阅读说明：本技术 一种低压有源配电网故障定位方法 (Low-voltage active power distribution network fault positioning method ) 是由 吴臻 苏毅方 巫水萍 郑志祥 徐凤乾 冯洋 乐健 吕慧珍 梁路 徐雷文 徐艺 于 2019-08-22 设计创作，主要内容包括：本发明公开了一种低压有源配电网故障定位方法,包括如下步骤：S1：故障信息采集；S2：建立畸变信息模型；S3：建立分层等效模型；S4：输出结果,所述的步骤S1中,在检测线路分段开关或联络开关处安装馈线终端单元,馈线终端单元搜集故障信息并上传至数据采集与监视控制系统,数据采集与监视控制系统中建立有畸变信息模型和分层等效模型。所述步骤S2中,畸变信息模型的建立遵循如下步骤：S21：故障检测信息编码；S22：畸变信息特性分类；S23：确定决策变量；S24：确定目标函数；S25：确定开关函数。该方法在信息发生畸变时,仍然能够定位故障区段,得出畸变类型和畸变位置,有效的保证信息发生畸变时低压配电网故障定位的准确率和容错度。(The invention discloses a fault positioning method for a low-voltage active power distribution network, which comprises the following steps of: s1: collecting fault information; s2: establishing a distortion information model; s3: establishing a layered equivalent model; s4: and outputting the result, wherein in the step S1, a feeder terminal unit is installed at the detection line section switch or the interconnection switch, the feeder terminal unit collects the fault information and uploads the fault information to the data acquisition and monitoring control system, and a distorted information model and a layered equivalent model are established in the data acquisition and monitoring control system. In step S2, the distortion information model is built according to the following steps: s21: encoding fault detection information; s22: classifying distortion information characteristics; s23: determining a decision variable; s24: determining an objective function; s25: a switching function is determined. According to the method, when the information is distorted, the fault section can still be positioned, the distortion type and the distortion position are obtained, and the accuracy and the fault tolerance of the fault positioning of the low-voltage distribution network when the information is distorted are effectively guaranteed.)

1. A fault positioning method for a low-voltage active power distribution network is characterized by comprising the following steps: the method comprises the following steps:

s1: collecting fault information;

s2: establishing a distortion information model;

s3: establishing a layered equivalent model;

s4: and outputting the result.

2. The method for locating the fault of the low-voltage active power distribution network according to claim 1, characterized in that: in the step S1, a feeder terminal unit is installed at the detection line section switch or the interconnection switch, the feeder terminal unit collects fault information and uploads the fault information to the data acquisition and monitoring control system, and a distorted information model and a layered equivalent model are established in the data acquisition and monitoring control system.

3. The method for locating the fault of the low-voltage active power distribution network according to claim 1, characterized in that: in step S2, the distortion information model is built according to the following steps:

s21: encoding fault detection information;

s22: classifying distortion information characteristics;

s23: determining a decision variable;

s24: determining an objective function;

s25: a switching function is determined.

4. A method for fault location in a low voltage active power distribution network according to claim 3, characterized in that: in step S21, the fault detection information encoding follows the following rules:

wherein M is_jNumbering feeder terminal units, I_jThe information is real-time monitoring information of the jth monitoring node.

5. A method for fault location in a low voltage active power distribution network according to claim 3, characterized in that: the distortion information characteristics are classified as follows:

class 1: i is_j＝0，D_j ¹＝0，D_j ²＝1；

Class 2: i is_j＝0，D_j ¹＝1，D_j ²＝1；

Class 3: i is_j＝1，D_j ¹＝0，D_j ²＝1；

Class 4: i is_j＝1，D_j ¹＝1，D_j ²＝1；

Class 5: i is_j＝-1，D_j ¹＝0，D_j ²＝1；

Category 6: i is_j＝-1，D_j ¹＝1，D_j ²＝1；

Wherein D is_j ¹First distortion factor, D, for the jth monitor node_j ²A second distortion factor for the jth monitor node.

6. A method for fault location in a low voltage active power distribution network according to claim 3, characterized in that: considering that the distribution line comprises N sections, when a fault occurs, defining the corresponding section state code as 1, otherwise, the code is 0; the specific coding rule of the decision variables is as follows:

where S denotes segment status information and D denotes distortion information.

7. A method for fault location in a low voltage active power distribution network according to claim 3, characterized in that: in a low-voltage distribution network, the problem of fault location can be equivalent to the problem of solving the minimum value of an objective function, and the objective function can be expressed by the following formula:

wherein: i is_jThe node monitoring information expected value considering the information distortion is calculated and obtained by decision variables; omega₁、ω₂、ω₃Are weight coefficients.

8. A method for fault location in a low voltage active power distribution network according to claim 3, characterized in that: in order to link the decision variables with the node monitoring information, a switching function of the decision variables and the expected values of the monitoring information needs to be constructed, and the switching function without considering the distortion information is represented by the following formula:

in the formula I_ju ^*(S)、I_jd ^*(S) an upstream switching function and a downstream switching function, respectively, demarcated by a monitoring node j(ii) a Pi is a logic or operator; n is a radical of_Su、N_SdThe number of the upstream and downstream power supplies of the monitoring node j is respectively; k_u、K_dThe power supply coefficients of the upstream and downstream of the monitoring node j are respectively, the power supply access is 1, otherwise, the power supply access is 0; s_j,Su、s_j,SdRespectively monitoring the state of the sections from the node j to the upstream power supply and from the node j to the downstream power supply; n is a radical of_u、N_dThe number of all sections of the upstream and downstream of the monitoring node j is respectively; s_j,u、s_j,dThe state information of all sections at the upstream and downstream of the monitoring node j is respectively.

9. A method for fault location in a low voltage active power distribution network according to claim 1, 2 or 3, characterized in that: the distortion information model converts the fault location problem after information distortion is considered into the solution of the unconstrained 0-1 integer programming problem, and a layered equivalent model is established to solve the unconstrained 0-1 integer programming problem, and the method comprises the following steps:

s31, establishing an outer layer equivalent model; the branch circuits in the section can be equivalently divided into active branch circuits and passive branch circuits according to whether the branch circuits have distributed power supplies or not, and the real-time monitoring information at the feeder line terminal unit M1 at the active branch circuit is equivalent to the real-time monitoring information I of the external line at the moment₁；

S32, establishing an inner layer equivalent model; defining: real-time monitoring information I₁When the value is 1, the external part is equivalent to a power supply; real-time monitoring information I₁And when the value is 0 or-1, the external equivalent is a load, so as to establish an inner layer equivalent model.

Technical Field

The invention relates to the technical field of power line fault diagnosis, in particular to a fault positioning method for a low-voltage active power distribution network.

Background

With the gradual increase of the power load and the large-scale access of Distributed Generation (DG) to the power grid, the structure and trend of the power distribution system tend to be more and more branched and complicated day by day, and the requirement of users on the power supply reliability is improved, and the fault location of the low-voltage power distribution network is gradually paid attention by researchers. The general measures for collecting the fault information of the power line are as follows: and a Feeder Terminal Unit (FTU) arranged at the line section switch or the interconnection switch uploads the collected fault information to a data acquisition and monitoring control System (SCADA), and fault positioning software is started to judge a fault section. However, errors such as packet loss, delay, noise interference and the like may occur in the low-voltage distribution network line fault information during the communication process between the feeder terminal unit and the control center, so that information distortion is caused, and the uploaded information cannot correctly reflect the fault condition.

The fault positioning method applied to the power distribution network at present comprises the following steps: the fault location and identification method based on the current phase angle difference value and the zero sequence current can be suitable for various power distribution network structures, has good adaptivity and robustness, but has higher requirements on the real-time performance and the reliability of measuring equipment, and has poorer fault tolerance.

Disclosure of Invention

The invention aims to solve the problem that the judgment of a background system on a fault position is inaccurate after the information of a fault line is distorted, and provides a fault positioning method for a low-voltage active power distribution network. The method effectively ensures the accuracy of fault location of the low-voltage active power distribution network, improves the reliability of the low-voltage power distribution network, and has important theoretical and practical application values.

In order to achieve the technical purpose, the invention provides a technical scheme that a fault positioning method for a low-voltage active power distribution network is characterized by comprising the following steps: the method comprises the following steps:

s1: collecting fault information;

s2: setting up a distortion information model;

s3: establishing a layered equivalent model;

s4: and outputting the result.

In the scheme, real-time state information of each branch node is acquired through a feeder terminal unit and uploaded to a data acquisition and monitoring control system, a distortion information model and a layered equivalent model are established in the data acquisition and monitoring control system, a fault location problem after information distortion is calculated is converted into a solution of an unconstrained 0-1 integer programming problem through the distortion information model, then a layered solution method in the layered equivalent model is adopted for solving, the fact that variable dimensionality of fault location of each layer is small can be guaranteed, and the accuracy of fault location is improved.

In the step S1, a feeder terminal unit is installed at the detection line section switch or the interconnection switch, the feeder terminal unit collects fault information and uploads the fault information to the data acquisition and monitoring control system, and a distorted information model and a layered equivalent model are established in the data acquisition and monitoring control system.

In step S2, the distortion information model is built according to the following steps:

s21: encoding fault detection information;

s22: classifying distortion information characteristics;

s23: determining a decision variable;

s24: determining an objective function;

s25: a switching function is determined.

In the scheme, firstly, fault detection information coding is carried out on fault information collected by a feeder line terminal unit, and as the fault information is mainly represented as over-current information, information distortion characteristics need to be classified in order to accurately account for the distortion condition of monitoring information; then coding the state information and the distortion information in the section, determining a decision variable, then defining an objective function, and solving the minimum value of the objective function; and finally, the decision variables are related to the node monitoring information, a switch function is determined, and the expected value of the node monitoring information during distortion is obtained through solving according to the switch function and the distortion information characteristic classification rule.

In step S21, the fault detection information encoding follows the following rules:

wherein M is_jNumbering feeder terminal units, I_jThe information is real-time monitoring information of the jth monitoring node.

In the scheme, overcurrent information is monitored through the feeder line terminal unit, when a line breaks down, the feeder line terminal unit can monitor fault current, the fault current flows in two directions due to the access of distributed power supplies such as photovoltaic power supplies, and the direction from a main power supply to a terminal load is specified to be the positive current direction.

The distortion information characteristics are classified as follows:

class 1: i is_j＝0，D_j ¹＝0，D_j ²＝1；

Class 2: i is_j＝0，D_j ¹＝1，D_j ²＝1；

Class 3: i is_j＝1，D_j ¹＝0，D_j ²＝1；

Class 4: i is_j＝1，D_j ¹＝1，D_j ²＝1；

Class 5: i is_j＝-1，D_j ¹＝0，D_j ²＝1；

Category 6: i is_j＝-1，D_j ¹＝1，D_j ²＝1；

Wherein D is_j ¹First distortion factor, D, for the jth monitor node_j ²A second distortion factor for the jth monitor node.

In the scheme, in order to accurately account for the distortion condition of the monitoring information, the information distortion characteristics need to be classified. In the communication process between the feeder terminal unit and the control center, errors such as packet loss, delay, noise interference and the like occur, and the caused information distortion can be mainly classified into six categories. The distortion is mainly false alarm or missing alarm, and the method of the invention sets a distortion factor D_j ¹、D_j ²And modeling the distortion condition of the state monitoring information of the jth monitoring node, namely determining the distortion type of the monitoring information by a distortion factor.

Considering that the distribution line comprises N sections, when a fault occurs, defining the corresponding section state code as 1, otherwise, the code is 0; the specific coding rule of the decision variables is as follows:

where S denotes segment status information and D denotes distortion information.

In the scheme, considering the occurrence of information distortion, the decision variable comprises section state information S and distortion information D, the distribution line comprises N sections, and when a fault occurs, the corresponding section state code is defined to be 1, otherwise, the code is 0, so that the dimension of the decision variable is 3N.

In a low-voltage distribution network, the problem of fault location can be equivalent to the problem of solving the minimum value of an objective function, and the objective function can be expressed by the following formula:

wherein: i is_jThe node monitoring information expected value considering the information distortion is calculated and obtained by decision variables; omega₁、ω₂、ω₃The weight coefficient is obtained through a system simulation test.

In the scheme, the first part in the target function reflects the approximation degree of the expected monitoring information and the monitoring information actually received by the control center, and is the main basis of fault positioning; the second part reflects the number of the fault sections, namely, the condition that the number of the fault sections is small is considered preferentially; the third part reflects the amount of information distortion, namely, the condition that the amount of information distortion is small is considered preferentially; the second part and the third part are auxiliary criteria for fault location, so that omega is provided₁＞＞ω₂And ω₁＞＞ω₃(ii) a Go through the simulation test, set omega₁＝20、ω₂＝0.6、ω₃＝1。

In order to link the decision variables with the node monitoring information, a switching function of the decision variables and the expected values of the monitoring information needs to be constructed, and the switching function without considering the distortion information is represented by the following formula:

in the formula I_ju ^*(S)、I_jd ^*(S) an upstream switching function and a downstream switching function which are respectively demarcated by the monitoring node j; pi is a logic or operator; n is a radical of_Su、N_SdThe number of the upstream and downstream power supplies of the monitoring node j is respectively; k_u、K_dThe power supply coefficients of the upstream and downstream of the monitoring node j are respectively, the power supply access is 1, otherwise, the power supply access is 0; s_j,Su、s_j,SdRespectively monitoring the state of the sections from the node j to the upstream power supply and from the node j to the downstream power supply; n is a radical of_u、N_dThe number of all sections of the upstream and downstream of the monitoring node j is respectively; s_j,u、s_j,dThe state information of all sections at the upstream and downstream of the monitoring node j is respectively.

In the scheme, in order to link the decision variable with the node monitoring information, a switch function of the decision variable and the monitoring information expected value needs to be constructed, and the node monitoring information expected value I in the distortion-free state can be obtained according to the state information by the aid of the switch function_j ^*. Meanwhile, the node monitoring information expected value I considering the distortion is known by the distortion information characteristic classification rule_j' can be composed of_j ^*And distortion factor D_j ¹、D_j ²Derived, the relationship is as follows:

thus I_j' is a function of all decision variables, i.e. I_j′＝f(S,D)。

The distortion information model converts the fault location problem after information distortion is considered into the solution of the unconstrained 0-1 integer programming problem, and a layered equivalent model is established to solve the unconstrained 0-1 integer programming problem, and the method comprises the following steps:

s31, establishing an outer layer equivalent model; the branch circuits in the section can be equivalently divided into active branch circuits and passive branch circuits according to whether the branch circuits have distributed power supplies or not, and the real-time monitoring information at the feeder line terminal unit M1 at the active branch circuit is equivalent to the real-time monitoring information I of the external line at the moment₁；

S32, establishing an inner layer equivalent model; defining: real-time monitoring information I₁When the value is 1, the external part is equivalent to a power supply; real-time monitoring information I₁And when the value is 0 or-1, the external equivalent is a load, so as to establish an inner layer equivalent model.

In the scheme, as the low-voltage distribution network has more branches, firstly establishing a fault equivalent model of a branch structure, specifically, establishing an outer layer equivalent model firstly, and when the outer layer fault positioning model judges that the equivalent section has a fault, carrying out external equivalence according to monitoring information of a demarcation point M1, namely when the monitoring information is 1, the external equivalence is a power supply; when the monitoring information is 0 or-1, the external equivalence is the load, an inner layer equivalent model is established, and through the equivalence, the variable dimensionality of fault location of each layer can be ensured to be small, and the accuracy of fault location is improved.

The invention has the beneficial effects that: 1. the method has the advantages that when the information is distorted, the fault section can still be positioned, and the distortion type and the distortion position can be obtained; 2. variable dimensionality is reduced by using a fault contradiction hypothesis and a layered equivalent model, and the accuracy and fault tolerance of low-voltage distribution network fault positioning when information is distorted are effectively guaranteed.

Drawings

Fig. 1 is a flowchart of a method for positioning a fault of a low-voltage active power distribution network according to this embodiment.

Fig. 2 is a schematic diagram of a multi-segment power distribution network in the method for locating a fault of a low-voltage active power distribution network according to the embodiment.

Fig. 3 is a schematic diagram of information distortion of a low-voltage active power distribution network fault location method according to an embodiment.

Fig. 4 is a diagram of a power distribution network branch mechanism of a low-voltage active power distribution network fault location method according to an embodiment.

Fig. 5 is a branch equivalent model diagram of a low-voltage active power distribution network fault location method according to an embodiment.

Fig. 6 is a flowchart of layered solution of a method for locating a fault in a low-voltage active power distribution network according to an embodiment.

Detailed Description

For the purpose of better understanding the objects, technical solutions and advantages of the present invention, the following detailed description of the present invention with reference to the accompanying drawings and examples should be understood that the specific embodiment described herein is only a preferred embodiment of the present invention, and is only used for explaining the present invention, and not for limiting the scope of the present invention, and all other embodiments obtained by a person of ordinary skill in the art without making creative efforts shall fall within the scope of the present invention.

As shown in fig. 1, it is a flowchart of a method for positioning a fault of a low-voltage active power distribution network in this embodiment, and includes the following steps:

s1: collecting fault information;

s2: establishing a distortion information model;

s3: establishing a layered equivalent model;

s4: and outputting the result.

In the embodiment, the real-time state information of each branch node is acquired through the feeder terminal unit and uploaded to the data acquisition and monitoring control system, a distortion information model and a layered equivalent model are established in the data acquisition and monitoring control system, the distortion information model converts the fault location problem after information distortion into the solution of an unconstrained 0-1 integer programming problem, and then the solution is performed by adopting a layered solution method in the layered equivalent model, so that the variable dimension of fault location of each layer can be ensured to be small, and the accuracy of fault location is improved.

Fig. 2 is a schematic diagram of a multi-segment power distribution network according to the method for locating a fault in a low-voltage active power distribution network of the present embodiment, where a segment head-end feeder terminal unit M is shown in the diagram₁-M₃The monitoring device is used for monitoring overcurrent information; feeder termination when line failure occursThe unit can monitor fault current, the fault current flows in two directions due to the connection of photovoltaic and other distributed power supplies, the direction from a main power supply to a terminal load is specified to be a current positive direction, and the coding rule of fault monitoring information is as follows:

in the formula: i is_jThe information is real-time monitoring information of the jth monitoring node.

In order to accurately count the distortion condition of the monitoring information, information distortion characteristics need to be classified, and in the communication process of a feeder line terminal unit and a control center, errors such as packet loss, delay, noise interference and the like can occur to cause information distortion, as shown in fig. 3, the information distortion is a schematic diagram of the information distortion of the low-voltage active power distribution network fault positioning method, specifically six types, and the distortion mainly refers to missing report or false report_j ¹、D_j ²Modeling the distortion condition of the state monitoring information of the jth monitoring node, namely determining the distortion type of the monitoring information by a distortion factor, as shown in table 1, the distortion characteristic classification model rule table is as follows:

TABLE 1 distortion characteristics classification model rule Table

The distribution line comprises N sections, when a fault occurs, the corresponding section state code is 1, otherwise, the code is 0. Considering the occurrence of information distortion, the decision variables of the method of the present invention include segment state information S and distortion information D, specifically coded as:

as can be seen from the above formula, the decision variable dimension of the method of the present invention is 3N.

In a low-voltage distribution network, the problem of fault location can be equivalent to the problem of solving the minimum value of an objective function, and after information distortion is considered, the objective function of fault location can be expressed as:

in the formula: i is_j' for the calculation of the expected value of the node monitoring information with information distortion, the expected value is calculated by a decision variable, omega₁、ω₂、ω₃As a weight coefficient, the first part in the objective function reflects the approximation degree of the expected monitoring information and the monitoring information actually received by the control center, and is the main basis for fault positioning; the second part reflects the number of faulty sections, i.e. prioritizes the case where the number of faulty sections is small. The third part reflects the amount of information distortion, namely, the condition that the amount of information distortion is small is considered preferentially; the second part and the third part are auxiliary criteria for fault location, so that omega is provided₁＞＞ω₂And ω₁＞＞ω₃Setting omega according to system simulation test₁＝20、ω₂＝0.6、ω₃＝1。

In order to link the decision variables with the node monitoring information, a switch function of the decision variables and the expected value of the monitoring information, a switch function I without considering the distortion information are constructed_j ^*(S) is represented by the following formula:

in the formula: i is_ju ^*(S)、I_jd ^*(S) an upstream switching function and a downstream switching function which are respectively demarcated by the monitoring node j; pi is a logic or operator; n is a radical of_Su、N_SdEach monitoring node jThe number of upstream and downstream power supplies; k_u、K_dThe power supply coefficients of the upstream and downstream of the monitoring node j are respectively, the power supply access is 1, otherwise, the power supply access is 0; s_j,Su、s_j,SdRespectively monitoring the state of the sections from the node j to the upstream power supply and from the node j to the downstream power supply; n is a radical of_u、N_dThe number of all sections of the upstream and downstream of the monitoring node j is respectively; s_j,u、s_j,dThe states of all sections upstream and downstream of the monitoring node j are respectively.

The expected value I of the node monitoring information in the case of no distortion can be obtained from the state information according to the formula switching function_j ^*. Meanwhile, the node monitoring information expected value I considering the distortion is known by the distortion information characteristic classification rule_j' can be composed of_j ^*And distortion factor D_j ¹、D_j ²Derived, the relationship is as follows:

according to the formula, I_j' is a function of all decision variables, i.e. I_j′＝f(S,D)。

The distortion information model converts the fault location problem after information distortion into the problem of solving unconstrained 0-1 integer programming, and the problem is solved by a plurality of algorithms, but the variable dimensionality is large due to the fact that a low-voltage power distribution network is mostly in a multi-branch radial structure and the number of sections is large, and the problem that the location accuracy is too low is often caused by adopting an optimization algorithm to directly solve the problem. If the optimization algorithm is improved, the problem of local convergence caused by large dimension is difficult to fundamentally avoid. Therefore, the method of the embodiment is improved from a solving method, a layered equivalent model is established to solve the unconstrained 0-1 integer programming problem, and the improved problem has lower requirements on the optimization algorithm.

In order to reduce the variable dimension of fault positioning, the method of the invention provides a layered solving method of a fault positioning model, namely, the whole fault positioning model is divided into multiple layers to be solved. Because the low-voltage distribution network has more branches, firstly, a fault equivalent model of a branch structure is established, as shown in fig. 4, the fault equivalent model is a distribution network branch mechanism diagram of a low-voltage active distribution network fault positioning method, as shown in fig. 5, the fault equivalent model diagram is a branch equivalent model diagram of the low-voltage active distribution network fault positioning method; according to the external equivalence principle, it can be equivalent to the structure of the active branch and the passive branch shown in fig. 4 according to whether there is a DG in the inner area of the segment line shown in fig. 3.

In fig. 4, port feeder termination unit M₁The monitoring information of (2) is equivalent monitoring information of the section, and the equivalent monitoring information participates in outer layer fault positioning.

When the outer layer fault positioning algorithm judges that the equivalent section has a fault, the fault is determined according to the demarcation point M₁The monitoring information of (2) is equivalent to the outside, namely when the monitoring information is 1, the outside is equivalent to a power supply; when the monitoring information is 0 or-1, the external equivalence is the load, an inner layer equivalent model is established, and through the equivalence, the variable dimensionality of fault location of each layer can be ensured to be small, and the accuracy of fault location is improved.

From the analysis, the value of the boundary node has a great influence on the accuracy of fault location, and when the information of the boundary node is distorted, false alarm or missing report of the fault is often caused, so that the boundary node should adopt monitoring equipment with higher reliability to ensure the accuracy of the key information.

According to the number of sections, fault location can be divided into multiple layers for solving, as shown in fig. 6, a layered solving flow chart of a low-voltage active power distribution network fault location method is provided, and a layered solving flow taking 2 layers as an example is provided; when the number of the sections is excessive, the inner layer model can be continuously partitioned and solved layer by layer, and accurate positioning of the fault section is realized.

The above-described embodiments are preferred embodiments of the present invention, and not intended to limit the scope of the present invention, which includes but is not limited to the embodiments described herein, and those skilled in the art can make various modifications, additions and substitutions to the described embodiments without departing from the spirit of the present invention or exceeding the scope of the appended claims.