current of each phase downstream of the fault point is:

[0429]

[0430] It can be seen that the current amplitude of the fault phase will change, but the direction will not change; the current amplitude of the healthy phase will change. When the distributed power supply at the end of the fault line reverses power to the network, the line phase current will reverse.

[0431] The study of a small current ground fault in an active distribution network with a neutral point grounded via an arc suppression coil is conducted. The fault phase is phase A. Taking a double-outlet system as an example, according to the superposition theorem, the original system can be regarded as the superposition of three single power supply systems, such as Figure 32 As shown in Figure 1, the arrows represent the direction of capacitive current flow. (a) is the original system, and (b), (c), and (d) are the corresponding three single-power systems.

[0432] The current of each phase at the head end of a healthy line is:

[0433]

[0434] It can be seen that the amplitude of the current in phases B and C of the healthy line will change, but the direction will not change, flowing from the bus to the line; the current amplitude of phase A of the healthy line will change, and when the distributed power supply of the healthy line reverses power to the network, the phase current of the healthy line will reverse.

[0435] The current of each phase at the head end of the fault line is:

[0436]

[0437] It can be seen that the amplitude of the three-phase current in the fault line will change, but the direction will not change, and it will always flow from the line to the bus.

[0438] Study the fault line and analyze the zero sequence current upstream and downstream of the fault point, such as Figure 33 As shown in Figure 2, the arrows represent the direction of capacitive current flow. (a) is the original system, and (b) and (c) are the corresponding two single-power systems.

[0439] in, and are the phase A, phase B and phase C currents upstream of the original line fault point respectively; and are the phase A, phase B, and phase C currents downstream of the original line fault point respectively; and Decompose the A-phase, B-phase and C-phase currents upstream of the fault point of line 1 respectively; and Decompose the A-phase, B-phase and C-phase currents downstream of the fault point of line 1 respectively; and Decompose the A-phase, B-phase and C-phase currents upstream of the fault point of line 2 respectively; and The A-phase, B-phase, and C-phase currents downstream of the fault point on line 2 are decomposed respectively.

[0440] The current of each phase upstream of the fault point is:

[0441]

[0442] The current of each phase downstream of the fault point is:

[0443]

[0444] It can be seen that the current amplitude of the healthy phase and the fault phase line upstream of the fault point will change, but the direction will not change; the current amplitude of the fault phase line downstream of the fault point will change. When the distributed power supply at the end of the fault line reverses power to the network, the line phase current will reverse.

[0445] Research on zero-sequence networks reveals that when distributed generation (DGs) are connected to the grid, the high-voltage side of the transformer at the grid connection point must utilize a delta or Y connection to ensure the system neutral point maintains its original grounding method. Otherwise, the ungrounded neutral point of the main grid loses its effectiveness in protecting personnel and equipment safety and improving grid power continuity. Furthermore, the State Grid Corporation of China's "QGDW1480-2015 Technical Regulations for the Connection of Distributed Generations to the Grid" requires that "the grounding method of DGs should be coordinated with that of the distribution network and meet the requirements for personnel and equipment safety and protection coordination." From both a safety and grid specification perspective, DGs in low-current grounded systems should be ungrounded. In an active distribution network, when a single-phase ground fault occurs, the zero-sequence network topology remains essentially the same as when no DGs are connected. Therefore, the addition of DGs does not alter the steady-state characteristics of the zero-sequence current.

[0446] (2) Analysis of the impact on transient characteristics

[0447] When a fault occurs in a rural / mountainous distribution network containing distributed generation (DG), both the system power supply and the DG (distributed generator) supply fault current to the fault point. This study analyzes the impact of DG integration on the transient characteristics of small-current ground faults in distribution networks, considering both DG type and integration location. DG is generally categorized as either rotary or inverter-type. Rotary DG primarily includes diesel generators and asynchronous motors, while inverter DG primarily includes photovoltaic and wind power generation.

[0448] When fault analysis is performed on a distribution network containing rotating DGs, the rotating DG fault model is usually equated with the traditional generator model. In a low-current grounding system, the DG is connected to the distribution network through a grid-connected transformer. The neutral point of the transformer distribution side is ungrounded or resonantly grounded. The DG does not affect the zero-mode loop. The fault model of the rotating DG can be used Figure 34 Inverter DG fault current is related to the inverter control strategy. When performing fault analysis on a distribution network containing inverter DG, the inverter DG is usually equivalent to a constant positive sequence current source, such as Figure 35 shown.

[0449] The transient characteristics of small current grounding faults under different DG access locations are analyzed. According to the different access locations of a single DG, it can be divided into three scenarios: ① DG is located on a healthy line, ② DG is located upstream of the fault point (between the fault point and the busbar), and ③ DG is located downstream of the fault point. The corresponding line mode equivalent circuit is as follows: Figure 36 As shown in (a). When the DG is located on a healthy line or upstream of the fault point, the equivalent impedance upstream of the fault point in equation (2) will be changed. When the DG is located downstream of the fault point, the equivalent impedance downstream of the fault point will be changed. Specifically, when the DG is located on a healthy line l n When , the equivalent impedance upstream of the obstacle becomes:

[0450] Z 1b =Z fb1 +Z s1 / / (Z L11 +Z d11 ) / / … / / [Z Ln11 +Z DG / / (Z Ln12 +Z dn1 )] (76)

[0451] When the DG is located upstream of the fault point, the equivalent impedance upstream of the fault point is:

[0452] Z 1b =Z fb12 +Z DG / / [Z fb11 / / Z S1 / / … / / (Z Ln11 +Z d11 ) / / …(Z Ln1 +Z dn1 )] (77)

[0453] When the DG is located downstream of the fault point, the equivalent impedance downstream of the fault point is:

[0454] Z 1a =Z fa1 +Z DG / / (Z fa12 +Z dj1 ) (78)

[0455] Where Z DG is the line mode equivalent impedance of DG and its grid-connected transformer; Z fa1 =Z fa11 +Z fa12 , Z fb1 =Z fb11 +Z fb12 , Z Ln1 =Z Ln11 +Z Ln12 .

[0456] Depend on Figure 36 It can be seen that although the DG access position may be different, it will change the system line mode structure and equivalent impedance.

[0457] Analysis shows that to further analyze the impact of DG access on the line-mode network, it is necessary to calculate the line-mode equivalent impedance of the DG and its grid-connected transformer:

[0458] Z DG =(Z T +Z S ) / / Z m (79)

[0459] Where Z T 、Z S 、Z m They represent the grid-connected transformer leakage impedance, DG equivalent impedance and grid-connected transformer excitation impedance respectively. T =

[0460] R T +jωL T , Z S =R S +jωL S , Z m =R m +jωL m Typically, the magnetizing impedance of a grid-connected transformer is several orders of magnitude larger than the leakage impedance.

[0461] When the DG is a rotating type DG, since its own equivalent impedance is much smaller than the grid-connected inverter, Z can be ignored. S , formula (79) is equivalent to:

[0462] Z DG ≈Z T (80)

[0463] This shows that the rotary DG acts as a transformer in the line mode network, and it can effectively control the line mode of the upstream, downstream or healthy line of the fault point.

[0464] The impact of equivalent impedance is also affected by the insertion position.

[0465] When the DG is an inverter type DG, it can be seen from the current source characteristics that the equivalent impedance of the DG is much larger than the transformer excitation impedance, that is, Z S =∞, so formula (79) can be equivalent to:

[0466] Z DG ≈Z m (81)

[0467] DG usually relies on grid-connected transformers to connect to the distribution network.

[0468] When considering the impact, the first thing to consider is the connection mode of the high voltage side (distribution system side) of the DG grid-connected transformer, that is, whether the DG grid-connected transformer is connected to the

[0469] A grounding point is introduced into the zero-mode network. Depending on whether a grounding point is introduced into the zero-mode network, DG grid-connected transformers can be divided into two categories: 1)

[0470] 1) Star connection with grounded neutral point on high voltage side; 2) Delta connection with ungrounded neutral point on high voltage side.

[0471] When the zero-mode network structure is changed, the zero-mode current will change. Figure 37 As shown, the typical Yn / △ connection mode is

[0472] For example, the influence of the star connection on the zero-mode network is explained. Assuming that a single-phase grounding fault occurs on line L2, the zero-mode equivalent network of the system is as follows: Figure 38 As shown in the figure, T1 is the main transformer, T2 is the DG grid-connected transformer, and Z T0 For grid-connected transformers

[0473] Neutral point grounding impedance. Z L10 、Z L20 、Z L30 are line mode impedances of lines L1 to L3 respectively; Z T20 is the line mode impedance of the grid-connected transformer; For this reason

[0474] Virtual voltage at the barrier point; Z DG0 =Z L30 +Z T20 +3Z T0 .

[0475] Depend on Figure 38 It can be seen that when the grid-connected transformer adopts the Yn / △ connection method, the system zero-mode current is:

[0476]

[0477] The zero-mode current flowing through L1 is:

[0478]

[0479] The zero-mode current provided by the DG and its grid-connected transformer is:

[0480]

[0481] When the high-voltage side of the DG grid-connected transformer adopts a star connection, a grounding point is introduced, and the zero-mode current of the fault line is jointly provided by the system main power supply and the DG. The DG connection location and the equivalent zero-sequence impedance of the grid-connected transformer both affect the magnitude of the system zero-mode current. However, the zero-mode current of the fault line is still the vector sum of the zero-mode currents of the non-fault lines. When the high-voltage side of the grid-connected transformer adopts a delta connection, the topology of the zero-mode network after DG connection is the same as before DG connection, and the distribution of transient electrical quantities in the zero-mode network will not be affected. In actual active distribution networks, the high-voltage side of the DG grid-connected transformer often adopts a delta connection. To be consistent with actual conditions, subsequent research and case analysis are carried out based on the premise that the high-voltage side of the DG grid-connected transformer is ungrounded.

[0482] If we analyze the impact of multiple DGs connected to the distribution network on fault characteristics, we can use the superposition theorem to equate the effects of multiple DGs to the effects of a single DG acting separately and then superimposed. It should be noted that when multiple DGs are connected to the distribution network simultaneously, the line-mode equivalent impedance upstream and downstream of the fault point may change. This is because some DGs may be located upstream of the fault point or on healthy lines, while some DGs may be located downstream of the fault point. At the same time, the difference in DG access capacity will also affect the size of the line-mode equivalent impedance. The above analysis shows that it is extremely difficult to design a method suitable for locating small current grounding fault sections in active distribution networks using phase current, phase voltage, or line-mode components.

[0483] 3.3 Evaluation of the impact of large-scale distributed power access on traditional fault protection methods

[0484] Traditional fault protection methods generally include two methods: phase sequence current amplitude differential protection and phase sequence current phase difference protection. This paper analyzes the impact of large-scale distributed power access on these two traditional fault protection methods.

[0485] (1) Analysis of differential protection based on phase sequence current amplitude

[0486] In order to explore the impact of large-scale distributed power on the fault location criteria currently proposed, a fault location criterion is constructed in PSCAD. Figure 39 The figure shows a typical distribution network with a source distribution network. This network, derived from an improved actual distribution network topology, features typical distribution network characteristics, including widespread load distribution, numerous branches, diverse line types, and distributed generators (DGs). Node 1 is the main network access point, and nodes 12, 21, and 25 are initially connected to distributed generation sources. The network load capacity is 5.6 MW. The lines utilize a centralized π-type equivalent circuit. Sections 14-18 are cable-connected, and the remainder are overhead lines.

[0487] Traditional differential protection criteria primarily select characteristic quantities based on phase sequence current amplitude and phase difference. The above system was constructed to verify the applicability of the above algorithm. Existing literature applies to longitudinal protection schemes for distribution networks containing distributed generation, primarily using positive sequence current amplitude differential comparison. Assuming that in any segment AB, ρ is the ratio of the current amplitudes at both ends of the segment, B and A, the following analysis shows:

[0488]

[0489] Where y is the distance between the fault point and end A. Given a constant equivalent potential and impedance on both sides, it's easy to prove that ρ increases monotonically with respect to y. That is, the closer the fault location is to end B, the greater ρ becomes. When y = 1, i.e., a three-phase short-circuit fault occurs at the end of the line, ρ is at its maximum. Based on this, the scalar product braking criterion is given as:

[0490]

[0491] The assignment of the new criterion coefficient must meet the requirements of the overall braking performance of the criterion. The assignment is based on the prediction of the fault severity according to the simplified DG model of the traditional power supply, and at the same time, sufficient reliability margin is left in the braking area and the action area to adapt to the problem of reduced protection sensitivity and reliability caused by the error of the DG equivalent model. It is set as:

[0492]

[0493] Sections 10-14 and 14-18 were selected as observation sections, with measurement devices installed at their nodes. Short-circuit faults were set at nodes 13 and 17, respectively, and the total penetration rate of distributed generation in the distribution network was adjusted. The test results are shown in Table 1.

[0494] Table 1 Performance of the standard product braking criterion under different DG penetration rates

[0495]

[0496] The simulation results shown in Table 1 show that as the DG penetration rate increases, the difference in positive-sequence current amplitudes between the two ends of the faulted section gradually decreases, becoming more similar to the positive-sequence current amplitudes in the non-faulted section. The simulation data also show that when the DG penetration rate is below 50%, the differential protection criterion can accurately locate the faulted section. However, as the penetration rate increases, the faulted section is missed at penetration rates above 50%, and the difference in the criterion between the faulted and non-faulted sections gradually converges. When the DG penetration rate is 60% or higher, the threshold calculation of the differential protection criterion proposed in existing literature has no real roots, because the DG positive-sequence impedance is greater than the main grid positive-sequence impedance, resulting in an unsolvable criterion (no solution after 0.26 seconds of the simulation at a penetration rate of 50%). This is primarily due to the fact that as the DG penetration rate increases, the difference in positive-sequence current amplitudes between the two ends of the faulted section gradually decreases, becoming more similar to the positive-sequence current amplitudes in the non-faulted section.

[0497] In addition, if the DG distribution is adjusted so that Figure 39 When DG3 moves to node 7, causing the number of DGs upstream of the section to exceed that downstream, the proposed criterion fails, and the threshold θ cannot be found as a real root. This is because the threshold value in this criterion depends on the positive-sequence impedance of each source in the distribution network during a fault, which is obtained by taking the inverse cosine function of the ratio of the positive-sequence impedances of the upstream and downstream sources in the section. Due to the low output of DGs, their positive-sequence impedance during a fault is generally greater than that at the main grid access point. However, when the upstream section contains a large number of distributed generators, the ratio of the positive-sequence impedances of the upstream and downstream sources is greater than 1, and the inverse cosine function has no solution. Therefore, the differential protection criterion proposed in the existing literature has limited applicability and cannot accurately locate the fault.

[0498] In summary, the differential protection criteria proposed in existing literature are based on the positive-sequence current amplitude at both ends of a section, and the threshold θ is constructed based on the principle that the positive-sequence impedance is larger during a DG short-circuit fault. Like most algorithms for segment location based on current amplitude, they are significantly affected by the DG penetration rate. Furthermore, due to the calculation method of the algorithm's threshold, the differential protection criteria proposed in existing literature are only applicable to sections where the capacity of various upstream power sources is greater than that of downstream power sources. Given the complex and variable structure of distribution networks, these conditions cannot be met, making these algorithms inapplicable.

[0499] Existing literature proposes a new protection scheme based on fault current amplitude differences. To ensure the sensitivity of internal faults and the reliability of protection during external faults and normal operation, a corresponding action-restraint characteristic equation is proposed. Specific method: Following the traditional scalar product restraint differential protection expression, the restraint equation that only reflects the current amplitude information is given as:

[0500]

[0501] In the formula, the independent variables are the fault current amplitudes on both sides; the coefficient k determines the braking characteristics and is assigned by the difference in current amplitude before and after the fault. The specific assignment method is:

[0502]

[0503] Where θ is defined as the equivalent phase angle, expressed as:

[0504]

[0505] In the formula, the current with subscript f corresponds to the fault current at the protection installation; the current with subscript n corresponds to the load current during normal operation. Clearly, θ has no actual physical meaning; it is instead an angle derived from the change in current amplitude before and after the fault. Similar verification has shown that DG penetration has a certain impact on the fault current amplitude difference protection schemes proposed in existing literature. When the penetration rate is below 50%, the criterion can accurately locate the fault. However, when the penetration rate is greater than 50%, the criterion threshold becomes unsolvable. At this point, if the criterion is set to zero, the algorithm will miss the fault section.

[0506] The fault current amplitude differential protection scheme proposed in existing literature essentially constructs a criterion based on the positive sequence current amplitude at both ends of the section as a characteristic quantity, and uses the fault current variation at both ends of the section to construct the criterion threshold. When the DG penetration rate is less than 50%, the main grid short-circuit current is greater than the DG short-circuit current. The threshold for the fault section can be calculated using a formula. The fault currents at both ends of the non-fault section are approximately the same, so the threshold provides a differentiated criterion result. When the DG penetration rate is greater than 50%, the main grid short-circuit current and the DG short-circuit current are approximately the same, and the criterion difference gradually decreases, which can easily lead to missed detections.

[0507] Further analysis shows that the DG type also affects the judgment criteria: the short-circuit current performance of motor-type DG is similar to that of the main grid. When a fault occurs, a short-circuit current of 3-8 times the rated current will be generated. Therefore, when the distribution network is composed entirely of motor-type DG, the impact of the penetration rate on the judgment criteria is more obvious. When a fault occurs, inverter-type DG only generates a short-circuit current of 1.2-2 times the rated current. Compared with motor-type DG, the penetration rate has less impact on its judgment criteria.

[0508] (2) Traditional phase sequence current phase difference protection analysis

[0509] Existing literature has proposed a characteristic of the change of the phase angle of the current phasor before and after the fault. The following is the sudden change of the positive sequence current phase angle and the direction of the total current phase mutation

[0510]

[0511] Where, is the steady-state current phasor value before the fault. Regardless of the direction of the initial load power, for an internal fault, the current phase angle mutation at both ends of the line is in opposite directions, while for an external fault, the current phase angle mutation is in the same direction. Studies have shown the relationship between the direction of the current phase angle mutation and the fault location. Assuming the current reference direction is from the node to the protected section, it is assumed that for an internal fault, the current phase angle mutation at both ends of the line is in opposite directions, while for an external fault, the current phase angle mutation is in the same direction. The phase angle range used is [-180°, 180°].

[0512] In the analysis of the impact of different types of distributed power sources on phase angle mutation, the current phase angle mutation is mainly related to the system impedance angle and the line impedance angle. When the distributed power penetration rate is low, the downstream connection of the section is mainly load, and the system equivalent impedance angle is small. Therefore, the current change phase angle is small. As the penetration rate increases, the system impedance angle downstream of the section gradually increases.

[0513] The segment location method based on phase angle mutation is theoretically unaffected by the type of distributed generation (DG) or its penetration rate, and exhibits good adaptability. As the DG penetration rate increases, the system equivalent impedance angle gradually increases, but still meets the criteria. When the DG penetration rate is above 50%, and the initial load power on some lines is low, the systems at both ends of the line are weakly connected, and the phase angle change relationship is less pronounced.

[0514] Existing literature proposes a directional overcurrent protection scheme based on the phase change of the positive sequence component of the distribution network current. This scheme first uses the phase angle difference of the positive sequence component of the current in the cycle before the fault and the cycle after the fault to determine the fault direction, and then determines the fault phase by comparing the difference between the maximum and minimum values ​​of the three-phase current absolute values ​​after the difference interval of 0.01s. The positive judgment criterion is The reverse criterion is in, is the phase angle difference between the positive sequence fault component of the protection current and the line current during normal system operation, and θ is the directional element blocking angle. This method is similar to the previous method and, after analysis, yields the same conclusion as criterion 1.

[0515] In summary, differential protection schemes based solely on segment current amplitude comparisons are not well suited for positioning distribution networks with high DG penetration or dual-terminal main grid access, as their primary criterion is that the short-circuit current on the main grid should be greater than the short-circuit current on the DG side. Differential protection schemes based solely on segment current phase differences are theoretically unaffected by DG penetration and type. However, when PMUs are not fully deployed—that is, when load distribution is included within the PMU segment—non-fault segments with high DG penetration are prone to phase differences exceeding 90°, leading to algorithm misjudgments. Furthermore, load fluctuations can easily cause loaded segments to exhibit phase difference characteristics similar to those of faulty segments, causing the algorithm to misjudge.

[0516] In actual distribution networks, economic constraints currently prevent measurement devices from achieving full system configuration. The presence of load within sections inevitably affects the performance of traditional fault detection criteria. Furthermore, due to natural conditions, DG output fluctuates significantly. Given high DG penetration, the same criterion must be adaptable to fault location at varying penetration rates. Furthermore, distribution network topologies vary, and power supply schemes are flexible and adaptable, making fixed setting methods often inadequate. Using PMUs (Pulsed Power Management Units) can generate accurate synchronized phasor data, allowing for accurate load-related feature extraction and adaptability to short-circuit fault location in distribution networks at any penetration rate.

[0517] For traditional radial distribution networks in rural / mountainous areas, the fault characteristics of four typical fault types are analyzed in detail. For typical low-current grounding faults in rural / mountainous areas, the steady-state and transient characteristics of single-phase grounding faults are analyzed for both ungrounded neutral points and grounded neutral points via arc suppression coils.

[0518] The impact of distributed generation (DG) integration on distribution network fault characteristics is related to DG type, fault location, DG neutral point wiring method, DG control method, DG location, and DG capacity. This paper analyzes the impact of large-scale DG integration on distribution network short-circuit fault characteristics. The influence of various factors, such as DG type, location, and fault impedance, on fault current amplitude and phase are systematically analyzed and summarized. The limitations of traditional short-circuit fault differential protection algorithms used in DG integration scenarios are discussed. Overall, an increase in the penetration rate of motor-type DGs (especially above 50%) will reduce the reliability of traditional amplitude criterion to the point of failure, and even eliminate the basis for setting the criterion itself. While the phase criterion is superior to the amplitude criterion, it is difficult to comprehensively consider the effects of DG switching, DG output variations, transition resistance changes, and load fluctuations within the zone, and to distinguish these from fluctuations in the fault section characteristics, when measurement equipment is not fully configured. The grounding method of DG in my country's small current grounding system is ungrounded. The structure of the positive-sequence network and the negative-sequence network will change due to the access of distributed power sources, which will cause the diagnostic criteria designed for the phase component to lose their original reliability; but the zero-sequence network will not be affected by the access of DG. Therefore, the diagnostic criteria designed based on the zero-sequence quantity can still maintain their original reliability in principle, but it is necessary to consider the reflection of the injected harmonics in the zero-sequence network after the access of DG.

Claims

1. A method for analyzing the fault characteristics of large-scale distributed power sources connected to rural distribution networks, characterized by: The following steps are involved: Step 1: Analyze the fault characteristics of rural distribution networks, including short-circuit fault characteristics, low-current grounding fault characteristics, and arc high-resistance fault characteristics; Step 2: Establish a fault equivalent model for distributed power sources, including an inverter-type distributed power source fault equivalent model and a small hydropower fault equivalent model that takes into account the excitation control response; Step 3: Analyze the fault characteristics of the rural distribution network with distributed generation access, including the analysis of short-circuit fault characteristics after distributed generation access, the analysis of small current grounding fault characteristics after distributed generation access, and the impact assessment of large-scale distributed generation access on traditional fault protection methods.

2. The method for analyzing fault characteristics of large-scale distributed power supply access to rural distribution network according to claim 1 is characterized in that: In step 1, the short circuit fault characteristics include three-phase short circuit, two-phase short circuit, single-phase ground short circuit, and two-phase ground short circuit; In a three-phase short circuit, for the steady-state current characteristics, the approximate calculation formula for the effective value of the three-phase short-circuit current of the rural distribution line is: Where U N is the system rated voltage; c is the voltage coefficient; cU N is the system equivalent voltage source voltage; Z s1 is the system positive sequence impedance after the medium voltage busbar of the substation; Z L1 R is the positive sequence impedance of the circuit formed by the line between the substation busbar and the fault point and the ground; k is the fault resistance; Record the rated short-circuit capacity S at the busbar k With rated line voltage U LP , the system positive sequence impedance is calculated as: The three-phase short-circuit current consists of a steady-state component and a transient component. The non-periodic component in the transient short-circuit current makes the effective value of the transient three-phase short-circuit current greater than the steady-state effective value. The expression of the transient three-phase short-circuit current is: Where: I m is the current amplitude when the system is operating normally; I pm is the current amplitude of the short-circuit cycle; α is the phase angle of the power supply voltage; It is the angle between the current and the circuit voltage when the system is operating normally; is the phase angle between the short-circuit current and the loop voltage; T a is the decay time constant of the non-periodic component current; the expression of the short-circuit current periodic component is: The expression of the non-periodic component of short-circuit current is: The three-phase short-circuit impulse current is related to the short-circuit phase angle and the grid time constant. The smaller the short-circuit phase angle, the larger the time constant and the higher the impulse current amplitude. In a two-phase short circuit, the positive sequence impedance and negative sequence impedance of the three-phase symmetrical line are equal; the medium voltage distribution network is far away from the system power supply, and the approximate calculation formula for the effective value of the two-phase short circuit current is: If the transition resistance is zero, then In a single-phase ground short circuit, the effective value of the short-circuit current is simplified to R n The neutral point grounding resistance of the main transformer; In a two-phase ground short circuit, the short-circuit currents of the two fault phases are equal, and the effective value calculation formula is: Where: Z1=Z S1 +Z L1 , Z0=Z S0 +Z L0 , a=e 120j is the operation factor; When two phases are short-circuited to ground, the effective value of the grounding current at the fault point is In a low-resistance grounded distribution network, Z0>>Z1, the above two equations are further simplified to This shows that when a two-phase ground short circuit occurs in a low-resistance grounded distribution network, the short-circuit current of the fault phase is basically the same as that when a two-phase short circuit occurs.

3. The method for analyzing fault characteristics of large-scale distributed power supply access to rural distribution network according to claim 1 is characterized in that: In step 1, the characteristics of the small current grounding fault include the steady-state characteristics of the single-phase grounding fault in the neutral point ungrounded system, the steady-state characteristics of the single-phase grounding fault in the neutral point resonant grounded distribution network, and the transient characteristics of the small current grounding fault. The steady-state characteristics of a single-phase grounding fault in a neutral point ungrounded system include: 1) When single-phase grounding occurs, the voltage relative to the fault drops to zero, and the voltage relative to the non-fault increases to the original value. times, which is the line voltage, and zero-sequence voltage appears in the entire system; 2) The zero-sequence current of the non-fault line is equal to the sum of the three-phase capacitance current to ground, and the direction of capacitive reactive power flows from the busbar to the line; 3) The zero-sequence current of the fault line is the sum of the capacitance current to ground of all non-fault lines, and the direction of capacitive reactive power flows from the line to the busbar; 4) The zero-sequence current of the non-fault line leads the zero-sequence voltage by 90°, and the zero-sequence current of the fault line lags the zero-sequence voltage by 90°; The steady-state characteristics of a single-phase ground fault in a neutral-grounded resonant distribution network include: The condition of the non-fault line is the same as that of a neutral-ungrounded system; the zero-sequence current is equal to the sum of the three-phase capacitive currents to ground during normal operation, and the capacitive reactive power flows from the busbar to the line; the zero-sequence current of the fault line is no longer equal to the sum of the zero-sequence currents of the non-fault lines, and its capacitive reactive power flows from the busbar to the line, just like in the non-fault line. The transient characteristics of a low-current grounding fault include: the initial fault phase angle, the grounding point transition resistance, and the feeder parameters. These three conditions affect the transient characteristics of the zero-mode current of a low-current grounding fault, and their effects on the fault transient current are coupled. The transient current of a single-phase grounding fault is divided into capacitive and inductive components. At the onset of the fault, the phase determines the ratio of the capacitive transient to the inductive transient, which in turn affects the amplitude of the transient current. When the fault occurs at the instant when the phase voltage approaches its maximum value, within the first transient half-wave, the transient zero-mode voltage has the opposite polarity to the zero-mode current of the fault line, but the same polarity as the transient zero-mode current of the non-fault line. In a neutral-ungrounded distribution network, the transient zero-mode current of the fault line flows from the line to the busbar, while the transient zero-mode current of the non-fault line flows from the busbar to the line. The derivatives of the transient zero-mode current and the zero-mode voltage of the fault line are always opposite in polarity, while the derivatives of the transient zero-mode current and the zero-mode voltage of the non-fault line are always the same in polarity.

4. The method for analyzing fault characteristics of large-scale distributed power supply access to rural distribution network according to claim 1 is characterized in that: In step 1, the arc high-resistance fault characteristics include steady-state characteristics—phase current, power frequency amplitude and phase of voltage, steady-state characteristics—power frequency zero-sequence voltage, power frequency amplitude of zero-sequence current, steady-state characteristics—low-order harmonic amplitude and phase, steady-state characteristics—randomness, transient characteristics—transient high-frequency signal, and transient characteristics—waveform morphology.

5. The method for analyzing fault characteristics of large-scale distributed power supply access to rural distribution network according to claim 1 is characterized in that: In step 2, the inverter-type distributed power supply fault equivalent model includes a distributed power supply fault analysis model considering constant power control and a distributed power supply fault analysis model considering low voltage ride-through; In the distributed power supply fault analysis model considering constant power control, PQ control is used when the distributed power supply inverter is connected to the distribution network and operated in grid-connected mode. PQ control controls the active power and reactive power output of the inverter. Under normal operating conditions, it maintains a constant active power output while also having a certain reactive power regulation capability. Under PQ control, the output voltage and frequency are determined by the grid. During grid-connected operation, the AC side of the inverter obtains the grid current and voltage, performs Park transformation, and decouples the current and voltage into active and reactive power to obtain the instantaneous active and reactive power output by the distributed power supply. Since the d-axis coincides with the grid-connected point voltage vector in Park transformation, we have Deduced: i dref and i qref is the given active current and reactive current, i d and i q It is the active component and reactive component of the inductor current. After the feedback current is compared with the given current, the SPWM modulation wave is obtained through PI regulation and inverse DQ transformation. The SPWM output controls the switch off to form a closed-loop control system, thereby controlling the inverter output current, and thus controlling the output active power and reactive power. The phase-locked voltage is the grid voltage u a 、u b 、u c , the phase-locked loop provides angle reference for DQ conversion and inverse DQ conversion; In the analysis model of distributed generation faults with low voltage ride-through, from the time the fault occurs until the voltage recovers to 0.9 pu, the reactive power output by the DG should track the voltage change at the grid connection point and meet the following requirements: Among them, u g.f is the per-unit voltage value of the distributed power access point after a fault occurs, I q.f is the reactive current value output by the distributed power supply after the fault, I N is the rated current value; in actual engineering, in order to provide sufficient reactive power support, u g.f When it is in the interval [0.2, 0.9], the coefficient is 1.5; therefore, the direct grid-connected distributed power supply considering low voltage ride-through is equivalent to a voltage-controlled current-controlled source; Considering the inverter's overcurrent capability and low voltage ride-through control, the output active current setting is: Among them, I d.f is the active power output when the system fails, P m is the active power output at the fault point, I max is the maximum current allowed to flow through the inverter. Therefore, considering the LVRT control strategy and inverter current limiting constraints during distribution network faults, as well as the provision of greater reactive power support when the voltage drops significantly, the relationship between the inverter output fault current and the grid connection point voltage is as follows: The active component of the fault current is superimposed with the grid connection point voltage vector, and δ represents the phase angle of the positive sequence voltage at the grid connection point; From the following formula, we can see that there is a boundary voltage u x Make the inverter output current just reach the output current amplitude limit; that is: Let P m =1,I max =2I N , then u x =0.537, the inverter reaches the current limiting constraint boundary; at this time, the phase angle of the fault current is: When u g.f When it is in the interval [0.537, 0.9], the inverter can adjust the reactive power output while ensuring that the active power output remains unchanged, and is in the grid-connected control stage; when u g.f When the inverter is in the interval [0.2, 0.537], it is constrained by the control strategy and the inverter current limiting condition, and cannot guarantee the active power output remains unchanged, and is in the low voltage ride-through control stage; When an asymmetric fault occurs, the active component of the fault output current is oriented toward the positive sequence voltage component of the grid connection point, that is, the inverter grid-connected distributed power supply is equivalent to a current source controlled by the positive sequence fault voltage of the grid connection point, that is:

6. The method for analyzing fault characteristics of large-scale distributed power supply access to rural distribution network according to claim 1, characterized in that: In step 2, the small hydropower fault equivalent model taking into account the excitation control response includes a simplified excitation system model and a small hydropower VCCS fault equivalent model; In the simplified model of the excitation system, U ref is the set terminal reference voltage; Ut is the terminal voltage; ΔU is the terminal voltage deviation value; E f is the stator excitation potential output to the generator; T A 、T B1 、T B2 、T C1 and T C2 K is the lead / lag time constant of each link; R and K A is the gain multiple of each link; V Rmax is the peak value of the generator excitation potential; Since the self-shunt static excitation system has a high response speed, the dynamic process of the excitation system can be ignored when calculating the fault current, and its excitation characteristic equation is simplified to: E f =min(K v (U ref -U t ),V Rmax ) Where: min(·,·) is the minimum function; Kv is the gain coefficient of the excitation system, and the gain coefficient of the small hydropower unit is between 30 and 150; In the small hydropower VCCS fault equivalent model, the steady-state d-axis and q-axis stator voltage and current relationship after the small hydropower unit fault is: Where: ud, uq and id, iq are the d-axis and q-axis components of the stator voltage and stator current respectively, Xd and Xq are the d-axis and q-axis components of the stator steady-state synchronous reactance respectively, Eq is the steady-state open-circuit potential, ra is the stator resistance, and the above variables are all per-unit values; Since the stator reactance of the synchronous generator is much larger than the stator resistance, ra≈0 is assumed in the analysis. In the steady-state process after a fault, there is no current in the damping windings D and Q, and only the excitation winding potential exists on the stator q-axis. Therefore, the stator steady-state open-circuit potential is equal to the stator excitation potential Ef, that is, Eq=Ef. After making the above assumptions, it is simplified to: The above equation shows that the fault output current of a small hydropower plant is related to the generator terminal voltage and the excitation potential. Since small hydropower plants are usually directly connected to the distribution network in a "T" connection and the connecting line is very short, it is approximately assumed that the small hydropower generator terminal voltage is equal to the voltage UPCC at the common connection point PCC. The small hydropower unit under fault ride-through is equivalent to a VCCS model in which the fault output current IG is controlled by the voltage at the PCC.

7. The method for analyzing fault characteristics of large-scale distributed power supply access to rural distribution network according to claim 1, characterized in that: In step 3, the short-circuit fault characteristic analysis after the distributed generation is connected specifically includes: 1) when a three-phase short-circuit fault occurs upstream of the feeder where the DG is located, the DG is decoupled from the system power supply side circuit, and the DG does not affect the short-circuit current supplied by the system; the connection of the DG does not affect the short-circuit current characteristics of the distribution network; 2) When the fault is not severe and the DG is outputting active current, the grid connection point voltage is affected by the DG output power, first increasing and then decreasing as the DG output power increases. When the fault is severe and the DG is only outputting reactive current, the grid connection point voltage is affected by the DG rated power, first increasing and then decreasing as the DG rated power increases. The change pattern of the grid connection point voltage is independent of the fault type and fault location. 3) When the fault is not severe and the DG is outputting active current, the short-circuit current flowing through the upstream feeder where the DG is located first decreases and then increases with the increase of the DG access capacity; the short-circuit positive sequence current flowing through the downstream feeder where the DG is located and the adjacent feeders first increases and then decreases with the increase of the DG access capacity; when the fault is severe and the DG is only outputting reactive current, the current flowing through all protections of the distribution network is only related to the DG rated power and has nothing to do with the DG output power; 4) Although the common bus voltage and the current of the fault branch adjacent to the DG first increase and then decrease with the increase of the DG access capacity, the impact of the access of the adjacent line DG on the common bus voltage and the fault branch fault current is very small due to the support of the system on the common bus voltage; 5) At the DC input end, the inverter generally adopts constant power control mode, and the output characteristics of the inverter have the characteristics of a current source; the pulse width modulation signal frequency of the inverter is several thousand hertz, and the response speed is only a few milliseconds. Therefore, ignoring the transient process of the inverter itself, the maximum output current is generally 1.2-1.5 times the rated current; 6) When a distribution network fault occurs, the IIDG will maintain power supply to the distribution network through the inverter before it is disconnected from the grid due to short-circuit protection. The output short-circuit current is related to the specific control strategy adopted during the fault stage. Taking the more commonly used constant power control as an example, when a distribution network fault occurs, the inverter maintains its pre-fault active power and reactive power output. When an asymmetric fault occurs, the inverter first calculates the positive sequence voltage at the grid connection point, and then calculates the short-circuit current that the inverter needs to output. If the grid connection point voltage is low, the calculated target output current exceeds its maximum allowable output current. The inverter will maintain the output current amplitude at the maximum current level, and the active and reactive power output of the inverter will also be proportionally reduced. If the grid connection point voltage is lower than the set threshold value, the inverter will stop outputting current.

8. The method for analyzing fault characteristics of large-scale distributed power supply access to rural distribution networks according to claim 1 is characterized in that: In step 3, the impact analysis on the steady-state fault characteristics specifically includes the impact analysis on the steady-state fault characteristics and the impact analysis on the transient characteristics.

9. The method for analyzing fault characteristics of large-scale distributed power supply access to rural distribution networks according to claim 1, characterized in that: In step 3, the impact assessment of large-scale distributed power access on traditional fault protection methods includes phase sequence current amplitude differential protection analysis and traditional phase sequence current phase difference protection analysis; Following the traditional scalar product braking differential protection expression, the braking equation that only reflects the current amplitude information is given as: In the formula, the independent variables are the fault current amplitudes on both sides; The coefficient k determines the braking characteristics and is assigned by the difference in current amplitude before and after the fault. The specific assignment method is: Where θ is defined as the equivalent phase angle, expressed as: Where, the current with subscript f corresponds to the fault current at the protection installation; the current with subscript n corresponds to the load current during normal operation; Positive sequence current phase angle mutation and the direction of the total current phase mutation Where, is the steady-state current phasor value before the fault.

Images