CN112964960B - Power grid fault diagnosis method based on multi-source data fusion of scheduling fault fingerprint library - Google Patents
Power grid fault diagnosis method based on multi-source data fusion of scheduling fault fingerprint library Download PDFInfo
- Publication number
- CN112964960B CN112964960B CN202110154576.3A CN202110154576A CN112964960B CN 112964960 B CN112964960 B CN 112964960B CN 202110154576 A CN202110154576 A CN 202110154576A CN 112964960 B CN112964960 B CN 112964960B
- Authority
- CN
- China
- Prior art keywords
- fault
- data
- fingerprint
- power grid
- fingerprint library
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000003745 diagnosis Methods 0.000 title claims abstract description 54
- 230000004927 fusion Effects 0.000 title claims abstract description 22
- 230000011664 signaling Effects 0.000 claims abstract description 91
- 238000010845 search algorithm Methods 0.000 claims abstract description 9
- 230000004224 protection Effects 0.000 claims description 91
- 238000013507 mapping Methods 0.000 claims description 20
- 230000009471 action Effects 0.000 claims description 17
- 230000008569 process Effects 0.000 claims description 13
- 238000004458 analytical method Methods 0.000 claims description 12
- 230000006870 function Effects 0.000 claims description 9
- 230000001960 triggered effect Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000004519 manufacturing process Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 4
- 230000009466 transformation Effects 0.000 claims description 4
- 238000009434 installation Methods 0.000 claims description 2
- 230000008901 benefit Effects 0.000 abstract description 7
- 238000004422 calculation algorithm Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000004088 simulation Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 2
- 238000002405 diagnostic procedure Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000007500 overflow downdraw method Methods 0.000 description 2
- 230000008521 reorganization Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 238000013398 bayesian method Methods 0.000 description 1
- 239000003990 capacitor Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/088—Aspects of digital computing
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/081—Locating faults in cables, transmission lines, or networks according to type of conductors
- G01R31/086—Locating faults in cables, transmission lines, or networks according to type of conductors in power transmission or distribution networks, i.e. with interconnected conductors
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
- Y04S10/52—Outage or fault management, e.g. fault detection or location
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Remote Monitoring And Control Of Power-Distribution Networks (AREA)
Abstract
The invention relates to a power grid fault diagnosis method based on multi-source data fusion of a dispatching fault fingerprint library, which comprises the following steps: step 1: acquiring data of a plurality of systems from a scheduling end; step 2: and establishing a fault local feature fingerprint library by using substation interval remote signaling data, and establishing a fault overall feature fingerprint library by using a power grid element. Step 3: and matching the local characteristic fingerprint library and the overall characteristic fingerprint library by using a matching search algorithm, searching out similar or matching values of the power grid fault remote signaling data received by the scheduling end and the data in the fingerprint library, and outputting corresponding fault identity information as a final output result of fault diagnosis. Compared with the prior art, the invention has the advantages of comprehensive and rapid fault diagnosis and the like.
Description
Technical Field
The invention relates to a power grid fault diagnosis method, in particular to a power grid fault diagnosis method based on multi-source data fusion of a dispatching fault fingerprint library.
Background
The fusion method for the fault wave recording system, the PMU system, the language image system and the meteorological system at the dispatching end comprises the following steps: combining radial basis function neural network and information fusion technology, fusing and electric quantity, PMU and SCADA system data, and realizing second-level diagnosis of cascading failure; the diagnosis method based on fuzzy integral information fusion and combined with the power flow and remote signaling information of the power grid is used for diagnosing and diagnosing the power grid faults; solving fault equipment and fault time sequence by using a method combining time sequence information and multi-source information, and improving fault tolerance of fault diagnosis; fully mining the application scene of multidimensional information containing SCADA, WAMS, FIS and other information sources in power system fault diagnosis, and providing a power grid fault diagnosis method based on time sequence matching of the multisource information. However, the above method has a certain limitation on the rapid identification of faults and the fusion of multiple systems and the interpretation of grid faults.
The fault diagnosis method utilizing the remote signaling and fault recording fusion comprises a D-S evidence theory fusion method, an artificial intelligence method and an inference model method: the multi-data source information fusion fault diagnosis method based on the improved D-S evidence theory; constructing a new wavelet neural network fault recognition model by utilizing lifting wavelets and PNN networks; an algorithm for performing fault diagnosis by using a quantum neural network is used for improving fault diagnosis tolerance; an intelligent power transmission network fault diagnosis method based on a knowledge grid technology; a method for diagnosing the faults of electric network by multi-agent technology and cooperative expert system. The study on the reasoning model mainly comprises: the subjective Bayesian method is put forward to exert the advantages of the Bayesian network in the reasoning aspect; a power grid fault diagnosis model based on a colored Petri network is established; a time Petri network fault diagnosis model considering a topological structure and a power system fault diagnosis method based on forward and reverse reasoning give out fault equipment or a fault range. The method can only integrate the information of the electric quantity and the switching value, and is difficult to display the panorama of the power grid fault information.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a power grid fault diagnosis method based on multi-source data fusion of a scheduling fault fingerprint library.
The aim of the invention can be achieved by the following technical scheme:
a power grid fault diagnosis method based on multi-source data fusion of a scheduling fault fingerprint library comprises the following steps:
step 1: acquiring data of a plurality of systems from a scheduling end;
step 2: establishing a fault local feature fingerprint library by using substation interval remote signaling data, and establishing a fault overall feature fingerprint library by using a power grid element;
step 3: and matching the local characteristic fingerprint library and the overall characteristic fingerprint library by using a matching search algorithm, searching out similar or matching values of the power grid fault remote signaling data received by the scheduling end and the data in the fingerprint library, and outputting corresponding fault identity information as a final output result of fault diagnosis.
Further, the data of the systems in the step 1 comprise SCADA, fault recording and PMU system data related to power grid operation and control; PMS, OMS system data relating to production management; video surveillance, dispatch telephony voice system data associated with operation and incident handling.
Further, the process of establishing the fault local feature fingerprint library in the step 2 specifically includes: dividing and encoding remote signaling data fields, mapping to a multidimensional data space according to a mapping transformation relation, and converting the fault diagnosis problem of remote signaling deflection data into the classification problem of sample data in the multidimensional space.
Further, the mapping transformation relation has a description formula as follows:
wherein A is 1 ...A n For n matrices of post-fault remote signalling binary data, c 1 ...c n For n encoded fault encoded data, f 1 ...f n For n fault coding mapping functions, a two-dimensional coding mode and a one-dimensional coding mode are adopted for the coding modes of the remote signaling binary data matrix and the fault coding mapping functions after faults.
Further, the process of establishing the fault overall feature fingerprint library in the step 2 specifically includes: defining logical multiplication and logical addition, and forming expected remote signaling values of the protection and circuit breaker according to protection logic of element faults, main protection, backup protection and failure protection, wherein the expected remote signaling values are used as data in a fault fingerprint overall characteristic fingerprint library.
Further, the description formula of the data in the fault fingerprint total feature fingerprint library is as follows:
wherein d k In the event of a failure of the component,desired telemetry value for main protection of faulty element, < >>Desired telemetry value for near backup protection of faulty element,>non-operational value for main protection of faulty element, < ->Desired telemetry value, Z (r ks ,d k ) Far backup protection range d for faulty element k A collection of adjacent elements, p (r ks ,d x ) To follow the power supply path from the faulty elementFar backup protection mounting to element d x All sets of circuit breakers>For the non-operational value of the t-th breaker in the set of all breakers, +.>Non-operational value for near-backup protection of faulty element,/->Is a circuit breaker c j Is claimed, +.>Is a circuit breaker c j Is not an operation value of R (c) j ) For all circuit-breakers c which can be actuated j Trip protection set r of (2) i To drive the circuit breaker c j I-th trip protection in trip protection set,/-th trip protection in trip protection set>Is a circuit breaker c j Is a remote signaling value.
Further, the matching process of the local feature fingerprint library in the step 3 by using the matching search algorithm includes: after the power grid fails, remote signaling data in intervals are collected and mapped to a three-dimensional space in a grouping and coding mode, fingerprint data with the smallest space distance and the smallest mismatch bit number of binary remote signaling data are found out from a local characteristic fingerprint library through a matching strategy, and a corresponding description formula is as follows:
wherein A (a, B, c) is three-dimensional data formed after encoding and mapping of grid fault interval remote signaling, B k (x, Y, z) is the space remote signaling coding mapping data in the kth local fingerprint database, Y' is the fault space remote signaling data, and Y (k) is the uncoded local fingerprint databaseIs a binary data of (a) in a memory.
Further, the matching process of the overall feature fingerprint library in the step 3 by using the matching search algorithm includes: performing bitwise exclusive OR operation on remote signaling data of corresponding elements in the overall characteristic fingerprint library by using remote signaling data triggered by failed elements after failure, and searching fingerprint data with minimum binary bit different numbers, wherein the corresponding description formula is as follows:
where r' (k) is remote signaling data of suspected fault elements in the power outage area after the power grid fault, r * (k) E is a matching result, and k and N are natural numbers for expected remote signaling data in the overall characteristic fingerprint database.
Further, the compiling of the fault identity in the step 3 includes preset fault fingerprint identity information and identity dynamic information, wherein the preset fault fingerprint identity information is fault information which can be determined in advance through a logic operation mode according to a coordination mode between protection of a relay protection principle and tripping logic after relay protection action after power grid faults, and the identity dynamic information is fault information which is output after analysis and calculation according to data of each system collected by each fault after power grid faults.
Compared with the prior art, the invention has the following advantages:
(1) The method forms a dispatching big data system by fusing a plurality of system data at a dispatching end, judges and corrects the validity of the data of each system by utilizing the advantages of the big data system, can identify bad data, diagnoses faults and defects of equipment and secondary circuits in each system, and searches hidden dangers in an automatic system.
(2) The method can also carry out deep use on the data of each system, and the panoramic information showing the faults accelerates the processing and recovery of the faults. Under the advantage of a large amount of data fused by multiple systems, the method has the advantages of being larger than the traditional fault diagnosis method, and the rapid and comprehensive diagnosis of faults is realized.
(3) According to the method, through matching of the local fingerprint library and the global fingerprint library, the similarity or matching value of the power grid fault remote signaling data received by the dispatching end and the data in the fingerprint library is found out, corresponding fault identity information is output as a final output result of fault diagnosis, and under the condition that the construction of a regulation large data platform is gradually and deeply regulated, the advantage of the large data can be fully utilized, so that the method has a good application prospect in the fault diagnosis of the power grid.
(4) The method constructs a dispatching fault fingerprint system of the power grid according to the interval remote signaling, the power grid element and the protection action remote signaling, establishes fault identity information in a multi-system fusion mode, serves as a diagnosis result of a fault fingerprint library, establishes a matching strategy of the fault fingerprint library, performs matching and searching on the fault fingerprint library, determines the fault identity of the power grid fault and performs panoramic display on the fault.
Drawings
FIG. 1 is an architecture diagram of fault fingerprint diagnostics provided by the present invention;
FIG. 2 is a power grid fault wiring diagram provided by the invention;
FIG. 3 is a diagram of a diagnostic process utilizing a dispatch fault fingerprint library provided by the present invention;
fig. 4 is a block diagram of an IEEE39 node system provided by the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
Along with the deep construction of a big data system at a dispatching end, the invention provides a method for constructing a dispatching fault fingerprint identification system to realize the fusion of the data of the multiple systems and the rapid diagnosis of the complex faults, aiming at the problems of the fusion of the data of the multiple systems, the deep use of the data, the rapid diagnosis and the treatment of the multiple complex faults of a power grid and the like. And constructing a dispatching data local feature fault fingerprint library by using remote signaling data of an inner space of a transformer substation, constructing a dispatching data overall feature fingerprint library by using remote signaling of primary equipment of a power grid and related protection and automatic devices and circuit breakers, and providing a matching search algorithm of the local feature fault fingerprint library and the overall feature fingerprint library of the fault fingerprint. And the fault data of each system are fused in a fault identity mode, and the diagnosis method of the power grid fault identity is rapidly determined through rapid matching of a fault fingerprint library. The validity of the diagnostic method proposed by the method of the present invention is illustrated by the IEEE39 node system.
As shown in fig. 1, the power grid fault diagnosis method based on multi-source data fusion of the scheduling fault fingerprint library comprises the following steps:
1) Acquiring data of a plurality of systems, such as SCADA, fault recording and PMU system data related to power grid operation and control, from a scheduling end; such as PMS, OMS systems, associated with production management, such as video surveillance, dispatch phone voice systems, associated with operation and accident handling, and associated with operating environments, such as lightning, weather systems, etc.
2) And establishing a fault local feature fingerprint library by using substation interval remote signaling data, and establishing a fault overall feature fingerprint library by using a power grid element.
3) And matching the local fingerprint library with the global fingerprint library, searching for similar or matched values of the power grid fault remote signaling data received by the scheduling end and the data in the fingerprint library, and outputting corresponding fault identity information as a final output result of fault diagnosis.
In the above method steps, in step 1), the building of the architecture for fault fingerprint diagnosis of the dispatching system requires building of a local fingerprint feature library of each interval inside the substation and a general fingerprint feature library of the power grid element. The operation of the transformer substation and the running mode of the power grid are determined, the fingerprint feature library is determined and is equivalent to the evaluation space of the power grid fault mode, each fault mode of the power grid is one fingerprint feature in the fingerprint feature library, and the reorganization feature value is obtained, namely the fault diagnosis result.
In the step 2) and the step 3), fault local feature fingerprint libraries are built by using substation interval remote signaling data, fault overall feature fingerprint libraries are built by using power grid elements, and two fingerprint libraries respectively use different building algorithms.
Establishing a code mapping function from remote signaling data to a fault space; defining and respectively representing logical multiplication and logical addition, and forming expected remote signaling values of the protection and circuit breaker according to protection logics such as element faults, main protection, backup protection, failure protection and the like, and taking the expected remote signaling values as data in a fault fingerprint overall feature library.
The local characteristic fingerprint library is established mainly aiming at all equipment in a transformer substation, and is divided into a main network local characteristic fingerprint library of 220kV and above and a distribution network local characteristic fingerprint library of 110kV and below according to different voltage levels; the global feature fingerprint library of the dispatching fault fingerprint library may need to perform logic operation on equipment elements and remote signaling data of a plurality of substations according to different fault types to form data in the fingerprint library; the data in the global fingerprint database are binary numbers, reflect the interrelation of the connection and protection actions between elements and the tripping of the circuit breaker under the conditions of single and complex faults, and reflect the global condition of the power grid.
The specific embodiment is as follows:
and (3) a step of: architecture for fault fingerprint diagnosis
1. Fault fingerprint integral architecture
The architecture for establishing fault fingerprint diagnosis of the dispatching system needs to establish a local fingerprint feature library of each interval in the transformer substation and an overall fingerprint feature library of the power grid element. The operation of the transformer substation and the running mode of the power grid are determined, the fingerprint feature library is determined and is equivalent to the evaluation space of the power grid fault mode, each fault mode of the power grid is one fingerprint feature in the fingerprint feature library, and the reorganization feature value is obtained, namely the fault diagnosis result.
With the deep construction of a large data regulation platform, a dispatching end can acquire data of a plurality of systems, such as SCADA, fault wave recording and PMU system data related to power grid operation and control, such as PMS and OMS systems related to production management, and related to operation and accident handling, such as video monitoring and dispatching telephone voice systems, and related to operation environments, such as lightning systems, weather systems and the like. The fault data in each system can be incorporated into a fault fingerprint diagnosis framework by constructing a fault identity and used as output information of fault panorama display.
In order to quickly find and match the fault fingerprint value with the highest similarity, a local feature matching algorithm and a whole feature recognition algorithm are respectively set, and after the power grid fails, the whole fingerprint library is searched to output the fault identity.
2. Establishment method of remote signaling data fault fingerprint library
(1) Establishment of fault fingerprint local characteristics
The remote signaling data fields are divided and encoded and then mapped to a multidimensional data space, which is mapped into:
wherein A is 1 ...A n For n matrices of post-fault remote signalling binary data, c 1 ...c n For n encoded fault encoded data, f 1 ...f n For n fault coding mapping functions, a two-dimensional coding mode and a one-dimensional coding mode are adopted for the coding modes of the remote signaling binary data matrix and the fault coding mapping functions after faults.
Through the above equation, the remote signaling of binary numbers into n-dimensional coding space (c 1 ,...,c n ) Therefore, the fault diagnosis problem using the remote signaling displacement data is converted into the classification problem of the sample data in the multidimensional space. In order to map to an n-dimensional coding space, a remote signaling binary data matrix a must be determined i And fault code mapping function f i That is, the coding mode of the remote signaling binary number is determined. Determination of A i 、f i There are two coding modes: two-dimensional coding scheme and one-dimensional coding scheme.
(2) Establishment of fault fingerprint overall characteristics
And after the element in the power grid fails, triggering the corresponding breaker to remove the failed element by the relevant relay protection action. Therefore, different relay protection actions and circuit breaker tripping remote signaling can be triggered after different elements fail. The fault fingerprint overall characteristic is mainly to establish an overall characteristic data space of all element faults and remote signaling data of trigger protection and circuit breakers in a power grid, and find out the closest group of data in the data space as a diagnosis result according to data matching of fault elements in a stop area and the triggered remote signaling data and the overall characteristic data space. The method for establishing the fault fingerprint overall feature library comprises the following steps:
definition of the definitionAnd->The logic multiplication and the logic addition are respectively represented, and expected remote signaling values of the protection and the circuit breaker are formed according to protection logics such as element faults, main protection, backup protection, failure protection and the like and are used as data in a fault fingerprint overall feature library. The following formula is shown:
wherein d k In the event of a failure of the component,desired telemetry value for main protection of faulty element, < >>Desired telemetry value for near backup protection of faulty element,>non-operational value for main protection of faulty element, < ->Desired telemetry value, Z (r ks ,d k ) Far backup protection range d for faulty element k A collection of adjacent elements, p (r ks ,d x ) To move along the power supply path from the failed component far backup protection installation to component d x All sets of circuit breakers>For the non-operational value of the t-th breaker in the set of all breakers, +.>Non-operational value for near-backup protection of faulty element,/->Is a circuit breaker c j Is claimed, +.>Is a circuit breaker c j Is not an operation value of R (c) j ) For all circuit-breakers c which can be actuated j Trip protection set r of (2) i To drive the circuit breaker c j I-th trip protection in trip protection set,/-th trip protection in trip protection set>Is a circuit breaker c j Is a remote signaling value.
(3) Establishment of power grid fault fingerprint library
And scheduling a fault fingerprint library local characteristic fingerprint library and a global characteristic fingerprint library. The establishment of the local characteristic fingerprint library is mainly carried out aiming at all devices in the substation. The method is divided into a main network local characteristic fingerprint library of 220kV and above and a distribution network local characteristic fingerprint library of 110kV and below according to different voltage levels. The device comprises a bus, a transformer, a circuit, a capacitor, a spare power automatic switching device and other elements, wherein each equipment element forms a sub-library of a local characteristic fingerprint library, and each sub-library contains codes of various fault remote signaling data. For example, the power transmission line sub-base contains remote signaling data such as main protection, backup protection, far jump, reclosing, PT disconnection and the like, and forms three field codes including fault process, protection action details and fault types, namely 159 kinds of the remote signaling data.
The global feature fingerprint library of the dispatching fault fingerprint library may need to perform logic operation on equipment elements and remote signaling data of a plurality of substations according to different fault types to form data in the fingerprint library. Such as single-stage breaker failure faults, binary data such as equipment in two substations and breaker failure remote signaling can be involved, and data in a bureau feature fingerprint library is formed through logic operation. Therefore, the data in the global fingerprint database are binary numbers, reflect the interrelation of the connection and protection actions between elements and the tripping of the circuit breaker under the conditions of single and complex faults, and reflect the global condition of the power grid.
3. Establishment of fault identity information
The fault identity is the output result of the diagnosis of the fault fingerprint. And matching the local fingerprint library with the global fingerprint library, searching for similar or matched values of the power grid fault remote signaling data received by the scheduling end and the data in the fingerprint library, and outputting corresponding fault identity information as a final output result of fault diagnosis. The fault identity fuses the original data and analysis result data related to the fault diagnosis of the system such as SCADA, PMU, fault wave recording, protection subsystem, PMS and the like. While the identity of the fault further comprises: accident voice information of dispatching telephone, image information of transformer substation equipment site, meteorological data information and the like.
The matching of the fault fingerprint is mainly to find out the fault identity information of the fault and output and display. The compiling of the fault identity comprises the following steps:
(1) Presetting fault fingerprint identity information
The preset fault fingerprint identity information refers to fault information which can be determined in advance through a logic operation mode according to a coordination mode between protection of a relay protection principle and a tripping strategy after relay protection action after power grid faults. For example, the device types include transformer, bus, line faults and the like, the fault types include single-phase, phase-to-phase and three-phase, the protection action process includes reclosing or not, main and backup protection action relation and the like, and the cascading fault types and the like.
(2) Identity dynamic information
The preset fault fingerprint identity information refers to the result which is output after analysis and calculation according to the data of each system collected by each fault, and cannot be determined in advance through a logic operation mode after the power grid has faults. For example, the fault distance of the PMU measurement system is calculated, the failure probability of the historical maintenance data and the fault data of the circuit breaker in the PMS system is calculated, the analysis and calculation of the real-time data of the power flow transfer are performed, the occurrence probability of the grid cascading faults is calculated, and the like. The data of the identity dynamic information can change according to different historical data and real-time fault data, and the data is filled in real time according to the calculated value of the data after the fault.
4. Matching strategy of fault fingerprint library
The matching strategy of the fault fingerprint library is to find the fault fingerprint of the power grid fault in the fingerprint local feature library and the global feature library and determine the identity of the fault.
The remote signaling data of the protection, reclosing and circuit breaker in a single interval are grouped according to the relay protection action and tripping logic, coded and mapped to a multi-bit data space to form a local characteristic fingerprint library. The matching of the fingerprint local feature library is to find out the closest fingerprint data in the fingerprint library under the condition that the power grid fault remote signaling is likely to be in error. Taking a power transmission line as an example, remote signaling data of equipment in an interval is divided into: the three groups of fault process words, protection action words and fault type words are coded and mapped into a three-dimensional space, and the matching strategy is as follows:
wherein A (a, B, c) is three-dimensional data formed after encoding and mapping of grid fault interval remote signaling, B k (x, Y, z) is the interval telemetry signaling code mapping data in the kth local fingerprint database, Y' is the fault interval telemetry signaling data, Y (k) is the uncoded binary data in the local fingerprint database,representing a logical exclusive or operation.
According to the formula, remote signaling data in intervals are collected after the power grid fails, block codes are mapped to a three-dimensional space, and the group of fingerprint data with the closest space distance and the minimum number of mismatch digits of binary remote signaling data is found out from a local fingerprint library through a matching strategy.
From the foregoing, it can be seen that the overall feature fingerprint database actually forms a combination manner of multiple remote signaling data according to protection actions and trip policy logic after assuming that the power grid element fails. The matching strategy of the overall characteristic fingerprint is to find out the closest combination mode in the fingerprint library as the overall characteristic fingerprint of the fault remote signaling according to the fault remote signaling. The matching formula is as follows:
where r' (k) is remote signaling data of suspected fault elements in the power outage area after the power grid fault, r * (k) E is a matching result, and k and N are natural numbers for expected remote signaling data in the overall characteristic fingerprint database.
The formula performs bitwise exclusive OR operation on remote signaling data of corresponding elements in the fingerprint library by using remote signaling data triggered by failed elements, and searches the group of fingerprint data with the least binary digits.
Therefore, the identity of the power grid fault can be determined and the result can be output through the search of the matching algorithm of the local feature library and the global feature library.
2. Practical cases
A method for diagnosing a grid fault using a dispatch fingerprint library will be described with reference to the grid wiring diagram shown in fig. 2. In fig. 2, the L1 transmission line fails to perform line protection, and the tripping circuit breakers B2 and B1 cut off the fault due to the failure protection of M1 caused by the failure of the B3 circuit breaker.
The diagnosis method of the dispatching fault fingerprint is shown in fig. 3, all fault remote signaling data of the fault and remote signaling data of the tripping transformer substation after the fault are captured, and then power failure area searching is carried out by using the remote signaling data after the fault. A trip circuit breaker with a voltage on one side and no voltage on one side is defined as a boundary circuit breaker. In the network topology, the area enclosed by the boundary circuit breaker, which contains the components and the switching circuit breaker, is a blackout area.
In the power failure area searching process, the boundary circuit breaker is used as a searching starting point to search, in the searching topological wiring diagram, the boundary circuit breaker is searched in a certain direction, namely, the searching is stopped, other directions are searched until the searching in all directions is completed, and finally, the boundary circuit breaker and the suspected fault element list are formed. All remote signaling data of the boundary circuit breaker are separated to form a local fault characteristic fingerprint library, the local fault characteristic library is matched, fingerprints formed by suspected fault elements are matched in an overall fault fingerprint library, the fault identity is finally determined, and multi-system fused fault identity information is output.
The matching modes of the fault fingerprint system are divided into two modes, namely local characteristic fingerprint matching and overall fingerprint matching, wherein the local characteristic fingerprint matching mainly aims at fingerprint matching of boundary breaker intervals, and the overall fingerprint matching mainly aims at fingerprint matching of equipment and protection and circuit breakers among multiple intervals.
3. Simulation verification result
To further verify the versatility and effectiveness of the methods herein, an example simulation analysis was performed using the IEEE39 node system shown in FIG. 4. For ease of analysis, the lines, circuit breakers and protections are all marked according to the bus numbers, for example, the line connecting bus B18 and B17 is marked L1817, the circuit breaker on the B18 side is marked CB1817, the circuit breaker on the opposite side is marked CB1718, the line main protections on both sides are marked L1817m and L1718m, respectively, and the main protection of bus B18 is marked RB18m. Considering the situation of various fault types such as error of alarm information, refusal of protection and circuit breaker, false operation and the like, the fault fingerprint diagnosis result is shown in table 1. Each remote signaling data is represented by 2-dimensional data, e.g., (R0203 s, 6) for L0203 line protection far backup protection actions, SOE time scale 6ms.
Two complex faults in the IEEE39 node power grid system are respectively simulated and analyzed by the fault barrier fingerprint identification system which is integrated with the multi-system analysis data and is provided in the specification, and the analysis data are shown in table 1. The fault 1 is: the B03 busbar has two-phase grounding faults, and the circuit breaker refuses to act, so that the remote backup actions of L0318, L0304 and L0302 are carried out, and finally the faults are cut off. The fault 2 is: the buses B03 and B14 fail, and CB0318 refuses to operate, and the faults are removed by the backup action far away from L0318.
The matching modes of the fault fingerprint system are divided into two modes, namely local characteristic fingerprint matching and overall fingerprint matching, wherein the local characteristic fingerprint matching mainly aims at fingerprint matching of boundary breaker intervals, and the overall fingerprint matching mainly aims at fingerprint matching of equipment and protection and circuit breakers among multiple intervals. In the simulation process, a power failure area search algorithm is utilized to find out an edge breaker and a suspected fault element, a local feature fingerprint and an overall feature fingerprint are formed through encoding, then a fault fingerprint library is matched through a matching algorithm, all matching values E of a fault 1 are 0, complete matching is achieved, and in a fault 2, the other matching values are 0 except the overall feature fingerprint is 1. The ensemble feature matching algorithm finds a set of unequal but closest fingerprint code values. And finally, outputting the fused fault identity values of the systems, and the detailed analysis results are shown in table 1.
Therefore, the corresponding fault identity can be found through the dispatching fault fingerprint identification system, analysis contents of other multiple systems are fused, and panoramic fault analysis data is displayed on a dispatching side.
Table 1: scheduling fault fingerprint diagnostic results
While the invention has been described with reference to certain preferred embodiments, it will be understood by those skilled in the art that various changes and substitutions of equivalents may be made and equivalents will be apparent to those skilled in the art without departing from the scope of the invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (7)
1. A power grid fault diagnosis method based on multi-source data fusion of a scheduling fault fingerprint library is characterized by comprising the following steps:
step 1: acquiring data of a plurality of systems from a scheduling end;
step 2: establishing a fault local feature fingerprint library by using substation interval remote signaling data, and establishing a fault overall feature fingerprint library by using a power grid element;
step 3: matching the local characteristic fingerprint library and the overall characteristic fingerprint library by using a matching search algorithm, searching for similar or matching values of power grid fault remote signaling data received by a scheduling end and data in the fingerprint library, and outputting corresponding fault identity information as a final output result of fault diagnosis;
the process of establishing the fault overall characteristic fingerprint library in the step 2 specifically comprises the following steps: defining logical multiplication and logical addition, and forming expected remote signaling values of the protection and circuit breaker according to protection logics of element faults, main protection, backup protection and failure protection, wherein the expected remote signaling values are used as data in a fault fingerprint overall characteristic fingerprint library;
the description formula of the data in the fault fingerprint total characteristic fingerprint library is as follows:
wherein d k In the event of a failure of the component,desired telemetry value for main protection of faulty element, < >>Desired telemetry value for near backup protection of faulty element,>non-operational value for main protection of faulty element, < ->Desired telemetry value, Z (r ks ,d k ) Far backup protection range d for faulty element k A collection of adjacent elements, p (r ks ,d x ) To follow the power supply path from the fault element far backup protection installation toElement d x All sets of circuit breakers>For the non-operational value of the t-th breaker in the set of all breakers, +.>Non-operational value for near-backup protection of faulty element,/->Is a circuit breaker c j Is claimed, +.>Is a circuit breaker c j Is not an operation value of R (c) j ) For all circuit-breakers c which can be actuated j Trip protection set r of (2) i To drive the circuit breaker c j I-th trip protection in trip protection set,/-th trip protection in trip protection set>Is a circuit breaker c j Is a remote signaling value.
2. The method for diagnosing a power grid fault by multi-source data fusion based on a scheduled fault fingerprint library according to claim 1, wherein the data of the systems in the step 1 comprises SCADA, fault logging and PMU system data related to power grid operation and control; PMS, OMS system data relating to production management; video surveillance, dispatch telephony voice system data associated with operation and incident handling.
3. The power grid fault diagnosis method based on multi-source data fusion of the scheduling fault fingerprint library according to claim 1, wherein the process of establishing the fault local feature fingerprint library in the step 2 specifically comprises the following steps: dividing and encoding remote signaling data fields, mapping to a multidimensional data space according to a mapping transformation relation, and converting the fault diagnosis problem of remote signaling deflection data into the classification problem of sample data in the multidimensional space.
4. The power grid fault diagnosis method based on multi-source data fusion of the scheduling fault fingerprint library according to claim 3, wherein the mapping transformation relation is described as the following formula:
wherein A is 1 ...A n For n matrices of post-fault remote signalling binary data, c 1 ...c n For n encoded fault encoded data, f 1 ...f n For n fault coding mapping functions, a two-dimensional coding mode and a one-dimensional coding mode are adopted for the coding modes of the remote signaling binary data matrix and the fault coding mapping functions after faults.
5. The power grid fault diagnosis method based on multi-source data fusion of the scheduled fault fingerprint library according to claim 1, wherein the matching process of the local feature fingerprint library by using the matching search algorithm in the step 3 comprises the following steps: after the power grid fails, remote signaling data in intervals are collected and mapped to a three-dimensional space in a grouping and coding mode, fingerprint data with the smallest space distance and the smallest mismatch bit number of binary remote signaling data are found out from a local characteristic fingerprint library through a matching strategy, and a corresponding description formula is as follows:
wherein A (a, B, c) is three-dimensional data formed after encoding and mapping of grid fault interval remote signaling, B k (x, Y, z) is the space remote signaling code mapping data in the kth local fingerprint database, Y' is the fault space remote signaling data, Y (k) is the uncoded local fingerprint databaseBinary data.
6. The power grid fault diagnosis method based on multi-source data fusion of the scheduled fault fingerprint library according to claim 1, wherein the matching process of the overall feature fingerprint library by using the matching search algorithm in the step 3 comprises the following steps: performing bitwise exclusive OR operation on remote signaling data of corresponding elements in the overall characteristic fingerprint library by using remote signaling data triggered by failed elements after failure, and searching fingerprint data with minimum binary bit different numbers, wherein the corresponding description formula is as follows:
where r' (k) is remote signaling data of suspected fault elements in the power outage area after the power grid fault, r * (k) E is a matching result, and k and N are natural numbers for expected remote signaling data in the overall characteristic fingerprint database.
7. The method for diagnosing the power grid fault by the multi-source data fusion based on the dispatching fault fingerprint library according to claim 1, wherein the compiling of the fault identities in the step 3 comprises preset fault fingerprint identity information and identity dynamic information, wherein the preset fault fingerprint identity information is fault information which can be determined in advance through a logical operation mode according to a coordination mode between protection of a relay protection principle and trip logic after relay protection action after power grid faults, and the identity dynamic information is fault information which is output after power grid faults through analysis and calculation according to data of each system collected by each fault.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110154576.3A CN112964960B (en) | 2021-02-04 | 2021-02-04 | Power grid fault diagnosis method based on multi-source data fusion of scheduling fault fingerprint library |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110154576.3A CN112964960B (en) | 2021-02-04 | 2021-02-04 | Power grid fault diagnosis method based on multi-source data fusion of scheduling fault fingerprint library |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112964960A CN112964960A (en) | 2021-06-15 |
| CN112964960B true CN112964960B (en) | 2024-04-12 |
Family
ID=76273819
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110154576.3A Active CN112964960B (en) | 2021-02-04 | 2021-02-04 | Power grid fault diagnosis method based on multi-source data fusion of scheduling fault fingerprint library |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112964960B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113671286A (en) * | 2021-08-09 | 2021-11-19 | 国网河北省电力有限公司检修分公司 | System and method for executing and testing safety measures of relay protection automatic device of intelligent station |
| CN114050613B (en) * | 2021-11-29 | 2023-10-27 | 国网湖南省电力有限公司 | Online identification and tracing method and system for power grid voltage transient event |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105093063A (en) * | 2015-08-06 | 2015-11-25 | 国家电网公司 | Online power grid fault diagnosis method based on multisource data characteristic unit combination judgment |
| CN108596449A (en) * | 2018-04-09 | 2018-09-28 | 南京邮电大学 | It is a kind of to consider distribution network reliability prediction technique of the weather to distribution network failure impact probability |
| CN109768877A (en) * | 2018-11-23 | 2019-05-17 | 国网上海市电力公司 | A kind of electric network failure diagnosis method based on space optimum code collection and DHNN error correction |
| CN110348114A (en) * | 2019-07-09 | 2019-10-18 | 电子科技大学 | A kind of non-precision fault recognition method of power grid completeness status information reconstruct |
| CN111366814A (en) * | 2020-03-31 | 2020-07-03 | 上海电力大学 | Power grid fault diagnosis method based on multi-source data and multi-dimensional fault coding space |
| CN111479231A (en) * | 2020-04-17 | 2020-07-31 | 西安交通大学 | Indoor fingerprint positioning method for millimeter wave large-scale MIMO system |
| CN111521905A (en) * | 2020-03-30 | 2020-08-11 | 国网上海市电力公司 | Power grid fault diagnosis method based on intelligent fusion of multi-dimensional data fault degrees |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107884707B (en) * | 2017-09-30 | 2018-12-21 | 杨启蓓 | High-voltage circuit-breaker switching on-off time on-Line Monitor Device and method |
-
2021
- 2021-02-04 CN CN202110154576.3A patent/CN112964960B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105093063A (en) * | 2015-08-06 | 2015-11-25 | 国家电网公司 | Online power grid fault diagnosis method based on multisource data characteristic unit combination judgment |
| CN108596449A (en) * | 2018-04-09 | 2018-09-28 | 南京邮电大学 | It is a kind of to consider distribution network reliability prediction technique of the weather to distribution network failure impact probability |
| CN109768877A (en) * | 2018-11-23 | 2019-05-17 | 国网上海市电力公司 | A kind of electric network failure diagnosis method based on space optimum code collection and DHNN error correction |
| CN110348114A (en) * | 2019-07-09 | 2019-10-18 | 电子科技大学 | A kind of non-precision fault recognition method of power grid completeness status information reconstruct |
| CN111521905A (en) * | 2020-03-30 | 2020-08-11 | 国网上海市电力公司 | Power grid fault diagnosis method based on intelligent fusion of multi-dimensional data fault degrees |
| CN111366814A (en) * | 2020-03-31 | 2020-07-03 | 上海电力大学 | Power grid fault diagnosis method based on multi-source data and multi-dimensional fault coding space |
| CN111479231A (en) * | 2020-04-17 | 2020-07-31 | 西安交通大学 | Indoor fingerprint positioning method for millimeter wave large-scale MIMO system |
Non-Patent Citations (2)
| Title |
|---|
| Research on intelligent diagnosis method of power grid fault components based on fault fingerprint technology;Xiao Fei;《2018 2nd IEEE Conference on Energy Internet and Energy System Integration (EI2)》;全文 * |
| 基于遥信数据故障编码技术及DHNN校正的电网故障诊断方法;肖飞;《电力系统保护与控制》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112964960A (en) | 2021-06-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106019084B (en) | Medium-voltage power distribution network disconnection fault diagnosis method based on power distribution and utilization data correlation | |
| CN112964960B (en) | Power grid fault diagnosis method based on multi-source data fusion of scheduling fault fingerprint library | |
| CN110807550B (en) | Distribution transformer overload recognition and early warning method based on neural network and terminal equipment | |
| CN113189451A (en) | Power distribution network fault positioning studying and judging method, system, computer equipment and storage medium | |
| CN109768877B (en) | Power grid fault diagnosis method based on space optimal coding set and DHNN error correction | |
| CN114254818B (en) | Low-voltage distribution network cascading failure early warning method based on risk assessment model | |
| CN109633369B (en) | Power grid fault diagnosis method based on multi-dimensional data similarity matching | |
| CN112561251B (en) | Power distribution network abnormal point detection method and device, computer equipment and storage medium | |
| CN109633370B (en) | Power grid fault diagnosis method based on fault information coding and fusion method | |
| CN103871004A (en) | Power distribution network failure cause analyzing method based on expert system and D-S evidence theory | |
| CN110095661B (en) | Distribution transformer high-voltage side open-phase fault first-aid repair method | |
| CN111366814A (en) | Power grid fault diagnosis method based on multi-source data and multi-dimensional fault coding space | |
| CN104616092A (en) | Distributed log analysis based distributed mode handling method | |
| CN110348690B (en) | Tree search based result query menu grid accident auxiliary decision making system and method | |
| CN111999591B (en) | Method for identifying abnormal state of primary equipment of power distribution network | |
| CN108399579B (en) | Intelligent analysis system for monitoring data signals of substation equipment | |
| CN108090674A (en) | The risk assessment of the area power grid method of operation and aid decision-making method and system | |
| Wang et al. | Novel hierarchical fault diagnosis approach for smart power grid with information fusion of multi-data resources based on fuzzy petri net | |
| CN106526465A (en) | High-voltage circuit breaker fault intelligent diagnosis method based on improved fuzzy Petri network | |
| Zhang et al. | Research on power grid fault diagnosis based on a quantitative representation of alarm information | |
| Ashouri et al. | A new approach for fault detection in digital relays-based power system using Petri nets | |
| Fei et al. | Research on intelligent diagnosis method of power grid fault components based on fault fingerprint technology | |
| CN107682232A (en) | A kind of electrical network business stream example code method | |
| Zong et al. | A method of fault diagnosis for secondary loop in intelligent substation based on Bayesian algorithm | |
| CN113569904B (en) | Bus wiring type identification method, system, storage medium and computing device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |