A self-healing method for power relay protection faults based on operational characteristic constraints
By identifying fault inheritance dependencies through pairwise verification and operational characteristic constraints, the problem of accurately judging the impact of abnormal protection function migration in existing technologies is solved, and efficient and reliable replacement of relay protection fault self-healing is achieved.
Patent Information
- Application Number
- CN202611425590.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-15
- Publication Date
- 2026-11-10
AI Technical Summary
Existing relay protection fault self-healing methods have difficulty accurately determining the impact of specific bearer dependencies on the original fault when migrating abnormal protection functions as a whole, which may lead to the continued formation or replacement of normal bearer resources.
By acquiring the original bearer relationship and synchronous data of the abnormal period of the anomaly protection function, pairwise verification of a single dependency to be verified is performed to identify fault inheritance dependencies, and effective alternative dependencies are determined based on operational characteristic constraints to reconstruct the bearer relationship.
It improves the pertinence of the fault self-healing process of relay protection and the reliability of the reconstruction results, reduces the interference of multiple dependency changes on the fault judgment results, and reduces unnecessary replacement of normal carrying resources.
Smart Images

Figure CN123073642A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power relay protection technology, and more specifically, to a method for self-healing faults in power relay protection based on operating characteristic constraints. Background Technology
[0002] With the development of smart substations and digital relay protection technology, relay protection functions are typically carried out by multiple links, including sampling, communication, protection calculation, and control output. When an abnormality occurs in the protection device or related secondary links, existing fault self-healing methods usually migrate the abnormal protection function to a backup protection device, backup computing resources, or backup communication link based on device alarms, communication status, or fault diagnosis results, and re-establish the corresponding data input and control output relationships to restore the normal operation of the relay protection function.
[0003] However, anomaly protection functions are often formed by multiple interrelated bearer dependencies. Existing overall migration methods typically change multiple acquisition, transmission, or processing dependencies simultaneously. When the original anomaly disappears or persists after migration, it is difficult to distinguish which specific bearer dependency actually affected the anomaly based on the protection operation results, and it is also difficult to determine which original dependencies can still be used. As a result, in subsequent self-healing processes, dependencies that would cause the original anomaly to recur may be retained, or normal bearer resources that were not affected by the anomaly may be replaced as well.
[0004] The above-disclosed technical solutions have at least the following technical problems: When the existing relay protection fault self-healing performs overall migration of abnormal protection functions, it will change multiple bearer dependencies at the same time, making it difficult to accurately determine the specific dependencies that cause the original abnormality to continue to form based on the protection operation results after migration. Summary of the Invention
[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a power relay protection fault self-healing method based on operational characteristic constraints. This method solves the problem that existing overall migration methods cannot accurately determine the effect of specific carrier dependencies on the continued formation of the original anomaly by performing pairwise verification of the dependencies to be verified with a single variable and identifying fault inheritance dependencies.
[0006] To achieve the above objectives, the present invention provides the following technical solution: The self-healing method for power relay protection faults based on operational characteristic constraints includes the following steps: obtaining the original bearer relationship of the abnormal protection function and the synchronous data of the abnormal period, and determining the dependencies to be verified and operational characteristic constraints involved in the formation of the anomaly judgment criteria; determining candidate alternative dependencies for each dependency to be verified according to the operational characteristic constraints, and constructing paired verification paths that differ only in the corresponding dependencies and have the same bearer relationship in the rest; driving the paired verification paths to run with the synchronous data of the abnormal period under the protection output limitation state, determining the fault inheritance dependency based on the difference in the reproduction of the original anomaly, and determining the corresponding effective alternative dependency from the candidate alternative dependencies; replacing the fault inheritance dependency based on the effective alternative dependency, reconstructing the original bearer relationship, and forming the target self-healing path after verification according to the operational characteristic constraints.
[0007] In a preferred embodiment, determining the unverified dependencies and operational characteristic constraints involved in the formation of the anomaly criterion includes: determining the target protection criterion corresponding to the anomaly state based on the synchronous data of the anomaly period; tracing back along the original bearer relationship to trace the formation and transmission relationship of the information required for the target protection criterion to obtain candidate dependencies; determining the unverified dependencies actually involved in the formation of the target protection criterion from the candidate dependencies based on the synchronous data of the anomaly period; and determining the corresponding operational characteristic constraints based on the role of each unverified dependency in the formation process of the target protection criterion.
[0008] In a preferred embodiment, determining candidate alternative dependencies for each dependency to be verified based on operational characteristic constraints includes: determining replacement boundaries based on the input / output positions of the dependency to be verified in the original bearer relationship; screening candidate dependencies from available protection bearer resources that can form corresponding protection information and do not reuse the original bearer resources within the replacement boundary; determining the boundary correction relationship between the candidate dependencies and the dependency to be verified based on the corresponding output information of the same system state before the anomaly occurred; correcting the candidate dependencies according to the boundary correction relationship; and determining candidate alternative dependencies based on operational characteristic constraints.
[0009] In a preferred embodiment, determining the replacement boundary based on the input and output positions of the dependency to be verified in the original bearer relationship includes: determining the nearest common input position that can provide the same input information upstream along the information transmission relationship of the dependency to be verified; determining the common output position where the output of the dependency to be verified first enters the common protection processing relationship downstream along the information output direction; determining a candidate replacement segment containing the dependency to be verified based on the nearest common input position and the common output position; and determining the nearest common input position and the common output position as the replacement boundary when the replacement of the candidate replacement segment does not change the input source outside the boundary and the downstream protection processing relationship.
[0010] In a preferred embodiment, the step of screening candidate dependencies from available protection bearer resources that can form corresponding protection information and do not reuse the original bearer resources within the replacement boundary includes: determining initial candidate dependencies that can form corresponding protection information from the set of available protection bearer resources based on the target protection information formed by the dependencies to be verified at the common output position; tracing the bearer resources used by each initial candidate dependency to form the target protection information within the replacement boundary and comparing them with the original bearer resources corresponding to the dependencies to be verified; excluding initial candidate dependencies that reuse the original bearer resources and determining the remaining initial candidate dependencies as candidate dependencies.
[0011] In a preferred embodiment, the construction of paired verification paths that correspond only to different dependencies and have the same other bearer relationships includes: using the replacement boundary of the dependency to be verified as a boundary, retaining the original bearer relationship outside the replacement boundary; connecting the dependency to be verified and the corresponding candidate alternative dependency to the two verification paths respectively, and connecting them to the same downstream protection processing position; correcting the output of the candidate alternative dependency according to the boundary correction relationship, so that the two verification paths meet the corresponding input conditions at the replacement boundary; unifying the protection settings, protection logic and initial operating state of the two verification paths to obtain paired verification paths that correspond only to different dependencies.
[0012] In a preferred embodiment, the step of correcting the output of the candidate substitution dependency according to the boundary correction relationship so that the two verification paths meet the corresponding input conditions at the substitution boundary includes: obtaining the amplitude, phase, and time correction parameters of the candidate substitution dependency relative to the dependency to be verified according to the boundary correction relationship; performing amplitude correction, phase compensation, and time alignment on the output information of the candidate substitution dependency based on the correction parameters; performing consistency verification between the corrected output information and the corresponding information of the dependency to be verified at the common output position; and using the corrected output information that meets the allowable deviation range as the input information for the candidate substitution dependency to be connected to the downstream protection processing relationship.
[0013] In a preferred embodiment, determining the corresponding effective alternative dependency from the candidate alternative dependencies includes: forming an original abnormal state sequence based on synchronous data during the abnormal period; driving paired verification paths under the same initial state to obtain protection state sequences corresponding to the two paths; comparing the protection states of the two paths along the original abnormal state sequence to determine the initial state differentiation position; tracking subsequent protection states based on the initial state differentiation position to determine the abnormal differentiation section; performing local dependency switching and re-verifying for the abnormal differentiation section; and determining the fault inheritance dependency and the corresponding effective alternative dependency based on the difference in abnormal continuation after the local switching.
[0014] In a preferred embodiment, the step of comparing the protection states of the two paths along the original abnormal state sequence to determine the first state differentiation position includes: establishing a state sequence corresponding to the original dependent path and the alternative path according to the state order and time position of the original abnormal state sequence; comparing the two state sequences sequentially, and determining the position where the preceding state is consistent but the current state is different for the first time as the candidate differentiation position; continuously comparing the protection states after the candidate differentiation position, and confirming the candidate differentiation position according to the continuous correspondence between the original dependent path and the original abnormal state sequence; and determining the earliest confirmed candidate differentiation position as the first state differentiation position.
[0015] In a preferred embodiment, forming the target self-healing path includes: determining the replacement positions in the original bearer relationship based on fault inheritance dependencies and corresponding effective alternative dependencies, and retaining the remaining available dependencies; connecting the corresponding effective alternative dependencies at each replacement position, reconstructing the connection relationship between the alternative dependencies and upstream and downstream dependencies according to the replacement boundary, and obtaining the reconstructed bearer relationship; correcting the output of the effective alternative dependencies according to the boundary correction relationship, and connecting it to the original downstream protection processing relationship; verifying the reconstructed bearer relationship according to the operating characteristic constraints, and determining the reconstructed bearer relationship that satisfies the operating characteristic constraints as the target self-healing path.
[0016] The technical effects and advantages of the power relay protection fault self-healing method based on operating characteristic constraints of this invention are as follows: This invention acquires the original bearer relationships and synchronous data of abnormal periods for abnormal protection functions. While maintaining consistency in other bearer relationships, it performs pairwise verification on single dependencies to be verified and their candidate alternative dependencies. Combining this with differences in the reproduction of the original anomaly, it determines fault inheritance dependencies and effective alternative dependencies. Then, it replaces and reconstructs bearer relationships only for dependencies verified to cause the original anomaly to continue. Compared to existing self-healing methods that migrate multiple acquisition, transmission, or processing stages, this invention reduces the interference of simultaneous changes in multiple dependencies on fault judgment results, reduces unnecessary replacement of normal bearer resources, and decreases the possibility of the reconstructed protection function continuing to inherit the original anomaly. This improves the targeting of dependency replacement and the reliability of reconstruction results during relay protection fault self-healing. Attached Figure Description
[0017] Figure 1 This is a flowchart illustrating the self-healing fault handling method for power relay protection based on operational characteristic constraints according to the present invention. Figure 2 This is a schematic diagram of the state differentiation and local switching verification of the paired verification path of the present invention; Figure 3 This is a schematic diagram showing the output comparison before and after boundary correction for candidate substitution dependencies in this invention. Detailed Implementation
[0018] 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.
[0019] Example 1, Figure 1 This invention presents a self-healing fault handling method for power relay protection based on operational characteristic constraints, comprising the following steps: S1, obtain the original bearer relationship of the anomaly protection function and the synchronous data of the anomaly period, and determine the unverified dependencies and operational characteristic constraints that participate in the formation of the anomaly judgment criteria; In this embodiment, the original bearing relationship refers to the information bearing relationship actually used by the abnormal protection function to form protection criteria and protection control results before an abnormality occurs. This includes the dependent links that the protection input information actually goes through in the process of forming the protection state or control output after being collected, transmitted and processed by the corresponding information source, as well as the information transmission relationship between each dependent link.
[0020] The dependent components include data source dependencies providing protection input information, transmission dependencies carrying corresponding information transmission, and processing dependencies using the information to form intermediate protection quantities or protection criteria. The original bearer relationship is based on the bearer relationship actually in operation when the anomaly occurs, rather than all available backup relationships in the protection system configuration.
[0021] Synchronized data during abnormal periods refers to a set of data obtained by associating the corresponding operational data of each dependent link in the original bearing relationship with a unified time benchmark within a predetermined time range before and after the time when the abnormal protection function is in an abnormal state. This data is used to reconstruct the changes in each input information and protection status during the abnormality period.
[0022] The abnormal period synchronization data includes electrical quantities, switch status, protection start status, lockout status, protection criterion status, and time information corresponding to each data related to the abnormal protection function; when there are communication quality status or processing status records in the original bearer relationship, the corresponding communication quality status or processing status is also included.
[0023] The dependency to be verified refers to the dependency link located in the original bearer relationship of the anomaly protection function, and whose information is actually involved in the formation of the original anomaly criterion. Therefore, it needs to be verified in pairs to determine whether it causes the continuation of the original anomaly.
[0024] The operational characteristic constraints refer to the operational conditions that need to be maintained after the dependency to be verified is replaced, so that the anomaly protection function can still form an effective protection criterion according to the original protection mechanism. They are used to constrain the determination of subsequent candidate replacement dependencies and the verification of the bearing relationship after reconstruction.
[0025] The determination of the unverified dependencies and runtime characteristic constraints involved in the formation of anomaly criteria includes: Based on the protection status or control result corresponding to the abnormal protection function when an abnormality occurs, the target protection criterion for forming an abnormal status is determined from the synchronous data during the abnormal period. Based on the input information and intermediate protection quantities used by the target protection criterion, the actual formation and transmission process of each input information is traced back from the target protection criterion along the original bearer relationship to obtain candidate dependencies that have a formation relationship with the target protection criterion; Based on the synchronized data during the abnormal period, determine whether each candidate dependency actually carries or processes information involved in the calculation of the target protection criterion during the formation of the target protection criterion, and determine the candidate dependencies corresponding to the information that actually participates as dependencies to be verified. Based on the role of each dependency to be verified in the formation process of the target protection criterion, the content preservation conditions, time correspondence conditions, processing time limit conditions, and connection conditions with the downstream protection processing of the corresponding input information are determined, thus forming the operational characteristic constraints corresponding to each dependency to be verified.
[0026] S2, based on runtime characteristic constraints, determine candidate alternative dependencies for each dependency to be verified, and construct paired verification paths that differ only in their corresponding dependencies and have the same bearing relationships in other aspects; In this embodiment, determining candidate alternative dependencies for each dependency to be verified based on runtime characteristic constraints includes: Based on the position of the dependency to be verified in the original bearer relationship, determine the position where the dependency to be verified receives input information and the position where it provides output information to the downstream protection processing stage, and obtain the replacement boundary of the dependency to be verified. From the available protection bearer resources, identify alternative dependencies that can form the same protection information as the dependency to be verified at the output location, and exclude alternative dependencies that continue to use the original bearer resources of the dependency to be verified within the replacement boundary; Extract normal operating data before the anomaly occurred from the synchronized data during the abnormal period, and obtain the corresponding output information of the dependency to be verified and each alternative dependency under the same system state. Based on the amplitude deviation, phase deviation, or time offset between the corresponding output information of the two, determine the boundary correction relationship of each candidate dependency relative to the dependency to be verified, and correct the output information of the candidate dependencies according to the boundary correction relationship. Candidate alternative dependencies are identified as those whose corrected output information satisfies the runtime characteristic constraints corresponding to the dependency to be verified.
[0027] The step of correcting the output information of the alternative dependencies according to the boundary correction relationship is as follows:
[0028] In the formula, The alternative dependency output after boundary correction; This is the magnitude correction amount for the alternative dependency relative to the dependency to be verified; This is the original alternative dependency output; The time alignment of the alternative dependency relative to the dependency to be verified; To complete the alternative dependency output after time alignment; This is the phase correction amount; The imaginary unit; Used to compensate for inherent phase differences.
[0029] The step of determining the location where the dependency to be verified receives input information and provides output information to the downstream protection processing stage based on its position in the original bearer relationship, thereby obtaining the replacement boundary of the dependency to be verified, includes: Tracing upstream along the information transmission relationship corresponding to the dependency to be verified, we determine the nearest common input location that can provide input information from the same abnormal time period to the dependency to be verified and its alternative dependencies. Tracing downstream along the output direction of the dependency to be verified, the location where the processing result of the dependency to be verified first enters the processing relationship used in conjunction with subsequent protection criteria is determined as the common output location. The carrying segment between the most recent common input position and the common output position, which contains the dependency to be verified and whose internal processing result is no longer re-formed by other dependencies outside the boundary, is determined as the candidate replacement segment of the dependency to be verified. When the candidate replacement segment is replaced, and the input information source upstream of the boundary and the protection processing relationship downstream of the boundary remain unchanged, the most recent common input position and common output position are determined as the replacement boundary to be verified.
[0030] Further, determining alternative dependencies from available protection bearer resources that can form the same protection information as the dependency to be verified at the output location, and excluding alternative dependencies that continue to use the original bearer resources of the dependency to be verified within the replacement boundary, includes: Obtain the protection bearer resources that are currently in normal operation and have not been isolated from faults, and form a set of available protection bearer resources based on the protection objects, bearer information types and their roles in the protection information transmission process of each protection bearer resource; Based on the target protection information provided by the unverified dependency to the downstream protection processing stage at the common output location, select available protection bearer resources from the set of available protection bearer resources that can form the same information type for the same protected object and whose output form can be accessed at the common output location to form initial candidate dependencies. Each initial candidate dependency is traced back to the data source, acquisition channel, communication link and processing node that it actually passed through to form the target protection information within the replacement boundary, and compared with the corresponding bearer resources actually used by the dependency to be verified within the replacement boundary. When the initial alternative dependency still uses any of the original bearer resources of the dependency to be verified within the replacement boundary, and the original bearer resource is located in the actual formation or transmission path of the target protection information, the initial alternative dependency is excluded. The initial alternative dependencies that do not reuse the original bearer resources and can form the target protection information at a common output location are determined as alternative dependencies.
[0031] Furthermore, the step of extracting normal operating data before the anomaly occurred from the synchronization data during the abnormal period, and obtaining the corresponding output information of the dependency to be verified and each alternative dependency under the same system state, includes: The overlapping time period during which both the dependency to be verified and the alternative dependency are in normal operating condition before the anomaly protection function is identified. According to a unified time reference, the data corresponding to the dependencies to be verified and the alternative dependencies in the overlapping time interval are time-matched, and the voltage, current and switching status data that characterize the operating state of the primary system are extracted simultaneously. Based on the voltage, current and switch status data at the corresponding time, a continuous data segment with consistent system topology and electrical operating quantity changes within the allowable range is selected as the comparison segment of the same state. Extract the corresponding output information formed at the common output position of the dependency to be verified and the alternative dependency from the same state comparison segment, and use it as comparison data to determine the inherent output difference between the two.
[0032] In a preferred embodiment, the comparison times between the dependency to be verified and the alternative dependencies are defined as follows: and When the following conditions are met:
[0033] and At that time, it is assumed that the two moments correspond to the same system state.
[0034] In the formula, For a moment and time The difference in system operating status at one time; For a moment The corresponding primary system voltage state vector can be composed of the phase voltages or line voltages related to the target protection function; For a moment The corresponding primary system current state vector; and These are the normalized reference values for voltage and current, respectively; and These represent the weights corresponding to the voltage state difference and the current state difference, respectively. ; The allowable difference threshold under the same state can be statistically determined based on the natural fluctuation range of primary electrical quantities in adjacent sampling sections under normal operating conditions, or pre-calibrated based on protection criteria to determine the allowable range of input changes. and They are time points and The corresponding primary system topology state can be composed of the open / closed states of relevant circuit breakers, disconnectors, and tie switches.
[0035] In this embodiment, the boundary correction relationship is used to eliminate the inherent output differences between the dependency to be verified and the alternative dependency under normal operating conditions caused by different acquisition, transmission, or processing links. Specifically, based on the amplitude ratio, phase difference, and time offset of the corresponding output information of the two dependencies within the same state comparison segment, the amplitude correction amount, phase correction amount, and time alignment amount of the alternative dependency relative to the dependency to be verified can be determined. The output information of the alternative dependency is then corrected using the correction amounts, so that the two paths have consistent output conditions or are within the allowable deviation range at the replacement boundary in subsequent pairwise verification.
[0036] The step of determining candidate alternative dependencies as those that satisfy the operational characteristic constraints corresponding to the dependencies to be verified, after correcting the output information, means using the operational characteristic constraints determined in S1 to re-verify the corrected candidate dependencies. When a candidate dependency can provide the information required for the target protection criterion, the corrected time-corresponding error is within the allowable range, the information formation and transmission time does not exceed the corresponding processing time limit, and it can maintain a correct connection with the original protection processing relationship downstream of the replacement boundary, then the candidate dependency is determined as a candidate alternative dependency.
[0037] In this embodiment, the construction corresponds only to paired verification paths with different dependencies but consistent other bearer relationships, including: Using the replacement boundary corresponding to the dependency to be verified as the dividing line, the information source relationship upstream of the replacement boundary and the protection processing relationship downstream of the replacement boundary in the original bearing relationship are preserved; The dependency to be verified is retained in one verification path, and the corresponding candidate alternative dependency is connected in another verification path, and the outputs of the dependency to be verified and the candidate alternative dependency are both connected to the same downstream protection processing location. According to the boundary correction relationship between the dependency to be verified and the candidate substitution dependency, the output of the candidate substitution dependency at the substitution boundary is corrected so that the two verification paths meet the same input conditions before entering the downstream protection processing relationship. By setting the two verification paths to the same protection settings, protection logic, and initial operating state, and limiting their actual protection control output, a pair of verification paths are obtained where the differences only exist between the dependency to be verified and the candidate alternative dependency.
[0038] Furthermore, the step of correcting the output of the candidate substitution dependency at the substitution boundary according to the boundary correction relationship corresponding to the dependency to be verified and the candidate substitution dependency, so that the two verification paths meet the same input conditions before entering the downstream protection processing relationship, includes: Read the magnitude correction, phase correction, and time alignment of the candidate alternative dependency relative to the dependency to be verified; Based on the amplitude correction amount and the phase correction amount, the amplitude conversion and phase compensation are performed on the corresponding electrical quantities formed at the common output position by the candidate substitution dependence to obtain the amplitude and phase correction results. Based on the time alignment amount, the time position of the amplitude and phase correction result is adjusted so that it corresponds to the output information of the dependency to be verified according to a unified time reference. The time-aligned output information is checked for consistency with the information type, data time, and value range of the dependency to be verified at the common output position. The correction result that meets the allowable deviation range is used as the input information for the downstream protection processing relationship of the candidate alternative dependency.
[0039] S3, under the condition of limited protection output, drives the pair verification path to run with synchronous data during abnormal periods, determines the fault inheritance dependency based on the difference in reproduction of the original anomaly, and determines the corresponding effective alternative dependency from the candidate alternative dependencies; In this embodiment, the protection output restricted state refers to the operating state that maintains the normal operation of the protection calculation, logic judgment, timing state and protection control result formation process in the paired verification path, while blocking the transmission of the formed protection control result to the actual circuit breaker or other primary execution equipment.
[0040] The protection output is restricted because the synchronous data during abnormal periods contains the actual electrical quantities and state changes that led to the original abnormality. If the actual protection output is directly driven by the verification path, it may trigger the circuit breaker again and change the current primary system operating state. However, if the protection calculation or protection logic is directly blocked, the true evolution results of each protection state during the abnormality criterion formation process cannot be obtained. Therefore, this embodiment only restricts the transmission of protection control results to the primary execution equipment, while retaining the internal calculation and state update process of the protection, so that the abnormal period data can drive the paired verification paths to complete the abnormality replay without affecting the current power grid operation.
[0041] The fault inheritance dependency refers to a dependency that, under the same abnormal time period input conditions, allows the evolution process of the protection state corresponding to the original abnormality to continue when the dependency is retained, while the abnormal state evolution process is interrupted at the position related to the dependency after the corresponding candidate replacement dependency is adopted, and further verification confirms that the abnormal interruption is caused by the replacement of the dependency.
[0042] The method of driving paired verification paths with synchronous data from abnormal periods under limited protection output, and determining fault inheritance dependencies based on the differences in the reproduction of the original anomaly, includes: Based on the synchronous data during the abnormal period and the original abnormal formation process, the protection status changes and their time positions corresponding to the original abnormality are extracted according to the formation order of protection input, intermediate criterion status, start or block status and protection control results, forming the original abnormal status sequence. Set the paired verification paths to the same initial protection state corresponding to the original anomaly before it occurred, and drive the original dependent path and the alternative path to run synchronously according to the original time order of the synchronous data during the anomaly period. Record the protection state changes in the two paths corresponding to the original anomaly state sequence. Compare the protection status of the two paths along the formation order of the original abnormal state sequence. Under the condition that the preceding protection status of the two paths is consistent, determine the time position and corresponding protection status of the first occurrence of different protection statuses, and obtain the first state differentiation position. Continue tracking the subsequent protection states of the two paths from the initial state differentiation position. When the original dependent path continues to form subsequent abnormal states in the same state order as the original abnormality, while the alternative path no longer forms corresponding subsequent abnormal states from the initial state differentiation position, the abnormal differentiation segment corresponding to the initial state differentiation position is determined. For anomalous differentiation segments, the pairwise validation paths are redriven. Before the anomalous differentiation segments, both paths use the dependency to be validated, and the alternative paths are switched to candidate alternative dependencies only within the anomalous differentiation segments. When the alternative path after a local switch no longer forms the subsequent state corresponding to the original anomaly from the abnormal differentiation section, while the original dependency path still forms the subsequent abnormal state according to the original abnormal state sequence, the dependency to be verified is determined as a fault inheritance dependency, and the corresponding candidate alternative dependency is determined as a valid alternative dependency.
[0043] The process involves setting the paired verification paths to the same initial protection state as before the original anomaly occurred, and synchronously driving the original dependent path and the alternative path according to the original time sequence of the synchronized data during the anomaly period. The changes in protection state corresponding to the original anomaly state sequence in each of the two paths are recorded, including: From the synchronous data of the abnormal period, determine the stable operating section before the start of the original abnormal state sequence, and extract the protection start state, blocking state, timing state and intermediate criterion state corresponding to the abnormal protection function at the end of the stable operating section to form the protection initial state. The initial protection state is loaded into the original dependent path and the alternative path respectively, and the two paths adopt the same protection settings, protection logic and state update time base. According to the original timestamp of each input information in the synchronous data during the abnormal period, the corresponding input information is provided to the original dependency path and the alternative path in sequence. The original dependency path forms the corresponding input through the dependency to be verified, and the alternative path forms the corresponding input after boundary correction through the candidate alternative dependency. The protection calculation process for the two paths is synchronously advanced using a unified time base. At the time position corresponding to each protection state in the original abnormal state sequence, the protection start-up state, blocking state, intermediate criterion state and protection control result of the two paths are recorded respectively, so as to obtain the protection state change sequence corresponding to the two paths.
[0044] Further, the protection states of the two paths are compared along the formation order of the original abnormal state sequence. Under the condition that the preceding protection states of the two paths remain consistent, the time position and corresponding protection state of the first occurrence of different protection states are determined, thus obtaining the first state differentiation position, including: According to the formation order and corresponding time position of each protection state in the original abnormal state sequence, the protection states generated by the original dependent path and the alternative path are respectively matched with the original abnormal state sequence to obtain the state correspondence sequence of the two paths. Starting from the initial state of the original abnormal state sequence, the corresponding protection states in the two paths are compared sequentially. The position where the previous corresponding state is consistent but the current corresponding state is inconsistent for the first time is determined as the candidate differentiation position. A predetermined number of subsequent protection states are continuously compared from the candidate differentiation position. When the original dependent path continues to correspond to the original abnormal state sequence, while the alternative path continues to deviate from the original abnormal state sequence, it is confirmed that the state difference corresponding to the candidate differentiation position is not an instantaneous state fluctuation. The earliest time position and protection status corresponding to the candidate differentiation position confirmed by continuous comparison are determined as the first state differentiation position.
[0045] The specific location of the initial state differentiation is as follows:
[0046] In the formula, This refers to the time position of the first state differentiation; The difference in protection status between the two verification paths at time t; The time interval for comparing protection status; The length of consecutive observations following the candidate differentiation location; The minimum number of valid states required to confirm continued differentiation; This is an indicator function; it takes the value 1 if the condition is true, and 0 otherwise. This is the threshold for determining state differentiation.
[0047] When no subsequent abnormal states are formed from the initial state differentiation position, the abnormal differentiation segment corresponding to the initial state differentiation position is determined, including: Starting from the initial state differentiation position, the subsequent protection states corresponding to the original dependent path and alternative path are obtained along the original abnormal state sequence. When the original dependent path continuously forms subsequent states corresponding to the original abnormal state sequence, while the alternative path does not form a corresponding state, the time range corresponding to the subsequent state is included in the state differentiation range. Continue to extend the state differentiation range along the original abnormal state sequence until the protection control result corresponding to the original abnormality is formed, or the two paths re-enter the same protection state and no longer show state differences corresponding to the original abnormality. The continuous time range from the initial state differentiation position to the termination position of the state differentiation range is defined as the abnormal differentiation segment.
[0048] The method of re-driving pairwise validation paths for anomalous differentiation segments, maintaining both paths using the dependency to be validated before the anomalous differentiation segment, and switching alternative paths to candidate alternative dependencies only within the anomalous differentiation segment, includes: The two verification paths are restored to the same initial protection state before the original anomaly occurred, and the drive is resynchronized according to the synchronous data during the anomaly period. Before reaching the abnormal differentiation section, both verification paths are made to form corresponding protection inputs through the dependencies to be verified, so as to keep the preceding protection state of the two paths consistent. When the starting position of the abnormal differentiation segment is reached, the dependency to be verified in the alternative path is switched to the corresponding candidate alternative dependency, and the time correspondence of the input information before and after the switch is maintained according to the boundary correction relationship. Within the abnormally differentiated segment, the original dependency path is kept to be verified, and the alternative path is kept to be a candidate alternative dependency. The subsequent protection status of the two paths is recorded respectively.
[0049] The step of determining the corresponding candidate alternative dependencies as valid alternative dependencies includes: When, after a local switch, the alternative path no longer forms subsequent abnormal states corresponding to the original abnormal state sequence from the abnormal differentiation segment, it is determined that the candidate alternative dependency can block the continued formation of the original abnormality. Continue tracking the alternative path to the end of the original abnormal state sequence, and determine whether the alternative path generates any new start-up, lockout, criterion action or protection control state that does not conform to the operating characteristic constraints during the period. When a candidate alternative dependency can prevent the original exception from continuing to form and does not introduce a new exception protection state, the candidate alternative dependency is determined as a valid alternative dependency corresponding to the fault inheritance dependency.
[0050] Figure 2 This illustrates the protection state change process of paired verification paths driven by synchronous data during anomaly periods. Before the anomaly occurs, the protection states of the original dependent path and the alternative path are basically consistent. During the anomaly evolution, state differences appear between the two paths starting from the first state differentiation point, forming corresponding anomaly differentiation segments. After a local dependency switch is implemented on the alternative path within the anomaly differentiation segment, the original dependent path continues to evolve according to the original anomaly state, while the alternative path no longer forms corresponding subsequent anomaly states. Therefore, while maintaining the consistency of the preceding protection state, it is possible to verify whether the continuous formation of the original anomaly is related to the dependency to be verified, providing a basis for determining fault inheritance dependencies and effective alternative dependencies.
[0051] Figure 3 The diagram illustrates the correspondence between the output of the dependency to be verified, the output of the candidate substitution dependency before correction, and the output of the candidate substitution dependency after correction. Within the same state comparison section, the output of the candidate substitution dependency before correction has inherent deviations in amplitude, phase, or time position relative to the dependency to be verified. After amplitude correction, phase compensation, and time alignment according to the boundary correction relationship, the output of the candidate substitution dependency basically corresponds to the reference output of the dependency to be verified. Through this boundary correction, the inherent differences of different acquisition, transmission, or processing links can be reduced on the subsequent pairwise verification results, enabling the two verification paths to have comparable input conditions at the substitution boundary.
[0052] S4 replaces faulty inherited dependencies with effective alternative dependencies, reconstructs the original bearing relationship, and forms the target self-healing path after verification based on runtime characteristic constraints.
[0053] In this embodiment, the reconstruction of the original bearer relationship means that, while keeping the remaining available dependencies and their connection relationships in the original bearer relationship unchanged except for the fault inheritance dependency, the fault inheritance dependency is replaced with the corresponding effective alternative dependency determined by S3, and the information transmission relationship between the effective alternative dependency and the upstream and downstream dependencies of the boundary is re-established according to the replacement boundary.
[0054] The target self-healing path refers to the reconstructed bearing relationship that, after completing the fault inheritance dependency replacement, can obtain the required input information, form protection criteria and output corresponding protection results according to the original protection mechanism of the abnormal protection function, and satisfies the operational characteristic constraints.
[0055] The process of replacing faulty inherited dependencies with effective alternative dependencies, reconstructing the original bearer relationship, and forming the target self-healing path after verification based on runtime characteristic constraints includes: Based on each fault inheritance dependency and its corresponding valid alternative dependency, determine the dependency positions that need to be replaced in the original bearing relationship, and retain the remaining available dependencies that have not been determined as fault inheritance dependencies. At each dependent location, the corresponding effective alternative dependency is connected. According to the replacement boundary, the connection between the effective alternative dependency and the upstream input relationship and the downstream protection processing relationship of the boundary is established to obtain the reconstructed bearing relationship. The output information formed at the replacement boundary by the effective substitution dependency is corrected according to the corresponding boundary correction relationship, and the corrected output information is connected to the downstream protection processing relationship retained in the original bearer relationship. Based on operational characteristic constraints, the protection input information, the time correspondence between each input information, the information formation and transmission time, and the connection status with downstream protection processing are verified in the reconstructed bearing relationship. When the reconstructed bearing relationship satisfies the operational characteristic constraints, the reconstructed bearing relationship is determined as the target self-healing path.
[0056] Furthermore, the step of accessing corresponding effective alternative dependencies at each dependent location, and establishing connections between effective alternative dependencies and upstream input relationships and downstream protection processing relationships according to the replacement boundaries to obtain the reconstructed bearer relationship includes: Based on the replacement boundary corresponding to the fault inheritance dependency, determine the input interface upstream of the boundary that provides input information to the fault inheritance dependency, and the processing interface downstream of the boundary that receives output information from the fault inheritance dependency; Disconnect the original connection between the fault inheritance dependency and the input interface and processing interface, while keeping the remaining bearer relationships outside the replacement boundary unchanged; Connect the input terminal corresponding to the effective alternative dependency to the input interface so that it receives the upstream input information corresponding to the fault inheritance dependency, and connect the output terminal of the effective alternative dependency to the processing interface so that its output continues to enter the original downstream protection processing relationship; Based on the transmission order and correspondence of each information item in the original bearer relationship, the information flow after access is verified. When the effective replacement dependency can continuously receive the upstream input of the boundary and output the corresponding protection information to the downstream of the boundary, a reconstructed bearer relationship is formed.
[0057] The verification of the reconstructed bearer relationship based on operational characteristic constraints refers to determining whether the reconstructed bearer relationship can fully provide the input information required for the formation of the target protection criterion, whether the time correspondence between each input information is within the allowable range, whether the time required for the input information to form the corresponding protection result after collection, transmission and protection processing exceeds the allowable processing time limit, and whether the output of the effective alternative dependency can be connected to the corresponding downstream protection processing relationship according to the original bearer relationship.
[0058] When the reconstructed bearer relationship can provide all the input information required for the target protection criterion, the time deviation between each input information is within the allowable range, the information formation and transmission time does not exceed the allowable processing time limit, and the connection between the effective alternative dependency and the downstream protection processing relationship is correct, the reconstructed bearer relationship is determined to meet the operating characteristic constraints and is identified as the target self-healing path; if any condition is not met, the corresponding reconstructed bearer relationship is not used as the target self-healing path.
[0059] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.
[0060] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.
[0061] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0062] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.
[0063] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.
[0064] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for self-healing fault handling in power relay protection based on operational characteristic constraints, characterized in that, Includes the following steps: Obtain the original bearer relationship and synchronous data of abnormal periods for the anomaly protection function, and determine the unverified dependencies and operational characteristic constraints that participate in the formation of anomaly judgment criteria; Based on runtime characteristic constraints, candidate alternative dependencies are determined for each dependency to be verified, and paired verification paths are constructed that differ only in the corresponding dependencies and have the same bearing relationships in other aspects. Under the condition of limited protection output, the pairwise verification path is driven by the synchronous data of the abnormal period. The fault inheritance dependency is determined according to the difference in the reproduction of the original anomaly, and the corresponding effective alternative dependency is determined from the candidate alternative dependencies. Based on the effective replacement of faulty inherited dependencies, the original bearing relationship is reconstructed, and the target self-healing path is formed after verification according to the constraints of the running characteristics.
2. The power relay protection fault self-healing method based on operating characteristic constraints according to claim 1, characterized in that, The determination of the unverified dependencies and runtime characteristic constraints involved in the formation of anomaly criteria includes: Determine the target protection criteria corresponding to the abnormal state based on the synchronized data during the abnormal period; By tracing the formation and transmission relationships of the information required for the target protection criteria along the original bearing relationship, candidate dependencies are obtained; Based on the synchronous data during abnormal periods, the dependencies to be verified are determined from the candidate dependencies that actually participate in the target protection criteria. Based on the role of each dependency to be verified in the formation process of the target protection criterion, the corresponding operational characteristic constraints are determined.
3. The power relay protection fault self-healing method based on operating characteristic constraints according to claim 2, characterized in that, The step of determining candidate alternative dependencies for each dependency to be verified based on runtime characteristic constraints includes: The replacement boundary is determined based on the input and output positions of the dependency to be verified in the original bearer relationship; Select alternative dependencies from available protection bearer resources that can form corresponding protection information and do not reuse or replace the original bearer resources within the boundary; Based on the corresponding output information of the same system state before the anomaly occurred, the boundary correction relationship between the alternative dependency and the dependency to be verified is determined. The alternative dependencies are corrected according to the boundary correction relationship, and candidate alternative dependencies are determined based on the runtime characteristic constraints.
4. The power relay protection fault self-healing method based on operating characteristic constraints according to claim 3, characterized in that, The step of determining the replacement boundary based on the input / output positions of the dependency to be verified in the original bearer relationship includes: Determine the nearest common input location that can provide the same input information by tracing the information transmission relationship of the dependency to be verified upstream; Determine the common output position of the dependent output to be verified for the first time in the common protection processing relationship by moving downstream along the information output direction; Candidate replacement segments containing dependencies to be verified are determined based on the most recent common input and common output locations; When the replacement segment is replaced without changing the input source outside the boundary and the downstream protection processing relationship, the most recent common input position and common output position are determined as the replacement boundary.
5. The power relay protection fault self-healing method based on operating characteristic constraints according to claim 4, characterized in that, The step of selecting alternative dependencies from available protected bearer resources that can form corresponding protection information and do not reuse or replace the original bearer resources within the boundary includes: Based on the target protection information formed by the dependencies to be verified at the common output location, initial candidate dependencies that can form corresponding protection information are determined from the set of available protection bearer resources; Trace the bearer resources used by each initial candidate dependency to form the target protection information within the replacement boundary, and compare them with the original bearer resources corresponding to the dependency to be verified. Exclude the initial alternative dependencies that reuse the original host resources, and determine the remaining initial alternative dependencies as alternative dependencies.
6. The power relay protection fault self-healing method based on operating characteristic constraints according to claim 5, characterized in that, The construction corresponds only to paired verification paths with different dependencies but consistent other bearer relationships, including: Using the replacement boundary of the dependency to be verified as the dividing line, the original bearing relationship outside the replacement boundary is preserved; The dependency to be verified and the corresponding candidate alternative dependency are connected to the same downstream protection processing location in the two verification paths respectively. The output of the candidate substitution dependency is corrected according to the boundary correction relationship so that the two verification paths satisfy the corresponding input conditions at the substitution boundary. By unifying the protection settings, protection logic, and initial operating status of the two verification paths, we obtain paired verification paths that correspond only to different dependencies.
7. The power relay protection fault self-healing method based on operating characteristic constraints according to claim 6, characterized in that, The step of correcting the output of the candidate substitution dependency based on the boundary correction relationship, so that the two verification paths satisfy the corresponding input conditions at the substitution boundary, includes: The magnitude, phase, and time correction parameters of the candidate alternative dependency relative to the dependency to be verified are obtained based on the boundary correction relationship. Based on the aforementioned correction parameters, amplitude correction, phase compensation, and time alignment are performed on the output information of the candidate substitution dependency. Perform a consistency check between the corrected output information and the corresponding information of the dependency to be verified at the common output position; The corrected output information that meets the allowable deviation range is used as the input information for the candidate alternative access downstream protection processing relationship.
8. The power relay protection fault self-healing method based on operating characteristic constraints according to claim 7, characterized in that, The step of determining the corresponding effective alternative dependency from the candidate alternative dependencies includes: The original abnormal state sequence is formed based on the synchronized data during the abnormal period; Drive the paired verification paths under the same initial state to obtain the protection state sequence corresponding to the two paths; The protection status of the two paths is compared along the original abnormal state sequence to determine the location of the first state differentiation; Based on the initial state differentiation location, subsequent protection status is tracked to determine abnormal differentiation segments; Implement local dependency switching and re-validate for abnormally differentiated segments; Based on the differences in abnormal continuation after partial switching, determine the fault inheritance dependencies and the corresponding effective alternative dependencies.
9. The power relay protection fault self-healing method based on operating characteristic constraints according to claim 8, characterized in that, The step of comparing the protection states of the two paths along the original abnormal state sequence to determine the initial state differentiation position includes: Establish the state sequences corresponding to the original dependent paths and alternative paths according to the state order and time position of the original abnormal state sequence; By comparing two state sequences in turn, the position where the previous state is the same but the current state is different for the first time is determined as the candidate differentiation position; The protection status after continuously comparing candidate differentiation positions is used to confirm candidate differentiation positions based on the continuous correspondence between the original dependency path and the original abnormal state sequence. The earliest confirmed candidate differentiation position is determined as the first state differentiation position.
10. The power relay protection fault self-healing method based on operating characteristic constraints according to claim 9, characterized in that, The formation of the target self-healing path includes: The replacement position in the original carrying relationship is determined based on the fault inheritance dependency and the corresponding valid replacement dependency, while the remaining usable dependencies are retained; At each replacement location, the corresponding valid alternative dependency is accessed, and the connection relationship between it and its upstream and downstream dependencies is reconstructed according to the replacement boundary to obtain the reconstructed bearing relationship; The output of the effective substitution dependency is corrected according to the boundary correction relationship and connected to the original downstream protection processing relationship; The reconstructed bearing relationship is verified based on the operational characteristic constraints, and the reconstructed bearing relationship that satisfies the operational characteristic constraints is determined as the target self-healing path.
