Distribution automation master station self-healing system suitable for complex scene

Abstract

The invention belongs to the technical field of power distribution, and particularly relates to a power distribution automation master station self-healing system suitable for a complex scene, which comprises a sensing layer, a network layer and an application layer, the sensing layer establishes a communication connection relation with the application layer based on the network layer, the sensing layer comprises a distributed intelligent terminal, a PMU and an intelligent electric meter, and the distributed intelligent terminal, the PMU and the intelligent electric meter collect states of main and auxiliary equipment in real time; the network layer comprises a wireless private network and wired communication and is used for data interaction between the sensing layer and the application layer; the application layer comprises a fault diagnosis module, a recovery decision module, a simulation verification module and an edge computing node. According to the scheme, a self-healing system is constructed through'multi-agent collaborative architecture + hybrid intelligent algorithm + edge computing ', and the core innovation points and effects are as follows: accurate fault diagnosis, shortening of fault recovery time, robustness enhancement and global optimization operation; and the technology is a multi-objective optimization model considering distributed power supply prediction, load characteristics and network constraints.

Description

Technical Field

[0001] This invention relates to the field of power distribution technology, specifically to a self-healing system for power distribution automation master stations suitable for complex scenarios. Background Technology

[0002] With the large-scale integration of distributed power sources (such as photovoltaic and wind power), energy storage systems, and flexible loads, modern power distribution networks exhibit complex characteristics of high penetration, strong uncertainty, and multi-timescale coupling. Traditional distribution automation systems, relying on manual intervention for fault handling (fault location → isolation → recovery taking minutes to hours), can no longer meet the demands for high-reliability power supply. Specifically, this manifests in:

[0003] Fault location is difficult: the reverse power supply from the distributed power source causes the fault current to change direction in many ways, and traditional overcurrent protection fails.

[0004] The recovery strategy is limited: it only supports predefined "daily daily" transfer schemes and cannot adapt to dynamic changes in network topology;

[0005] Lack of global optimization: The fault recovery process does not take into account the absorption of distributed energy resources, load priority, and network losses.

[0006] Strong communication dependency: Centralized architecture is prone to system paralysis when communication is interrupted, and its self-healing ability is weak.

[0007] Inaccurate fault diagnosis: Existing fault location algorithms have a high false alarm rate (up to 20% or more) when multiple distributed power sources are connected, and it is difficult to distinguish between transient and permanent faults.

[0008] Low efficiency in recovery decision-making: Traditional optimization algorithms (such as genetic algorithms and particle swarm optimization algorithms) take a long time (minutes) to compute when dealing with distribution networks containing a large number of distributed power sources, which cannot meet the requirements for rapid self-healing.

[0009] Insufficient system robustness: When communication is interrupted or local equipment malfunctions, the self-healing function may fail, leading to an expansion of the power outage area.

[0010] Coordination across multiple time scales is challenging: the time scales of distributed power generation output fluctuations (from seconds to minutes) and distribution network fault recovery (from milliseconds to seconds) are mismatched, making it difficult for traditional control methods to achieve coordinated optimization. Summary of the Invention

[0011] The purpose of this invention is to provide a self-healing system for power distribution automation master stations suitable for complex scenarios, so as to solve the problems mentioned in the background art.

[0012] To achieve the above objectives, the present invention provides the following technical solution: a self-healing system for distribution automation master stations suitable for complex scenarios, comprising:

[0013] The perception layer, network layer, and application layer;

[0014] The perception layer establishes a communication connection between the network layer and the application layer;

[0015] The perception layer includes distributed intelligent terminals, PMUs, and smart meters, which collect the status of main and distribution equipment in real time.

[0016] The network layer includes wireless private networks and wired communication, and is used for data interaction between the perception layer and the application layer.

[0017] The application layer includes fault diagnosis, recovery decision-making, simulation verification modules, and edge computing nodes.

[0018] Preferably, the wireless private network includes a 5G communication module and an industrial Internet of Things (IoT).

[0019] Preferably, the wired communication includes optical fiber and EPON.

[0020] Preferably, the distributed intelligent terminal, PMU, and smart meter collect multi-source data and fuse the multi-source data, specifically based on weighted Kalman filtering and anomaly detection based on isolated forest to clean the multi-source data.

[0021] Preferably, the application layer performs fusion positioning based on multi-source data collected from distributed smart terminals, PMUs, and smart meters, using a fusion positioning GNN-GRU hybrid network;

[0022] Inputs: node voltage phasors, branch current rate of change, and switch state sequence;

[0023] Output: Failure probability matrix P(f n |x1,x2,…,x m );

[0024] Fault type identification: Support Vector Machine.

[0025] Preferably, the application layer further includes a self-healing decision and optimization module;

[0026] Based on multi-source data collected from distributed smart terminals, PMUs, and smart meters, the status of main and distribution equipment is obtained, and the optimal coordination strategy between the main and distribution equipment is derived by analyzing the status of the main and distribution equipment.

[0027] Preferably, the self-healing decision and optimization module pre-stores a mapping table of the status of the main and auxiliary equipment and the coordination strategy of the main and auxiliary equipment. The obtained status of the main and auxiliary equipment is compared with the mapping table to obtain the optimal coordination strategy of the main and auxiliary equipment.

[0028] Preferably, when the application layer outputs the optimal coordination strategy instruction between the master and slave devices to the execution end:

[0029] The instructions are based on the GOOSE message of the IEC 61850 standard.

[0030] Simulation verification: The digital twin platform uses time-domain simulation to verify the feasibility of the optimal coordination strategy between the main and auxiliary equipment.

[0031] Compared with the prior art, the beneficial effects of the present invention are:

[0032] This solution constructs a self-healing system through a "multi-agent collaborative architecture + hybrid intelligent algorithm + edge computing," with the following core innovations and effects:

[0033] Precise Fault Diagnosis: Technology: A multi-feature fusion fault location algorithm that integrates wide-area measurement data (SCADA / PMU) with edge computing. Results: Improved fault location accuracy, reduced location time to less than 100ms, and the ability to distinguish between transient and permanent faults.

[0034] Rapid self-healing decision-making: Technology: A hybrid intelligent optimization algorithm based on graph neural networks (GNN) and model predictive control (MPC). Effect: Reduced fault recovery time.

[0035] Enhanced Robustness: Technology: Distributed multi-agent architecture + edge-center collaboration mechanism, supporting local autonomy during communication interruptions. Effect: Critical workloads can still achieve large-scale self-healing recovery even during communication interruptions, improving system availability.

[0036] Global Optimization Operation: Technology: A multi-objective optimization model considering distributed power generation prediction, load characteristics, and network constraints. Results: Reduced network losses and increased distributed energy absorption rate after fault recovery. Attached Figure Description

[0037] Figure 1 This is a system logic block diagram of the present invention. Detailed Implementation

[0038] 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.

[0039] In the description of this invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0040] Example 1:

[0041] Please see Figure 1 This invention provides a technical solution: a self-healing system for distribution automation master stations suitable for complex scenarios, comprising:

[0042] The perception layer, network layer, and application layer;

[0043] The perception layer establishes a communication connection between the network layer and the application layer;

[0044] The perception layer includes distributed intelligent terminals, PMUs, and smart meters, which collect the status of main and distribution equipment in real time.

[0045] Distributed intelligent terminals are core equipment for power distribution network automation. They are used to monitor and control primary power distribution equipment (such as circuit breakers, load switches, and distribution transformers), and support fault detection, isolation, and self-healing.

[0046] Status collection object

[0047] Switchgear: Circuit breaker / load switch opening and closing status, position signals, energy storage status, mechanical life cycles, etc.

[0048] Electrical parameters: Real-time data such as three-phase voltage, current, active / reactive power, power factor, and frequency.

[0049] Environmental parameters: temperature, humidity, SF6 gas pressure (for gas-insulated equipment), etc.

[0050] Acquisition Principle

[0051] Sensor interface:

[0052] Current and voltage signals are acquired by electromagnetic / electronic instrument transformers (such as CT / PT), and then converted into digital quantities (ADC sampling) by signal conditioning modules (amplification and filtering).

[0053] Switch position signals are acquired through microswitches / stroke sensors, and environmental parameters are monitored through pressure transmitters / temperature sensors.

[0054] Fault detection logic:

[0055] Based on threshold judgments such as overcurrent, zero-sequence current, and voltage surge, and combined with timing logic (such as fault duration), fault types such as short circuit and grounding are identified, and alarms or automatic control commands are triggered.

[0056] The principle of PMU (Phasor Measurement Unit) for acquiring the status of power distribution equipment

[0057] The PMU is based on synchronous phasor technology (GPS / BeiDou timing) and is used for high-precision measurement of voltage, current phasors and frequency in power systems, supporting dynamic process analysis (such as transient stability and harmonic monitoring).

[0058] Status collection object

[0059] High-precision electrical parameters:

[0060] Voltage / current phasors (amplitude, phase, with an accuracy of within 0.1%), frequency, and rate of change of frequency (df / dt).

[0061] Transient characteristics such as harmonic components (amplitude and phase of 2nd-50th harmonics), interharmonics, and voltage sags / sags.

[0062] Synchronization timestamp: Data with nanosecond-level synchronization clock (based on IEEE C37.118 standard).

[0063] Acquisition Principle

[0064] Synchronous sampling:

[0065] By locking the sampling time using a synchronization clock module (such as a GPS receiver), it is ensured that all PMU data in the network are collected under the same time reference (synchronization error < 1 μs).

[0066] Fourier transform (DFT) or dynamic phasor algorithm is used to perform spectral analysis on the sampled signal to extract the fundamental phasor and harmonic components.

[0067] Smart meters are terminals that allow users to interact with the power distribution network. They are used to measure electrical energy and collect status data from low-voltage power distribution equipment (such as user incoming line switches and branch lines).

[0068] Status collection object

[0069] Metering data: active / reactive energy, demand, time-of-use pricing data, etc.

[0070] Equipment status: switch open / closed status, residual current (leakage current), overvoltage / undervoltage, phase loss alarm, etc.

[0071] User-side parameters: voltage sag / boost records, harmonic distortion rate (THD), load curves, etc.

[0072] Acquisition Principle

[0073] Sampling and Measurement:

[0074] Voltage and current signals are acquired by using dedicated metering chips (such as ADE7758 and ATT7022), and power and energy are calculated using digital multipliers with an accuracy of 0.5S or higher.

[0075] Built-in non-volatile memory (EEPROM) stores historical data (such as daily / monthly frozen data and event logs).

[0076] Status monitoring logic:

[0077] Abnormal states are detected by threshold comparison (e.g., an alarm is triggered when the voltage is >1.2Un or <0.8Un).

[0078] Residual current monitoring is based on a zero-sequence current transformer (RCT). When the detected leakage current exceeds the set value (e.g., 30mA), the control switch trips.

[0079] The network layer includes wireless private networks (including 5G communication modules, industrial IoT, etc.) and wired communication (including optical fiber, EPON, etc.). The network layer is used for data interaction between the perception layer and the application layer.

[0080] The application layer includes fault diagnosis, recovery decision-making, simulation verification modules, and edge computing nodes.

[0081] Self-healing logic relationship

[0082] Fault detection: Based on distributed intelligent terminals (DTU / FTU), PMU, and smart meters, the system monitors the change rate of node voltage and branch current in real time using wide-area measurement data, and triggers a fault alarm when the threshold is exceeded.

[0083] Fault location: The GNN-GRU model combines switch state sequences and electrical characteristics to calculate the fault probability of each node and locate the fault area;

[0084] Type identification: SVM determines the fault type (transient / permanent). If it is a transient fault, it attempts to reclose.

[0085] Recovery Decision: The MISOCP model generates a network reconstruction scheme, and MPC coordinates the output of distributed power sources to meet multiple objectives of optimization;

[0086] Execution verification: The digital twin platform verifies the feasibility of the solution by issuing instructions through GOOSE messages;

[0087] Edge autonomy: When communication is interrupted, edge nodes execute preset rules based on local measurements to maintain power supply to critical loads.

[0088] The distributed intelligent terminal, PMU, and smart meter collect multi-source data and fuse the multi-source data. Specifically, the multi-source data is cleaned based on weighted Kalman filtering and anomaly detection based on isolated forest.

[0089] Multi-source data fusion model: Weighted Kalman filter

[0090] formula:

[0091] in,

[0092] K k For Kalman gain, P k To estimate the error covariance, Z k Here, H represents the measured value, and H is the measurement matrix.

[0093] Weighted Kalman filtering improves upon standard Kalman filtering through the following steps:

[0094] Weighting

[0095] Objective: To assign different weights to observed data or state prediction errors to reduce the impact of outliers.

[0096] method:

[0097] Weights are calculated based on the residuals (the difference between observed and predicted values), for example:

[0098] If the residual exceeds the threshold, the data is considered an outlier and the weight is set to be close to 0.

[0099] The weight of normal data is set to 1 or dynamically adjusted according to the Gaussian kernel function.

[0100] Common weighting functions include the Huber function, Cauchy function, and Geman-McClure function, which are used to balance fitting accuracy and anti-interference ability.

[0101] The application layer performs fusion positioning based on multi-source data collected from distributed smart terminals, PMUs, and smart meters, using a fusion positioning GNN-GRU hybrid network.

[0102] Inputs: node voltage phasors, branch current rate of change, and switch state sequence;

[0103] Output: Failure probability matrix P(f n |x1,x2,…,x m );

[0104] Fault type identification: Support Vector Machine (SVM).

[0105] In power systems, node voltage phasors, branch current rates of change, and switch state sequences are key characteristics reflecting the system's operating state. Among them:

[0106] Node voltage phasors: contain amplitude and phase information, and directly reflect system stability;

[0107] Branch current change rate: characterizes the sudden change in current during fault transient process;

[0108] Switch state sequence: records the dynamic changes in the power grid topology (such as protection actions after a fault).

[0109] Objective: To design a hybrid network that integrates graph neural networks (GNNs) and gated recurrent units (GRUs) to model the power grid topology (graph structure) and time-series dynamic characteristics (sequence data), and output a fault probability matrix P(fn|x1,x2,…,xm), which is the fault probability of each node / branch fn given a historical observation sequence x1,…,xm.

[0110] The application layer also includes a self-healing decision-making and optimization module;

[0111] Based on multi-source data collected from distributed smart terminals, PMUs, and smart meters, the status of main and distribution equipment is determined, and the optimal coordination strategy between the main and distribution equipment is derived by analyzing the status of the main and distribution equipment.

[0112] The self-healing decision and optimization module pre-stores a mapping table between the status of the main and auxiliary equipment and the coordination strategy of the main and auxiliary equipment. It compares the obtained status of the main and auxiliary equipment with the mapping table to obtain the optimal coordination strategy of the main and auxiliary equipment.

[0113] Establish a mapping table between the status of main and auxiliary equipment and their coordination strategies.

[0114] Define the equipment and status type

[0115] Main equipment includes transformers, circuit breakers, busbars, etc., which play a crucial role in control and transmission within the power distribution system. The status of main equipment includes normal operation, overload, fault tripping, and maintenance. Taking a transformer as an example, during normal operation, all parameters are stable; under overload, this manifests as increased temperature and excessive load rate; fault tripping may be caused by short circuits, grounding faults, etc.; and maintenance status indicates that the equipment is undergoing upkeep.

[0116] Distribution equipment, such as fuses, disconnect switches, and power distribution lines, is responsible for distributing electrical energy to various points of use. The status of distribution equipment includes normal power supply, line breakage, fuse blown, and poor contact. For example, in a power distribution line, the current and voltage are stable during normal power supply. A line breakage will cause a power outage, while a blown fuse indicates a possible overload or short circuit fault.

[0117] Analyze cooperation relationships and strategies

[0118] Normal operation coordination: Clarify how the main equipment and distribution equipment work together under normal conditions. For example, when a transformer is operating normally, its output voltage and capacity must meet the power demand of the downstream distribution lines and equipment. Circuit breakers, disconnect switches, etc., should be kept closed or open according to the established wiring methods and operating rules to ensure the orderly transmission of electrical energy.

[0119] Anomaly and Fault Coordination: Corresponding coordination strategies should be developed for various anomalies and faults occurring in the main and distribution equipment. When the main equipment circuit breaker trips due to a fault, the fuses of the downstream distribution equipment should selectively blow according to the magnitude of the fault current and the protection time limit to isolate the fault area and prevent the fault range from expanding. If the distribution equipment line is disconnected, the protection device of the main equipment should be able to detect abnormal changes in current and voltage in a timely manner and take corresponding control measures, such as issuing an alarm signal or initiating a backup line switchover.

[0120] Define mapping rules and table structure

[0121] Rule Formulation: Based on the logical relationship between equipment status and coordination strategies, formulate clear mapping rules. For example, when the main equipment transformer is overloaded and a certain line of the distribution equipment has a high load, the mapping coordination strategy is to adjust the transformer tap changer and transfer some of the load of the distribution equipment line to other lines.

[0122] Table Design: Mapping tables typically include columns for master device status, slave device status, and coordination strategy. Presenting them in tabular form facilitates intuitive viewing and management.

[0123] Data collection and validation

[0124] Data collection: Obtain actual operating data and historical fault data of main and distribution equipment through power distribution automation systems and equipment monitoring devices to provide a basis for the development of mapping tables. This data includes information such as equipment voltage, current, temperature, and switch status.

[0125] Verification and Optimization: Based on actual operational experience and simulation tests, the mapping table is verified and optimized. During the actual operation of the power distribution system, the effects of implementing the coordination strategy according to the mapping table are observed, such as whether the fault isolation time is shortened and whether the equipment operation is more stable. The mapping table is then adjusted and improved based on the feedback results.

[0126] Use the mapping table of main and auxiliary equipment status and coordination strategy

[0127] Real-time monitoring and status acquisition: Utilizing distribution automation systems, smart meters, sensors, and other equipment, the operating status of main and distribution equipment is monitored in real time. Various parameters and status information of the equipment are acquired and matched with the status types in a mapping table. For example, monitoring parameters such as transformer temperature and load rate determines whether it is in an overload state, and detecting the current and voltage of distribution lines determines whether the line is supplying power normally.

[0128] Find the mapping strategy: Once the current status of the main and distribution equipment is determined, quickly find the corresponding coordination strategy in the mapping table. If the main circuit breaker trips due to a fault and the downstream fuse does not blow, the corresponding handling strategy can be obtained from the mapping table. That is, check whether the fuse is damaged. If it is normal, troubleshoot and repair the main circuit breaker. If the fuse is damaged, replace the fuse and try to restore power.

[0129] Execution and Feedback: Following the coordination strategy provided in the mapping table, operators or automated control systems perform corresponding operations on the main and auxiliary equipment. During execution, changes in equipment status are continuously monitored, and the results are fed back to relevant personnel or the system. If the equipment returns to normal operation after the strategy is executed, the process is recorded; if the problem is not resolved, the equipment status is re-examined, the cause is further analyzed, and it is considered whether the mapping table needs to be updated or supplemented.

[0130] Training and Maintenance: Training on the use of mapping tables will be provided to power distribution system operation and maintenance personnel, ensuring they are familiar with the coordination strategies and operating procedures corresponding to various equipment states. This will guarantee their accurate and rapid application of the mapping tables in practical work. Furthermore, as power distribution equipment is upgraded, system structure changes, and operational experience accumulates, the mapping tables will be regularly maintained and updated to ensure their effectiveness and applicability.

[0131] When the application layer outputs the optimal coordination strategy instruction between the master and slave devices to the execution end:

[0132] The instructions are based on the GOOSE message of the IEC 61850 standard.

[0133] Simulation verification: The digital twin platform uses time-domain simulation to verify the feasibility of the optimal coordination strategy between the main and auxiliary equipment.

[0134] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or basic characteristics. Therefore, the embodiments should be considered exemplary and non-limiting in all respects. The scope of the invention is defined by the appended claims rather than the foregoing description. Therefore, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention, and no reference numerals in the claims should be construed as limiting the scope of the claims.

[0135] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A self-healing system for power distribution automation master stations suitable for complex scenarios, characterized in that, include: The perception layer, network layer, and application layer; The perception layer establishes a communication connection between the network layer and the application layer; The perception layer includes distributed intelligent terminals, PMUs, and smart meters, which collect the status of main and distribution equipment in real time. The network layer includes wireless private networks and wired communication, and is used for data interaction between the perception layer and the application layer. The application layer includes fault diagnosis, recovery decision-making, simulation verification modules, and edge computing nodes.

2. The self-healing system for distribution automation master stations suitable for complex scenarios according to claim 1, characterized in that: The private wireless network includes 5G communication modules and industrial IoT.

3. The self-healing system for distribution automation master stations suitable for complex scenarios according to claim 1, characterized in that: The wired communication includes optical fiber and EPON.

4. The self-healing system for distribution automation master stations suitable for complex scenarios according to claim 1, characterized in that: The distributed intelligent terminal, PMU, and smart meter collect multi-source data and fuse the multi-source data. Specifically, the multi-source data is cleaned based on weighted Kalman filtering and anomaly detection based on isolated forest.

5. A self-healing system for distribution automation master stations suitable for complex scenarios according to claim 1, characterized in that: The application layer performs fusion positioning based on multi-source data collected from distributed smart terminals, PMUs, and smart meters, using a fusion positioning GNN-GRU hybrid network. Inputs: node voltage phasors, branch current rate of change, and switch state sequence; Output: Failure probability matrix P(f n |x1,x2,…,x m ); Fault type identification: Support Vector Machine.

6. A self-healing system for distribution automation master stations suitable for complex scenarios according to claim 1, characterized in that: The application layer also includes a self-healing decision-making and optimization module; Based on multi-source data collected from distributed smart terminals, PMUs, and smart meters, the status of main and distribution equipment is determined, and the optimal coordination strategy between the main and distribution equipment is derived by analyzing the status of the main and distribution equipment.

7. A self-healing system for distribution automation master stations suitable for complex scenarios according to claim 6, characterized in that: The self-healing decision and optimization module pre-stores a mapping table between the status of the main and auxiliary equipment and the coordination strategy of the main and auxiliary equipment. It compares the obtained status of the main and auxiliary equipment with the mapping table to obtain the optimal coordination strategy of the main and auxiliary equipment.

8. A self-healing system for distribution automation master stations suitable for complex scenarios according to claim 7, characterized in that: When the application layer outputs the optimal coordination strategy instruction between the master and slave devices to the execution end: The instructions are based on the GOOSE message of the IEC 61850 standard. Simulation verification: The digital twin platform uses time-domain simulation to verify the feasibility of the optimal coordination strategy between the main and auxiliary equipment.

Images