CN-122001091-A - Self-adaptive intelligent distribution network distribution automation system for distribution network

CN122001091ACN 122001091 ACN122001091 ACN 122001091ACN-122001091-A

Abstract

The invention discloses a self-adaptive intelligent distribution network distribution automation system for a distribution network, which relates to the technical field of power grid distribution systems, wherein a calculation module introduces a block-level risk confidence scoring model to calculate the risk confidence score of each physical node, the physical nodes are classified and sequenced according to the scoring result, continuous sections with scores higher than a preset safety threshold are screened out to serve as preferential treatment target areas, the whole network is divided into a plurality of control execution units according to a feeder section and a control domain, each control execution unit containing the preferential treatment target area applies for a micro-granularity operation lock, a level difference isolation module performs level difference isolation and protection logic operation, and all opening and closing instructions and protection parameters are adjusted to be synchronized to related terminals in real time through a 5G network. The system effectively improves the fault handling speed and accuracy of the power distribution network in a complex operation environment, and reduces malfunction and refusal risks.

Inventors

ZHANG XIAOYI
XIAO ZHIYING
WU ZHIYANG

Assignees

珠海市阿普顿电气有限公司

Dates

Publication Date: 20260508
Application Date: 20260410

Claims (10)

1. The self-adaptive intelligent distribution network distribution automation system for the distribution network is characterized by comprising a construction module, a comparison module, a calculation module and a level difference isolation module; The construction module is used for acquiring a real-time operation situation data pool in a target power distribution network operation environment, wherein the operation situation data pool comprises a physical mapping set and a control intention releasing set, the physical mapping set comprises installation position coordinates of a sectionalizing switch, a contact switch and a protection relay, a feeder section number, and an upstream and downstream electric connection node identifier, and the control intention releasing set comprises a remote control/remote adjustment instruction issued by a main station for centralized control, an autonomous control action triggered by an on-site feeder automation logic and an inter-node negotiation instruction when an intelligent distributed cooperative control strategy is started; The comparison module is used for carrying out item-by-item cross comparison on the physical mapping set and the control intention release set to construct a bidirectional state comparison matrix; The calculation module is used for introducing a block-level risk confidence scoring model to calculate the risk confidence score of each physical node, grading and sorting the physical nodes according to the scoring result, screening out continuous sections with scores higher than a preset safety threshold as preferential treatment target areas, dividing the whole network into a plurality of control execution units according to feeder sections and control domains, and applying a micro-granularity operation lock to each control execution unit containing the preferential treatment target areas; And the step difference isolation module is used for carrying out step difference isolation and protection logic when the micro-granularity operation lock is obtained and fails, and regulating all opening and closing instructions and protection parameters to be synchronized to related terminals in real time through a 5G network.
2. The adaptive intelligent distribution network distribution automation system for the distribution network of claim 1 wherein the computing module introduces a block level risk confidence score model to compute a risk confidence score for each physical node: based on telemetry data and telemetry signaling data, extracting characteristic quantities reflecting abnormal states of equipment and lines, and converting the characteristic quantities into risk factors; the various risk factors are aggregated according to node attribution in a block-level risk confidence scoring model to form an original risk feature vector of the node; After the original risk feature vector is obtained, setting an influence weight for each risk factor by a block-level risk confidence scoring model; Carrying out normalization processing on the count or the amplitude of each risk factor, and mapping the count or the amplitude to a unified score interval; accumulating the normalized factor scores according to the weights to obtain initial risk confidence scores of the nodes; And introducing an environment correction coefficient to adjust the initial score up and down to form a block-level risk confidence score.
3. The adaptive intelligent distribution network power distribution automation system for a distribution network of claim 2, wherein the calculation module ranks the physical nodes according to the scoring result, and screens out continuous sections with scores higher than a preset safety threshold as a preferential treatment target area: after risk confidence scoring of the physical nodes of the whole network is completed, sorting is carried out according to scoring results to form a node risk sequence from high to low, and the sorting results are used for hierarchical management; Dividing the nodes into a safety zone, an early warning zone and a high-risk zone according to a preset safety threshold; The high-risk area node distribution is scanned on the topological structure, and a plurality of high-risk nodes which are electrically connected with each other and are distributed continuously are combined into a preferential treatment target area.
4. An adaptive intelligent distribution automation system for a distribution network as set forth in claim 3 wherein the computing module divides the overall network into a plurality of control execution units according to feeder segments and control domains, and applies for a micro-granularity operation lock for each control execution unit containing a priority handling target region: after the identification of the preferential treatment target area is completed, dividing the whole network into a plurality of control execution units according to the established feeder line section division and control domain boundary of the power distribution network, wherein each unit comprises a plurality of feeder line sections, corresponding switches, protection devices and power distribution terminals; for a control execution unit containing a priority treatment target area, initiating a request of a microparticle operation lock to a dispatching and control system, wherein the request comprises a switch preemption right, a protection fixed value temporary modification right and a feeder automation logic short-time disabling right, and the request logic is as follows: determining all equipment lists in the target area, wherein the equipment lists need to be subjected to state change or parameter adjustment in the follow-up strategy execution, and checking the current control right state of the equipment lists; For the equipment which is occupied by other processes or in an uncontrollable state, marking conflict and returning application failure information, and restarting after the conflict is resolved; for conflict-free equipment, a lock request message is generated, the type of locked resources, the number of target equipment, the locking validity period and unlocking conditions are definitely determined, and the target area identification and risk confidence score abstract calculated at this time are added; After the security authentication, the lock request is issued to the corresponding equipment control service or the regional controller, and the lock registration is completed and the confirmation is fed back.
5. The adaptive intelligent distribution network distribution automation system for a distribution network according to claim 2, wherein the risk factor generation logic is as follows: For each physical node, collecting a voltage sampling sequence of the physical node in a set statistical period, detecting whether a sudden drop event with the amplitude lower than a normal operation lower limit and the duration exceeding a set threshold value occurs, and accumulating corresponding voltage sudden drop risk counts once; Carrying out mutation detection on the current sampling sequence, identifying sudden increase or sudden decrease events with the change rate exceeding a set slope and the duration, and accumulating current mutation risk counts; Counting the times of opening and closing actions of a sectionalizing switch and a connecting switch associated with a node in a period, judging whether the operation is abnormally frequently operated or not by combining action intervals and reason codes, and accumulating the high-frequency action risk count of the switch; And counting the abnormal starting times of the protection relay by combining the protection triggering record.
6. The self-adaptive intelligent distribution network distribution automation system for a distribution network according to claim 1, wherein the comparison module performs item-by-item cross comparison on a physical mapping set and a control intention release set to construct a bidirectional state comparison matrix: Traversing each node record of the physical mapping set, searching all target nodes in the control intention releasing set according to the number and the acquisition time of the node, and sending out instruction records with the time earlier than or equal to the acquisition time to form a candidate instruction set; For each instruction in the candidate instruction set, judging whether the execution feedback time is before the acquisition time so as to determine whether the instruction has an execution influence before the acquisition time; if the state of a certain node in the physical mapping set is consistent with the expected execution state of the corresponding instruction in the control intention releasing set and the execution feedback is confirmed to be achieved, judging that the states are consistent; if the states are inconsistent or lack of execution feedback, marking as state deviation; Constructing a bidirectional state comparison matrix, wherein the row dimension of the bidirectional state comparison matrix corresponds to space nodes in a physical mapping set, the column dimension corresponds to an intention instruction in a control intention release set, and the matrix element value is obtained through the matching degree between the comprehensive index quantization nodes and the instruction and related operation quality parameters.
7. The adaptive intelligent distribution network distribution automation system for a distribution network of claim 6, wherein the element generation of the bidirectional state comparison matrix comprises the following steps: Checking whether the actual state of the node at the acquisition time accords with the expected execution state of the corresponding instruction, if so, giving a high matching degree mark, and if partially accords or does not accord, reducing the matching degree grade according to the deviation type; Analyzing whether a link from the instruction sending to the feedback executing is interrupted or overtime, if the communication link packet loss exists, the feedback is lost or the equipment cannot respond to the instruction due to the locking state, judging that the communication link packet loss is unreachable or partially reachable, and recording the reachability level in the element value; The communication quality indexes of the power distribution terminal of the node in the instruction period are called, wherein the communication quality indexes comprise link delay, packet loss rate and retransmission times, and if the delay exceeds a preset threshold or the packet loss rate is higher than an allowable upper limit, a communication reliability degradation factor is introduced into an element value; inquiring the history execution records of the nodes and the similar instructions, counting the successful execution proportion, integrating the history success rate into the element value as a weighting coefficient, and combining the matching degree, the reachability, the communication reliability and the history success rate into a comprehensive score according to a preset priority rule to obtain a matrix element value.
8. The adaptive intelligent distribution network distribution automation system for a distribution network of claim 1, wherein the physical mapping set is established according to an actual spatial layout and an electrical coupling relation of field devices; the control intent release set is derived from a historical and currently validated fault handling plan library, run mode switch instruction sequence.
9. The adaptive intelligent distribution network distribution automation system for a distribution network of claim 1, wherein the level difference isolation module performs level difference isolation and protection logic: When the micro-granularity operation lock is obtained and a fault occurs, the sectionalizing switch A closest to the power supply side of the fault point is opened, and the sectionalizing switch B closest to the load side is kept to lose voltage and not opened; After a set delay, trying to close the sectional switch A, if a permanent fault is detected, accelerating to open and positively locking, and meanwhile, opening the sectional switch B after a short-time incoming call is detected; and the contact switch is switched on to realize reverse power supply, other sectionalized switches are switched on in sequence, and differential protection or overcurrent protection logic is dynamically selected according to state estimation and a power flow calculation result under a ring network/radiation network mixed scene, so that misoperation or refusal operation is avoided.
10. The self-adaptive intelligent distribution network distribution automation system for the distribution network of claim 9, wherein the step difference isolation module retrieves an operation lock list to confirm that the preemption of the tie switch is authorized, checks that the voltage states at two sides of the tie switch meet the synchronous or quasi-synchronous conditions, and if the conditions are met, sends a closing instruction to enable a sound power supply to supply power to the load side of an original fault section, sequentially closes other section switches according to a feeder section and a control domain level, and gradually restores the power supply range of the sound section.

Description

Self-adaptive intelligent distribution network distribution automation system for distribution network Technical Field The invention relates to the technical field of power grid distribution systems, in particular to a self-adaptive intelligent distribution network distribution automation system for a distribution network. Background In the power supply of the alternating current transmission of more than 750 kilovolts, a large-scale power grid safety guarantee and defense system, an intelligent scheduling system and the like, the timely discovery and accurate treatment of faults directly relate to the continuous power supply of the power supply reliability and important loads, the power distribution network is complex in structure, various in topology and different in communication condition, the operation working condition is often accompanied by multi-source disturbance and mode switching, the traditional power distribution automation system is difficult to sense the operation situation of the whole network in real time, accurately identify the consistency of the physical state and the control intention of equipment, potential risks cannot be effectively evaluated, and the priority treatment section is locked, so that the problems of incomplete isolation, path recovery conflicts, false operation or refusal operation and the like are easily caused under the complex fault scene, the power failure time is prolonged, and the influence range is enlarged. In view of urgent requirements of modern distribution networks on high reliability, high efficiency and intelligent treatment capability, particularly when the severe conditions of ring network and radiation network mixing, communication quality fluctuation, multi-fault concurrency and the like are faced, an adaptive intelligent distribution automation system and method capable of integrating global situation sensing, double-domain mapping comparison, risk feature extraction and level difference isolation control are required to be provided, so that full-closed-loop targeting management and control from risk identification to accurate execution are realized, and the accuracy, instantaneity and self-adaptive capability of fault treatment are remarkably improved. Disclosure of Invention The invention aims to provide a self-adaptive intelligent distribution network distribution automation system for a distribution network, which aims to solve the problem of the defects in the background technology. In order to achieve the purpose, the invention provides the technical scheme that the self-adaptive intelligent distribution network distribution automation system for the distribution network comprises a construction module, a comparison module, a calculation module and a level difference isolation module; the construction module is used for acquiring a real-time operation situation data pool in an operation environment of the target power distribution network, wherein the operation situation data pool comprises a physical mapping set and a control intention releasing set, and the physical mapping set and the control intention releasing set are sent to the comparison module; the comparison module is used for carrying out item-by-item cross comparison on the physical mapping set and the control intention release set, constructing a bidirectional state comparison matrix, and sending the bidirectional state comparison matrix to the calculation module; The calculation module is used for introducing a block-level risk confidence scoring model to calculate risk confidence scores of each physical node, grading and sorting the physical nodes according to scoring results, screening out continuous sections with scores higher than a preset safety threshold as preferential treatment target areas, dividing the whole network into a plurality of control execution units according to feed line sections and control domains, applying a micro-granularity operation lock to each control execution unit containing the preferential treatment target areas, and sending the micro-granularity operation lock to the grading difference isolation module; And the step difference isolation module is used for carrying out step difference isolation and protection logic when the micro-granularity operation lock is obtained and fails, and regulating all opening and closing instructions and protection parameters to be synchronized to related terminals in real time through a 5G network. Preferably, the calculation module introduces a block-level risk confidence score model to calculate a risk confidence score for each physical node: based on telemetry data and telemetry signaling data, extracting characteristic quantities reflecting abnormal states of equipment and lines, and converting the characteristic quantities into risk factors; the various risk factors are aggregated according to node attribution in a block-level risk confidence scoring model to form an original risk feature vector of the node; After