An electrical automation power distribution maintenance process management method and system

Abstract

The application relates to the technical field of power distribution network operation and maintenance management, and discloses an electrical automation power distribution maintenance process management method and system. The method analyzes an instruction sequence of a to-be-executed maintenance task, identifies a topological state switching instruction causing a numerical jump of an adjacency matrix and a topological steady-state operation instruction maintaining the constant of the matrix by comparing an expected state of a controlled device with a network frame connection attribute, extracts a state switching instruction contained in each task to calculate a conflict coupling factor, sets a job starting time offset for mutually exclusive tasks, drives a network frame structure change action to avoid a safety protection period, and realizes accurate peak-shaving during transient switching and parallel scheduling in a steady-state interval of the maintenance process, solves a job conflict problem caused by the disconnection of management logic and a physical topology, and improves the scheduling certainty of the power distribution maintenance process and the system throughput performance within a safety boundary.

Description

Technical Field

[0001] This invention relates to an electrical automation power distribution maintenance process management method and system, belonging to the field of power distribution network operation and maintenance management technology. Background Technology

[0002] Currently, the operation and maintenance of electrical automation distribution networks typically adopts a planned management model. Maintenance work orders are generated by extracting equipment status information, and tasks are arranged according to the manpower and time windows of the work teams to ensure the reliability of power supply. With the large-scale grid connection of distributed energy, the distribution network structure is showing a highly dynamic trend. The switching actions involved in maintenance work are no longer just isolated power outage events, but logical breakpoints that change the impedance distribution of branches and affect the power flow direction. Existing process management usually treats maintenance tasks as independent management units, making it difficult to perceive the topological interference caused by the work actions. When maintenance tasks for multiple areas are issued in parallel, bypass overloads are often caused by uncontrolled power flow transfer at the physical execution level.

[0003] To alleviate the aforementioned conflicts, adopting a linear increase in monitoring point density or calculating the entire network power flow not only generates significant communication delays and computational overhead, but also creates coordination obstacles due to phase asynchrony between the maintenance execution end and the management decision-making end. This leads to a rigid constraint where the maintenance process struggles to balance workload and safety boundaries. For example, Chinese invention patent application CN105069587A discloses a relay protection verification device based on mobile internet, which uses a handheld terminal to standardize the distribution of work instructions and automatically generate test reports, improving the level of information management in on-site operations. However, in actual power distribution maintenance conditions, this solution belongs to a closed-loop feedback mode of instruction execution. Management stops at the standardization of business processes and does not delve into the constraints of physical network nodes. When faced with multiple concurrent maintenance tasks, this solution cannot identify the power flow reconstruction path triggered by the instantaneous switching action, making it difficult to calculate the conflict coupling factor between tasks. This results in insufficient determinism in process orchestration under dynamic physical constraints, making it impossible to achieve parallel scheduling of maintenance resources while ensuring safety boundaries.

[0004] Therefore, how to construct a spatiotemporal exclusion mechanism between maintenance tasks based on the adjacency relationship of the network topology, and realize the orderly arrangement of asynchronous maintenance tasks under dynamic physical constraints, has become the technical problem to be solved by this invention. Summary of the Invention

[0005] To address the problems mentioned in the background art, the technical solution of the present invention is as follows: A method for managing electrical automation power distribution maintenance processes, comprising the following steps:

[0006] Step S101: Obtain the initial topology distribution data of the power distribution network and map it into an adjacency matrix that reflects the connection relationship of power distribution equipment;

[0007] Step S102: Parse the work instruction text associated with the maintenance task to be performed, extract the operation instruction sequence of each maintenance task, and identify the expected state of the maintained equipment corresponding to each operation instruction and the physical object identifier corresponding to the operation instruction.

[0008] Step S103: Compare the expected state of the maintained equipment with the initial connection attributes of the corresponding physical object identifier in the adjacency matrix. Define the operation instruction that causes the matrix element in the adjacency matrix to change its value as the topology state switching instruction, and define the operation instruction that keeps the adjacency matrix element constant as the topology steady-state operation instruction.

[0009] Step S104: Determine the priority based on the task type of each maintenance task, extract the topology state switching instructions contained in different maintenance tasks, and calculate the conflict coupling factor between different maintenance tasks through the adjacency matrix.

[0010] Step S105: When the conflict coupling factor exceeds the preset mutual exclusion threshold, determine the execution start time of the higher priority maintenance task and calculate the job start time offset of the lower priority maintenance task so as to drive the job start time of the lower priority maintenance task to avoid the preset safety protection period in which the execution start time is located.

[0011] Step S106: When different maintenance tasks are all in the corresponding topological steady-state operation instruction execution stage, cancel the exclusion constraints between different maintenance tasks and perform parallel scheduling.

[0012] Preferably, the method further includes the following steps: Step S201, obtaining real-time fault alarm information of the distribution network, and mapping the controlled equipment corresponding to the real-time fault alarm information to a risk-occupying task with a preset highest conflict coupling factor; Step S202, searching the adjacency matrix for a set of nodes that have electrical logical association with the risk-occupying task, and marking the maintenance tasks located in the node set as disturbed maintenance tasks; Step S203, performing timing decoupling on the disturbed maintenance tasks, delaying the triggering time of the topology state switching instruction of the disturbed maintenance tasks until the fault clearing signal corresponding to the risk-occupying task is obtained, and simultaneously driving other maintenance tasks located outside the node set to execute according to the original timing.

[0013] Preferably, the method further includes the following steps: Step S301, statistically analyzing the actual execution time and standard execution time of historical maintenance tasks, and quantifying the execution deviation dispersion of the execution process relative to the preset plan; Step S302, calculating the deviation correction coefficient based on the execution deviation dispersion. Deviation correction factor The calculation rules are as follows: ,in, As a preset adjustment factor, The actual execution time of the historical preservation task. The standard execution time for historical maintenance tasks; step S303, using the deviation correction coefficient. The length of the preset safety protection period can be adjusted in real time to correct process timing drift caused by uncertainties in the on-site working environment.

[0014] Preferably, step S106 specifically includes: step S401, identifying the resource occupancy list of each maintenance task during the topology steady-state operation instruction execution phase, the resource occupancy list including human resource identifiers and tool identifiers; step S402, determining whether the resource occupancy lists of different maintenance tasks overlap on the spatial access path; step S403, if there is no overlap in the spatial access path, releasing the mutual exclusion logic of different maintenance tasks during the topology steady-state operation instruction execution phase.

[0015] Preferably, the calculation of the conflict coupling factor between different maintenance tasks in step S104 specifically includes: step S501, based on the graph theory search algorithm, calculating the power flow reconfiguration path triggered by the topology state switching command contained in each maintenance task in the adjacency matrix; step S502, extracting the common node elements of the two power flow reconfiguration paths, and determining the conflict coupling factor according to the weight level of the common node elements in the adjacency matrix, and the priority is also determined based on the estimated impact range of the maintenance task and the power consumption priority of the distribution area to which it belongs.

[0016] Preferably, the step S102 of parsing the work instruction text associated with the maintenance task to be performed specifically includes: step S601, performing semantic parsing on the work instruction text to identify the operation predicate and physical object identifier of each operation instruction; step S602, mapping the current state of the device corresponding to the physical object identifier according to the operation predicate to generate the expected state of the device to be maintained.

[0017] Preferably, step S103 specifically includes: step S701, locating the matrix coordinates corresponding to the physical object identifier in the adjacency matrix; step S702, performing a logical XOR operation between the expected state of the maintained equipment and the initial connection value at the matrix coordinates to determine whether the operation instruction belongs to the action that triggers a change in the grid structure.

[0018] Preferably, step S105 specifically includes: step S801, obtaining the estimated execution step length of a high-priority maintenance task when executing a topology state switching command; step S802, summing the estimated execution step length with the length of a preset safety protection period to obtain the offset of the start time of the operation.

[0019] Preferably, the method further includes the following steps: Step S901, after each maintenance task is completed, obtain the real-time topology connection information of the distribution network; Step S902, update the adjacency matrix online based on the real-time topology connection information, as the reference topology for scheduling subsequent maintenance tasks.

[0020] An electrical automation power distribution maintenance process management system, comprising:

[0021] The data acquisition module is used to acquire the initial topology distribution data of the power distribution network and map it into an adjacency matrix that reflects the connection relationship of power distribution equipment.

[0022] The text parsing module is used to parse the work instruction text associated with the maintenance task to be performed, extract the operation instruction sequence of each maintenance task, and identify the expected state of the maintained equipment corresponding to each operation instruction and the physical object identifier corresponding to the operation instruction.

[0023] The instruction classification module is used to compare the expected state of the maintained equipment with the initial connection attributes of the corresponding physical object identifier in the adjacency matrix. Operation instructions that cause a numerical jump in the matrix elements in the adjacency matrix are identified as topology state switching instructions, and operation instructions that maintain the constant value of the adjacency matrix elements are identified as topology steady-state operation instructions.

[0024] The conflict analysis module is used to determine the priority of each maintenance task based on its task type, extract the topology state switching instructions contained in different maintenance tasks, and calculate the conflict coupling factor between different maintenance tasks through the adjacency matrix.

[0025] The timing orchestration module is used to determine the execution start time of higher priority maintenance tasks and calculate the job start time offset of lower priority maintenance tasks when the conflict coupling factor exceeds the preset mutual exclusion threshold, so as to drive the job start time of lower priority maintenance tasks to avoid the preset safety protection period in which the execution start time is located.

[0026] The parallel scheduling module is used to cancel the exclusion constraints between different maintenance tasks and perform scheduling control when different maintenance tasks are all in the corresponding topological steady-state operation instruction execution stage.

[0027] Compared with the prior art, the beneficial effects of the present invention are:

[0028] 1. In the electrical automation power distribution maintenance process, by transforming the physical topology adjacency attributes of the power distribution network into logical constraint resources for maintenance process management, an asynchronous process orchestration mechanism based on topology correlation is established. The system determines the topology exclusion weight by calculating the logical hop count of maintenance task nodes in the adjacency matrix, and performs time-series shifting on task sets with high conflict probability accordingly. This mechanism eliminates the risk of power flow reconfiguration overload caused by maintenance actions from a logical perspective, enabling the process management system to perceive the physical safety boundary of the power grid, solving the problem of unplanned maintenance interruption caused by the disconnect between management logic and physical topology in the traditional scheduling mode, and improving the determinism of power distribution maintenance operations in the spatiotemporal dimensions.

[0029] 2. The system utilizes a fault topology shadowing mechanism to improve the operational continuity of the distribution network under extreme disturbance conditions. By instantiating the physical node corresponding to the real-time fault signal into a virtual task node with the highest exclusion weight, and quickly determining the electrical logic coverage in the adjacency matrix, the system achieves precise separation of risk-intersecting tasks. This local process decoupling method ensures that other maintenance tasks located outside the topology shadow domain can continue to operate in their original order, changing the extensive handling mode of global process circuit breaking in the face of sudden faults in the existing technology, and improving the anti-disturbance stability of the maintenance management system.

[0030] 3. The solution introduces a process execution damping adjustment mechanism to offset execution friction caused by the physical environment. By extracting the execution entropy parameter of completed tasks, the deviation dispersion of the actual execution time relative to the standard time is quantified. The damping coefficient generated by the linear mapping function is used to implement weighted correction of the start interval of subsequent tasks to be executed. This closed-loop feedback mechanism effectively digests the cumulative phase misalignment caused by uncertainties in the field environment, prevents the safety time window from physically overlapping due to time drift, and ensures that the topology exclusion constraint maintains logical fidelity throughout the entire long-cycle maintenance sequence. Attached Figure Description

[0031] Figure 1 This is a flowchart of the power distribution maintenance process management method for topology state comparison according to the present invention;

[0032] Figure 2 This is a multi-level logical architecture diagram of the electrical automation power distribution maintenance process management system of the present invention.

[0033] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0034] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0035] An electrical automation power distribution maintenance process management method includes the following steps:

[0036] Step S101: Obtain the initial topology distribution data of the power distribution network and map it into an adjacency matrix that reflects the connection relationship of power distribution equipment;

[0037] Step S102: Parse the work instruction text associated with the maintenance task to be performed, extract the operation instruction sequence of each maintenance task, and identify the expected state of the maintained equipment corresponding to each operation instruction and the physical object identifier corresponding to the operation instruction.

[0038] Step S103: Compare the expected state of the maintained equipment with the initial connection attributes of the corresponding physical object identifier in the adjacency matrix. Define the operation instruction that causes the matrix element in the adjacency matrix to change its value as the topology state switching instruction, and define the operation instruction that keeps the adjacency matrix element constant as the topology steady-state operation instruction.

[0039] Step S104: Determine the priority based on the task type of each maintenance task, extract the topology state switching instructions contained in different maintenance tasks, and calculate the conflict coupling factor between different maintenance tasks through the adjacency matrix.

[0040] Step S105: When the conflict coupling factor exceeds the preset mutual exclusion threshold, determine the execution start time of the higher priority maintenance task and calculate the job start time offset of the lower priority maintenance task so as to drive the job start time of the lower priority maintenance task to avoid the preset safety protection period in which the execution start time is located.

[0041] Step S106: When different maintenance tasks are all in the corresponding topological steady-state operation instruction execution stage, cancel the exclusion constraints between different maintenance tasks and perform parallel scheduling.

[0042] Preferably, the method further includes the following steps: Step S201, obtaining real-time fault alarm information of the distribution network, and mapping the controlled equipment corresponding to the real-time fault alarm information to a risk-occupying task with a preset highest conflict coupling factor; Step S202, searching the adjacency matrix for a set of nodes that have electrical logical association with the risk-occupying task, and marking the maintenance tasks located in the node set as disturbed maintenance tasks; Step S203, performing timing decoupling on the disturbed maintenance tasks, delaying the triggering time of the topology state switching instruction of the disturbed maintenance tasks until the fault clearing signal corresponding to the risk-occupying task is obtained, and simultaneously driving other maintenance tasks located outside the node set to execute according to the original timing.

[0043] Preferably, the method further includes the following steps: Step S301, statistically analyzing the actual execution time and standard execution time of historical maintenance tasks, and quantifying the execution deviation dispersion of the execution process relative to the preset plan; Step S302, calculating the deviation correction coefficient based on the execution deviation dispersion. Deviation correction factor The calculation rules are as follows: ,in, As a preset adjustment factor, The actual execution time of the historical preservation task. The standard execution time for historical maintenance tasks; step S303, using the deviation correction coefficient. The length of the preset safety protection period can be adjusted in real time to correct process timing drift caused by uncertainties in the on-site working environment.

[0044] Preferably, step S106 specifically includes: step S401, identifying the resource occupancy list of each maintenance task during the topology steady-state operation instruction execution phase, the resource occupancy list including human resource identifiers and tool identifiers; step S402, determining whether the resource occupancy lists of different maintenance tasks overlap on the spatial access path; step S403, if there is no overlap in the spatial access path, releasing the mutual exclusion logic of different maintenance tasks during the topology steady-state operation instruction execution phase.

[0045] Preferably, the calculation of the conflict coupling factor between different maintenance tasks in step S104 specifically includes: step S501, based on the graph theory search algorithm, calculating the power flow reconfiguration path triggered by the topology state switching command contained in each maintenance task in the adjacency matrix; step S502, extracting the common node elements of the two power flow reconfiguration paths, and determining the conflict coupling factor according to the weight level of the common node elements in the adjacency matrix, and the priority is also determined based on the estimated impact range of the maintenance task and the power consumption priority of the distribution area to which it belongs.

[0046] Preferably, the step S102 of parsing the work instruction text associated with the maintenance task to be performed specifically includes: step S601, performing semantic parsing on the work instruction text to identify the operation predicate and physical object identifier of each operation instruction; step S602, mapping the current state of the device corresponding to the physical object identifier according to the operation predicate to generate the expected state of the device to be maintained.

[0047] Preferably, step S103 specifically includes: step S701, locating the matrix coordinates corresponding to the physical object identifier in the adjacency matrix; step S702, performing a logical XOR operation between the expected state of the maintained equipment and the initial connection value at the matrix coordinates to determine whether the operation instruction belongs to the action that triggers a change in the grid structure.

[0048] Preferably, step S105 specifically includes: step S801, obtaining the estimated execution step length of a high-priority maintenance task when executing a topology state switching command; step S802, summing the estimated execution step length with the length of a preset safety protection period to obtain the offset of the start time of the operation.

[0049] Preferably, the method further includes the following steps: Step S901, after each maintenance task is completed, obtain the real-time topology connection information of the distribution network; Step S902, update the adjacency matrix online based on the real-time topology connection information, as the reference topology for scheduling subsequent maintenance tasks.

[0050] An electrical automation power distribution maintenance process management system, comprising:

[0051] The data acquisition module is used to acquire the initial topology distribution data of the power distribution network and map it into an adjacency matrix that reflects the connection relationship of power distribution equipment.

[0052] The text parsing module is used to parse the work instruction text associated with the maintenance task to be performed, extract the operation instruction sequence of each maintenance task, and identify the expected state of the maintained equipment corresponding to each operation instruction and the physical object identifier corresponding to the operation instruction.

[0053] The instruction classification module is used to compare the expected state of the maintained equipment with the initial connection attributes of the corresponding physical object identifier in the adjacency matrix. Operation instructions that cause a numerical jump in the matrix elements in the adjacency matrix are identified as topology state switching instructions, and operation instructions that maintain the constant value of the adjacency matrix elements are identified as topology steady-state operation instructions.

[0054] The conflict analysis module is used to determine the priority of each maintenance task based on its task type, extract the topology state switching instructions contained in different maintenance tasks, and calculate the conflict coupling factor between different maintenance tasks through the adjacency matrix.

[0055] The timing orchestration module is used to determine the execution start time of higher priority maintenance tasks and calculate the job start time offset of lower priority maintenance tasks when the conflict coupling factor exceeds the preset mutual exclusion threshold, so as to drive the job start time of lower priority maintenance tasks to avoid the preset safety protection period in which the execution start time is located.

[0056] The parallel scheduling module is used to cancel the exclusion constraints between different maintenance tasks and perform scheduling control when different maintenance tasks are all in the corresponding topological steady-state operation instruction execution stage.

[0057] Example 1: In the peak summer operation and maintenance of urban power distribution networks with high penetration of distributed energy access, the dispatch system concurrently receives expansion and renovation tasks for specific main bus sections and insulator replacement tasks for adjacent feeders. The work instruction texts associated with both tasks have an overall duration of up to 4 hours. Following a time-window-based planned process management logic, the system only verifies the sufficiency of human resources for the maintenance teams in both locations, failing to recognize the physical fact that a power outage on the main bus would force a power flow shift to adjacent feeders. This leads to the insulator replacement work, which is in a parallel state, facing the risk of bypass overload. The dispatching end adopts a global process circuit breaker strategy, serializing concurrent tasks to reduce maintenance resource throughput. In the processing flow of parsing the work instruction texts associated with the maintenance tasks to be executed and defining the instruction attributes, the system calls a natural language processing algorithm based on a hidden Markov model to scan the text character sequence. This algorithm sets the sampling frequency to 500Hz in the underlying processing flow, uses a sliding sampling buffer of 256 characters, and sets a 50% window overlap. The overlap rate ensures the integrity of the operation predicate boundary. The core instruction is locked by statistically analyzing the path weight of the feature character sequence in the state transition probability matrix. The operation predicate with state referential attributes and the physical object identifier with network topology attributes are matched and extracted. Based on the preloaded device state association table, the operation predicate is converted into a binary value representing the expected state of the maintained device. The disconnection action is mapped to the value 0 and the closure action is mapped to the value 1. Based on the physical object identifier, the corresponding matrix coordinates are located in the adjacency matrix of the power distribution equipment connection relationship. The initial connection value representing the initial connection attribute at the matrix coordinate is read. The controller calculates the logical XOR operation result of the expected state of the maintained device and the initial connection value. When the logical XOR operation result is a value of 1, it is determined that the operation instruction drives a substantial change in the physical network connection relationship and is defined as a topology state switching instruction. When the logical XOR operation result is a value of 0, it is determined that the connection attribute of the physical device is constant before and after the operation instruction is implemented and is defined as a topology steady-state operation instruction that maintains the value of the adjacency matrix elements.

[0058] The electrical automation power distribution maintenance process management method of the present invention acquires the initial topology distribution data of the power distribution network, maps the initial topology distribution data into an adjacency matrix reflecting the connection relationship of power distribution equipment, parses the work instruction text associated with the maintenance task to be performed, extracts the operation instruction sequence including switching, voltage testing, grounding, and physical replacement, and identifies the expected state of the maintained equipment and the physical object identifier corresponding to each operation instruction; by comparing the initial connection attributes of the expected state of the maintained equipment with the corresponding physical object identifier in the adjacency matrix, the operation instruction that causes a numerical jump in the matrix element in the adjacency matrix is ​​defined as a topology state switching instruction, and the operation instruction that maintains the constant adjacency matrix element is defined as a topology steady-state operation instruction; based on the graph theory search algorithm, the power flow reconstruction path caused by the topology state switching instruction contained in each maintenance task in the adjacency matrix is ​​calculated, and the common node elements are extracted and the conflict coupling factor representing the probability of process interference is determined according to the weight level of the common node elements in the adjacency matrix.

[0059] When the conflict coupling factor between two maintenance tasks exceeds a preset mutual exclusion threshold, the system determines that the higher-priority task has a higher priority based on the estimated impact range of the main busbar segment expansion task. It then locks the execution start time of the higher-priority maintenance task, retrieves the estimated execution step size when executing the topology state switching command, adds the estimated execution step size to the length of the preset safety protection period, calculates the operation start time offset of the lower-priority insulator replacement task, and adjusts the operation start time of the lower-priority maintenance task to avoid the preset safety protection period where the execution start time falls. The system also identifies the two maintenance tasks under topology steady-state operation commands. The resource occupancy list during the execution phase, in the case that there is no overlap in the spatial access paths, removes the mutual exclusion logic of different maintenance tasks in the topology steady-state operation instruction execution phase, triggering parallel scheduling control. The above-mentioned asynchronous orchestration mechanism based on instruction attribute stripping transforms the risk of overlap in physical space into peak-shifting decoupling of transient actions on the time axis, enabling multiple maintenance processes in a specific electrical adjacency area to achieve parallel scheduling in the topology steady-state operation phase while maintaining the safety boundary of power grid operation. This avoids work interruptions caused by dynamic interference of physical topology and maintains the process scheduling determinism and concurrent throughput efficiency of the distribution network during intensive maintenance periods.

[0060] Example 2: In the test case of intensive maintenance and scheduling concurrency performance of a high-penetration distribution network, a simulation platform was constructed using the physical network topology data of a 10kV distribution network in a certain city. The Newton-Raphson alternating solution model was used to calculate the power flow state of network nodes. To simulate an industrial communication environment, Gaussian white noise with a signal-to-noise ratio of 20dB was injected into the received remote signaling data stream, and random network delay parameters following a Poisson distribution with a mean of 50ms were superimposed. This was used to construct the original input data stream that maps the initial connection attributes of physical devices, and a preset mutual exclusion threshold was set to define the probability of process interference. ,in, A dimensionless parameter representing interference tolerance; when the apparent power of the shared node elements in the power flow reconfiguration path approaches the upper limit of the rated capacity of the corresponding equipment, a preset mutual exclusion threshold is controlled. The value tends towards the lower limit of the range; under the heavy-load network distribution conditions of this test case, the preset mutual exclusion threshold is determined based on the above calculation logic. It is 0.65.

[0061] The system inputs a test sequence containing 100 randomly generated distribution network maintenance tasks into the simulation platform. The system parses the work instruction text for each maintenance task and extracts the underlying operation instructions. By comparing the characteristics of the main line defect elimination tasks and the adjacent feeder grid connection tasks in the task sequence, the system calculates the conflict coupling factor between the two maintenance tasks. It is 0.72, of which, A dimensionless numerical variable representing the degree of physical interference; when the conflict coupling factor Greater than the preset mutual exclusion threshold Based on the estimated impact range, the main line maintenance task is determined to have a higher priority. The estimated execution step size of 15 minutes when the main line maintenance task responds to the topology state switching command is extracted. The estimated execution step size is added to the preset safety protection period of 10 minutes to calculate the operation start time offset of the adjacent feeder grid connection task as 25 minutes. The operation start time of the adjacent feeder grid connection task is shifted 25 minutes on the time axis. At the same time, it is identified that the spatial access paths of the two maintenance tasks do not overlap during the topology steady-state operation phase. Thus, the mutual exclusion logic is released and parallel scheduling is triggered, and the sample group of this invention is established. A control group 1 using a planned time window serial scheduling strategy is established simultaneously, and a control group 2 driving the global time axis shift without removing the topology steady-state operation command is established. The preset mutual exclusion threshold is set. The lower limit constraint group and the upper limit allowable group are set to 0.25 and 0.95 respectively.

[0062] The global maintenance throughput and the number of bypass overload alarms within the test period were extracted as evaluation indicators. The global maintenance throughput of control group 1 was 12 items / h with no bypass overload alarms, and the global maintenance throughput of control group 2 was 18 items / h with no bypass overload alarms. The sample group of this invention increased the global maintenance throughput to 34 items / h without triggering bypass overload alarms. The lower limit constraint group caused the global maintenance throughput to drop to 14 items / h due to the serial queuing of a large number of maintenance tasks. The upper limit unrestricted group achieved a global maintenance throughput of 38 items / h, but triggered 7 bypass overload alarms at the moment of power flow reconstruction. The nonlinear performance degradation trend after exceeding the optimal parameter range and the data comparison of each control group confirmed that the instruction attribute stripping and asynchronous orchestration mechanism combined with the specific parameter working window met the safety margin of distribution network power flow reconstruction and the concurrent throughput resource utilization requirements of maintenance tasks under the condition of data disturbance.

[0063] Example 3: In the intensive operation of complex urban ring network power distribution systems, the dispatch center faces physical limitations in handling cross-regional concurrent maintenance processes due to a lack of quantitative dimensions for interference risks. Relying on the basic network overlap attributes cannot accurately reflect the physical limits of load transfer, causing processes with safe concurrency conditions to enter a serial waiting sequence due to exclusion judgment. The electrical automation power distribution maintenance process management method of this invention acquires network operation data containing node rated capacity and real-time load parameters, integrates the network operation data into an adjacency matrix reflecting the connection relationship of power distribution equipment, and makes the non-zero matrix elements of the adjacency matrix carry specific values ​​representing the remaining capacity of the line. After parsing the maintenance task to be executed and determining the topology state switching command that causes the matrix element value to jump, the system uses the disconnected node pointed to by the topology state switching command as the search starting point, and uses a breadth-first search algorithm in the adjacency matrix to traverse adjacent nodes with non-zero characteristic values ​​layer by layer until reaching a standby network node with connectivity conditions. The entire node sequence traversed by the search is extracted to generate a power flow reconfiguration path, and the conflict analysis module calculates the conflict coupling factor of the maintenance task. The breadth-first search algorithm is invoked to traverse the adjacency matrix. The physical object identifier pointed to by the topology state switching instruction is selected as the starting anchor point for the path search. The search proceeds level by level along the branches representing electrical connectivity represented by the non-zero elements of the matrix until a backup power node with connectivity is obtained. The sequence of searched nodes is extracted to form a power flow reconfiguration path. The common node elements that intersect and overlap between two power flow reconfiguration paths are identified. The load current data of the common node elements within the sampling period is read, and the ratio of the load current to the rated current carrying capacity of the common node elements is calculated. The ratio is defined as the weight level of the common node elements. All weight levels The cumulative operation yields the conflict coupling factor. Weighting levels Coupling factors with conflict All of them are dimensionless variables.

[0064] The system extracts the shared node elements where two power flow reconfiguration paths intersect and overlap. It retrieves the estimated superimposed apparent power carried by each shared node element during topology state transitions, and determines the weight level of that shared node element by calculating its quotient with its rated capacity. The system aggregates the weight levels of all shared node elements and calculates their sum, outputting this sum as the conflict coupling factor. ,in, A dimensionless numerical variable representing the degree of physical interference; in this test condition, a power flow reconfiguration path containing three shared node elements was extracted, and the weight levels of each shared node element were calculated to be 0.2, 0.31, and 0.42, respectively. The system accumulates the above three values ​​and outputs the conflict coupling factor. The value is 0.93. The system compares the output value with the set mutual exclusion threshold. When the value is greater than the mutual exclusion threshold, the system drives the start time of the lower priority maintenance task to generate an offset on the time axis. The quantitative calculation of graph theory search and capacity ratio calculation constructs a mapping path from the underlying physical network topology variables to the upper-level process scheduling control parameters. This provides mathematical criteria for the spatiotemporal staggered scheduling of maintenance processes, reduces the concurrent throughput loss caused by experience evaluation, and maintains the process scheduling determinism of the management system in the dynamic evolution environment of the distribution network topology.

[0065] Example 4: In the pre-deployment commissioning of a system targeting an urban ring network distribution area, the dispatch center accesses a historical dataset of the benchmark network structure reflecting seasonal environmental parameters and equipment service life. The system parses the physical nameplate factory parameters of each distribution node in this dataset and extracts the initial rated apparent power. Simultaneously, it collects the node environmental operating temperature baseline and the main transformer insulation aging assessment index for 30 consecutive natural days. The controller calculates the derating correction value of the initial rated apparent power based on the aforementioned environmental variables. The specific mathematical logic is to construct the effective dynamic capacity. Calculation formula ,in, The initial apparent power rating is indicated on the physical nameplate. This represents the temperature derating factor mapped from the ambient operating temperature baseline. This represents the conversion factor for the service life of equipment determined based on the insulation aging assessment index. The unit is kVA. The system traverses all node elements in the distribution network and outputs the corresponding effective dynamic capacity matrix.

[0066] The controller loads the aforementioned output effective dynamic capacity matrix into the topology simulation sandbox and injects a stress test data stream containing multiple concurrent maintenance tasks into the topology simulation sandbox. The system increases the concurrent issuance density of topology state switching commands in the stress test data stream on a cycle-by-cycle basis, and monitors the apparent power step peak of shared node elements on the power flow reconfiguration path in real time. When the apparent power step peak of any shared node element first reaches the effective dynamic capacity corresponding to that node... When the system reaches the critical boundary, it extracts the conflict coupling factor of the concurrent task composition mapping at that moment. The conflict coupling factor The value is offset in the negative direction by a margin of 5%, based on which the preset mutual exclusion threshold for this power distribution area is generated. The initial fixed parameters, the aforementioned extreme value boundary detection operation, writes the baseline scheduling parameters adapted to local hardware constraints into the system memory when the process management system is connected to a new physical environment.

[0067] Example 5: In the pre-deployment parameter calibration of a newly connected distribution network area, the dispatch center retrieves the switch action status records for twelve consecutive natural months from the historical operation database, filters out the actual action time data that matches the identifier of the physical object to be maintained to construct a sample sequence, and the controller calculates the statistical mean and standard deviation of the sample sequence. The statistical mean is set as the estimated execution step size of the physical device in response to the topology state switching command, and a preset safety protection period is established at the same time. calibration calculation formula ,in, Represents the statistical mean of the sample sequence. The standard deviation of the sample sequence is represented by... This indicates the dimensionless safety margin coefficient mapped based on the local power grid reliability benchmark. and as well as The dimensions of all values ​​are min; based on this calibration formula, the specific values ​​are output and the time boundary parameters in the local control database are updated.

[0068] The system loads the calibrated estimated execution step size and preset safety protection period into the process orchestration module. When a stress test data stream containing multiple concurrent maintenance tasks is received, the scheduling kernel extracts the calibration parameters associated with each task and generates a staggered projection sequence along the time axis. It simultaneously monitors the apparent power step peak of common node elements within overlapping time slices. If the test feedback indicates that the apparent power of all common node elements is within the safe capacity range, the system determines that the current calibration parameters meet local operating physical constraints and solidifies the parameter matrix into the production environment. This offline calibration and data filling procedure constructs a mapping path from the historical operating characteristics of the underlying equipment to the upper-level process scheduling parameters. This provides a statistically based data benchmark for the system's asynchronous orchestration mechanism in a network environment, reducing the risk of concurrent interference caused by the discretization of action time due to equipment mechanical aging. The timing orchestration module quantitatively determines the preset safety protection period. The benchmark is based on data collected from the power distribution terminal, including ambient temperature and equipment insulation aging, and then calculated using the formula... Calculate the dynamic capacity of shared node elements , The initial rated capacity indicated on the equipment nameplate. To determine the temperature derating factor based on ambient temperature sensor readings. To calculate the performance loss factor based on the service life, The unit is kVA. The actual execution time of similar switching actions is sampled from historical databases. The mean of the sample sequence is taken as the estimated execution step size. The estimated execution step size is then compared with the dynamic capacity. The corresponding safety margin times are added together, and the output drive timing is shifted by the preset safety protection period. Deviation correction factor Based on the actual execution time of historical maintenance tasks With standard execution time The degree of deviation, within the preset safety protection period of the scheduling cycle. Length update.

[0069] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0070] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for managing electrical automation power distribution maintenance processes, characterized in that, Includes the following steps: Step S101: Obtain the initial topology distribution data of the power distribution network and map it into an adjacency matrix that reflects the connection relationship of power distribution equipment; Step S102: Parse the work instruction text associated with the maintenance task to be performed, extract the operation instruction sequence of each maintenance task, and identify the expected state of the maintained equipment corresponding to each operation instruction and the physical object identifier corresponding to the operation instruction. Step S103: Compare the expected state of the maintained equipment with the initial connection attributes of the corresponding physical object identifier in the adjacency matrix. Define the operation instruction that causes the matrix element in the adjacency matrix to change its value as the topology state switching instruction, and define the operation instruction that keeps the adjacency matrix element constant as the topology steady-state operation instruction. Step S104: Determine the priority based on the task type of each maintenance task, extract the topology state switching instructions contained in different maintenance tasks, and calculate the power flow reconfiguration path triggered by the topology state switching instructions contained in each maintenance task in the adjacency matrix based on the graph theory search algorithm; extract the common node elements of the two power flow reconfiguration paths, and determine the conflict coupling factor based on the weight level of the common node elements in the adjacency matrix. The priority is also determined based on the estimated impact range of the maintenance task and the power consumption priority of the distribution area to which it belongs. Step S105: When the conflict coupling factor exceeds the preset mutual exclusion threshold, determine the execution start time of the higher priority maintenance task and calculate the job start time offset of the lower priority maintenance task so as to drive the job start time of the lower priority maintenance task to avoid the preset safety protection period in which the execution start time is located. Step S106: When different maintenance tasks are all in the corresponding topological steady-state operation instruction execution stage, cancel the exclusion constraints between different maintenance tasks and perform parallel scheduling.

2. The electrical automation power distribution maintenance process management method according to claim 1, characterized in that, The method further includes the following steps: Step S201, obtaining real-time fault alarm information of the distribution network and mapping the controlled equipment corresponding to the real-time fault alarm information to a risk-occupying task with a preset highest conflict coupling factor; Step S202, searching the adjacency matrix for a set of nodes that have electrical logical association with the risk-occupying task and marking the maintenance tasks located in the node set as disturbed maintenance tasks; Step S203, performing timing decoupling on the disturbed maintenance tasks, delaying the triggering time of the topology state switching instruction of the disturbed maintenance tasks until the fault clearing signal corresponding to the risk-occupying task is obtained, and simultaneously driving other maintenance tasks located outside the node set to execute according to the original timing.

3. The electrical automation power distribution maintenance process management method according to claim 1, characterized in that, The method further includes the following steps: Step S301, statistically analyzing the actual execution time and standard execution time of historical maintenance tasks, and quantifying the execution deviation dispersion of the execution process relative to the preset plan; Step S302, calculating the deviation correction coefficient based on the execution deviation dispersion. Deviation correction factor The calculation rules are as follows: ,in, As a preset adjustment factor, The actual execution time of the historical preservation task. The standard execution time for historical maintenance tasks; step S303, using the deviation correction coefficient. The length of the preset safety protection period can be adjusted in real time to correct process timing drift caused by uncertainties in the on-site working environment.

4. The electrical automation power distribution maintenance process management method according to claim 1, characterized in that, Step S106 specifically includes: Step S401, identifying the resource occupancy list of each maintenance task during the topology steady-state operation instruction execution phase, the resource occupancy list including human resource identifiers and tool identifiers; Step S402, determining whether the resource occupancy lists of different maintenance tasks overlap on the spatial access path; Step S403, if there is no overlap in the spatial access path, releasing the mutual exclusion logic of different maintenance tasks during the topology steady-state operation instruction execution phase.

5. The electrical automation power distribution maintenance process management method according to claim 1, characterized in that, The step S102 of parsing the work instruction text associated with the maintenance task to be performed specifically includes: step S601, performing semantic parsing on the work instruction text to identify the operation predicate and physical object identifier of each operation instruction; step S602, mapping the current state of the equipment corresponding to the physical object identifier according to the operation predicate to generate the expected state of the equipment to be maintained.

6. The electrical automation power distribution maintenance process management method according to claim 1, characterized in that, Step S103 specifically includes: Step S701, locating the matrix coordinates corresponding to the physical object identifier in the adjacency matrix; Step S702, performing a logical XOR operation between the expected state of the maintained equipment and the initial connection value at the matrix coordinates to determine whether the operation instruction belongs to the action that triggers a change in the grid structure.

7. The electrical automation power distribution maintenance process management method according to claim 1, characterized in that, Step S105 specifically includes: Step S801, obtaining the estimated execution step length of the high-priority maintenance task when executing the topology state switching command; Step S802, summing the estimated execution step length with the length of the preset safety protection period to obtain the offset of the start time of the operation.

8. The electrical automation power distribution maintenance process management method according to claim 1, characterized in that, The method also includes the following steps: Step S901, after each maintenance task is completed, obtain the real-time topology connection information of the distribution network; Step S902, update the adjacency matrix online based on the real-time topology connection information, as the reference topology for scheduling subsequent maintenance tasks.

9. An electrical automation power distribution maintenance process management system, used to implement the electrical automation power distribution maintenance process management method of claim 1, characterized in that, include: The data acquisition module is used to acquire the initial topology distribution data of the power distribution network and map it into an adjacency matrix that reflects the connection relationship of power distribution equipment. The text parsing module is used to parse the work instruction text associated with the maintenance task to be performed, extract the operation instruction sequence of each maintenance task, and identify the expected state of the maintained equipment corresponding to each operation instruction and the physical object identifier corresponding to the operation instruction. The instruction classification module is used to compare the expected state of the maintained equipment with the initial connection attributes of the corresponding physical object identifier in the adjacency matrix. Operation instructions that cause a numerical jump in the matrix elements in the adjacency matrix are identified as topology state switching instructions, and operation instructions that maintain the constant value of the adjacency matrix elements are identified as topology steady-state operation instructions. The conflict analysis module is used to determine the priority of each maintenance task based on its task type, extract the topology state switching instructions contained in different maintenance tasks, and calculate the power flow reconfiguration path caused by the topology state switching instructions contained in each maintenance task in the adjacency matrix based on the graph theory search algorithm. It also extracts the common node elements of the two power flow reconfiguration paths and determines the conflict coupling factor based on the weight level of the common node elements in the adjacency matrix. The priority is also determined based on the estimated impact range of the maintenance task and the power consumption priority of the distribution area to which it belongs. The timing orchestration module is used to determine the execution start time of higher priority maintenance tasks and calculate the job start time offset of lower priority maintenance tasks when the conflict coupling factor exceeds the preset mutual exclusion threshold, so as to drive the job start time of lower priority maintenance tasks to avoid the preset safety protection period in which the execution start time is located. The parallel scheduling module is used to cancel the exclusion constraints between different maintenance tasks and perform scheduling control when different maintenance tasks are all in the corresponding topological steady-state operation instruction execution stage.

Images