CN110412428B

CN110412428B - Power distribution network time representation method based on time sequence constraint network

Info

Publication number: CN110412428B
Application number: CN201910820659.4A
Authority: CN
Inventors: 李鹏; 袁智勇; 于力; 徐全; 白浩; 汪悦颀; 焦在滨
Original assignee: Xian Jiaotong University; Research Institute of Southern Power Grid Co Ltd
Current assignee: Xian Jiaotong University; Research Institute of Southern Power Grid Co Ltd
Priority date: 2019-08-29
Filing date: 2019-08-29
Publication date: 2020-08-04
Anticipated expiration: 2039-08-29
Also published as: CN110412428A

Abstract

The invention discloses a power distribution network time representation method based on a time sequence constraint network. According to the method, the precedence relationship between time points is described by combining the interval algebraic theory, the expression of fault criteria and the description of fault situations are obtained, and the problems that the existing power distribution network fault diagnosis method based on fault time sequence information is inaccurate in diagnosis result due to the fact that time represents too accurately and the redundancy of the time sequence information is too low are solved; the method can quickly and accurately realize fault diagnosis, and greatly improves the reliability and safety of system operation.

Description

Power distribution network time representation method based on time sequence constraint network

Technical Field

The invention belongs to the field of time sequence constraint network application, and particularly relates to a power distribution network time representation method based on a time sequence constraint network.

Background

The SCADA system is used as a data acquisition and monitoring control system which is most widely applied in a power system and has the most mature technical development, and after a fault occurs in a power distribution network, the collected related protection breaker information is uploaded to a dispatching center in a form of soe (sequence event) record. Monitoring personnel need to complete fault diagnosis rapidly through the uploaded information in a short time, find abnormal conditions of equipment in time and improve the safety and reliability of the power distribution network. However, the existing information with time scales in the power distribution network is accurate to the moment, and although the accuracy seems to be high, the problem of measurement and calculation errors exists in the use, so that the accuracy of the diagnosis result obtained by the power distribution network fault diagnosis method based on the fault time sequence information is not high.

The time point constraint and time distance constraint representation method in the time sequence constraint network widens the definition of time and time period, and different time representation methods can be obtained according to different time characteristics in the power distribution network. The interval algebra theory of Allen describes the time sequence relationship between time intervals by an interval algebra method. The method defines 13 non-intersecting and joint complete (JEPD) temporal relations by using the relation between 2 time interval endpoints, completely covers all relations between time points and time distances, and is beneficial to the expression of fault criteria and the description of fault situations in fault diagnosis.

In the analysis, the existing power distribution network fault diagnosis method based on the fault time sequence information has the problems that time representation is too accurate, and the redundancy of the time sequence information is too low, so that the diagnosis result is inaccurate. Therefore, in consideration of the time characteristics in the power distribution network, it is necessary to invent a method for representing the time of the power distribution network with higher redundancy.

Disclosure of Invention

The invention aims to provide a power distribution network time representation method based on a time sequence constraint network, which aims to overcome the defects in the prior art and can realize that when a power distribution network fails, time point constraint and time distance constraint in the time sequence constraint network are utilized to replace time and time periods in time scale information of the power distribution network, so as to obtain a new time representation method; the invention can improve the accuracy of fault diagnosis of the power distribution network and has important practical significance in the aspects of shortening power failure time, reducing economic loss and the like.

In order to achieve the purpose, the invention adopts the following technical scheme:

a power distribution network time representation method based on a time sequence constraint network comprises the following steps:

step 1, determining the time type of known time in a power distribution network;

step 2, respectively representing by time points and time distances according to different time types;

step 3, determining the relation between the time needing to be expressed and the known time;

step 4, writing the time required to be expressed in the power distribution network into a time expression form based on a time sequence network according to 3 operations of time point and time distance constraints;

step 5, judging whether the relation between the time in the power distribution network needs to be judged;

step 6, if the relation between the time in the power distribution network needs to be judged, extracting the time point needing to be judged; otherwise, the following steps are not needed to be continued;

step 7, determining the satisfied temporal relation among the time points;

and 8, judging whether the obtained time points are equal according to the temporal relation met between the time points.

Further, the time types in the power distribution network in step 1 are divided into time instants and time periods.

Further, the time points and the time distances in step 2 are represented as follows:

the time points may be divided into a definite time point and an indefinite time point. The indeterminate time t is a variable defining a time interval t (t) ═ t^-，t⁺]Which represents a constraint of the uncertain point in time t, i.e. t ∈ T (t); t^-And t⁺Respectively, the start point and the end point of T (t). When t is^-＝t⁺Then, t is a certain time point.

The temporal distance may be divided into a deterministic temporal distance and an indeterminate temporal distance. Uncertainty time distance d (t)_i，t_j) Is a variable, define

Representing the uncertainty time distance d (t)_i，t_j) Constraint of (2), i.e. d (t)_i,t_j)∈D(t_i,t_j)；

And

respectively represent the interval D (t)_i，t_j) The start point and the end point of (c). When in use

When, d (t)_i,t_j) Is a determined time distance.

Further, the 3 operations of time point and time distance constraint in step 4 are as follows:

(1) and (3) solving the constraint of the subsequent event time point: knowing T (T)_i) And D (t)_i,t_j) Calculating the time point constraint T (T) of the subsequent event j of the event i_j). From t_j＝t_i+d(t_i,t_j) The following can be obtained:

(2) and (3) solving the constraint of precursor event time points: knowing T (T)_j) And D (t)_i,t_j) Solving the time point constraint T (T) of the precursor event i of the event j_i). From t_i＝t_j-d(t_i,t_j) The following can be obtained:

(3) superposition of time-distance constraints: the number of events i, j,k is respectively at t_i,t_j,t_k(t_i≤t_j≤t_K) Occurring one after the other, D (t) being known_i,t_j) And D (t)_j，t_k) Solving for the time point t_iAnd t_kTime distance d (t) therebetween_i，t_k) Of (3) is performed. From d (t)_i,t_k)＝d(t_i,t_j)+d(t_j,t_k) The following can be obtained:

further, 13 kinds of interval algebras of Allen in step 7 are mutually exclusive and have a joint completion (JEPD) temporal relationship as follows:

further, the new definition of the time points equal in step 8 is as follows:

if T₁And T₂Any one of the following conditions is satisfied: during (T)₁，T₂),Contains(T₁，T₂),Overlaps(T₁，T₂),Overlapped-by(T₁，T₂),Starts(T₁，T₂),Started-by(T₁，T₂),Finishes(T₁，T₂),Finished-by(T₁，T₂),Equals(T₁，T₂) We call T₁＝T₂(ii) a Otherwise, the two time points are not equal.

Compared with the prior art, the invention has the following beneficial technical effects:

the method utilizes time point constraint and time distance constraint in a time sequence constraint network to replace time and time periods in time mark information of the power distribution network to obtain a new time representation method; the invention describes the precedence relationship between time points by combining the interval algebra theory, obtains the expression of fault criteria and the description of fault situations, solves the problems that the existing power distribution network fault diagnosis method based on fault time sequence information is inaccurate in diagnosis result due to too accurate time representation and too low redundancy of time sequence information, can quickly and accurately realize fault diagnosis, and greatly increases the reliability and safety of system operation.

Drawings

FIG. 1 is a topological diagram of a standard 13-node ungrounded neutral distribution network;

FIG. 2 is a diagram of a power distribution network simulation circuit;

fig. 3 is a flow chart of a method for representing time of a power distribution network according to the present invention.

Detailed Description

The following describes the implementation of the present invention in further detail with reference to the accompanying drawings:

the invention relates to a power distribution network time representation method based on a time sequence constraint network, which specifically comprises the following steps:

firstly, a power distribution network simulation circuit diagram as shown in fig. 2 is established based on a standard 13-node neutral point ungrounded power distribution network topological diagram as shown in fig. 1, the length of the circuit is marked in the diagram, the transformer and the load parameters are shown in table 1, and each section of the circuit is provided with complete three-section type current protection. Faults are set simultaneously on the lines 6, 7, 9 in the distribution network. Determining known protection action model time parameters in the power distribution network: for the

end lines

6, 8, 10 and 11, the action time limit of the current I section is 30 ms; for the lines 7 and 9, the action time limit of the current I section is 40ms, the action time limit of the current II section is 0.5s, and the action time limit of the current III section is 2 s; for the circuit 5, the action time limit of the current I section is 50ms, the action time limit of the current II section is 0.6s, and the action time limit of the current III section is 2.1 s; the action time limit of the circuit breaker is 40 ms; known breaker information: the line 6 breaker trips at time 1.091s, the line 7 breaker trips at time 1.100s, and the line 9 breaker trips at time 1.561 s.

TABLE 1 Transformer and load parameters

Secondly, according to the flow chart of the time representation method of the power distribution network as shown in fig. 3, firstly, time points and time distances are respectively represented according to different time types: for the

end lines

6, 8, 10 and 11, the current I section has an action time limit [30,50] ms; for the circuits 7 and 9, the action time limit of the current I section is [40,60] ms, the action time limit of the current II section is [480,520] ms, and the action time limit of the current III section is [1950,2050] ms; for the circuit 5, the action time limit of the current I section is [50,70] ms, the action time limit of the current II section is [580,620] ms, and the action time limit of the current III section is [2050,2150] ms; the action time limit of the circuit breaker is [40,60] ms, the circuit breaker of the line 6 is tripped at the time point [1091,1091] ms, the circuit breaker of the line 7 is tripped at the time point [1100,1100] ms, and the circuit breaker of the line 9 is tripped at the time point [1561,1561] ms.

Thirdly, to obtain the association relation among multiple faults, the moment when the fault occurs needs to be obtained in the diagnosis criterion, the current known time is the tripping time point of the circuit breaker, and the relation between the time needing to be expressed and the known time is determined: the time when the fault occurs is the time of fault tripping, the action time limit of the breaker and the protection action time limit.

Writing the time required to be expressed in the power distribution network into a time expression form based on a time sequence network according to 3 operations of time point and time distance constraints;

the 3 operations of the time point and time distance constraint are as follows:

(1) and (3) solving the constraint of the subsequent event time point: knowing T (T)_i) And D (t)_i，t_j) Calculating the time point constraint T (T) of the subsequent event j of the event i_j). From t_j＝t_i+d(t_i,t_j) The following can be obtained:

(2) and (3) solving the constraint of precursor event time points: knowing T (T)_j) And D (t)_i，t_j) Solving the time point constraint T (T) of the precursor event i of the event j_i). From t_i＝t_j-d(t_i,t_j) The following can be obtained:

(3) superposition of time-distance constraints: event i, j, k are at t_i，t_j，t_k(t_i≤t_j≤t_k) Occurring one after the other, D (t) being known_i，t_j) And D (t)_j，t_k) Solving for the time point t_iAnd t_kTime distance d (t) therebetween_i，t_k) Of (3) is performed. From d (t)_i,t_k)＝d(t_i,t_j)+d(t_j,t_k) The following can be obtained:

and fifthly, judging the incidence relation of the multiple faults and judging the relation between the fault time of each fault in the power distribution network.

Extracting time points needing to be judged, namely the fault occurrence time of each fault: point in time of occurrence of fault

Time point of occurrence of fault two

Time point of occurrence of fault three

Seventhly, determining a temporal relation satisfied between the time points: overlapped-by (T (T)_s1)，T(t_s2))，Starts(T(t_s2)，T(t_s3))；

Eighthly, judging whether the obtained time points are equal: t (T)_s1)＝T(t_s2)，T(t_s2)＝T(t_s3)；

And ninthly, bringing the time points into a criterion of successive faults to obtain a final diagnosis result: all three faults are consecutive faults with a common fault source, as set.

Analyzing the simulation diagnosis result to obtain: the method can improve the diagnosis criterion on the basis of the original diagnosis method and more accurately diagnose the incidence relation between the faults.

Claims

1. A power distribution network time representation method based on a time sequence constraint network is characterized by comprising the following steps:

step 2, respectively representing different time types by using time points and time distances;

step 3, determining a constraint relation between the target time and the known time;

step 4, writing the target time in the power distribution network into a time representation form based on a time sequence network according to the operation of time point and time distance constraint;

step 6, if the relation between the time in the power distribution network needs to be judged, extracting the time point needing to be judged, and entering step 7; otherwise, ending;

step 7, determining the satisfied temporal relation among the time points;

2. The method for representing the time of the power distribution network based on the timing constraint network as claimed in claim 1, wherein the time types in the power distribution network in step 1 are divided into time moments and time periods.

3. The method for representing time of the power distribution network based on the time sequence constraint network as claimed in claim 1, wherein the time points and the time distances in step 2 are represented as follows:

the time points are divided into a determined time point and an uncertain time point, the uncertain time point t is a variable, and a time interval T (t) is defined as [ t ]^-，t⁺]Which represents a constraint of the uncertain point in time t, i.e. t ∈ T (t); t^-And t⁺Respectively representing the starting point and the end point of T (t), when t is^-＝t⁺When, t is a determined time point;

the time distance is divided into a definite time distance and an indefinite time distance, the indefinite time distance d (t)_i，t_j) Is a variable, define

Representing the uncertainty time distance d (t)_i，t_j) Constraint of (2), i.e. d (t)_i，t_j)∈D(t_i，t_j)；

And

respectively represent the interval D (t)_i，t_j) When starting and ending, when

When, d (t)_i，t_j) Is a determined time distance.

4. The method for representing time of the power distribution network based on the time sequence constraint network as claimed in claim 3, wherein the operation of time point and time distance constraint in step 4 is as follows:

(1) and (3) solving the constraint of the subsequent event time point: knowing T (T)_i) And D (t)_i，t_j) Calculating the time point constraint T (T) of the subsequent event j of the event i_j) From t_j＝t_i+d(t_i，t_j) Obtaining:

(2) and (3) solving the constraint of precursor event time points: knowing T (T)_j) And D (t)_i，t_j) Solving the time point constraint T (T) of the precursor event i of the event j_i) From t_i＝t_j-d(t_i，t_j) Obtaining:

(3) superposition of time-distance constraints: event i, j, k are at t_i，t_j，t_k(t_i≤t_j≤t_k) Occurring one after the other, D (t) being known_i，t_j) And D (t)_j，t_k) Solving for the time point t_iAnd t_kTime distance d (t) therebetween_i，t_k) Is constrained by d (t)_i，t_k)＝d(t_i，t_j)+d(t_j，t_k) Obtaining:

5. the method according to claim 1, wherein the temporal relationship satisfied between the time points in step 7 is represented as follows:

6. the method according to claim 5, wherein the time points in step 8 are defined as follows:

if T₁And T₂Any one of the following temporal relationships is satisfied: during (T)₁，T₂)，Contains(T₁，T₂)，Overlaps(T₁，T₂)，Overlapped-by(T₁，T₂)，Starts(T₁，T₂)，Started-by(T₁，T₂)，Finishes(T₁，T₂)，Finished-by(T₁，T₂)，Equals(T₁，T₂) Then called T₁＝T₂(ii) a Otherwise, the two time points are not equal.