A high-fault-tolerance fault section location method for distribution network based on multi-source information fusion

Abstract

The present invention discloses a method for locating fault sections in a distribution network with high fault tolerance based on multi-source information fusion. The method constructs a matrix of feeder circuit switches, distribution transformers, and substation meters based on the distribution network topology, and divides network ports using circuit breakers as demarcation points. The credibility levels of feeder terminal units (FTUs), distribution transformer terminal units (TTUs), and substation meters are divided, and the telemetered quantities of low-credibility FTUs and TTUs are verified and corrected using telemetry information. The suspected fault area is determined using the dimensionality reduction solution space of high-credibility FTUs, TTUs, and substation meters. Distorted information sequences of FTUs, TTUs, and substation meters are obtained, and the basic distribution probability is solved using a Bayesian estimation model. This method can accurately locate the fault section even when measurement data at the distribution terminal is lost or distorted on a large scale, demonstrating excellent fault tolerance.

Description

Technical Field

[0001] The present invention relates to the technical field of AC distribution network fault location, and in particular to a distribution network high-fault-tolerance fault section location method based on multi-source information fusion. Background Art

[0002] Compared to fault location methods based on telemetry from measurement devices, methods based on telesignaling have the advantages of simple principles and ease of implementation, and have been a research hotspot in recent years. However, since most measurement devices in distribution networks are installed outdoors, telesignaling information is easily lost or distorted, posing a challenge to fault location based on telesignaling. Fault segment location methods based on telesignaling can be divided into direct and indirect methods. Direct methods, such as matrix methods and linked list methods, directly determine the fault segment based on network topology and fault information. While simple in principle, they are practical and efficient, but suffer from poor fault tolerance and lack versatility. Indirect methods, such as matrix methods and linked list methods, transform the fault segment location problem into a mathematical optimization problem and then solve it using intelligent algorithms. While they offer some degree of fault tolerance, their results are unstable and time-consuming.

[0003] Existing solutions have improved fault-segment location methods based on intelligent algorithms from both a modeling and solution perspective, resulting in improvements in positioning accuracy, efficiency, and fault tolerance. However, when large-scale data loss or distortion occurs at distribution terminals due to communication or environmental factors, these improved methods struggle to maintain a balanced balance of positioning accuracy, efficiency, and fault tolerance. Therefore, it is necessary to research solutions for high-fault-tolerance fault-segment location in distribution networks based on multi-source information fusion. Summary of the Invention

[0004] The purpose of the present invention is to provide a high-fault-tolerant fault section location method for distribution network based on multi-source information fusion. This method can still accurately locate the fault section when the measurement data of the distribution terminal is lost or distorted on a large scale, and has good fault-tolerant performance.

[0005] The purpose of the present invention is achieved through the following technical solutions:

[0006] A method for locating a fault section in a distribution network with high fault tolerance based on multi-source information fusion, the method comprising:

[0007] Step 1: Construct a matrix of feeder circuit switches, distribution transformers, and substation meters based on the distribution network topology, and divide the network ports using circuit breakers as dividing points.

[0008] Step 2: Classify the reliability levels of feeder terminal units (FTUs), distribution transformer terminal units (TTUs), and substation meters into low, medium, and high reliability levels. Use telemetry information to verify and correct the telemetering quantities of low-reliability FTUs and TTUs.

[0009] Step 3: Use the dimensionality reduction solution space of high-confidence FTU, TTU, and substation meters to determine the suspected fault area;

[0010] Step 4: Obtain the distorted information sequence of FTU, TTU, and substation meters, and solve the basic distribution probability based on the Bayesian estimation model;

[0011] Step 5: Determine whether the fault section located by the FTU is consistent with the fault port located by the TTU and the substation meter. If they are consistent, output the fault location result. Otherwise, use the telemetry information of the FTU to verify and correct the telemetry of the medium-reliability FTU.

[0012] Step 6: Locate the fault section again. If the fault section location result is consistent with the fault port location result, output the fault location result. Otherwise, use the telemetry information of the TTU to verify and correct the telemetry of the medium-reliability TTU.

[0013] Step 7: Locate the fault port again. If the fault port location result is consistent with the fault section location result, output the fault location result. Otherwise, return to step 6 to continue verifying and correcting the high-reliability TTU.

[0014] It can be seen from the technical solution provided by the present invention that the above method measures the authenticity of the telesignaling information by defining the credibility of the measuring device, and thereby reduces the dimension of the solution space, speeds up the positioning speed, and reduces the occurrence of misjudgment; at the same time, through the fusion of multi-source data and the verification of telesignaling and telemetry, the fault tolerance performance of the positioning method can be improved, avoiding positioning errors caused by large-scale loss or distortion of measurement data. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 A flow chart of a method for locating a high-fault-tolerance fault section in a distribution network based on multi-source information fusion provided by an embodiment of the present invention;

[0017] Figure 2 This is a schematic diagram of the structure of a dual-power radial distribution network according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments, and do not constitute a limitation of the present invention. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making creative efforts shall fall within the scope of protection of the present invention.

[0019] like Figure 1 FIG2 is a flow chart of a method for locating a high-fault-tolerance fault section in a distribution network based on multi-source information fusion according to an embodiment of the present invention. The method includes:

[0020] Step 1: Construct a matrix of feeder circuit switches, distribution transformers, and substation meters based on the distribution network topology, and divide the network ports using circuit breakers as dividing points.

[0021] In this step, if Figure 2 The figure shows a schematic diagram of the structure of the dual-power radial distribution network according to an embodiment of the present invention. A substation may contain multiple smart meters. Substation meters are defined here to represent the status information of the smart meters in the substation. In order to reflect the hierarchical relationship between switches, distribution transformers, and substation meters in the feeder loop of the radial distribution network, a matrix of feeder loop switches, distribution transformers, and substation meters is constructed according to the distribution network topology. Compared with the system capacity of the distribution network, the capacity of the distributed power source is generally small, and its fault current will not be greater than the protection setting value. The forward overcurrent alarm is used as the basis for encoding the status information of the feeder terminal unit (Feeder Terminal Units, FTU), as shown in Table 1:

[0022] Table 1 FTU information coding

[0023]

[0024] The feeder is divided into individual ports using the circuit breaker as the dividing point. When a fault occurs in a section within a port, the circuit breaker at the port will operate, and the transformer terminal units (TTUs) and smart meters within the port will lose power. The information coding of the TTUs and meters is shown in Tables 2 and 3:

[0025] Table 2 TTU information coding

[0026]

[0027]

[0028] Table 3 Electricity meter information coding

[0029]

[0030] Step 2: Classify the reliability levels of feeder terminal units (FTUs), distribution transformer terminal units (TTUs), and substation meters into low, medium, and high reliability levels. Use telemetry information to verify and correct the telemetering quantities of low-reliability FTUs and TTUs.

[0031] In this step, the specific process of dividing the credibility levels of the feeder terminal unit FTU, distribution transformer terminal unit TTU, and substation meter is as follows:

[0032] The present invention measures the authenticity of measurement device information by defining credibility. The following provides the credibility definitions for FTUs, TTUs, and substation meters. For FTUs, based on the distribution pattern of forward fault current in radial distribution networks, when a forward overcurrent alarm is issued by a FTU in a feeder, all preceding FTUs in the feeder loop switch matrix should also issue overcurrent alarms. Therefore, the credibility of the FTU can be determined based on the proportion of forward overcurrent alarms in the preceding FTUs. A FTU that meets the following formula is considered to have high credibility:

[0033]

[0034] Where, F 过 is the number of forward overcurrent alarms in the previous FTU; F 过 is the total number of previous FTUs; K 阈值 The credibility threshold is set, and the present invention takes 0.5;

[0035] For an FTU with an overcurrent status of zero in a feeder with a forward overcurrent alarm, the FTU can be identified as a low-reliability FTU based on whether there is a high-reliability FTU behind it in the feeder switch matrix. If so, the FTU is identified as a high-reliability FTU; otherwise, it is identified as a low-reliability FTU.

[0036] For the FTU at the head end of the feeder and the FTU at the end of the feeder that does not give a positive overcurrent alarm, it is impossible to classify their reliability levels and they are considered to be medium reliability FTUs.

[0037] For TTUs, under normal circumstances, the power failure alarm signals of TTUs within a port are consistent. If the power failure status of the distribution transformers within a port is inconsistent, the TTUs with a larger proportion within the port are highly reliable, and the others are low reliable. If there is only one TTU within a port, the TTU is medium reliable.

[0038] According to the power failure pattern of distribution transformers in the feeder circuit, when a distribution transformer fails to generate a power failure alarm, all subsequent distribution transformers in the same row as the distribution transformer in the feeder distribution transformer matrix should also fail to generate a power failure alarm. Therefore, the reliability of the distribution transformer TTU can be judged based on the proportion of power failure alarms in the subsequent distribution transformers. A TTU that meets the following formula is a high-reliability TTU:

[0039]

[0040] Where, T 过 F is the number of power failure alarms in the following TTU; 过 is the total number of TTUs, T 阈值 The credibility threshold is set to 0.5 in the present invention.

[0041] The definition of the credibility of the substation meter is similar to that of the TTU and will not be repeated here.

[0042] In addition, when the telemetering information reported by a certain FTU or TTU is in doubt, the telemetering information sent by it can be used for verification and correction, specifically:

[0043] For FTU, a phase current I x For example, define F i *

[0044]

[0045] Where, I set It is the fixed value set in FTU;

[0046] At this time, if the FTU telemetering quantity F i With F i *If they are inconsistent, let F i =F i *;

[0047] For TTU, a phase current U x For example, define T i *

[0048]

[0049] At this time, if the telemetering quantity T i With T i *If they are inconsistent, let T i =T i *.

[0050] Step 3: Use the dimensionality reduction solution space of high-confidence FTU, TTU, and substation meters to determine the suspected fault area;

[0051] In this step, the shortest path from the high-confidence FTU to the main power supply is calculated and the suspected fault area G determined by the high-confidence FTU is obtained. F , which is:

[0052] G F =P F1 ∪P F2 ∪…P Fi

[0053] Where: PFi is the area contained by the shortest path from the i-th high-confidence FTU to the main power supply;

[0054] The section of the port containing the high-reliability TTU is the suspected fault area G T , which is:

[0055] G T =P T1 ∪P T2 ∪...P Ti

[0056] Where: P Ti is the area contained by the i-th port containing a high-reliability TTU;

[0057] In summary, the suspected fault area after dimensionality reduction of the solution space is:

[0058] G=G F ∪G T ∪G DU

[0059] Where: G DU The area covered by the ports containing high-reliability area meters.

[0060] Step 4: Obtain the distorted information sequence of FTU, TTU, and substation meters, and solve the basic distribution probability based on the Bayesian estimation model;

[0061] In this step, when a fault occurs, the probability of occurrence of the fault hypothesis variable is estimated based on the alarm information reported by the measurement device. The hypothesis variable with the highest probability is the positioning result.

[0062] For a distribution network with n FTUs or TTUs or port meters, x feeder sections, and m ports, the fault hypothesis variables are {f1, f2, …, f x} or {f1,f2,…,f m}, the alarm information variables of the measuring device are {S1, S2, ..., S n}, to facilitate modeling, the following approximate conditions are processed:

[0063] 1) The probability of distortion of alarm information reported by the same type of measurement devices is the same;

[0064] 2) The probability of multiple faults occurring on the same branch line is zero;

[0065] 3) All feeder sections have the same probability of failure;

[0066] According to the Bayesian estimation model, the basic distribution probability of the section fault hypothesis variable is:

[0067]

[0068] Where: P(f i ) represents the probability of failure in feeder section j; P(S1, S2, …, S n |f j ) represents the probability of obtaining an overcurrent alarm message from a known FTU when a fault occurs in section j;

[0069] For different feeder sections, P(f i ) and P(S1,S2,…,S n ) are all equal, so P(f j |S1,S2,…,S n ) depends only on P(S1,S2,…,S n |f j ); Similarly, for the fault port hypothesis variable, the solution of its distribution probability is similar, except that the probability of failure of different ports is related to the number of segments contained in the port and is not necessarily equal, so P(f j |S1,S2,…,S n ) depends on P(S1,S2,…,S n |f j ) and P(f i );

[0070] P(S1,S2,…,S n |f j ) can be expanded to obtain

[0071]

[0072] Where: M0 is the number of 0s in the actual alarm information matrix; M1 is the number of 1s in the actual alarm information matrix; L is the number of missed alarms of the measurement device; P L represents the probability of missed alarm information of the measuring device; W is the number of false alarms of the measuring device; P W Indicates the probability of false alarm of the measuring device;

[0073] Defining logical operations When X ij >0, Y ij =1, otherwise Y ij =0;

[0074] The number of missed reports and false reports in FTU reporting can be calculated as follows:

[0075] The expected alarm information sequence is:

[0076]

[0077] Where: L is the segment state sequence; SF is the feeder switch causal relationship matrix; C K is a diagonal matrix with the switch sequence values as diagonal elements;

[0078] The distortion information sequence is:

[0079] J F =G F * -G F

[0080] Where: G F * is the actual alarm information sequence of FTU;

[0081] The number of values 1 in the distortion information sequence is the number of false positives; the number of values -1 is the number of missed negatives;

[0082] For TTUs and port meters, to adapt the location algorithm to changes in the distribution network topology, the feeder switch disconnection information must first be converted into equivalent port fault information and defined as a virtual port state sequence. The expected alarm information sequence is:

[0083]

[0084] Where: D is the port state sequence; D* is the diagonal matrix composed of virtual port state sequences as diagonal elements; S T is the distribution transformer causal relationship matrix;

[0085] The distortion information sequence is:

[0086] J T =G T * -G T

[0087] The calculation of the distortion information sequence of the port meter is similar to that of the TTU and will not be repeated here.

[0088] Since the positioning results of the TTU and the port meter are both ports, in order to improve fault tolerance, this embodiment uses the DS evidence theory to fuse the positioning results of the two.

[0089] Step 5: Determine whether the fault section located by the FTU is consistent with the fault port located by the TTU and the substation meter. If they are consistent, output the fault location result. Otherwise, use the telemetry information of the FTU to verify and correct the telemetry of the medium-reliability FTU.

[0090] Step 6: Locate the fault section again. If the fault section location result is consistent with the fault port location result, output the fault location result. Otherwise, use the telemetry information of the TTU to verify and correct the telemetry of the medium-reliability TTU.

[0091] Step 7: Locate the fault port again. If the fault port location result is consistent with the fault section location result, output the fault location result. Otherwise, return to step 6 to verify and correct the high-reliability TTU.

[0092] It should be noted that the contents not described in detail in the embodiments of the present invention belong to the prior art known to those skilled in the art.

[0093] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by any person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims. The information disclosed in the background technology section of this article is only intended to deepen the understanding of the overall background technology of the present invention, and should not be regarded as an admission or any form of implication that the information constitutes prior art already known to those skilled in the art.

Claims

1. A method for locating fault sections in a distribution network with high fault tolerance based on multi-source information fusion, characterized in that: The method comprises: Step 1: Construct a matrix of feeder circuit switches, distribution transformers, and substation meters based on the distribution network topology, and divide the network ports using circuit breakers as dividing points. Step 2: Classify the reliability levels of feeder terminal units (FTUs), distribution transformer terminal units (TTUs), and substation meters into low, medium, and high reliability levels. Use telemetry information to verify and correct the telemetering quantities of low-reliability FTUs and TTUs. In step 2, the specific process of dividing the credibility levels of the feeder terminal unit FTU, distribution transformer terminal unit TTU, and substation meter is as follows: For FTUs, according to the distribution law of forward fault current in radial distribution networks, when a certain FTU in a feeder generates a forward overcurrent alarm, all preceding FTUs in the feeder loop switch matrix should generate overcurrent alarms. Therefore, the reliability of the FTU can be determined based on the proportion of forward overcurrent alarms in the preceding FTUs. A FTU that meets the following formula is considered a high-reliability FTU: Where, F 过 is the number of forward overcurrent alarms in the previous FTU; F 总 is the total number of previous FTUs; K 阈值 is the set credibility threshold; For an FTU with an overcurrent status of zero in a feeder with a forward overcurrent alarm, the FTU can be identified as a low-reliability FTU based on whether there is a high-reliability FTU behind it in the feeder switch matrix. If so, the FTU is identified as a high-reliability FTU; otherwise, it is identified as a low-reliability FTU. For the FTU at the head end of the feeder and the FTU at the end of the feeder that does not give a positive overcurrent alarm, it is impossible to classify their reliability levels and they are considered to be medium reliability FTUs. For TTUs, under normal circumstances, the power failure alarm signals of TTUs within a port are consistent. If the power failure status of the distribution transformers within a port is inconsistent, the TTUs with a larger proportion within the port are highly reliable, and the others are low reliable. If there is only one TTU within a port, the TTU is medium reliable. According to the power failure pattern of distribution transformers in the feeder circuit, when a distribution transformer fails to generate a power failure alarm, all subsequent distribution transformers in the same row as the distribution transformer in the feeder distribution transformer matrix should also fail to generate a power failure alarm. Therefore, the reliability of the distribution transformer TTU can be judged based on the proportion of power failure alarms in the subsequent distribution transformers. A TTU that meets the following formula is a high-reliability TTU: Where, T 失 is the number of power failure alarms in the following TTU; T 总 is the total number of TTUs, T 阈值 is the set credibility threshold, which is 0.5; Step 3: Use the dimensionality reduction solution space of high-confidence FTU, TTU, and substation meters to determine the suspected fault area; Step 4: Obtain the distorted information sequence of FTU, TTU, and substation meters, and solve the basic distribution probability based on the Bayesian estimation model; Step 5: Determine whether the fault section located by the FTU is consistent with the fault port located by the TTU and the substation meter. If they are consistent, output the fault location result. Otherwise, use the telemetry information of the FTU to verify and correct the telemetry of the medium-reliability FTU. Step 6: Locate the fault section again. If the fault section location result is consistent with the fault port location result, output the fault location result. Otherwise, use the telemetry information of the TTU to verify and correct the telemetry of the medium-reliability TTU. Step 7: Locate the fault port again. If the fault port location result is consistent with the fault section location result, output the fault location result. Otherwise, return to step 6 to continue verifying and correcting the high-reliability TTU.

2. The method for locating fault sections in a distribution network with high fault tolerance based on multi-source information fusion according to claim 1 is characterized in that: The process of step 1 is specifically as follows: To reflect the hierarchical relationship between switches, distribution transformers, and substation meters in the radial distribution network feeder circuit, a matrix of feeder circuit switches, distribution transformers, and substation meters is constructed based on the distribution network topology. Compared with the system capacity of the distribution network, the capacity of distributed power sources is generally small, and their fault current will not exceed the protection setting value. The forward overcurrent alarm is used as the basis for encoding the status information of the feeder terminal unit. The feeder is divided into various ports with the circuit breaker as the dividing point. When a fault occurs in the section within the port, the circuit breaker at the port will operate, and the distribution terminal unit and smart meter in the port will lose power.

3. The method for locating fault sections in a distribution network with high fault tolerance based on multi-source information fusion according to claim 1 is characterized in that: In step 2, the process of verifying and correcting the telemetered values of low-reliability FTUs and TTUs using telemetry information is as follows: For FTU, a phase current I x For example, define F i * Where, I set It is the fixed value set in FTU; At this time, if the FTU telemetering quantity F i With F i *If they are inconsistent, let F i =F i *; For TTU, a phase current U x For example, define T i * At this time, if the telemetering quantity T i With T i *If they are inconsistent, let T i =T i *.

4. The method for locating fault sections in a distribution network with high fault tolerance based on multi-source information fusion according to claim 1 is characterized in that: The process of step 3 is specifically as follows: The shortest path from the high-confidence FTU to the main power supply can be calculated by the union operation to obtain the suspected fault area G determined by the high-confidence FTU. F , which is: G F =P F1 ∪P F2 ∪…P Fi Where: P Fi is the area contained by the shortest path from the i-th high-confidence FTU to the main power supply; The section of the port containing the high-reliability TTU is the suspected fault area G T , which is: G T =P T1 ∪P T2 ∪…P Ti Where: P Ti is the area contained by the i-th port containing a high-reliability TTU; In summary, the suspected fault area after dimensionality reduction of the solution space is: G=G F ∪G T ∪G DU Where: G DU The area covered by the ports containing high-reliability area meters.

5. The method for locating fault sections in distribution networks with high fault tolerance based on multi-source information fusion according to claim 1, characterized in that: The process of step 4 is specifically as follows: When a fault occurs, the probability of occurrence of the fault hypothesis variable is estimated based on the alarm information reported by the measuring device. The hypothesis variable with the highest probability is the positioning result. For a distribution network with n FTUs or TTUs or port meters, x feeder sections, and m ports, the fault hypothesis variable is {f 1, f 2,…, f x } or {f 1, f 2,…, f m }, the alarm information variable of the measuring device is {S 1, S 2,…, S n }, to facilitate modeling, the following approximate conditions are processed: 1) The probability of distortion of alarm information reported by the same type of measurement devices is the same; 2) The probability of multiple faults occurring on the same branch line is zero; 3) All feeder sections have the same probability of failure; According to the Bayesian estimation model, the basic distribution probability of the section fault hypothesis variable is: Where: P(f i ) represents the probability of failure in feeder section j; P(S1, S2, …, S n |f j ) represents the probability of obtaining an overcurrent alarm message from a known FTU when a fault occurs in section j; For different feeder sections, P(f i ) and P(S1,S2,…,S n ) are all equal, so P(f j |S1,S2,…,S n ) depends only on P(S1,S2,…,S n |f j ); Similarly, for the fault port hypothesis variable, the solution of its distribution probability is similar, except that the probability of failure of different ports is related to the number of segments contained in the port and is not necessarily equal, so P(f j |S1,S2,…,S n ) depends on P(S1,S2,…,S n |f j ) and P(f i ); P(S1,S2,…,S n |f j ) can be expanded to obtain Where: M0 is the number of 0s in the actual alarm information matrix; M1 is the number of 1s in the actual alarm information matrix; L is the number of missed alarms of the measurement device; P L represents the probability of missed alarm information of the measuring device; W is the number of false alarms of the measuring device; P W Indicates the probability of false alarm of the measuring device; Defining logical operations When X ij >0, Y ij =1, otherwise Y ij =0; The number of missed reports and false reports in FTU reporting can be calculated as follows: The expected alarm information sequence is: Where: L is the segment state sequence; S F is the feeder switch causal relationship matrix; C K is a diagonal matrix with the switch sequence values as diagonal elements; The distortion information sequence is: J F =G F * -G F Where: G F * is the actual alarm information sequence of FTU; The number of values 1 in the distortion information sequence is the number of false positives; the number of values -1 is the number of missed negatives; For TTUs and port meters, to adapt the location algorithm to changes in the distribution network topology, the feeder switch disconnection information must first be converted into equivalent port fault information and defined as a virtual port state sequence. The expected alarm information sequence is: Where: D is the port state sequence; D* is the diagonal matrix composed of virtual port state sequences as diagonal elements; S T is the distribution transformer causal relationship matrix; The distortion information sequence is: J T =G T * -G T Where: G T * is the actual fault area; The calculation of the distortion information sequence of the port meter is similar to that of the TTU.

Images