Networked micro-grid hierarchical protection strategy optimization method and system in non-communication environment

Abstract

The application belongs to the technical field of power system protection and control, and particularly relates to a networked micro-grid hierarchical protection strategy optimization method and system in a non-communication environment. The method first performs local fault isolation based on electrical information of protection units arranged at each micro-grid and a public grid connection point, then performs regional electrical information inference on the opposite side boundary protection unit based on electrical information of the local boundary protection unit, and performs regional fault isolation based on the regional electrical information inference result. Finally, the method performs global electrical information inference on the networked micro-grid based on electrical information of the protection unit at the public grid connection point, inputs the global electrical information inference result into a strategy optimization model, and updates the local fault isolation strategy and the regional fault isolation strategy by solving the model. The application can realize reliable fault isolation of the networked micro-grid in the non-communication environment by constructing a hierarchical protection logic based on local electrical information sensing.

Description

Technical Field

[0001] This invention belongs to the field of power system protection and control technology, specifically relating to an optimization method and system for hierarchical protection strategy of networked microgrids in a non-communication environment. Background Technology

[0002] A microgrid (MG) is an autonomous small-scale power system composed of distributed energy resources (DER), energy storage devices, loads, and local control systems. It can operate in grid-connected or islanded modes. Networked microgrids (NMGs) refer to a cluster formed by multiple microgrids interconnected via interconnecting lines (such as AC / DC tie lines), offering advantages such as energy sharing, improved power supply reliability, and resilience. Networked microgrids are electrically connected to the main grid and exchange power through a point of common coupling (PCC). To ensure the safe and stable operation of networked microgrids, effective protection systems are required to quickly isolate faulty areas.

[0003] Traditional distribution network protection relies on communication-assisted centralized or adaptive strategies, such as regional coordinated tripping using GOOSE (Generic Object Oriented Substation Event) messages based on the IEC 61850 protocol. However, such solutions heavily depend on highly reliable, low-latency communication infrastructure, which is difficult to deploy in networked microgrids with no or weak communication. Existing microgrid protection solutions for environments with no or weak communication mainly include the following approaches:

[0004] 1. Overcurrent protection based on local measurement. Inverse-definite-minimum-time (IDMT) overcurrent relays are used to protect the feeder, but they are difficult to handle problems such as bidirectional power flow and large fluctuations in short-circuit current levels;

[0005] 2. Directional overcurrent protection. It uses power direction as a criterion to distinguish between internal and external faults, but its sensitivity is insufficient in islanded mode due to the low fault current amplitude.

[0006] 3. Voltage / frequency over-limit protection. This method serves as a backup protection measure, but it has a slow response and cannot accurately locate the fault.

[0007] The aforementioned methods are mostly designed for single microgrids and do not consider the challenges brought about by the interconnection of multiple microgrids in networked microgrids, such as topological dynamics, complexity of fault current paths, and frequent switching of operating modes. Especially under conditions without communication, there is a lack of a hierarchical protection architecture that can balance selectivity, speed, and adaptability. Summary of the Invention

[0008] The purpose of this invention is to address the aforementioned problems in the prior art by providing a method and system for optimizing the hierarchical protection strategy of a networked microgrid in a communication-free environment, enabling reliable fault isolation of the networked microgrid system without any communication infrastructure.

[0009] To achieve the above objectives, the technical solution of the present invention is as follows:

[0010] In a first aspect, this invention proposes an optimization method for hierarchical protection strategy of networked microgrids in a non-communication environment, the hierarchical protection strategy optimization method comprising:

[0011] S1. In a networked microgrid, protection units are installed inside each microgrid, at both ends of the interconnection lines between microgrids, and at the common grid connection point. Based on the electrical information of the protection units installed inside each microgrid and at the common grid connection point, local faults in the local microgrid are identified and local fault isolation is performed based on the current local fault isolation strategy.

[0012] S2. Let the protection unit set on the side of the interconnection line closer to the local microgrid be the local boundary protection unit, and the protection unit set on the side closer to the opposite microgrid be the opposite boundary protection unit. Based on the electrical information of the local boundary protection unit, perform regional electrical information inference on the opposite boundary protection unit. Based on the inferred regional electrical information inference result and the current regional fault isolation strategy, perform regional fault isolation.

[0013] S3. Based on the electrical information of the protection unit at the common grid connection point, perform global electrical information inference on the networked microgrid. Input the inferred global electrical information inference results into the strategy optimization model constructed with the optimization objective of maximizing power supply reliability. Solve the strategy optimization model to update the local fault isolation strategy and the regional fault isolation strategy.

[0014] S1 includes:

[0015] S11. Protection units are installed inside each microgrid, at both ends of the interconnection lines between microgrids, and at the common grid connection point in the networked microgrid. Based on the electrical information collected by the protection units installed inside each microgrid and at the common grid connection point, the current operating mode of the networked microgrid is identified.

[0016] S12. Based on the combination of primary and secondary criteria corresponding to the current operating mode, calculate the comprehensive fault confidence score of each local protection unit within the local microgrid, compare whether the comprehensive fault confidence score is greater than the preset fault threshold, and if so, determine that there is a local fault in the local protection unit and isolate the local fault by initiating the tripping action of the local protection unit.

[0017] S11 includes:

[0018] If the frequency deviation and voltage deviation collected by the protection unit deployed at the photovoltaic grid connection point inside the microgrid are greater than the corresponding preset threshold, the networked microgrid is determined to be in an islanded state; if the voltage collected by the protection unit deployed at the common grid connection point drops significantly, the networked microgrid is determined to be in a grid-connected state; otherwise, the networked microgrid is determined to be in a transitional state.

[0019] S12 includes:

[0020] When the networked microgrid is in an islanded state, the comprehensive fault confidence score is calculated using a combination of primary and secondary criteria. The primary criterion in the first combination of primary and secondary criteria is the rate of change of current, and the secondary criterion is the voltage drop depth. If the comprehensive fault confidence score calculated using the first combination of primary and secondary criteria is greater than the preset fault threshold, it is determined that there is a local fault.

[0021] When the networked microgrid is in grid-connected state, the comprehensive fault confidence score is calculated using the second main and auxiliary criterion combination. The main criterion in the second main and auxiliary criterion combination is the measured impedance, and the auxiliary criterion is the change in fundamental current. If the comprehensive fault confidence score calculated using the second main and auxiliary criterion combination is greater than the preset fault threshold, it is determined that there is a local fault.

[0022] When the networked microgrid is in a transitional state, the comprehensive fault confidence score is calculated by using both the first primary and secondary criteria combination and the second primary and secondary criteria combination. If the comprehensive fault confidence score calculated by using any primary and secondary criteria combination is greater than the preset fault threshold, it is determined that there is a local fault.

[0023] The combined fault confidence score is obtained by weighted fusion of the main and auxiliary criteria using the following formula:

[0024] ;

[0025] In the above formula, The overall fault confidence score; , These refer to the primary criterion and the secondary criterion in the primary and secondary criterion combination; , These are the weights corresponding to the primary criterion and the secondary criterion, respectively.

[0026] S2 includes:

[0027] Consider a local microgrid in a networked microgrid. Connected to the opposite microgrid via interconnection lines Connection, based on local microgrids Local boundary protection unit on one side The measured electrical information is for the microgrid located on the opposite side. One side of the opposite boundary protection unit By performing regional electrical information inference, the protection units of the opposite boundary are obtained. The results of regional electrical information inference;

[0028] According to the opposite boundary protection unit Based on the regional electrical information inference results, and the following regional fault isolation strategy, the microgrid on the opposite side is analyzed. Perform area fault isolation:

[0029] Regional fault isolation strategy one:

[0030] If the local boundary protection unit The measured impedance decrease, the side boundary protection unit If the voltage amplitude and phase angle are normal, and the power flow direction is normal, then an external through-fault is determined to have occurred, and a blocking delay is initiated. The blocking delay refers to prohibiting the local boundary protection unit from operating for a preset duration. and the opposite boundary protection unit If the circuit breaker trips and continues to monitor, and the external through-fault does not disappear after a preset monitoring period, the interlock will be lifted.

[0031] Regional fault isolation strategy two:

[0032] If the local boundary protection unit With the opposite boundary protection unit If both voltage drops and power flows in the opposite direction, then the local microgrid is identified. In the event of an internal fault, the local microgrid should be contacted. The fault electrical distance sequencing triggers the tripping action of the boundary protection unit;

[0033] Regional fault isolation strategy three:

[0034] First determine if it is a local boundary protection unit. The measured impedance decrease, the side boundary protection unit If the voltage amplitude and phase angle are normal, and the power flow is normal, then first initiate the blocking delay, and then continue to determine whether the impedance trajectory exceeds the maximum allowable measured impedance and whether the microgrid on the opposite side... If the voltage amplitude and phase angle change abruptly, it is determined that an external through-fault has developed into the area, and the local boundary protection unit is immediately activated. and the opposite boundary protection unit The tripping action.

[0035] S3 includes:

[0036] S31. Let the global state vector be... ,in , The data collected by the protection unit at the PCC point are respectively the first... Voltage and phase angle of a microgrid The number of microgrids in the networked microgrid is given; using the active and reactive power injected into each microgrid as collected by the protection unit at the PCC point as input, the optimal estimate of the global state vector is obtained by minimizing the measurement residuals using the least squares method. The expression for the measurement residual is:

[0037] ;

[0038] In the above formula, To measure residuals; Inject a column vector consisting of active and reactive power into each microgrid collected at the PCC point; For As input, the estimated active and reactive power of each microgrid is obtained by solving the nonlinear power balance equations of the entire network based on the power grid topology and impedance. It is a diagonal positive definite matrix, whose diagonal elements are the weights of the PCC measurement, usually taken as the reciprocal of the variance of the corresponding measurement error;

[0039] S32. Construct a strategy optimization model with the goal of maximizing power supply reliability. The objective function of the strategy optimization model is:

[0040] ;

[0041] In the above formula, The objective function of the strategy optimization model; , All are weighting coefficients; This represents the total amount of electricity that the system would not supply due to power outages within a given year. ; This represents the average number of power outage hours for users within the system within a given year. ;

[0042] Optimal estimation based on global state vector Solve the above strategy optimization model and output the updated local fault isolation strategy and regional fault isolation strategy.

[0043] Secondly, this invention proposes a hierarchical protection strategy optimization system for networked microgrids in a non-communication environment, the hierarchical protection strategy optimization system comprising:

[0044] The local protection module is used to install protection units within each microgrid, at both ends of the interconnection lines between microgrids, and at the common grid connection point in a networked microgrid. Based on the electrical information of the protection units installed within each microgrid and at the common grid connection point, it identifies local faults in the local microgrid and performs local fault isolation based on the current local fault isolation strategy.

[0045] The regional coordination module is used to designate the protection unit set on the side of the interconnection line closer to the local microgrid as the local boundary protection unit and the protection unit set on the side closer to the opposite microgrid as the opposite boundary protection unit. Based on the electrical information of the local boundary protection unit, it infers the regional electrical information of the opposite boundary protection unit. Based on the inferred regional electrical information and the current regional fault isolation strategy, it performs regional fault isolation.

[0046] The global decision module is used to infer global electrical information of the networked microgrid based on the electrical information of the protection unit at the common grid connection point. The inferred global electrical information is then input into the strategy optimization model constructed with the goal of maximizing power supply reliability. The strategy optimization model is solved to update the local fault isolation strategy and the regional fault isolation strategy.

[0047] The local protection module includes an operating mode recognition subunit and a local fault judgment subunit.

[0048] The operation mode identification subunit is used to identify the current operation mode of the networked microgrid by setting up protection units inside each microgrid, at both ends of the interconnection lines between microgrids, and at the common grid connection point.

[0049] The local fault judgment subunit is used to calculate the comprehensive fault confidence score of each local protection unit in the local microgrid according to the combination of primary and secondary criteria corresponding to the current operating mode, compare whether the comprehensive fault confidence score is greater than the preset fault threshold, and if so, determine that there is a local fault in the local protection unit and isolate the local fault by initiating the tripping action of the local protection unit.

[0050] The operation mode identification subunit is used to determine that the networked microgrid is in an islanded state when the frequency deviation and voltage deviation collected by the protection unit deployed at the photovoltaic grid connection point inside the microgrid are greater than the corresponding preset threshold; and to determine that the networked microgrid is in a grid-connected state when the voltage collected by the protection unit deployed at the common grid connection point drops significantly; otherwise, it is determined that the networked microgrid is in a transitional state.

[0051] The local fault judgment subunit is used to calculate the comprehensive fault confidence score by using a first main and auxiliary criterion combination when the networked microgrid is in an islanded state. The main criterion in the first main and auxiliary criterion combination is the current change rate and the auxiliary criterion is the voltage drop depth. If the comprehensive fault confidence score calculated by using the first main and auxiliary criterion combination is greater than the preset fault threshold, it is determined that there is a local fault.

[0052] The local fault judgment subunit is used to calculate the comprehensive fault confidence score by using the second main and auxiliary criteria combination when the networked microgrid is in grid-connected state. The main criterion in the second main and auxiliary criteria combination is the measured impedance, and the auxiliary criterion is the change in fundamental current. If the comprehensive fault confidence score calculated by the second main and auxiliary criteria combination is greater than the preset fault threshold, it is determined that there is a local fault.

[0053] The local fault judgment subunit is used to calculate the comprehensive fault confidence score by using a combination of the first primary and secondary criteria and a combination of the second primary and secondary criteria when the networked microgrid is in a transitional state. If the comprehensive fault confidence score calculated by using any combination of primary and secondary criteria is greater than the preset fault threshold, it is determined that there is a local fault.

[0054] The local fault judgment subunit is used to perform a weighted fusion of the main and auxiliary criteria combinations to obtain a comprehensive fault confidence score according to the following formula:

[0055] ;

[0056] In the above formula, The overall fault confidence score; , These refer to the primary criterion and the secondary criterion in the primary and secondary criterion combination; , These are the weights corresponding to the primary criterion and the secondary criterion, respectively.

[0057] The regional coordination module includes a regional inference subunit and a regional fault isolation subunit.

[0058] The region inference subunit is used to set up local microgrids in a networked microgrid. Connected to the opposite microgrid via interconnection lines Connection, based on local microgrids Local boundary protection unit on one side The measured electrical information is for the microgrid located on the opposite side. One side of the opposite boundary protection unit By performing regional electrical information inference, the protection units of the opposite boundary are obtained. The results of regional electrical information inference;

[0059] The regional fault isolation subunit is used to determine the opposite boundary protection unit. Based on the regional electrical information inference results, and the following regional fault isolation strategy, the microgrid on the opposite side is analyzed. Perform area fault isolation:

[0060] Regional fault isolation strategy one:

[0061] If the local boundary protection unit The measured impedance decrease, the side boundary protection unit If the voltage amplitude and phase angle are normal, and the power flow direction is normal, then an external through-fault is determined to have occurred, and a blocking delay is initiated. The blocking delay refers to prohibiting the local boundary protection unit from operating for a preset duration. and the opposite boundary protection unit If the circuit breaker trips and continues to monitor, and the external through-fault does not disappear after a preset monitoring period, the interlock will be lifted.

[0062] Regional fault isolation strategy two:

[0063] If the local boundary protection unit With the opposite boundary protection unit If both voltage drops and power flows in the opposite direction, then the local microgrid is identified. In the event of an internal fault, the local microgrid should be contacted. The fault electrical distance sequencing triggers the tripping action of the boundary protection unit;

[0064] Regional fault isolation strategy three:

[0065] First determine if it is a local boundary protection unit. The measured impedance decrease, the side boundary protection unit If the voltage amplitude and phase angle are normal, and the power flow is normal, then first initiate the blocking delay, and then continue to determine whether the impedance trajectory exceeds the maximum allowable measured impedance and whether the microgrid on the opposite side... If the voltage amplitude and phase angle change abruptly, it is determined that an external through-fault has developed into the area, and the local boundary protection unit is immediately activated. and the opposite boundary protection unit The tripping action.

[0066] The global decision-making module includes a global state inference subunit and a parameter optimization subunit.

[0067] The global state inference subunit is used to set the global state vector. ,in , The data collected by the protection unit at the PCC point are respectively the first... Voltage and phase angle of a microgrid The number of microgrids in the networked microgrid is given; using the active and reactive power injected into each microgrid as collected by the protection unit at the PCC point as input, the optimal estimate of the global state vector is obtained by minimizing the measurement residuals using the least squares method. The expression for the measurement residual is:

[0068] ;

[0069] In the above formula, To measure residuals; Inject a column vector consisting of active and reactive power into each microgrid collected at the PCC point; For As input, the estimated active and reactive power of each microgrid is obtained by solving the nonlinear power balance equations of the entire network based on the power grid topology and impedance. It is a diagonal positive definite matrix, whose diagonal elements are the weights of the PCC measurement, usually taken as the reciprocal of the variance of the corresponding measurement error;

[0070] The parameter optimization subunit is used to construct a strategy optimization model with the goal of maximizing power supply reliability, based on the optimal estimation of the global state vector. Solve the above strategy optimization model to output the updated local fault isolation strategy and regional fault isolation strategy; the objective function of the strategy optimization model is:

[0071] ;

[0072] In the above formula, The objective function of the strategy optimization model; , All are weighting coefficients; This represents the total amount of electricity that the system would not supply due to power outages within a given year. ; This represents the average number of power outage hours for users within the system within a given year. .

[0073] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0074] 1. The hierarchical protection strategy optimization method described in this invention constructs a hierarchical protection logic based on local electrical information perception, including a local protection layer, a regional coordination layer, and a global decision layer. Each layer makes independent judgments and decisions based on locally collected electrical quantities, thereby achieving reliable fault isolation of the networked microgrid system under the condition of no communication infrastructure. This improves the selectivity, speed, and adaptability of the system in communication-constrained environments. Each microgrid can act independently and also work together to maintain the overall stability of the system.

[0075] 2. The hierarchical protection strategy optimization method described in this invention can avoid false tripping / failure to trip caused by switching of operating modes (grid-connected / islanded) and drastic fluctuations in the amplitude and direction of fault current. Attached Figure Description

[0076] Figure 1 This is a flowchart of the hierarchical protection strategy optimization method described in this invention.

[0077] Figure 2 The topology diagram of the networked microgrid system used in the example is shown.

[0078] Figure 3 This is a schematic diagram of the structure of the hierarchical protection strategy optimization system described in this invention. Detailed Implementation

[0079] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings.

[0080] The proposed optimization method for hierarchical protection strategy of networked microgrids in a non-communication environment is applicable to networked microgrid systems consisting of multiple interconnected AC, DC, or hybrid microgrids, where multiple microgrids are connected by interconnecting lines. However, relying solely on local fault isolation strategies can easily lead to: erroneous disconnection of interconnecting lines (incorrectly isolating healthy microgrids); failure to operate or delayed operation (due to the complexity of fault current paths and the insensitivity of local criteria); and non-selective tripping (multiple protections operating simultaneously, expanding the power outage area). Therefore, this invention sets up a local protection layer, a regional coordination layer, and a global decision-making layer in the networked microgrid system. Protection units are deployed at key electrical nodes of the local microgrid (such as within the microgrid, at both ends of the interconnecting lines between microgrids, and at PCC points). These protection units collect electrical information from their own installation points and issue tripping commands to the circuit breakers (or solid-state switches) they control to isolate faults. The electrical information includes three-phase voltage, three-phase current, system frequency, and active / reactive power, etc. Protection units deployed within a local microgrid (e.g., distributed energy sources, critical load branches, microgrid internal buses / feeders) are called local protection units. Local protection units at distributed energy sources refer to those at the outputs of photovoltaic inverters, wind turbines, and energy storage converters, respectively. Local protection units at critical load branches refer to those at the outputs of critical load feeders. Local protection units at microgrid internal buses / feeders refer to those at the microgrid's internal tie lines or sectionalizing switches. Protection units deployed at both ends of the interconnection lines between the local microgrid and adjacent microgrids are called boundary protection units. In the absence of communication, the local protection layer first identifies local faults by collecting electrical information through local measurements from the local protection units within the local microgrid and using operating mode (grid-connected / islanded) switching criteria. Initial fault isolation is then achieved by initiating tripping actions of the local protection units. The regional coordination layer infers the state of the opposite microgrid based on electrical information collected locally by the boundary protection units on one side of the local microgrid. Since each interconnecting line can autonomously and independently determine the fault location, a regional fault isolation strategy is generated based on preset logic to achieve communication-free collaborative protection. The regional fault isolation strategy allows, accelerates, or blocks the tripping behavior of boundary protection units already activated by the local protection layer, achieving selective coordination. For example, when an internal fault occurs in microgrid A, the local protection unit within microgrid A will trip to isolate the fault point. The state of the opposite microgrid is inferred from the electrical information of the boundary protection units on the interconnecting line. If the opposite microgrid is in a healthy state, the tripping command of the boundary protection unit on that interconnecting line is actively blocked, thereby maintaining the structural integrity and energy mutual support capability of the entire networked microgrid. If the fault in microgrid A affects the opposite microgrid, the tripping of the boundary protection unit on the interconnecting line is allowed.The present invention also deploys a global decision layer at the common grid connection point (PCC) of the networked microgrid. The global decision layer infers the global operating status based solely on local measurements and offline topology models, and generates offline optimization parameter packages to periodically improve the performance of local fault isolation strategies and regional fault isolation strategies.

[0081] Example 1:

[0082] See Figure 1 An optimization method for hierarchical protection strategy of networked microgrids in a non-communication environment is proposed, which is carried out in the following steps:

[0083] S1. In a networked microgrid, protection units are installed inside each microgrid, at both ends of the interconnection lines between microgrids, and at the common grid connection point. Based on the electrical information of the protection units installed inside each microgrid and at the common grid connection point, local faults in the local microgrid are identified and local fault isolation is performed based on the current local fault isolation strategy.

[0084] Specifically, S1 includes:

[0085] S11. Protection units are installed within each microgrid, at both ends of the interconnection lines between microgrids, and at the common grid connection point in the networked microgrid. Based on the electrical information collected by the protection units installed within each microgrid and at the common grid connection point, the current operating mode of the networked microgrid is identified.

[0086] If the frequency deviation and voltage deviation collected by the protection unit deployed at the photovoltaic grid connection point inside the microgrid are greater than the corresponding preset threshold (e.g., frequency deviation greater than 0.5Hz, voltage deviation greater than 10%), the networked microgrid is determined to be in an islanded state.

[0087] If the voltage collected by the protection unit deployed at the public grid connection point drops significantly (e.g., drops to 70% of the rated voltage), the networked microgrid is determined to be in grid-connected state.

[0088] Otherwise, the networked microgrid is determined to be in a transitional state.

[0089] S12. Based on the combination of primary and secondary criteria corresponding to the current operating mode, calculate the comprehensive fault confidence score of each local protection unit within the local microgrid, compare whether the comprehensive fault confidence score is greater than the preset fault threshold, and if so, determine that there is a local fault in the local protection unit and isolate the local fault by initiating the tripping action of the local protection unit.

[0090] When the networked microgrid is in an islanded state, the comprehensive fault confidence score is calculated using the first primary and secondary criteria combination. The primary criterion in the first primary and secondary criteria combination is the current change rate, and the secondary criterion is the voltage drop depth. If the comprehensive fault confidence score calculated using the first primary and secondary criteria combination is greater than the preset fault threshold (e.g., 0.65) and the duration is ≥100ms, then a local fault is determined to exist, and the tripping action of the local protection unit is initiated.

[0091] When the networked microgrid is in grid-connected state, the comprehensive fault confidence score is calculated using the second main and auxiliary criteria combination. The main criterion in the second main and auxiliary criteria combination is the measured impedance, and the auxiliary criterion is the change in fundamental current. If the comprehensive fault confidence score calculated using the second main and auxiliary criteria combination is greater than the preset fault threshold (e.g., 0.7) and the duration is ≥200ms, then a local fault is determined to exist, and the tripping action of the local protection unit is initiated.

[0092] When the networked microgrid is in a transitional state, the comprehensive fault confidence score is calculated using both the first primary and secondary criteria combination and the second primary and secondary criteria combination to determine whether a local fault exists. If no local fault is detected within 200ms, the networked microgrid is forcibly treated as an island state, and only the first primary and secondary criteria combination is used to calculate the comprehensive fault confidence score and determine whether a local fault exists.

[0093] First, normalize the combination of primary and secondary criteria. The normalization calculation formula for the first combination of primary and secondary criteria is as follows:

[0094] ; ;

[0095] In the above formula, , These refer to the primary criterion and the secondary criterion in the primary and secondary criterion combination; The rate of change of current; This is the sensitivity coefficient. It can be adjusted according to the type of distributed energy source; for example, for a photovoltaic inverter-dominated system... The value is 1.0, which is the case for diesel generator-driven systems. The value is 0.8; This refers to the rated operating current during steady-state operation. This refers to the voltage drop depth. This is the rated voltage for steady-state operation; It is the lowest effective value of the three-phase voltage within a short time window after the fault occurs (usually 1-3 power frequency cycles, i.e., 20-60ms).

[0096] The normalized calculation formula for the second primary and secondary criterion combination is:

[0097] ; ;

[0098] In the above formula, Maximum permissible measurement impedance; To measure impedance during a short circuit; This refers to the change in fundamental current at power frequency.

[0099] Then, the normalized combination of primary and secondary criteria is weighted and fused according to the following formula to obtain the comprehensive fault confidence score:

[0100] ;

[0101] In the above formula, The overall fault confidence score; , These are the weights corresponding to the primary and secondary criteria, respectively. For the first combination of primary and secondary criteria... , The values ​​are 0.3 and 0.7 respectively. For the combination of the second primary and secondary criteria, , The values ​​are 0.6 and 0.4 respectively.

[0102] S2. Let the protection unit set on the side of the interconnection line closer to the local microgrid be the local boundary protection unit, and the protection unit set on the side closer to the opposite microgrid be the opposite boundary protection unit. Based on the electrical information of the local boundary protection unit, perform regional electrical information inference on the opposite boundary protection unit. Based on the inferred regional electrical information inference result and the current regional fault isolation strategy, perform regional fault isolation.

[0103] Specifically, S2 includes:

[0104] Consider a local microgrid in a networked microgrid. Connected to the opposite microgrid via interconnection lines Connection, based on local microgrid Local boundary protection unit on one side The measured electrical information is used for the microgrid on the opposite side. Perform side boundary protection unit on one side By performing regional electrical information inference, the protection units of the opposite boundary are obtained. The results of regional electrical information inference; specifically, for local boundary protection units. The following electrical information can be measured:

[0105] ;in Represents the electrical information matrix; , These are from local boundary protection units Flow to the opposite boundary protection unit Active and reactive power; , Local boundary protection units , opposite side boundary protection unit The voltage amplitude; This is the noise vector; It is a nonlinear power flow function;

[0106] When the local microgrid For AC microgrids, neglecting the line-to-ground admittance, the local boundary protection unit is calculated using the following formula. , opposite side boundary protection unit phase angle , : , ,in For local microgrids With the opposite microgrid Reactance of AC lines between; when the local microgrid When it is a DC microgrid, from the local boundary protection unit Flow to the opposite boundary protection unit reactive power and local boundary protection units , opposite side boundary protection unit phase angle , All are 0, from the local boundary protection unit Flow to the opposite boundary protection unit active power The following relationship must be satisfied: ,in For local microgrids With the opposite microgrid The DC resistance of the power supply lines between them;

[0107] Further construct the boundary node state vector The optimal estimate of the boundary node state vector is obtained by weighted least squares estimation (WLS). : ,in The measurement weight matrix is ​​the inverse of the measurement error covariance matrix;

[0108] To prevent estimation distortion caused by sensor malfunction or strong interference, a standardized residual test is used: first, the residual vector is calculated. And generate residual vector Standardized residuals for each dimension ,in It is the first The standard deviation of each measurement data, if Then determine the first One measurement data point is considered abnormal; this measurement data point will be removed from... Remove from the middle, re-perform WLS estimation, the measurement data refers to the electrical information matrix. In , , , ;

[0109] The optimal estimation of the boundary node state vector Converse boundary protection unit The voltage amplitude and phase angle serve as the boundary protection unit on the opposite side. Based on the regional electrical information inference results, and according to the opposite boundary protection unit Based on the regional electrical information inference results, and the following regional fault isolation strategy, the microgrid on the opposite side is analyzed. Perform area fault isolation:

[0110] Regional fault isolation strategy one:

[0111] If the local boundary protection unit The measured impedance decrease, the side boundary protection unit If the voltage amplitude and phase angle are normal, and the power flow is normal, then an external through-fault (such as a fault in the upstream power grid) is determined to have occurred, and a blocking delay is initiated. The blocking delay refers to prohibiting the opposite boundary protection unit from operating for a preset time. The system trips and continuously monitors the system, preventing erroneous isolation of healthy microgrids through a blocking delay, thereby maintaining the energy mutual support capability of NMGs. If the external through-fault does not disappear after a preset time, the blocking is lifted to prevent the fault from spreading to the healthy microgrid.

[0112] Regional fault isolation strategy two:

[0113] If the local boundary protection unit With the opposite boundary protection unit If both voltage drops and power flows in the opposite direction, then the local microgrid is identified. In the event of an internal fault, the local microgrid should be contacted. The fault electrical distance is prioritized, and the tripping actions of other boundary protection units are initiated sequentially from nearest to farthest. For example, consider the microgrid on the opposite side. With another microgrid Interconnection, in the boundary protection unit After a 20ms delay following the trip (tripping action completion time must be ≤15ms), the microgrid on the opposite side will be used. For the local microgrid, further determine the microgrid on the opposite side. microgrids If a voltage drop and reverse power flow occur on the interconnecting lines, the boundary protection unit on that interconnecting line will trip. By employing this stepped timing coordination, selective isolation can be achieved, minimizing the scope of the power outage.

[0114] Regional fault isolation strategy three:

[0115] First determine if it is a local boundary protection unit. The measured impedance decrease, the side boundary protection unit If the voltage amplitude and phase angle are normal, and the power flow is normal, then first initiate the blocking delay, and then continue to determine whether the impedance trajectory exceeds the maximum allowable measured impedance and whether the microgrid on the opposite side... If the voltage amplitude and phase angle change abruptly, it is determined that an external through-fault has developed into the area, and the local boundary protection unit is immediately activated. and the opposite boundary protection unit The tripping action must be completed within 0.1 seconds from the detection of a sudden change in voltage amplitude and phase angle to the tripping action, in order to avoid failure to trip due to initial blockage and improve safety.

[0116] S3. Perform global electrical information inference on the networked microgrid, input the obtained global electrical information inference results into the strategy optimization model constructed with the optimization objective of maximizing power supply reliability, solve the strategy optimization model and output the optimal local fault isolation strategy and regional fault isolation strategy.

[0117] Specifically, S3 includes:

[0118] S31. Let the global state vector be... ,in , The data collected by the protection unit at the PCC point are respectively the first... Voltage and phase angle of a microgrid The number of microgrids in a networked microgrid is defined. Using the active and reactive power injected into each microgrid at the PCC point as outputs, the optimal estimate of the global state vector is obtained by minimizing the measurement residuals and solving using the least squares method. The expression for the measurement residual is:

[0119] ;

[0120] In the above formula, This is the global state vector. , The number of microgrids in a networked microgrid. To measure residuals; Inject a column vector consisting of active and reactive power into each microgrid collected at the PCC point; For As input, the estimated active and reactive power of each microgrid is obtained by solving the nonlinear power balance equations of the entire network based on the power grid topology and impedance. It is a diagonal positive definite matrix, whose diagonal elements are the weights of the PCC measurement, usually taken as the reciprocal of the variance of the corresponding measurement error;

[0121] S32. Construct a strategy optimization model with the goal of maximizing power supply reliability. The objective function of the strategy optimization model is:

[0122] ;

[0123] In the above formula, The objective function of the policy optimization model is minimized. To maximize power supply reliability; , All are weighting coefficients; This represents the total amount of electricity lost due to power outages within a given year, expressed in kWh. ; This indicates the average number of power outage hours per user within the system per year, expressed in hours per user per year. The model includes physical constraints and information constraints. The physical constraints mean that all protection settings must be within a reasonable range, while the information constraints mean that the information required for optimization must be based on information available without communication.

[0124] The optimal estimate of the obtained global state vector The input is fed into the strategy optimization model, and the parameters in the local fault isolation strategy and the regional fault isolation strategy are updated by solving the above strategy optimization model; the solution method can adopt particle swarm optimization algorithm or genetic algorithm, and the index value ( as well as The calculations can be based on industry standards such as the "DLT 1563-2016 Guidelines for Reliability Assessment of Medium Voltage Distribution Networks". The resulting parameter optimization results are packaged into a parameter configuration file. The regional fault isolation strategy includes the following parameters: blocking delay, measurement weight matrix, etc. The local fault isolation strategy includes the following parameters: in island mode Value, maximum permissible measured impedance Weights corresponding to the primary and secondary criteria and The parameter configuration file is periodically distributed via a local maintenance interface (such as RS485, USB, or Bluetooth) (e.g., during off-peak hours in the early morning each day), or triggered upon detection of a system topology change, without interfering with real-time protection. Through real-time sensing and rolling parameter adjustments, the protection performance is autonomously optimized, ensuring the isolation strategy is always at its optimal operating point. All adjustments are completed within 1ms, without affecting the speed of protection.

[0125] S3 also includes optimal estimation based on the global state vector. The failure probability of each node is calculated according to the following formula, and vulnerable nodes (i.e., vulnerable microgrids) are identified based on the failure probability:

[0126] ;

[0127] In the above formula, For microgrids The probability of failure; , These are all weighted coefficients, which can be set according to the preference for voltage drop and phase angle shift in the actual application scenario; , These represent the optimal estimates of the global state vector. Optimal estimation of voltage amplitude and phase angle; This is the rated voltage for steady-state operation; The reference phase angle generally represents the node phase angle value of the system in steady state.

[0128] Optimal estimation based on global state vector A probabilistic assessment of the failure probability of all system nodes directly reflects the likelihood of different nodes failing, helping maintenance personnel quickly locate faults. The failure impact value is calculated by multiplying the node's load importance by its failure probability. Nodes are then sorted in descending order based on their failure impact values. Nodes ranked higher are identified as vulnerable nodes. Failures of vulnerable nodes will have a greater impact on the system; therefore, they should receive more attention. Measures such as adjusting protection strategies or increasing monitoring frequency can mitigate the risk. This provides guidance for the allocation of maintenance resources.

[0129] Performance verification:

[0130] 1. System Topology

[0131] This example is based on, for example Figure 1The networked microgrid system shown is an AC system, where the physical connections are AC lines and the virtual connections are DC lines (also serving as interconnection lines between microgrids). PU represents the deployed protection unit. This networked microgrid system includes the following microgrids: an AC microgrid (node ​​A) with 50kW photovoltaic power, a 30kW diesel generator, and 40kW load (400VAC system); a DC microgrid (node ​​B) with 40kW wind power, 50kWh energy storage, and 30kW load (500VDC system); and a hybrid microgrid (node ​​C) with 30kW photovoltaic power and 25kW load (AC / DC hybrid). The tie line on the DC side is 0.05Ω; the maximum switching capacity of the microgrid is 120kW; the nominal line voltage of the AC system is 400V (corresponding to a phase voltage of 230V and a frequency of 50Hz). The nominal DC voltage of the DC system is 500V. The above three microgrids are connected to a common bus via a star topology and then connected to the 10kV distribution network via the PCC point.

[0132] 2. The deployment of protection units and initialization of key parameters in the networked microgrid system are shown in Table 1.

[0133] Table 1. Protection Unit Deployment and Key Parameter Initialization in Networked Microgrid Systems

[0134]

[0135] 3. Fault events

[0136] DC microgrid internal inter-electrode short circuit (t=0s): Set the DC bus of the DC microgrid (50m away from the interconnection converter) to experience an inter-electrode short circuit. The system is in grid-connected mode, with wind power output of 35kW, energy storage discharge of 10kW, and load of 28kW.

[0137] 4. The layered protection operation using the method described in this invention includes the following five stages:

[0138] Phase 1: The local protection layer collects local electrical information and makes a preliminary fault judgment (t=0-8ms). The results are shown in Table 2.

[0139] Table 2 Results of Local Electrical Information Collection and Preliminary Fault Assessment

[0140]

[0141] Phase 2: The regional coordination layer performs regional electrical information inference and protection coordination (t=8-50ms), and the results are shown in Table 3.

[0142] Table 3 Results of Regional Electrical Information Inference and Protection Coordination

[0143]

[0144] Phase 3: Global decision-making layer strategy optimization (t=10s), the results are shown in Table 4.

[0145] Table 4. Optimization Results of Global Decision-Making Layer

[0146]

[0147] Phase 4: Adaptive adjustment of the closed loop, the results are shown in Table 5.

[0148] Table 5 Adaptive Adjustment Closed-Loop Results

[0149]

[0150] Phase 5: System recovery (t=100ms-5min), the results are shown in Table 6.

[0151] Table 6 System Recovery Status

[0152]

[0153] Regarding fault isolation range, traditional fault isolation schemes require disconnecting the entire DC microgrid and interconnecting lines, while the method described in this invention can accurately isolate faulty branches, significantly reducing the power outage range. In terms of power supply assurance in non-faulty areas, traditional fault isolation schemes are prone to misjudgment leading to the loss of interconnection support for AC / hybrid microgrids, while the method described in this invention maintains interconnection line connections through boundary state inference, improving power supply reliability. Regarding action selectivity, traditional fault isolation schemes are prone to mistakenly disconnecting healthy lines under through-fault conditions, while the method described in this invention relies on the preset logic of the regional coordination layer to accurately locate the fault source, avoiding mistaken disconnection. Crucially, traditional fault isolation schemes rely on communication links, and protection functions fail when communication is interrupted, while the method described in this invention achieves collaborative decision-making based solely on local electrical quantity sensing and preset rules, enabling operation without communication. This effectively extends the application scenarios of NMGs to island microgrids, disaster emergencies, remote areas, and other environments with weak or interrupted communication, significantly improving the survivability and power supply resilience of NMGs under extreme conditions.

[0154] Example 2:

[0155] See Figure 3A hierarchical protection strategy optimization system for networked microgrids in a non-communication environment is disclosed. The system includes a local protection module, a regional coordination module, and a global decision-making module. The local protection module is used to install protection units within each microgrid, at both ends of the interconnection lines between microgrids, and at the common grid connection point. Based on the electrical information of the protection units within each microgrid and at the common grid connection point, it identifies local faults in the local microgrid and performs local fault isolation based on the current local fault isolation strategy. Specifically, the local protection module includes an operating mode identification subunit and a local fault judgment subunit. The unit, namely the operation mode identification subunit, is used to identify the current operation mode of the networked microgrid by setting up protection units within each microgrid, at both ends of the interconnection lines between microgrids, and at the common grid connection point, based on the electrical information collected by the protection units set up within each microgrid and at the common grid connection point. The local fault judgment subunit is used to calculate the comprehensive fault confidence score of each local protection unit within the local microgrid according to the combination of primary and secondary criteria corresponding to the current operation mode, compare whether the comprehensive fault confidence score is greater than a preset fault threshold, and if so, determine that the local protection unit has a local fault, and activate it. The local protection unit trips to isolate local faults. Specifically, the operation mode identification subunit determines that the networked microgrid is in an islanded state when the frequency and voltage deviations collected by the protection unit deployed at the photovoltaic grid connection point within the microgrid exceed the corresponding preset thresholds; it determines that the networked microgrid is in a grid-connected state when the voltage collected by the protection unit deployed at the common grid connection point drops significantly; otherwise, it determines that the networked microgrid is in a transitional state. Specifically, the local fault judgment subunit calculates the comprehensive fault confidence score using a combination of the first primary and secondary criteria when the networked microgrid is in an islanded state. In the first primary-secondary criterion combination, the primary criterion is the rate of change of current, and the secondary criterion is the voltage sag depth. If the comprehensive fault confidence score calculated using the first primary-secondary criterion combination is greater than a preset fault threshold, a local fault is determined to exist. Specifically, the local fault judgment subunit is used to calculate the comprehensive fault confidence score using a second primary-secondary criterion combination when the networked microgrid is in a grid-connected state. The primary criterion in the second primary-secondary criterion combination is the measured impedance, and the secondary criterion is the fundamental current change. If the comprehensive fault confidence score calculated using the second primary-secondary criterion combination is greater than a preset fault threshold, a local fault is determined to exist. Specifically, the local fault judgment subunit is used to calculate the comprehensive fault confidence score using both the first and second primary-secondary criterion combinations when the networked microgrid is in a transition state. If the comprehensive fault confidence score calculated using either primary-secondary criterion combination is greater than a preset fault threshold, a local fault is determined to exist. Specifically, the local fault judgment subunit is used to obtain the comprehensive fault confidence score by weighted fusion of the primary and secondary criterion combinations according to the following formula:

[0156] ;

[0157] In the above formula, The overall fault confidence score; , These refer to the primary criterion and the secondary criterion in the primary and secondary criterion combination; , These are the weights corresponding to the primary criterion and the secondary criterion, respectively.

[0158] The regional coordination module is used to designate protection units located on the side of the interconnecting line closest to the local microgrid as local boundary protection units and protection units located on the side closest to the opposite microgrid as opposite boundary protection units. Based on the electrical information of the local boundary protection units, it infers regional electrical information for the opposite boundary protection units. Based on the inferred regional electrical information and the current regional fault isolation strategy, it performs regional fault isolation. Specifically, the regional coordination module includes a regional inference subunit and a regional fault isolation subunit. The regional inference subunit is used to designate local microgrids within the networked microgrid... Connected to the opposite microgrid via interconnection lines Connection, based on local microgrids Local boundary protection unit on one side The measured electrical information is for the microgrid located on the opposite side. One side of the opposite boundary protection unit By performing regional electrical information inference, the protection units of the opposite boundary are obtained. The regional electrical information inference results; specifically, the regional fault isolation subunit is used to infer the results based on the opposite boundary protection unit. Based on the regional electrical information inference results, and the following regional fault isolation strategy, the microgrid on the opposite side is analyzed. Perform area fault isolation:

[0159] Regional fault isolation strategy one:

[0160] If the local boundary protection unit The measured impedance decrease, the side boundary protection unit If the voltage amplitude and phase angle are normal, and the power flow direction is normal, then an external through-fault is determined to have occurred, and a blocking delay is initiated. The blocking delay refers to prohibiting the local boundary protection unit from operating for a preset duration. and the opposite boundary protection unit If the circuit breaker trips and continues to monitor, and the external through-fault does not disappear after a preset monitoring period, the interlock will be lifted.

[0161] Regional fault isolation strategy two:

[0162] If the local boundary protection unit With the opposite boundary protection unit If both voltage drops and power flows in the opposite direction, then the local microgrid is identified. In the event of an internal fault, the local microgrid should be contacted. The fault electrical distance sequencing triggers the tripping action of the boundary protection unit;

[0163] Regional fault isolation strategy three:

[0164] First determine if it is a local boundary protection unit. The measured impedance decrease, the side boundary protection unit If the voltage amplitude and phase angle are normal, and the power flow is normal, then first initiate the blocking delay, and then continue to determine whether the impedance trajectory exceeds the maximum allowable measured impedance and whether the microgrid on the opposite side... If the voltage amplitude and phase angle change abruptly, it is determined that an external through-fault has developed into the area, and the local boundary protection unit is immediately activated. and the opposite boundary protection unit The tripping action.

[0165] The global decision-making module is used to infer global electrical information of the networked microgrid based on the electrical information of the protection unit at the common grid connection point. The inferred global electrical information is then input into a strategy optimization model constructed with maximizing power supply reliability as the optimization objective. The strategy optimization model is solved to update the parameters in the local fault isolation strategy and the regional fault isolation strategy. Specifically, the global decision-making module includes a global state inference subunit and a parameter optimization subunit. The global state inference subunit is used to set the global state vector. ,in , The data collected by the protection unit at the PCC point are respectively the first... Voltage and phase angle of a microgrid The number of microgrids in the networked microgrid is given; using the active and reactive power injected into each microgrid as collected by the protection unit at the PCC point as input, the optimal estimate of the global state vector is obtained by minimizing the measurement residuals using the least squares method. The expression for the measurement residual is:

[0166] ;

[0167] In the above formula, To measure residuals; Inject a column vector consisting of active and reactive power into each microgrid collected at the PCC point; For As input, the estimated active and reactive power of each microgrid is obtained by solving the nonlinear power balance equations of the entire network based on the power grid topology and impedance. It is a diagonal positive definite matrix, whose diagonal elements are the weights of the PCC measurement, usually taken as the reciprocal of the variance of the corresponding measurement error;

[0168] Specifically, the parameter optimization subunit is used to construct a strategy optimization model with the goal of maximizing power supply reliability, based on the optimal estimation of the global state vector. Solve the above strategy optimization model to output the updated local fault isolation strategy and regional fault isolation strategy; the objective function of the strategy optimization model is:

[0169] ;

[0170] In the above formula, The objective function of the strategy optimization model; , All are weighting coefficients; This represents the total amount of electricity that the system would not supply due to power outages within a given year. ; This represents the average number of power outage hours for users within the system within a given year. .

[0171] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program goods. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0172] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0173] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0174] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0175] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. An optimization method for hierarchical protection strategy of networked microgrids in a non-communication environment, characterized in that: The method for optimizing the hierarchical protection strategy includes: S1. In a networked microgrid, protection units are installed inside each microgrid, at both ends of the interconnection lines between microgrids, and at the common grid connection point. Based on the electrical information of the protection units installed inside each microgrid and at the common grid connection point, local faults in the local microgrid are identified and local fault isolation is performed based on the current local fault isolation strategy. S2. Let the protection unit set on the side of the interconnection line closer to the local microgrid be the local boundary protection unit, and the protection unit set on the side closer to the opposite microgrid be the opposite boundary protection unit. Based on the electrical information of the local boundary protection unit, perform regional electrical information inference on the opposite boundary protection unit. Based on the inferred regional electrical information inference result and the current regional fault isolation strategy, perform regional fault isolation. S3. Based on the electrical information of the protection unit at the common grid connection point, perform global electrical information inference on the networked microgrid. Input the inferred global electrical information inference results into the strategy optimization model constructed with the optimization objective of maximizing power supply reliability. Solve the strategy optimization model to update the local fault isolation strategy and the regional fault isolation strategy.

2. The method for optimizing hierarchical protection strategy of networked microgrids in a non-communication environment according to claim 1, characterized in that: S1 includes: S11. Protection units are installed inside each microgrid, at both ends of the interconnection lines between microgrids, and at the common grid connection point in the networked microgrid. Based on the electrical information collected by the protection units installed inside each microgrid and at the common grid connection point, the current operating mode of the networked microgrid is identified. S12. Based on the combination of primary and secondary criteria corresponding to the current operating mode, calculate the comprehensive fault confidence score of each local protection unit within the local microgrid, compare whether the comprehensive fault confidence score is greater than the preset fault threshold, and if so, determine that there is a local fault in the local protection unit and isolate the local fault by initiating the tripping action of the local protection unit.

3. The method for optimizing hierarchical protection strategy of networked microgrids in a non-communication environment according to claim 2, characterized in that: S11 includes: If the frequency deviation and voltage deviation collected by the protection unit deployed at the photovoltaic grid connection point inside the microgrid are greater than the corresponding preset threshold, the networked microgrid is determined to be in an islanded state; if the voltage collected by the protection unit deployed at the common grid connection point drops significantly, the networked microgrid is determined to be in a grid-connected state; otherwise, the networked microgrid is determined to be in a transitional state. S12 includes: When the networked microgrid is in an islanded state, the comprehensive fault confidence score is calculated using a combination of primary and secondary criteria. The primary criterion in the first combination of primary and secondary criteria is the rate of change of current, and the secondary criterion is the voltage drop depth. If the comprehensive fault confidence score calculated using the first combination of primary and secondary criteria is greater than the preset fault threshold, it is determined that there is a local fault. When the networked microgrid is in grid-connected state, the comprehensive fault confidence score is calculated using the second main and auxiliary criterion combination. The main criterion in the second main and auxiliary criterion combination is the measured impedance, and the auxiliary criterion is the change in fundamental current. If the comprehensive fault confidence score calculated using the second main and auxiliary criterion combination is greater than the preset fault threshold, it is determined that there is a local fault. When the networked microgrid is in a transitional state, the comprehensive fault confidence score is calculated by using both the first primary and secondary criteria combination and the second primary and secondary criteria combination. If the comprehensive fault confidence score calculated by using any primary and secondary criteria combination is greater than the preset fault threshold, it is determined that there is a local fault. The combined fault confidence score is obtained by weighted fusion of the main and auxiliary criteria using the following formula: ； In the above formula, The overall fault confidence score; , These refer to the primary criterion and the secondary criterion in the primary and secondary criterion combination; , These are the weights corresponding to the primary criterion and the secondary criterion, respectively.

4. The method for optimizing hierarchical protection strategy of networked microgrids in a non-communication environment according to claim 1, characterized in that: S2 includes: Consider a local microgrid in a networked microgrid. Connected to the opposite microgrid via interconnection lines Connection, based on local microgrids Local boundary protection unit on one side The measured electrical information is for the microgrid located on the opposite side. One side of the opposite boundary protection unit By performing regional electrical information inference, the protection units of the opposite boundary are obtained. The results of regional electrical information inference; According to the opposite boundary protection unit Based on the regional electrical information inference results, and the following regional fault isolation strategy, the microgrid on the opposite side is analyzed. Perform area fault isolation: Regional fault isolation strategy one: If the local boundary protection unit The measured impedance decrease, the side boundary protection unit If the voltage amplitude and phase angle are normal, and the power flow direction is normal, then an external through-fault is determined to have occurred, and a blocking delay is initiated. The blocking delay refers to prohibiting the local boundary protection unit from operating for a preset duration. and the opposite boundary protection unit If the circuit breaker trips and continues to monitor, and the external through-fault does not disappear after a preset monitoring period, the interlock will be lifted. Regional fault isolation strategy two: If the local boundary protection unit With the opposite boundary protection unit If both voltage drops and power flows in the opposite direction, then the local microgrid is identified. In the event of an internal fault, the local microgrid should be contacted. The fault electrical distance sequencing triggers the tripping action of the boundary protection unit; Regional fault isolation strategy three: First determine if it is a local boundary protection unit. The measured impedance decrease, the side boundary protection unit If the voltage amplitude and phase angle are normal, and the power flow is normal, then first initiate the blocking delay, and then continue to determine whether the impedance trajectory exceeds the maximum allowable measured impedance and whether the microgrid on the opposite side... If the voltage amplitude and phase angle change abruptly, it is determined that an external through-fault has developed into the area, and the local boundary protection unit is immediately activated. and the opposite boundary protection unit The tripping action.

5. The method for optimizing hierarchical protection strategy of networked microgrids in a non-communication environment according to claim 1, characterized in that: S3 includes: S31. Let the global state vector be... ,in , The data collected by the protection unit at the PCC point are respectively the first... Voltage and phase angle of a microgrid The number of microgrids in the networked microgrid is given; using the active and reactive power injected into each microgrid as collected by the protection unit at the PCC point as input, the optimal estimate of the global state vector is obtained by minimizing the measurement residuals using the least squares method. The expression for the measurement residual is: ； In the above formula, To measure residuals; Inject a column vector consisting of active and reactive power into each microgrid collected at the PCC point; For As input, the estimated active and reactive power of each microgrid is obtained by solving the nonlinear power balance equations of the entire network based on the power grid topology and impedance. It is a diagonal positive definite matrix, whose diagonal elements are the weights of the PCC measurement, usually taken as the reciprocal of the variance of the corresponding measurement error; S32. Construct a strategy optimization model with the goal of maximizing power supply reliability. The objective function of the strategy optimization model is: ； In the above formula, The objective function of the strategy optimization model; , All are weighting coefficients; This represents the total amount of electricity that the system would not supply due to power outages within a given year. ; This represents the average number of power outage hours for users within the system within a given year. ; The number of tripping actions for all protection units; Optimal estimation based on global state vector Solve the above strategy optimization model and output the updated local fault isolation strategy and regional fault isolation strategy.

6. A hierarchical protection strategy optimization system for networked microgrids in a non-communication environment, characterized in that: The hierarchical protection strategy optimization system includes: The local protection module is used to install protection units within each microgrid, at both ends of the interconnection lines between microgrids, and at the common grid connection point in a networked microgrid. Based on the electrical information of the protection units installed within each microgrid and at the common grid connection point, it identifies local faults in the local microgrid and performs local fault isolation based on the current local fault isolation strategy. The regional coordination module is used to designate the protection unit set on the side of the interconnection line closer to the local microgrid as the local boundary protection unit and the protection unit set on the side closer to the opposite microgrid as the opposite boundary protection unit. Based on the electrical information of the local boundary protection unit, it infers the regional electrical information of the opposite boundary protection unit. Based on the inferred regional electrical information and the current regional fault isolation strategy, it performs regional fault isolation. The global decision module is used to infer global electrical information of the networked microgrid based on the electrical information of the protection unit at the common grid connection point. The inferred global electrical information is then input into the strategy optimization model constructed with the goal of maximizing power supply reliability. The strategy optimization model is solved to update the local fault isolation strategy and the regional fault isolation strategy.

7. The hierarchical protection strategy optimization system for networked microgrids in a non-communication environment as described in claim 6, characterized in that: The local protection module includes an operating mode recognition subunit and a local fault judgment subunit. The operation mode identification subunit is used to identify the current operation mode of the networked microgrid by setting up protection units inside each microgrid, at both ends of the interconnection lines between microgrids, and at the common grid connection point. The local fault judgment subunit is used to calculate the comprehensive fault confidence score of each local protection unit in the local microgrid according to the combination of primary and secondary criteria corresponding to the current operating mode, compare whether the comprehensive fault confidence score is greater than the preset fault threshold, and if so, determine that there is a local fault in the local protection unit and isolate the local fault by initiating the tripping action of the local protection unit.

8. The hierarchical protection strategy optimization system for networked microgrids in a non-communication environment according to claim 7, characterized in that: The operation mode identification subunit is used to determine that the networked microgrid is in an islanded state when the frequency deviation and voltage deviation collected by the protection unit deployed at the photovoltaic grid connection point inside the microgrid are greater than the corresponding preset threshold; and to determine that the networked microgrid is in a grid-connected state when the voltage collected by the protection unit deployed at the common grid connection point drops significantly. Otherwise, the networked microgrid is determined to be in a transitional state; The local fault judgment subunit is used to calculate the comprehensive fault confidence score by using a first main and auxiliary criterion combination when the networked microgrid is in an islanded state. The main criterion in the first main and auxiliary criterion combination is the current change rate and the auxiliary criterion is the voltage drop depth. If the comprehensive fault confidence score calculated by using the first main and auxiliary criterion combination is greater than the preset fault threshold, it is determined that there is a local fault. The local fault judgment subunit is used to calculate the comprehensive fault confidence score by using the second main and auxiliary criteria combination when the networked microgrid is in grid-connected state. The main criterion in the second main and auxiliary criteria combination is the measured impedance, and the auxiliary criterion is the change in fundamental current. If the comprehensive fault confidence score calculated by the second main and auxiliary criteria combination is greater than the preset fault threshold, it is determined that there is a local fault. The local fault judgment subunit is used to calculate the comprehensive fault confidence score by using a combination of the first primary and secondary criteria and a combination of the second primary and secondary criteria when the networked microgrid is in a transitional state. If the comprehensive fault confidence score calculated by using any combination of primary and secondary criteria is greater than the preset fault threshold, it is determined that there is a local fault. The local fault judgment subunit is used to perform a weighted fusion of the main and auxiliary criteria combinations to obtain a comprehensive fault confidence score according to the following formula: ； In the above formula, The overall fault confidence score; , These refer to the primary criterion and the secondary criterion in the primary and secondary criterion combination; , These are the weights corresponding to the primary criterion and the secondary criterion, respectively.

9. The hierarchical protection strategy optimization system for networked microgrids in a non-communication environment according to claim 6, characterized in that: The regional coordination module includes a regional inference subunit and a regional fault isolation subunit. The region inference subunit is used to set up local microgrids in a networked microgrid. Connected to the opposite microgrid via interconnection lines Connection, based on local microgrids Local boundary protection unit on one side The measured electrical information is for the microgrid located on the opposite side. One side of the opposite boundary protection unit By performing regional electrical information inference, the protection units of the opposite boundary are obtained. The results of regional electrical information inference; The regional fault isolation subunit is used to determine the opposite boundary protection unit. Based on the regional electrical information inference results, and the following regional fault isolation strategy, the microgrid on the opposite side is analyzed. Perform area fault isolation: Regional fault isolation strategy one: If the local boundary protection unit The measured impedance decrease, the side boundary protection unit If the voltage amplitude and phase angle are normal, and the power flow direction is normal, then an external through-fault is determined to have occurred, and a blocking delay is initiated. The blocking delay refers to prohibiting the local boundary protection unit from operating for a preset duration. and the opposite boundary protection unit If the circuit breaker trips and continues to monitor, and the external through-fault does not disappear after a preset monitoring period, the interlock will be lifted. Regional fault isolation strategy two: If the local boundary protection unit With the opposite boundary protection unit If both voltage drops and power flows in the opposite direction, then the local microgrid is identified. In the event of an internal fault, the local microgrid should be contacted. The fault electrical distance sequencing triggers the tripping action of the boundary protection unit; Regional fault isolation strategy three: First determine if it is a local boundary protection unit. The measured impedance decrease, the side boundary protection unit If the voltage amplitude and phase angle are normal, and the power flow is normal, then first initiate the blocking delay, and then continue to determine whether the impedance trajectory exceeds the maximum allowable measured impedance and whether the microgrid on the opposite side... If the voltage amplitude and phase angle change abruptly, it is determined that an external through-fault has developed into the area, and the local boundary protection unit is immediately activated. and the opposite boundary protection unit The tripping action.

10. The hierarchical protection strategy optimization system for networked microgrids in a non-communication environment according to claim 6, characterized in that: The global decision-making module includes a global state inference subunit and a parameter optimization subunit. The global state inference subunit is used to set the global state vector. ,in , The data collected by the protection unit at the PCC point are respectively the first... Voltage and phase angle of a microgrid The number of microgrids in a networked microgrid; Using the active and reactive power injected into each microgrid collected by the protection unit at the PCC point as input, the optimal estimate of the global state vector is obtained by minimizing the measurement residuals using the least squares method. The expression for the measurement residual is: ； In the above formula, To measure residuals; Inject a column vector consisting of active and reactive power into each microgrid collected at the PCC point; For As input, the estimated active and reactive power of each microgrid is obtained by solving the nonlinear power balance equations of the entire network based on the power grid topology and impedance. It is a diagonal positive definite matrix, whose diagonal elements are the weights of the PCC measurement, usually taken as the reciprocal of the variance of the corresponding measurement error; The parameter optimization subunit is used to construct a strategy optimization model with the goal of maximizing power supply reliability, based on the optimal estimation of the global state vector. Solve the above strategy optimization model to output the updated local fault isolation strategy and regional fault isolation strategy; the objective function of the strategy optimization model is: ； In the above formula, The objective function of the strategy optimization model; , All are weighting coefficients; This represents the total amount of electricity that the system would not supply due to power outages within a given year. ; This represents the average number of power outage hours for users within the system within a given year. .

Images