Relay protection analysis method and system for distributed photovoltaic access to distribution network

Abstract

The present invention discloses a distributed photovoltaic access distribution network relay protection analysis method and system, which belongs to the field of photovoltaic grid-connected technology, and specifically includes: injecting a dynamic resonance detection signal on the AC side of the converter of the photovoltaic grid-connected system, with the frequency matching the natural resonance frequency of the target section; analyzing the energy focusing effect of the dynamic resonance detection signal in the distribution network to generate a resonance energy distribution spectrum; when the energy density of a certain section in the resonance energy distribution spectrum exceeds a preset threshold, adjusting the photovoltaic output current waveform to a discontinuous waveform including periodic intermittent current interruption; constructing a space-time mapping model of the fault point based on the reflection delay difference of the waveform in the fault section, and analytically determining the fault location; according to the output result of the space-time mapping model, controlling the photovoltaic side circuit breaker and the grid side relay to perform a selective tripping action based on waveform phase synchronization; the present invention improves the sensitivity and reliability of the distributed photovoltaic access distribution network relay protection.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic grid-connected technology, and in particular to a method and system for analyzing relay protection of distributed photovoltaic access to a distribution network. Background Art

[0002] With the widespread adoption of distributed photovoltaic power generation in distribution networks, the topology and energy flow characteristics of these networks have undergone significant changes. Traditional passive relay protection technology is no longer able to meet the demands for rapid fault detection and precise isolation in complex operating scenarios. Active detection protection technology has emerged as a result. By injecting specific signals into the grid, it captures system response characteristics to identify potential faults, providing a new technical approach for relay protection and becoming a research hotspot in the field of smart grid protection.

[0003] Existing active detection protection technologies primarily detect faults by injecting fixed-frequency detection signals into the distribution network and analyzing the signal's propagation and reflection characteristics within the network. Some solutions utilize harmonic signal injection to determine the fault location based on the distribution patterns of harmonic currents; others employ pulse injection, locating faults based on the attenuation and delay characteristics of the pulse signal. These methods have improved fault detection to a certain extent and provide a practical foundation for active protection.

[0004] However, fixed-frequency detection signals are difficult to adapt to the real-time changing impedance characteristics of the distribution network, and cannot form effective resonance with the natural resonant frequency of a specific section of the power grid, resulting in insufficient extraction of fault signal features. In the fault location process, the dynamic changes in the power grid topology and the energy transfer characteristics during the fault development process are not fully considered, making it impossible to achieve advanced warning and effective isolation of the fault, affecting the power supply reliability and operational stability of the distribution network. Summary of the Invention

[0005] The purpose of the present invention is to provide a distributed photovoltaic access distribution network relay protection analysis method and system to solve the following technical problems:

[0006] Fixed-frequency detection signals are difficult to adapt to the real-time changing impedance characteristics of the distribution network, and cannot form effective resonance with the natural resonant frequency of a specific section of the power grid, resulting in insufficient extraction of fault signal features. In the fault location process, the dynamic changes in the power grid topology and the energy transfer characteristics during the development of the fault are not fully considered, making it impossible to achieve advanced warning and effective isolation of the fault.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] The relay protection analysis method for distributed photovoltaic access to the distribution network includes the following steps:

[0009] Injecting a dynamic resonance detection signal on the AC side of the inverter of the photovoltaic grid-connected system, wherein the frequency of the dynamic resonance detection signal is adaptively adjusted according to the real-time impedance characteristics of the distribution network so that the frequency matches the natural resonant frequency of the target section;

[0010] By analyzing the energy focusing effect of the dynamic resonance detection signal in the distribution network, a resonance energy distribution spectrum reflecting the fault characteristics is generated;

[0011] When the energy density of a certain section in the resonance energy distribution spectrum exceeds a preset threshold, the fault current reconstruction mode on the photovoltaic side is triggered, and the photovoltaic output current waveform is adjusted to a discontinuous waveform with periodic intermittent current interruption;

[0012] Based on the reflection delay difference of the intermittent interruption waveform in the fault section, a time-space mapping model of the fault point is constructed, and the fault location is determined by analyzing the time-space mapping model;

[0013] According to the output results of the spatiotemporal mapping model, the photovoltaic side circuit breaker and the grid side relay are controlled to perform selective tripping actions based on waveform phase synchronization.

[0014] As a further solution of the present invention: the dynamic resonance detection signal is specifically:

[0015] A random pulse sequence is embedded in the modulation wave of the inverter of the photovoltaic grid-connected system, wherein the width of the random pulse sequence is negatively correlated with the current fundamental frequency period of the distribution network;

[0016] By monitoring the slight fluctuations of the distribution network impedance in real time and dynamically adjusting the interval time of the random pulse sequence, the frequency of the dynamic resonance detection signal forms a dynamic envelope around the natural resonant frequency of the distribution network;

[0017] The frequency coverage range of the dynamic envelope is from a preset low-frequency octave band to a preset high-frequency octave band of the fundamental frequency of the distribution network, and the shape of the dynamic envelope is reshaped in real time according to the voltage harmonic distortion rate of the adjacent nodes.

[0018] As a further solution of the present invention: the generation process of the resonance energy distribution spectrum is:

[0019] Deploy bipolar magnetic field sensors at the headend and end of each branch line of the distribution network to capture the electromagnetic field vortex characteristics generated by the propagation of dynamic resonance detection signals;

[0020] Extract the magnetic field rotation direction and magnetic field intensity gradient information from the electromagnetic field vortex characteristics and calculate the energy absorption coefficient of each line section;

[0021] The energy absorption coefficient is correlated with the dynamic envelope frequency in three dimensions to generate a three-dimensional map with frequency dimension, spatial dimension and energy density dimension as coordinate axes, and abnormal sections with sudden increase in energy density are marked.

[0022] As a further solution of the present invention: the fault current reconstruction mode specifically includes:

[0023] After detecting a sudden increase in energy density, the maximum power point tracking control on the photovoltaic DC side is switched to a chaotic modulation mode, and the photovoltaic output current exhibits an intermittent interruption characteristic by introducing a nonlinear carrier disturbance.

[0024] The duration of the intermittent interruption is positively correlated with the sudden increase in energy density, and the interruption interval is synchronized with the dynamic envelope frequency;

[0025] At the same time, a reverse compensation current is superimposed on the AC side of the inverter of the photovoltaic grid-connected system, so that the electromagnetic transient wave generated at the moment of current interruption forms a traceable reflection mark.

[0026] As a further solution of the present invention: the method for constructing the spatiotemporal mapping model is:

[0027] The equivalent electrical distance of the electromagnetic transient wave front propagation path is calculated by using the reflection time difference between the fault section and the non-fault section at the moment of power outage. The dynamic displacement trajectory of the fault point is identified by comparing the drift of the equivalent electrical distance in different power outage cycles. The dynamic displacement trajectory is matched with the preset power grid topology delay database to determine the physical grid coordinates of the fault point.

[0028] As a further solution of the present invention: the specific logic of the selective tripping action based on waveform phase synchronization is:

[0029] After the physical grid coordinates of the fault point are determined, the voltage phase variation characteristics of the grid nodes corresponding to the physical grid coordinates within the latest preset number of fundamental frequency cycles are extracted;

[0030] Control the PV-side circuit breaker to perform a pre-breakdown operation within a preset angle range before the voltage phase crosses zero, artificially creating a local arc to enhance the fault characteristics;

[0031] The action timing of the grid-side relay is synchronously adjusted so that the tripping moment of the grid-side relay is strictly aligned with the peak value of the electromagnetic radiation of the pre-breakdown arc, thereby achieving the spatiotemporal focusing of the protection action.

[0032] As a further solution of the present invention: the pre-breakdown operation specifically includes:

[0033] Within a preset angle interval before the voltage phase crosses zero, a high-frequency induced current is injected into the contacts of the photovoltaic-side circuit breaker to ionize the dielectric in the contact gap in advance. The amplitude gradient of the high-frequency induced current is dynamically adjusted based on the electrical properties of the physical grid coordinates of the fault point to match the ionization intensity with the fault severity. The optical signal spectrum characteristics during the ionization process are recorded as a basis for secondary verification of the fault type.

[0034] As a further solution of the present invention: after the first tripping action is completed, a low-frequency sweep signal is injected into the isolated section to monitor the harmonic regeneration phenomenon of the low-frequency sweep signal in the adjacent section;

[0035] By analyzing the frequency offset and amplitude attenuation slope of the harmonic regeneration signal, a fault energy transfer chain model is constructed. When the fault energy transfer chain model shows that the energy transfer chain length exceeds the preset safety value, the full-domain impedance reshaping program is automatically started to forcibly change the resonant frequency distribution of the non-fault section.

[0036] As a further solution of the present invention: the full-network impedance reshaping process specifically includes:

[0037] By coordinating the output phases of the inverter and energy storage device of the photovoltaic grid-connected system, a controllable standing wave interference phenomenon is artificially created in the distribution network; the position of the standing wave antinode is adjusted to cover the key nodes of the fault energy transfer chain, and the low impedance characteristics of the node area are used to absorb the residual fault energy; after the residual fault energy is absorbed, the original frequency distribution of the distribution network is gradually restored, and the stability of the distribution network is verified by full-band scanning of the dynamic resonance detection signal.

[0038] The present invention also includes a distributed photovoltaic access distribution network relay protection analysis system for implementing the above-mentioned distributed photovoltaic access distribution network relay protection analysis method, including:

[0039] A signal injection module is used to inject a dynamic resonance detection signal on the AC side of the inverter of the photovoltaic grid-connected system. The frequency of the dynamic resonance detection signal is adaptively adjusted according to the real-time impedance characteristics of the distribution network to match the natural resonant frequency of the target section;

[0040] A spectrum generating module, configured to generate a resonance energy distribution spectrum reflecting fault characteristics by analyzing the energy focusing effect of the dynamic resonance detection signal in the distribution network;

[0041] The waveform adjustment module is used to trigger the fault current reconstruction mode on the photovoltaic side when the energy density of a certain section in the resonance energy distribution spectrum exceeds a preset threshold, and adjust the photovoltaic output current waveform to a discontinuous waveform with periodic intermittent current interruption;

[0042] The fault location module is used to construct a time-space mapping model of the fault point based on the reflection delay difference of the intermittent current interruption waveform in the fault section, and determine the fault location through the time-space mapping model analysis;

[0043] The tripping control module is used to control the photovoltaic side circuit breaker and the grid side relay to perform selective tripping actions based on waveform phase synchronization according to the output results of the time-space mapping model.

[0044] Beneficial effects of the present invention:

[0045] The present invention injects a frequency-adaptive dynamic resonance detection signal on the AC side of the photovoltaic grid-connected system converter to match the natural resonant frequency of the target section, and uses the energy focusing effect to generate a resonance energy distribution map. This solves the problems that traditional fixed-frequency detection signals are difficult to adapt to grid impedance changes and insufficient fault signal feature extraction. When the energy density in the map exceeds the threshold, the photovoltaic side fault current reconstruction mode is triggered, and the output current is adjusted to a discontinuous waveform with periodic intermittent current interruption. A reverse compensation current is superimposed to form a reflection mark, enhancing the recognition of fault features. A spatiotemporal mapping model is constructed based on the reflection delay difference of the intermittent current interruption waveform in the fault section, and the fault location is determined in combination with the grid topology delay database, overcoming the defect that existing methods cannot consider dynamic changes in grid topology. By controlling the pre-breakdown operation of the photovoltaic side circuit breaker and the operation timing of the grid-side relay, selective tripping based on waveform phase synchronization is achieved, solving the problem of inaccurate fault removal. After the first trip, a low-frequency scanning signal is injected to construct a fault energy transfer chain model, and a full-domain impedance reshaping program is initiated to change the resonant frequency distribution, absorb residual fault energy, and prevent fault propagation, comprehensively improving the sensitivity, accuracy, and reliability of relay protection for distributed photovoltaic access to the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The present invention will be further described below with reference to the accompanying drawings.

[0047] Figure 1 It is a flow chart of the relay protection analysis method for distributed photovoltaic access to distribution network of the present invention. DETAILED DESCRIPTION

[0048] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0049] See also Figure 1 As shown, the present invention is a distributed photovoltaic access distribution network relay protection analysis method, comprising the following steps:

[0050] First, a dynamic resonance detection signal is injected into the AC side of the inverter of the photovoltaic grid-connected system. The system monitors the impedance changes of the distribution network in real time and uses an adaptive frequency adjustment algorithm to dynamically adjust the detection signal frequency. Specifically, by online calculating the natural resonant frequency of each section of the distribution network, the frequency of the injected dynamic resonance detection signal is accurately matched with the target section. In actual operation, a random pulse sequence is embedded in the modulation wave of the inverter of the photovoltaic grid-connected system. The width of this sequence is negatively correlated with the current fundamental frequency period of the distribution network. By tracking the tiny fluctuations of the distribution network impedance in real time, the interval time of the random pulse sequence is dynamically adjusted, so that the detection signal frequency forms a dynamic envelope around the natural resonant frequency of the distribution network. Its frequency coverage range is from the preset low-frequency octave band to the preset high-frequency octave band of the distribution network fundamental frequency, and the envelope shape is reshaped in real time according to the voltage harmonic distortion rate of the adjacent nodes, thereby ensuring that the detection signal can effectively stimulate the resonance effect of the power grid.

[0051] Next, bipolar magnetic field sensors deployed at the beginning and end of each branch line of the distribution network are used to capture the electromagnetic field vortex characteristics generated when the dynamic resonance detection signal propagates. The magnetic field rotation direction and magnetic field intensity gradient information in the electromagnetic field vortex characteristics are extracted, and the energy absorption coefficient of each line section is calculated using an energy analysis algorithm. Subsequently, the energy absorption coefficient is three-dimensionally correlated with the dynamic envelope frequency to generate a three-dimensional resonance energy distribution map with frequency, spatial, and energy density dimensions as coordinate axes. The map can intuitively display the energy distribution status of the detection signal in the power grid and automatically mark abnormal sections with sudden increases in energy density, providing a visual basis for fault diagnosis.

[0052] Once the energy density in a specific section of the resonant energy distribution spectrum exceeds a preset threshold, the system immediately triggers the fault current reconstruction mode on the photovoltaic side. At this point, the maximum power point tracking control on the photovoltaic DC side is switched to chaotic modulation mode. By introducing nonlinear carrier perturbations, the photovoltaic output current exhibits intermittent current interruption characteristics. The duration of the intermittent current interruption is positively correlated with the magnitude of the sudden increase in energy density, and the current interruption interval is synchronized with the dynamic envelope frequency. Simultaneously, a reverse compensation current is superimposed on the AC side of the photovoltaic grid-connected system's converter, causing the electromagnetic transient waves generated at the moment of current interruption to form a traceable reflection marker, further enhancing the fault characteristics.

[0053] Then, based on the difference in reflection delay between the intermittent outage waveform in the fault section and the non-fault section, a spatiotemporal mapping model of the fault point was constructed. The equivalent electrical distance of the electromagnetic transient wave front propagation path was calculated using the reflection time difference between different sections of the electromagnetic transient wave generated at the moment of outage. By comparing the drift of the equivalent electrical distance during different outage cycles, the dynamic displacement trajectory of the fault point was identified. This dynamic displacement trajectory was then matched with a pre-defined grid topology delay database to accurately determine the physical grid coordinates of the fault point.

[0054] Finally, based on the physical grid coordinates of the fault point output by the time-space mapping model, the system extracts the voltage phase variation characteristics of the grid node corresponding to these coordinates within the most recent preset number of fundamental frequency cycles. The system controls the photovoltaic circuit breaker to perform a pre-breakdown operation within a preset angle interval before the voltage phase crosses zero, injecting a high-frequency induced current into the circuit breaker contacts to ionize the dielectric in the contact gap in advance. Based on the electrical properties of the physical grid coordinates of the fault point, the system dynamically adjusts the amplitude gradient of the high-frequency induced current to match the ionization intensity with the severity of the fault. Simultaneously, the grid-side relay's action sequence is synchronously adjusted to ensure that the grid-side relay's tripping moment is strictly aligned with the peak electromagnetic radiation of the pre-breakdown arc. This achieves selective tripping based on waveform phase synchronization, quickly and accurately isolating the fault section and ensuring the safe and stable operation of the distribution network.

[0055] In a preferred embodiment of the present invention, the dynamic resonance detection signal is specifically:

[0056] The dynamic resonance detection signal utilizes an innovative pulse train modulation strategy to achieve deep coupling with the distribution network state. Specifically, a random pulse sequence is embedded in the modulation wave of the PV grid-connected system's converter. The width of this sequence follows a strict negative correlation mechanism—linked in real time to the current fundamental frequency period of the distribution network. When the fundamental frequency period shortens, the pulses narrow, and vice versa, they widen, ensuring the signal's dynamic adaptability. The system is equipped with a high-precision impedance monitoring module that captures subtle fluctuations in the distribution network's impedance at a millisecond sampling rate. An adaptive timing adjustment algorithm dynamically adjusts the intervals between the random pulse sequences. This adjustment process is based on real-time measurement of the distribution network's natural resonant frequency, aligning the detection signal frequency with the natural resonant frequency, forming a dynamically changing frequency envelope. This envelope's frequency range spans from a preset low-frequency octave to a high-frequency octave of the distribution network's fundamental frequency (e.g., 0.5-3 times the fundamental frequency). Its shape is not fixed but reshapes in real time based on the harmonic distortion of the voltage at adjacent nodes. When the distortion rate increases, the envelope automatically widens to enhance signal coverage; when the distortion rate decreases, the focus shrinks to increase energy density, ensuring that the detection signal always stimulates the grid resonance effect in the optimal form.

[0057] In another preferred embodiment of the present invention, the generation of the resonance energy distribution map relies on multi-dimensional electromagnetic signature capture and 3D modeling technology. Bipolar magnetic field sensor arrays are deployed at key nodes (headends and tailends) of each branch line in the distribution network. These sensors utilize fluxgate and Hall effect composite sensing technology to simultaneously detect dynamic changes in magnetic field intensity and direction, accurately capturing the electromagnetic field vortex characteristics generated during the propagation of dynamic resonance detection signals. An edge computing unit processes sensor data in real time, extracting the magnetic field rotation direction (clockwise / counterclockwise) and magnetic field intensity gradient information. This gradient information reflects the energy decay rate in space. An energy absorption coefficient calculation model is established based on electromagnetic field theory, incorporating parameters such as the rate of change of magnetic field intensity and angular velocity into the calculation to derive the energy absorption coefficient for each line section. A 3D data mapping algorithm is then used to correlate the energy absorption coefficient, dynamic envelope frequency, and line spatial coordinates. An intuitive 3D map is generated, with frequency as the vertical axis, spatial position as the horizontal and vertical axes, and energy density as color / height coding. The map has an intelligent marking function. When the energy density of a certain area exceeds the threshold set based on historical data and operating experience, the abnormal section is automatically highlighted, providing a visual basis for fault location.

[0058] In another preferred embodiment of the present invention, the fault current reconstruction mode specifically includes:

[0059] The fault current reconstruction mode utilizes a dual enhancement mechanism of chaotic modulation and electromagnetic signature. Upon detecting a sudden increase in energy density in the resonant energy distribution spectrum, the system immediately initiates a switch in the PV DC-side control strategy, converting conventional maximum power point tracking (MPPT) control to a chaotic modulation mode. This mode introduces a nonlinear carrier perturbation algorithm to induce periodic intermittent current interruptions in the PV output current. The duration of the interruption is strictly positively correlated with the magnitude of the sudden increase in energy density. For example, for every 10% increase in energy density, the interruption duration is extended by 20%. The interruption interval is synchronized with the dynamic envelope frequency to ensure that the interruption period is coordinated with changes in the detection signal frequency. Furthermore, a reverse compensation current injection module is deployed on the AC side of the PV grid-connected system's converter. Based on a fault transient analysis model, this module calculates the characteristics of the electromagnetic transient waves generated at the moment of interruption and generates a matching reverse compensation current. The injected compensation current is superimposed on the interruption transient wave to form a unique reflection signature. This signature, which contains characteristic encodings of frequency, phase, and amplitude, facilitates precise fault location identification using traveling wave analysis techniques, significantly improving fault feature recognition and location accuracy.

[0060] In another preferred embodiment of the present invention, the method for constructing the spatiotemporal mapping model is:

[0061] The construction of the spatiotemporal mapping model integrates electromagnetic transient analysis and dynamic trajectory matching technology to achieve high-precision fault location. The system captures electromagnetic transient waves generated by intermittent interruptions in photovoltaic output current in real time. Transient monitoring devices deployed at key nodes in the distribution network collect reflection signals from the transient waves in both faulty and non-faulty sections. Based on traveling wave transmission theory, the system utilizes the reflection time difference between different sections, combined with line parameters and wave velocity calculation models, to accurately calculate the equivalent electrical distance along the electromagnetic transient wavefront propagation path. Taking into account factors such as arc dynamics and line parameter fluctuations during fault development, the system continuously monitors multiple interruption cycles, comparing the drift of the equivalent electrical distance within each cycle. The data is smoothed using a Kalman filter algorithm to identify the dynamic displacement trajectory of the fault point over time. Simultaneously, a time delay database containing information such as grid topology, line length, and impedance parameters is established. The dynamic displacement trajectory is pattern-matched with the theoretical time delay data in the database. An optimization algorithm is used to determine the optimal matching solution, ultimately determining the physical grid coordinates of the fault point, with an error within 10% of the actual line node spacing.

[0062] In another preferred embodiment of the present invention, the specific logic of the selective tripping action based on waveform phase synchronization is:

[0063] The selective tripping action based on waveform phase synchronization utilizes a multi-step coordinated control strategy to improve fault isolation efficiency and accuracy. Once the physical grid coordinates of the fault point are determined, the system immediately retrieves real-time monitoring data from the grid node corresponding to that coordinate, extracting voltage phase variation characteristics within a preset number (e.g., 5-10) of fundamental frequency cycles, and achieving high-precision phase tracking using phase-locked loop technology. For photovoltaic circuit breakers, a pre-breakdown operation is triggered within a preset angle range (e.g., 15°-30°) just before the voltage phase crosses zero. This program uses a dedicated control module to inject a high-frequency induced current into the circuit breaker contacts. The current frequency is set to several kilohertz to tens of kilohertz, accelerating the ionization of the contact gap dielectric by leveraging the skin effect. Simultaneously, the amplitude gradient of the high-frequency induced current is dynamically adjusted based on electrical properties such as the line voltage level and calculated short-circuit current associated with the physical grid coordinates of the fault point. For example, for high-voltage and high-short-circuit current faults, the current amplitude is automatically increased to enhance ionization intensity, ensuring that the degree of ionization accurately matches the fault severity.

[0064] In a preferred embodiment of the present invention, the pre-breakdown operation specifically includes:

[0065] During the pre-breakdown operation, the system deploys spectral monitoring equipment near the circuit breaker to record the optical signal generated by the ionization of the contact gap in real time, and analyzes the spectral characteristics of the optical signal through spectral analysis technology. Different fault types (such as metallic short circuits and arc-to-ground short circuits) will generate optical signals with specific spectral fingerprints during the ionization process. The system compares the collected spectral characteristics with the preset fault spectrum database as a basis for secondary verification of the fault type, effectively reducing the risk of false protection operation. At the same time, the grid-side relay control unit adjusts the relay's action time based on the electromagnetic radiation signal characteristics generated by the pre-breakdown arc through a precise timing control algorithm to ensure that its tripping action is strictly aligned with the peak of the pre-breakdown arc electromagnetic radiation. This spatiotemporal focused protection action strategy can shorten the fault isolation time to within a few fundamental frequency cycles, significantly reducing the scope and duration of the fault's impact on the distribution network.

[0066] In another preferred embodiment of the present invention, to prevent the spread of fault energy and triggering a chain reaction, the system immediately initiates secondary monitoring and active defense mechanisms after the initial tripping action is completed and the faulty section is successfully isolated. An intelligent signal injection device injects a low-frequency sweeping signal into the isolated section. This signal frequency is set outside the normal operating frequency range of the distribution network (e.g., 5-20 Hz) to avoid interference with normal operating signals. Wideband monitoring sensors deployed along the line capture harmonic regeneration—i.e., variations in the signal's frequency components caused by abnormal line parameters, electromagnetic coupling, and other factors—as the low-frequency sweeping signal propagates through adjacent sections.

[0067] The system uses a time-frequency analysis algorithm to deeply analyze the harmonic regeneration signal, accurately calculating the signal's frequency offset (the difference between the actual frequency and the injected frequency) and amplitude attenuation slope (the rate at which the signal amplitude decays with propagation distance). Based on these characteristic parameters, combined with the distribution network topology and line electrical parameters, a fault energy transfer chain model is constructed. This model graphically presents the propagation path and intensity changes of fault energy in the network. Each node represents a key location in the power grid, and the connecting lines represent the direction and intensity of energy transfer. When the energy transfer chain length calculated by the model exceeds the preset safety value set based on factors such as the scale of the power grid and the tolerance of the equipment, the system determines that there is a risk of fault propagation and immediately triggers the full-network impedance reshaping program.

[0068] In a preferred embodiment of the present invention, the network-wide impedance reshaping procedure specifically includes:

[0069] During the implementation of the full-network impedance reshaping process, the system collaboratively controls the output phases of the photovoltaic grid-connected system's inverters and energy storage devices through the energy management platform. Leveraging the rapid control capabilities of power electronics, a controllable standing wave interference phenomenon is artificially created in the distribution network. By precisely calculating the output phase and amplitude of each device, the antinodes of the standing wave are precisely positioned at the critical nodes of the fault energy transmission chain—the hubs or weak links in energy transmission. The low impedance characteristics of the node regions efficiently absorb residual fault energy, acting as "energy traps" to block the fault propagation path.

[0070] During the residual fault energy absorption process, the system continuously monitors the voltage and current waveforms and energy flow data at each node. When it detects that the abnormal energy level in the network has dropped below a safe threshold, it initiates a frequency recovery procedure. This procedure incrementally adjusts the output parameters of the photovoltaic and energy storage devices to restore the distribution network's frequency distribution to its original operating state. Once the recovery is complete, the system re-injects a full-band dynamic resonance detection signal to scan the entire distribution network. By analyzing the detection signal's propagation characteristics, energy distribution, and network response, the system verifies the distribution network's operational stability and fault mitigation capabilities, ensuring the grid's return to a safe and reliable state.

[0071] The present invention also includes a distributed photovoltaic access to a distribution network relay protection analysis system, which is used to implement the above-mentioned distributed photovoltaic access to a distribution network relay protection analysis method, including:

[0072] A signal injection module is used to inject a dynamic resonance detection signal on the AC side of the inverter of the photovoltaic grid-connected system. The frequency of the dynamic resonance detection signal is adaptively adjusted according to the real-time impedance characteristics of the distribution network to match the natural resonant frequency of the target section;

[0073] A spectrum generating module, configured to generate a resonance energy distribution spectrum reflecting fault characteristics by analyzing the energy focusing effect of the dynamic resonance detection signal in the distribution network;

[0074] The waveform adjustment module is used to trigger the fault current reconstruction mode on the photovoltaic side when the energy density of a certain section in the resonance energy distribution spectrum exceeds a preset threshold, and adjust the photovoltaic output current waveform to a discontinuous waveform with periodic intermittent current interruption;

[0075] The fault location module is used to construct a time-space mapping model of the fault point based on the reflection delay difference of the intermittent current interruption waveform in the fault section, and determine the fault location through the time-space mapping model analysis;

[0076] The tripping control module is used to control the photovoltaic side circuit breaker and the grid side relay to perform selective tripping actions based on waveform phase synchronization according to the output results of the time-space mapping model.

[0077] The above is a detailed description of an embodiment of the present invention. However, the content described is only a preferred embodiment of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. The relay protection analysis method for distributed photovoltaic access to distribution network is characterized by: The following steps are involved: Injecting a dynamic resonance detection signal on the AC side of the inverter of the photovoltaic grid-connected system, wherein the frequency of the dynamic resonance detection signal is adaptively adjusted according to the real-time impedance characteristics of the distribution network so that the frequency matches the natural resonant frequency of the target section; By analyzing the energy focusing effect of the dynamic resonance detection signal in the distribution network, a resonance energy distribution spectrum reflecting the fault characteristics is generated; When the energy density of a certain section in the resonance energy distribution spectrum exceeds a preset threshold, the fault current reconstruction mode on the photovoltaic side is triggered, and the photovoltaic output current waveform is adjusted to a discontinuous waveform with periodic intermittent current interruption; Based on the reflection delay difference of the intermittent interruption waveform in the fault section, a time-space mapping model of the fault point is constructed, and the fault location is determined by analyzing the time-space mapping model; According to the output of the spatiotemporal mapping model, the PV-side circuit breaker and the grid-side relay are controlled to perform selective tripping based on waveform phase synchronization. The generation process of the resonance energy distribution spectrum is as follows: Deploy bipolar magnetic field sensors at the headend and end of each branch line of the distribution network to capture the electromagnetic field vortex characteristics generated by the propagation of dynamic resonance detection signals; Extract the magnetic field rotation direction and magnetic field intensity gradient information from the electromagnetic field vortex characteristics and calculate the energy absorption coefficient of each line section; The energy absorption coefficient is correlated with the dynamic envelope frequency in three dimensions to generate a three-dimensional map with the frequency dimension, spatial dimension and energy density dimension as coordinate axes, and abnormal sections with sudden increases in energy density are marked; The fault current reconstruction mode specifically includes: After detecting a sudden increase in energy density, the maximum power point tracking control on the photovoltaic DC side is switched to a chaotic modulation mode, and the photovoltaic output current exhibits an intermittent interruption characteristic by introducing a nonlinear carrier disturbance. The duration of the intermittent interruption is positively correlated with the sudden increase in energy density, and the interruption interval is synchronized with the dynamic envelope frequency; At the same time, a reverse compensation current is superimposed on the AC side of the inverter of the photovoltaic grid-connected system, so that the electromagnetic transient wave generated at the moment of current interruption forms a traceable reflection mark.

2. The distributed photovoltaic access distribution network relay protection analysis method according to claim 1 is characterized in that: The dynamic resonance detection signal is specifically: A random pulse sequence is embedded in the modulation wave of the inverter of the photovoltaic grid-connected system, wherein the width of the random pulse sequence is negatively correlated with the current fundamental frequency period of the distribution network; By monitoring the slight fluctuations of the distribution network impedance in real time and dynamically adjusting the interval time of the random pulse sequence, the frequency of the dynamic resonance detection signal forms a dynamic envelope around the natural resonant frequency of the distribution network; The frequency coverage range of the dynamic envelope is from a preset low-frequency octave band to a preset high-frequency octave band of the fundamental frequency of the distribution network, and the shape of the dynamic envelope is reshaped in real time according to the voltage harmonic distortion rate of the adjacent nodes.

3. The distributed photovoltaic access distribution network relay protection analysis method according to claim 1 is characterized in that: The construction method of the space-time mapping model is: The equivalent electrical distance of the electromagnetic transient wave front propagation path is calculated by using the reflection time difference between the fault section and the non-fault section at the moment of power outage. The dynamic displacement trajectory of the fault point is identified by comparing the drift of the equivalent electrical distance in different power outage cycles. The dynamic displacement trajectory is matched with the preset power grid topology delay database to determine the physical grid coordinates of the fault point.

4. The distributed photovoltaic access distribution network relay protection analysis method according to claim 1 is characterized in that: The specific logic of the selective tripping action based on waveform phase synchronization is: After the physical grid coordinates of the fault point are determined, the voltage phase variation characteristics of the grid nodes corresponding to the physical grid coordinates within the latest preset number of fundamental frequency cycles are extracted; Control the PV-side circuit breaker to perform a pre-breakdown operation within a preset angle range before the voltage phase crosses zero, artificially creating a local arc to enhance the fault characteristics; The action timing of the grid-side relay is synchronously adjusted so that the tripping moment of the grid-side relay is strictly aligned with the peak value of the electromagnetic radiation of the pre-breakdown arc, thereby achieving the spatiotemporal focusing of the protection action.

5. The distributed photovoltaic access distribution network relay protection analysis method according to claim 4 is characterized in that: The pre-breakdown operation specifically includes: Within a preset angle interval before the voltage phase crosses zero, a high-frequency induced current is injected into the contacts of the photovoltaic-side circuit breaker to ionize the dielectric in the contact gap in advance. The amplitude gradient of the high-frequency induced current is dynamically adjusted based on the electrical properties of the physical grid coordinates of the fault point to match the ionization intensity with the fault severity. The optical signal spectrum characteristics during the ionization process are recorded as a basis for secondary verification of the fault type.

6. The distributed photovoltaic access distribution network relay protection analysis method according to claim 1 is characterized in that: After the first tripping action is completed, a low-frequency sweep signal is injected into the isolated section to monitor the harmonic regeneration phenomenon of the low-frequency sweep signal in the adjacent sections; By analyzing the frequency offset and amplitude attenuation slope of the harmonic regeneration signal, a fault energy transfer chain model is constructed. When the fault energy transfer chain model shows that the energy transfer chain length exceeds the preset safety value, the full-domain impedance reshaping program is automatically started to forcibly change the resonant frequency distribution of the non-fault section.

7. The distributed photovoltaic access distribution network relay protection analysis method according to claim 6 is characterized in that: The full network impedance reshaping procedure specifically includes: By coordinating the output phases of the inverter and energy storage device of the photovoltaic grid-connected system, a controllable standing wave interference phenomenon is artificially created in the distribution network; the position of the standing wave antinode is adjusted to cover the key nodes of the fault energy transfer chain, and the low impedance characteristics of the node area are used to absorb the residual fault energy; after the residual fault energy is absorbed, the original frequency distribution of the distribution network is gradually restored, and the stability of the distribution network is verified by full-band scanning of the dynamic resonance detection signal.

8. A distributed photovoltaic access distribution network relay protection analysis system, used to implement a distributed photovoltaic access distribution network relay protection analysis method according to any one of claims 1 to 7, characterized in that: include: A signal injection module is used to inject a dynamic resonance detection signal on the AC side of the inverter of the photovoltaic grid-connected system. The frequency of the dynamic resonance detection signal is adaptively adjusted according to the real-time impedance characteristics of the distribution network to match the natural resonant frequency of the target section; A spectrum generating module, configured to generate a resonance energy distribution spectrum reflecting fault characteristics by analyzing the energy focusing effect of the dynamic resonance detection signal in the distribution network; The waveform adjustment module is used to trigger the fault current reconstruction mode on the photovoltaic side when the energy density of a certain section in the resonance energy distribution spectrum exceeds a preset threshold, and adjust the photovoltaic output current waveform to a discontinuous waveform with periodic intermittent current interruption; The fault location module is used to construct a time-space mapping model of the fault point based on the reflection delay difference of the intermittent current interruption waveform in the fault section, and determine the fault location through the time-space mapping model analysis; The tripping control module is used to control the photovoltaic side circuit breaker and the grid side relay to perform selective tripping actions based on waveform phase synchronization according to the output results of the time-space mapping model.

Images