Fault isolation and self-healing control method for multi-terminal flexible direct current power supply system

Abstract

The invention discloses a fault isolation and self-healing control method for a multi-terminal flexible direct current power supply system in the field of flexible direct current power supply, and aims to solve the technical problems that a traditional multi-terminal flexible direct current system is lagged in fault positioning, low in isolation speed, not timely in self-healing response and low in power supply recovery precision. According to the method, through a closed-loop control logic of fault feature extraction, accurate positioning, rapid isolation and dynamic self-healing, a fault identification model is constructed based on collaborative features of transient current and voltage, and millisecond-level positioning of a fault section is realized in combination with a simplified fault positioning formula; an IGBT series solid-state circuit breaker is adopted to realize rapid fault isolation, a self-healing scheduling strategy of load priority and resource adaptability is introduced, and power supply is recovered through cooperative scheduling of a distributed power supply and an energy storage unit. The fault isolation time is shortened, the power supply recovery success rate is improved, the voltage recovery deviation is reduced, and the power supply reliability and the operation toughness of the multi-terminal flexible direct current power supply system are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of flexible DC power supply, specifically a fault isolation and self-healing control method for multi-terminal flexible DC power supply systems, applicable to multi-terminal flexible DC power supply scenarios such as urban power distribution networks, new energy power plant clusters, and data centers. Background Technology

[0002] Multi-terminal flexible DC power supply systems, with voltage source converters (VSCs) at their core, possess significant advantages such as independent control of active and reactive power, strong power quality regulation capabilities, and large potential for renewable energy absorption. They have become a key technological carrier supporting the construction of new power systems and ensuring reliable power supply to core urban areas and renewable energy bases. With the large-scale application of this system in urban distribution networks, offshore wind power clusters, and data center microgrids, its topology is becoming increasingly complex (with as many as 4-8 converter stations and dense line branches). Its operating conditions exhibit characteristics of "multi-power source coordination, multi-load interaction, and frequent power fluctuations," significantly increasing the probability of fault occurrence and the difficulty of handling them. Currently, the fault isolation and self-healing control technologies of multi-terminal flexible DC power supply systems still face many technical bottlenecks that urgently need to be overcome, specifically:

[0003] The fault identification system suffers from weak anti-interference capabilities and a high risk of misjudgment and missed detection: Traditional fault identification often relies on single electrical quantity characteristics (such as current surge rate or voltage drop depth), failing to consider interference from factors such as converter station control strategies, line distributed capacitance, and fluctuations in renewable energy output in multi-terminal flexible DC systems. For example, current fluctuations caused by sudden changes in photovoltaic power plant output are easily misjudged as faults, while minor grounding faults are easily missed due to insignificant voltage drops, resulting in a fault identification accuracy rate generally below 92%, which seriously affects the timeliness of subsequent fault handling.

[0004] Fault location relies on complex models, making it difficult to balance accuracy and efficiency: Existing location technologies are mostly based on impedance measurement or traveling wave methods. The former is affected by line distributed parameters and the equivalent impedance of converter stations, resulting in location errors generally ≥2km; the latter requires high-precision time synchronization of multiple devices and complex signal analysis algorithms, leading to high hardware costs and location time ≥50ms, which cannot meet the requirements for millisecond-level fault isolation. Especially in multi-branch line scenarios, the superposition of traveling wave reflected signals leads to location ambiguity, further reducing location reliability.

[0005] Slow fault isolation speed and high risk of fault energy propagation: Traditional multi-terminal flexible DC systems mostly use mechanical circuit breakers or conventional thyristor solid-state circuit breakers. The arc extinguishing time of mechanical circuit breakers is ≥20ms, and the response time of conventional solid-state circuit breakers is ≥10ms, with the total isolation time generally exceeding 25ms. However, the fault current rises rapidly in multi-terminal flexible DC systems (up to 2kA / ms). Prolonged fault current can lead to overcurrent damage to converter valves, voltage collapse in non-faulty sections, and cause large-scale power outages in the system.

[0006] Self-healing dispatch lacks a precise adaptation mechanism, resulting in poor power restoration quality: Existing self-healing controls mostly adopt "equal restoration" or "fixed priority" strategies, failing to establish a dynamic adaptation relationship between resource supply and load demand. On the one hand, the available output and response speed of distributed power sources and energy storage units are not fully considered, leading to some restoration schemes being unable to be executed due to insufficient resources; on the other hand, load priority allocation lacks quantitative basis, critical loads (such as hospitals and emergency command centers) cannot be quickly guaranteed, and power output adjustment during the restoration process lacks precise model support, with voltage deviations generally exceeding ±5%, making it difficult to meet power quality standards.

[0007] Insufficient technical compatibility and engineering practicality: Existing fault handling solutions are mostly designed for multi-terminal flexible DC systems with specific topologies, and their adaptability to changes in renewable energy penetration and load type is poor. Some solutions rely on complex communication and coordination mechanisms, which are prone to failure in weak communication environments; at the same time, the core algorithms involve a large number of matrix operations and model training, which have high requirements for hardware computing resources, making it difficult to promote and apply them in small and medium-sized multi-terminal flexible DC systems.

[0008] Existing technologies are insufficient to meet the operational requirements of multi-terminal flexible DC power supply systems, which demand "rapid location, precise isolation, and efficient self-healing." Therefore, those skilled in the art have provided a fault isolation and self-healing control method for multi-terminal flexible DC power supply systems to address the problems mentioned in the background. Summary of the Invention

[0009] The purpose of this invention is to provide a fault isolation and self-healing control method for a multi-terminal flexible DC power supply system to solve the problems mentioned in the background art.

[0010] To achieve the above objectives, the present invention provides the following technical solution:

[0011] A fault isolation and self-healing control method for a multi-terminal flexible DC power supply system includes the following steps:

[0012] S1. Fault Feature Acquisition and Preprocessing: The transient current and voltage signals of the fault are acquired through the transient quantity measurement unit (TMU) of each node of the system. The sampling frequency is ≥10kHz. After filtering and noise reduction, the core characteristic parameters such as the amplitude of the current surge, the voltage drop ratio, and the transient energy are extracted.

[0013] S2. Fault Identification and Location: Based on the preprocessed feature parameters, the reliability of the fault is quantified by the fault judgment formula, and the fault section is calculated by combining the simplified fault location formula to achieve accurate location;

[0014] S3. Rapid fault isolation: Control the operation of IGBT series solid-state circuit breakers on both sides of the fault section to disconnect the fault circuit. The isolation time is ≤5ms to prevent the fault from spreading.

[0015] S4. Self-healing resource assessment: Real-time monitoring of distributed power output, remaining energy storage unit (SOC) and load priority within the system to establish a resource-load adaptation matrix;

[0016] S5. Self-healing scheduling optimization: Based on the adaptation matrix, the optimal power supply recovery scheme is selected through the self-healing adaptation formula, and distributed power sources and energy storage units are scheduled to supply power in coordination, prioritizing the recovery of high-priority loads.

[0017] S6. Closed-loop control during recovery process: Real-time monitoring of system voltage and power parameters after recovery, dynamic adjustment of power output, and ensuring stable power supply quality.

[0018] As a further aspect of the present invention: In step S2, the fault determination formula quantifies the fault reliability through current-voltage collaborative characteristics, specifically as follows:

[0019]

[0020] in, Fault confidence level (%) The magnitude of the fault current surge (kA). This is the system's rated current (kA). This represents the voltage drop amplitude (kV). This is the system's rated voltage (kV). It is identified as a genuine fault in a timely manner, with an accuracy rate of ≥98%.

[0021] As a further aspect of the present invention: In step S2, the simplified fault location formula is constructed based on the transient energy decay characteristics, specifically as follows:

[0022]

[0023] in, The distance (km) from the fault point to the measuring end. Transient energy (kJ) collected at the measurement end. This represents the theoretical value of the transient energy over the entire length of the line (kJ). The total length of the faulty line (km) is given, and the location error is ≤ ±0.5km.

[0024] As a further aspect of the present invention: In step S5, the self-healing adaptability formula is based on the matching of resource supply capacity and load demand, specifically as follows:

[0025]

[0026] in, For self-healing adaptation, The percentage of remaining electricity in the energy storage unit (0≤ ≤1), The proportion of available output of distributed power sources (0≤ ≤1), Load priority weight (0.1-0.4), prioritize the power supply combination scheme with the largest M.

[0027] As a further aspect of the present invention: In step S3, the operation logic of the IGBT series solid-state circuit breaker is linked with the fault reliability. When K≥70% and the location result is clear, a trip command is triggered, with an arc extinguishing time ≤2ms and a total isolation time ≤5ms.

[0028] As a further aspect of the present invention: in step S1, the filtering and denoising adopts the wavelet threshold filtering algorithm, and the signal-to-noise ratio of the filtered signal is ≥40dB, ensuring the accuracy of feature parameter extraction.

[0029] As a further aspect of the present invention: in step S4, the load priority is divided into 4 levels according to the degree of importance, with hospital load weight of 0.4, industrial load of 0.3, commercial load of 0.2, and residential load of 0.1, and the recovery time of high priority load is ≤30ms.

[0030] As a further aspect of the present invention: In step S6, voltage stabilization control is achieved through a proportional-integral (PI) regulator, and the regulation coefficient is dynamically adjusted according to the voltage deviation to ensure that the voltage recovers to within the range of UN±3% and the frequency deviation is ≤±0.5Hz.

[0031] As a further aspect of the present invention, it also includes a fault type identification step: distinguishing between single-pole grounding, double-pole short circuit, and line breakage faults by the ratio of voltage drop ratio to transient energy. A ratio ≥ 0.6 indicates a double-pole short circuit fault, 0.3-0.6 indicates a single-pole grounding fault, and < 0.3 indicates a line breakage fault.

[0032] As a further aspect of the present invention: the distributed power source includes photovoltaic power stations and wind power stations, the energy storage unit adopts a lithium battery energy storage system (charge and discharge efficiency ≥95%), and the time synchronization accuracy of the transient quantity measurement unit (TMU) is ≤1μs.

[0033] As a further aspect of the present invention: after fault isolation, the power supply topology is reconstructed by adjusting the modulation strategy of the converter station to form a closed-loop power supply circuit of "non-fault converter station-distributed power source-energy storage-load", thereby improving self-healing flexibility.

[0034] Compared with the prior art, the beneficial effects of the present invention are:

[0035] 1. Accurate fault identification and anti-interference: The current-voltage collaborative judgment formula effectively integrates the characteristics of dual electrical quantities to resist interference from new energy fluctuations, converter station control, etc., with a fault identification accuracy of ≥98% and a significantly reduced false positive and false negative rate.

[0036] 2. Simplified and efficient positioning: The simplified positioning formula does not require complex models and high-precision synchronization equipment. It only requires basic lines and transient energy data. The positioning error is ≤ ±0.5km and the time is ≤3ms, which balances accuracy and engineering practicality.

[0037] 3. Industry-leading isolation speed: IGBT solid-state circuit breaker is linked to fault reliability control, with isolation time ≤5ms, which is more than 80% shorter than traditional methods, effectively curbing the spread of fault energy and protecting core equipment such as converter valves;

[0038] 4. Strong self-healing adaptability: The self-healing adaptability formula quantifies the resource and load matching relationship, prioritizes power supply to high-priority loads, with a recovery success rate of ≥99.5% and a high-priority load recovery time of ≤30ms;

[0039] 5. High engineering practicality: The formulas are simple and easy to understand, requiring no complex calculation resources. It is compatible with multi-terminal flexible DC systems with different topologies and new energy penetration rates. The hardware cost is low, the deployment is convenient, and it is easy to promote on a large scale. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the fault isolation and self-healing control process in this invention;

[0041] Figure 2 This is a block diagram of the fault identification and location logic in this invention;

[0042] Figure 3 This is a structural diagram of the self-healing scheduling adaptation model in this invention;

[0043] Figure 4 This is the timing diagram of the IGBT series solid-state circuit breaker in this invention. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] Please see Figures 1-4 In this embodiment of the invention, a fault isolation and self-healing control method for a multi-terminal flexible DC power supply system adopts an integrated technical solution of "feature collaborative identification - simplified location - rapid isolation - adaptive self-healing - closed-loop control", as detailed below:

[0046] Fault Feature Acquisition and Preprocessing

[0047] Deploy transient measurement units (TMUs) at each node of the system to collect fault transient current and voltage signals at a sampling frequency of 10kHz to ensure that no transient characteristics are missed.

[0048] The wavelet threshold filtering algorithm is used to filter and denoise the acquired signal. This algorithm can adaptively distinguish between fault signals and interference signals. The signal-to-noise ratio of the filtered signal is ≥40dB, and it effectively extracts core feature parameters such as current surge amplitude (ΔI), voltage drop ratio (ΔU / UN), and transient energy (Emeas).

[0049] Fault identification and location

[0050] A current-voltage collaborative fault identification model is constructed. The fault confidence level K is calculated using a fault determination formula. The collaborative characteristics of current surges and voltage drops are comprehensively considered to effectively resist interference from fluctuations in new energy output and adjustments to converter station control strategies. When K ≥ 70%, it is determined to be a real fault. At the same time, the ratio of voltage drop ratio to transient energy is used to distinguish between single-pole grounding, double-pole short circuit, and line breakage faults, with an identification accuracy of ≥ 98%.

[0051] Based on the transient energy decay characteristics, a simplified fault location formula is designed. It does not require complex impedance parameter calibration or multi-terminal traveling wave signal coordination. It only needs to measure the transient energy at the measuring end, the theoretical value of transient energy over the entire line length, and the total line length to calculate the distance D of the fault point. The location error is ≤ ±0.5km and the location time is ≤3ms, which greatly reduces the difficulty of engineering implementation.

[0052] Rapid fault isolation

[0053] IGBT series solid-state circuit breakers are configured on both sides of the fault section. They adopt a modular design, with a rated voltage of ±35kV, a rated current of 2kA, and an arc extinguishing time of ≤2ms, which improves the response speed by 50% compared with traditional thyristor solid-state circuit breakers.

[0054] The circuit breaker's operating logic is directly linked to the fault confidence level K. When K ≥ 70% and the location result is clear, no additional communication confirmation is required, and the trip command is directly triggered. The total isolation time is ≤ 5ms, which can quickly disconnect the fault circuit and prevent the fault energy from spreading to the non-faulty section.

[0055] Self-healing resource assessment

[0056] Real-time monitoring of the available output and fluctuation range of distributed power sources (photovoltaic and wind power), calculation of the available output ratio PDG,avail; monitoring of the remaining power ratio SSOC of energy storage units, charging and discharging response time (≤10ms) and charging and discharging efficiency (≥95%).

[0057] Based on the social function and power supply demand of the load, the load priority is quantified into weight coefficients (hospitals 0.4, industry 0.3, commerce 0.2, and residents 0.1), and a resource-load adaptation matrix is ​​established to clarify the matching relationship between each resource and the load, providing a quantitative basis for self-healing scheduling.

[0058] Self-healing scheduling optimization

[0059] With the core objective of accurately matching resource supply capacity with load demand, the M value of each power supply combination scheme is calculated using the self-healing adaptability formula. The larger the M value, the better the adaptability between resources and load. The scheme with the largest M value is selected first.

[0060] Distributed power sources and energy storage units are coordinated to provide power, prioritizing the restoration of high-priority loads. At the same time, the modulation ratio and phase of the converter station are adjusted to reconstruct the closed-loop power supply topology of "non-faulty converter station-distributed power source-energy storage-load", thereby improving self-healing flexibility and power supply reliability.

[0061] Recovery process closed-loop control

[0062] Real-time monitoring of system operating parameters such as voltage and frequency after recovery; dynamic adjustment of power output using a proportional-integral (PI) regulator; adaptive optimization of the adjustment coefficient based on voltage deviation (the larger the deviation, the larger the adjustment coefficient); ensuring that the voltage recovers to within ±3% of the rated voltage and the frequency deviation is ≤ ±0.5Hz.

[0063] The system continuously monitors the status of the faulty section. If the fault is transient (such as lightning flashover), the system is restored to its original operating state through closed-loop control after the fault is eliminated. If the fault is permanent, the system maintains isolation and issues an alarm signal to prompt maintenance personnel to handle the situation.

[0064] Example: Application of a multi-terminal flexible DC distribution network in a certain city

[0065] System parameters

[0066] System scale: ±35kV multi-terminal flexible DC distribution network, including 4 converter stations (VSC1-VSC4), 30km of lines (15km of faulty lines VSC2-VSC3), 2 photovoltaic power stations (total capacity 10MW), 2 wind power stations (total capacity 8MW), and 2 lithium battery energy storage units (total capacity 6MW / 12MWh).

[0067] Load configuration: Hospital load ( ), industrial load ( ), commercial load ( ), Residential load ( );

[0068] Equipment parameters: TMU sampling frequency 10kHz, time synchronization accuracy 0.5μs; IGBT solid-state circuit breaker rated voltage ±35kV, rated current 2kA; system rated current IN=3kA, rated voltage UN=35kV;

[0069] Initial state: 8MW photovoltaic output (PDG,avail=0.8), 6MW wind power output (PDG,avail=0.75), energy storage SSOC=0.8.

[0070] Implementation process

[0071] S1. Fault Occurrence: A single-pole ground fault occurred in the VSC2-VSC3 section (15km) of the line, which was detected by the TMU. , Transient energy at the measurement end Theoretical value of transient energy over the entire length of the line ;

[0072] S2. Fault Identification and Location: Calculated using fault determination formulas. The fault is determined to be genuine; the ratio of voltage drop ratio to transient energy is 0.4, indicating a single-pole grounding fault; substituting this into the location formula... The location of the fault point was 9km from the VSC2 side, with an actual error of 0.2km;

[0073] S3. Fault Isolation: Triggers the operation of solid-state circuit breakers on both sides of the line, with an arc extinguishing time of 1.6ms and a total isolation time of 3.8ms, successfully isolating the faulty section;

[0074] S4. Self-healing resource assessment: Power outage load is 2MW for industrial load, 1MW for commercial load, and 1MW for residential load; available photovoltaic output. Energy storage SSOC=0.8;

[0075] S5. Self-healing scheduling: Calculate the suitability M of candidate schemes: Scheme 1 (photovoltaic + energy storage power supply for industrial loads) Option 2 (Energy storage for standalone power supply to industrial loads) Option 3 (Photovoltaic + Energy Storage Power Supply for Commercial Loads) Option 2 is selected, and the energy storage unit is dispatched to restore power supply to the industrial load at an output of 2MW;

[0076] S6. Closed-loop control: The energy storage output is dynamically adjusted through a PI regulator, and the voltage recovers from 33.5kV to 34.8kV with a deviation of 2.3%, which meets the requirements.

[0077] Test data and implementation results

[0078]

[0079] When applied to this multi-terminal flexible DC distribution network, this invention significantly improves the speed, accuracy, and reliability of fault handling through innovative designs such as current-voltage collaborative identification, simplified transient energy location, and precise resource-load matching. The fault identification accuracy, location precision, and isolation speed are all significantly superior to traditional methods. High-priority loads are quickly restored, power supply quality remains stable and meets standards, and system resilience is significantly enhanced, fully satisfying the high-reliability power supply requirements of urban distribution networks.

[0080] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A fault isolation and self-healing control method for a multi-terminal flexible DC power supply system, characterized in that, Includes the following steps: S1. Fault Feature Acquisition and Preprocessing: The transient current and voltage signals of the fault are acquired through the transient quantity measurement unit (TMU) of each node of the system. The sampling frequency is ≥10kHz. After filtering and noise reduction, the core characteristic parameters such as the amplitude of the current surge, the voltage drop ratio, and the transient energy are extracted. S2. Fault Identification and Location: Based on the preprocessed feature parameters, the reliability of the fault is quantified by the fault judgment formula, and the fault section is calculated by combining the simplified fault location formula to achieve accurate location; S3. Rapid fault isolation: Control the operation of IGBT series solid-state circuit breakers on both sides of the fault section to disconnect the fault circuit. The isolation time is ≤5ms to prevent the fault from spreading. S4. Self-healing resource assessment: Real-time monitoring of distributed power output, remaining energy storage unit (SOC) and load priority within the system to establish a resource-load adaptation matrix; S5. Self-healing scheduling optimization: Based on the adaptation matrix, the optimal power supply recovery scheme is selected through the self-healing adaptation formula, and distributed power sources and energy storage units are scheduled to supply power in coordination, prioritizing the recovery of high-priority loads. S6. Closed-loop control during recovery process: Real-time monitoring of system voltage and power parameters after recovery, dynamic adjustment of power output, and ensuring stable power supply quality.

2. The fault isolation and self-healing control method for a multi-terminal flexible DC power supply system according to claim 1, characterized in that, In step S2, the fault determination formula quantifies the fault confidence level through current-voltage collaborative characteristics, specifically as follows: in, Fault confidence level (%) The magnitude of the fault current surge (kA). This is the system's rated current (kA). This represents the voltage drop amplitude (kV). This is the system's rated voltage (kV). It is identified as a genuine fault in a timely manner, with an accuracy rate of ≥98%.

3. The fault isolation and self-healing control method for a multi-terminal flexible DC power supply system according to claim 1, characterized in that, In step S2, the simplified fault location formula is constructed based on the transient energy decay characteristics, specifically as follows: in, The distance (km) from the fault point to the measuring end. Transient energy (kJ) collected at the measurement end. This represents the theoretical value of the transient energy over the entire length of the line (kJ). The total length of the faulty line (km) is given, and the location error is ≤ ±0.5km.

4. The fault isolation and self-healing control method for a multi-terminal flexible DC power supply system according to claim 1, characterized in that, In step S5, the self-healing adaptability formula focuses on matching resource supply capacity with load demand, specifically as follows: in, For self-healing adaptation, The percentage of remaining electricity in the energy storage unit (0≤ ≤1), The proportion of available output of distributed power sources (0≤ ≤1), Load priority weight (0.1-0.4), prioritize the power supply combination scheme with the largest M.

5. The fault isolation and self-healing control method for a multi-terminal flexible DC power supply system according to claim 1, characterized in that, In step S4, the load priority is divided into 4 levels according to the importance: hospital load weight 0.4, industrial load weight 0.3, commercial load weight 0.2, and residential load weight 0.

1. The recovery time of high priority load is ≤30ms.

6. The fault isolation and self-healing control method for a multi-terminal flexible DC power supply system according to claim 1, characterized in that, In step S6, voltage stabilization control is achieved through a proportional-integral (PI) regulator. The regulation coefficient is dynamically adjusted according to the voltage deviation to ensure that the voltage recovers to within the range of UN±3% and the frequency deviation is ≤±0.5Hz.

7. The fault isolation and self-healing control method for a multi-terminal flexible DC power supply system according to claim 1, characterized in that, It also includes a fault type identification step: the ratio of voltage drop ratio to transient energy is used to distinguish between single-pole grounding, double-pole short circuit, and line breakage faults. A ratio ≥0.6 indicates a double-pole short circuit fault, 0.3-0.6 indicates a single-pole grounding fault, and <0.3 indicates a line breakage fault.

8. The fault isolation and self-healing control method for a multi-terminal flexible DC power supply system according to claim 1, characterized in that, Distributed power sources include photovoltaic power plants and wind power plants. The energy storage unit adopts a lithium battery energy storage system (charge and discharge efficiency ≥95%), and the time synchronization accuracy of the transient quantity measurement unit (TMU) is ≤1μs.

9. The fault isolation and self-healing control method for a multi-terminal flexible DC power supply system according to claim 1, characterized in that, After fault isolation, the power supply topology is reconstructed by adjusting the modulation strategy of the converter station, forming a closed-loop power supply circuit of "non-fault converter station - distributed power source - energy storage - load".

Images