Power distribution network fault diagnosis method, device, equipment, storage medium and program product

Abstract

The invention relates to a power distribution network fault diagnosis method and device, equipment, a storage medium and a program product, and the method comprises the steps: calling a target feature extraction model to carry out the feature extraction of the current fault data of a power distribution network under the condition that a fault occurs in the power distribution network, and obtaining the features of the current fault data; determining a target reference fault data feature matched with the current fault data feature from a plurality of predetermined reference fault data features; each reference fault data feature corresponds to one fault type; and obtaining a fault type corresponding to the target reference fault data feature as a current fault type of the power distribution network. By adopting the method, the accuracy of the determined fault type of the power distribution network can be improved.

Description

Technical Field

[0001] The present application relates to the field of electric power technology, and in particular to a distribution network fault diagnosis method, device, equipment, storage medium and program product. Background Art

[0002] As more and more distributed power sources and new loads are widely connected to the distribution network, the topology of the distribution network is becoming increasingly complex, which makes it extremely difficult for maintenance personnel to analyze distribution network faults. At present, the fault diagnosis of the distribution network is mainly carried out by the dispatching center to analyze the collected protection and circuit breaker action information to determine the type of fault and the faulty component in the fault area. However, in the case of refusal to operate or malfunction of the protection device and circuit breaker, the relay protection action information collected by the dispatching center may not be completely accurate, which may lead to the final determination of the distribution network fault type is not accurate, and then, it will lead to unstable operation of the distribution network due to the failure to repair the fault in a timely and accurate manner.

[0003] Therefore, how to improve the accuracy of the determined distribution network fault type has become an urgent problem to be solved. Summary of the invention

[0004] The embodiments of the present application provide a distribution network fault diagnosis method, apparatus, device, storage medium and program product, which can improve the accuracy of the determined distribution network fault type.

[0005] In a first aspect, an embodiment of the present application provides a distribution network fault diagnosis method, the method comprising:

[0006] When a fault occurs in the distribution network, the target feature extraction model is called to extract features from the current fault data of the distribution network to obtain the features of the current fault data;

[0007] Determining a target reference fault data feature that matches the current fault data feature from a plurality of pre-determined reference fault data features; each reference fault data feature corresponds to a fault type;

[0008] The fault type corresponding to the target reference fault data feature is obtained as the current fault type of the distribution network.

[0009] In one of the embodiments, multiple reference fault data features are determined in the following manner: based on a pre-constructed distribution network model, the fault data corresponding to the distribution network under multiple fault types are simulated; and the target feature extraction model is called to determine the reference fault data features corresponding to each fault type based on the fault data under the fault type.

[0010] In one embodiment, the distribution network model includes a first line, a second line and a bus; based on a pre-constructed distribution network model, fault data corresponding to the distribution network under multiple fault types are simulated, including: determining multiple fault types corresponding to the first line and the second line in the pre-constructed distribution network model; obtaining line information of the first line and the second line respectively whose distance from the bus is less than or equal to a preset distance threshold under each fault type, the line information including voltage amplitude information, voltage phase information, current amplitude information and current phase information; integrating the line information of the first line and the second line respectively under each fault type to obtain the fault data corresponding to each fault type.

[0011] In one of the embodiments, a target feature extraction model is called to determine a reference fault data feature corresponding to each fault type based on the fault data of the fault type, including: integrating the fault data of each fault type to obtain a fault data matrix of the fault type; normalizing the fault data matrix to obtain a normalized fault data matrix; and calling the target feature extraction model to perform feature extraction on the normalized fault data matrix to obtain a reference fault data feature corresponding to the fault type.

[0012] In one of the embodiments, the method further includes: traversing each integer in a preset model parameter interval, and inputting the normalized fault data matrix into a feature extraction model whose model parameters are each integer, to obtain multiple predicted fault data feature matrices; wherein each predicted fault data feature matrix is ​​a predicted fault data feature matrix corresponding to the fault type when the model parameters of the feature extraction model are each integer; determining the loss value between each predicted fault data feature matrix and the normalized fault data matrix; and from multiple loss values, selecting the model parameter corresponding to the minimum loss value as the model parameter in the target feature extraction model.

[0013] In one of the embodiments, a target reference fault data feature that matches a current fault data feature is determined from a plurality of predetermined reference fault data features, including: based on the plurality of predetermined reference fault data features and the current fault data feature, determining the probability that the current fault data corresponds to each reference fault data feature through a preset probability determination strategy; and selecting, from multiple probabilities, a reference fault data feature corresponding to a maximum probability as a target reference fault data feature that matches the current fault data feature.

[0014] In a second aspect, the present application provides a distribution network fault diagnosis device, the device comprising:

[0015] The feature extraction module is used to call the target feature extraction model to extract features of the current fault data of the distribution network in the event of a fault in the distribution network, so as to obtain features of the current fault data;

[0016] A determination module, used to determine a target reference fault data feature that matches the current fault data feature from a plurality of pre-determined reference fault data features; each reference fault data feature corresponds to a fault type;

[0017] The acquisition module is used to obtain the fault type corresponding to the target reference fault data feature as the current fault type of the distribution network.

[0018] In a third aspect, the present application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0019] When a fault occurs in the distribution network, the target feature extraction model is called to extract features from the current fault data of the distribution network to obtain the features of the current fault data;

[0020] Determining a target reference fault data feature that matches the current fault data feature from a plurality of pre-determined reference fault data features; each reference fault data feature corresponds to a fault type;

[0021] The fault type corresponding to the target reference fault data feature is obtained as the current fault type of the distribution network.

[0022] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the following steps are implemented:

[0023] When a fault occurs in the distribution network, the target feature extraction model is called to extract features from the current fault data of the distribution network to obtain the features of the current fault data;

[0024] Determining a target reference fault data feature that matches the current fault data feature from a plurality of pre-determined reference fault data features; each reference fault data feature corresponds to a fault type;

[0025] The fault type corresponding to the target reference fault data feature is obtained as the current fault type of the distribution network.

[0026] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0027] When a fault occurs in the distribution network, the target feature extraction model is called to extract features from the current fault data of the distribution network to obtain the features of the current fault data;

[0028] Determining a target reference fault data feature that matches the current fault data feature from a plurality of pre-determined reference fault data features; each reference fault data feature corresponds to a fault type;

[0029] The fault type corresponding to the target reference fault data feature is obtained as the current fault type of the distribution network.

[0030] In the above-mentioned distribution network fault diagnosis method, device, equipment, storage medium and program product, the computer equipment can call the target feature extraction model to extract the current fault data of the distribution network to obtain the current fault data feature when the distribution network fails; determine the target reference fault data feature that matches the current fault data feature from a plurality of pre-determined reference fault data features; each reference fault data feature corresponds to a fault type; obtain the fault type corresponding to the target reference fault data feature as the current fault type of the distribution network. Using this method, when a distribution network fails, the target reference fault data feature that matches the current fault data feature can be determined by comparing the matching degree between the current fault data feature and a plurality of pre-determined reference fault features. Since each reference fault feature corresponds to a fault type, the fault type corresponding to the target reference fault data feature is used as the current fault type of the distribution network, thereby improving the accuracy of the determined distribution network fault type. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 This is a schematic diagram of an application scenario of a distribution network fault diagnosis method provided by an embodiment of the present application;

[0033] Figure 2 It is a schematic diagram of a flow chart of a distribution network fault diagnosis method provided in an embodiment of the present application;

[0034] Figure 3 It is a flowchart of another method for diagnosing a fault in a distribution network provided in an embodiment of the present application;

[0035] Figure 4 is a schematic diagram of a distribution network model provided in an embodiment of the present application;

[0036] Figure 5This is a flow chart of another method for diagnosing a fault in a distribution network provided in an embodiment of the present application;

[0037] Figure 6 It is a schematic diagram of a loss function curve provided in an embodiment of the present application;

[0038] Figure 7 It is a structural schematic diagram of a distribution network fault diagnosis device provided in an embodiment of the present application;

[0039] Figure 8 It is a structural diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0041] The following introduces the application scenarios of the distribution network fault diagnosis method provided in the embodiments of the present application.

[0042] See also Figure 1 , Figure 1 Schematic diagram of an application scenario of a distribution network fault diagnosis method provided by an embodiment of the present application. Figure 1 As shown, the computer device 101 ( Figure 1 In the figure, the computer device 101 is drawn as an example of a terminal device) and a database server 102, wherein data can be transmitted between the computer device 101 and the database server 102 via a network.

[0043] When a fault occurs in the distribution network, the computer device 101 can obtain a target feature extraction model and a plurality of predetermined reference fault data features from the database server 102, and then call the target feature extraction model to extract features from the current fault data of the distribution network to obtain the current fault data features; determine a target reference fault data feature that matches the current fault data feature from the predetermined plurality of reference fault data features; each reference fault data feature corresponds to a fault type; and obtain the fault type corresponding to the target reference fault data feature as the current fault type of the distribution network. By adopting this method, when a fault occurs in the distribution network, the target reference fault data feature that matches the current fault data feature can be determined by comparing the matching degree between the current fault data feature and the predetermined plurality of reference fault features. Since each reference fault feature corresponds to a fault type, the fault type corresponding to the target reference fault data feature is used as the current fault type of the distribution network, thereby improving the accuracy of the determined distribution network fault type.

[0044] Optionally, the computer device 101 may be a terminal device or a server. The terminal device mentioned here may include but is not limited to: a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart watch, a smart TV, a smart car terminal, etc. The server mentioned here may be an independent physical server, or a server cluster or a distributed system composed of multiple physical servers.

[0045] See also Figure 2 , Figure 2 1 is a flow chart of a method for diagnosing a distribution network fault provided by an embodiment of the present application. The method may be executed by a computer device (for example, the computer device 101 described above). Figure 2 As shown, the distribution network fault diagnosis method may include but is not limited to the following steps:

[0046] S201. When a fault occurs in the distribution network, a target feature extraction model is called to extract features of current fault data of the distribution network to obtain features of the current fault data.

[0047] In an optional embodiment, the computer device calls the target feature extraction model to perform feature extraction on the current fault data of the distribution network to obtain the current fault data characteristics, which may include: integrating the current fault data to obtain the current fault data matrix; calling the target feature extraction network to perform feature extraction on the current fault data matrix to obtain the current fault data characteristics.

[0048] Optionally, the target feature extraction model may be a convolutional neural network after determining model parameters.

[0049] S202: Determine a target reference fault data feature that matches the current fault data feature from a plurality of pre-determined reference fault data features; each reference fault data feature corresponds to a fault type.

[0050] In an optional implementation manner, the fault types corresponding to the multiple reference fault data may include: A-phase fault, AB-phase short circuit, AB-phase ground short circuit, three-phase ground short circuit, and three-phase short circuit.

[0051] In an optional embodiment, a computer device determines a target reference fault data feature that matches a current fault data feature from a plurality of predetermined reference fault data features, which may include: determining the probability that the current fault data corresponds to each reference fault data feature through a preset probability determination strategy based on the plurality of predetermined reference fault data features and the current fault data feature; and selecting, from multiple probabilities, a reference fault data feature corresponding to a maximum probability as a target reference fault data feature that matches the current fault data feature.

[0052] S203. Obtain a fault type corresponding to the target reference fault data feature as the current fault type of the distribution network.

[0053] Exemplarily, assuming that the fault type corresponding to the target reference fault data feature is A-phase short circuit, the computer device can determine that the current fault type of the distribution network is A-phase short circuit.

[0054] In an embodiment of the present application, the computer device can call the target feature extraction model to extract features from the current fault data of the distribution network to obtain the current fault data features when a fault occurs in the distribution network; determine the target reference fault data features that match the current fault data features from a plurality of pre-determined reference fault data features; each reference fault data feature corresponds to a fault type; obtain the fault type corresponding to the target reference fault data features as the current fault type of the distribution network. Using this method, when a fault occurs in the distribution network, the target reference fault data features that match the current fault data features can be determined by comparing the matching degree between the current fault data features and a plurality of pre-determined reference fault features. Since each reference fault feature corresponds to a fault type, the fault type corresponding to the target reference fault data features can be used as the current fault type of the distribution network, thereby improving the accuracy of the determined distribution network fault type.

[0055] See also Figure 3 , Figure 3 FIG. 1 is a flow chart of another method for diagnosing a distribution network fault provided by an embodiment of the present application. Figure 2 Compared with the distribution network fault diagnosis method shown in Figure 3 The method shown also describes how the computer device determines multiple reference fault data characteristics. Figure 3 As shown, the distribution network fault diagnosis method may include but is not limited to the following steps:

[0056] S301. Based on a pre-built distribution network model, simulate and obtain fault data corresponding to various fault types of the distribution network.

[0057] In an optional embodiment, the distribution network model includes a first line, a second line and a bus; the computer device simulates and obtains fault data corresponding to the distribution network under multiple fault types based on a pre-constructed distribution network model, which may include: determining multiple fault types corresponding to the first line and the second line in the pre-constructed distribution network model; obtaining line information of the first line and the second line respectively whose distance from the bus is less than or equal to a preset distance threshold under each fault type, the line information including voltage amplitude information, voltage phase information, current amplitude information and current phase information; integrating the line information of the first line and the second line respectively under each fault type to obtain the fault data corresponding to each fault type.

[0058] For example, the distribution network model pre-built by the computer equipment can be found in Figure 4 , Figure 4 Schematic diagram of a distribution network model provided in an embodiment of the present application. Figure 4 As shown, the distribution network model includes the first line L1, the second line L2, the third line L3 and the bus M. The specific parameters of the line are positive sequence impedance R1=0.03986Ω / km, ωL1=0.478Ω / km; zero sequence impedance R0=0.398Ω / km, ωL0=1.273Ω / km. The first line L1 is 3kM long, the second line L2 is 4kM long, the third line L3 is 3kM long, the load 1 is 3MW, and the load 2 is 2MW.

[0059] In this implementation, the computer device can determine that the duration of the fault is 0.3s. The multiple fault types may include the following five: A phase short circuit, then the computer device can determine that the current fault type of the distribution network is A phase short circuit. The fault transition resistance is: 0Ω to 10Ω, increasing by 0.1Ω. The fault location is 1km away from the busbar between the first line L1 and the second line L2.

[0060] Exemplarily, the computer device obtains line information of the first line and the second line whose distance from the bus is less than or equal to the preset distance threshold under each fault type, which may include: voltage amplitude information and phase information and current amplitude information and phase information of the first line L1 and the second line L2 close to the bus M side under each fault type. The three-phase voltage amplitude information obtained for the first line L1 can be recorded as U 1,A , U 1,B , U 1,C , the three-phase voltage phase information obtained for the first line L1 can be recorded as U 1,Aφ , U 1,Bφ , U 1,Cφ , the three-phase current amplitude information obtained for the first line L1 can be recorded as I 1,A ,I 1,B ,I 1,C , the phase information of the three-phase current obtained for the first line L1 can be recorded as I 1,Aφ ,I 1,Bφ ,I 1,Cφ ; The three-phase voltage amplitude information obtained for the second line L2 can be recorded as U 2,A , U 2,B , U 2,C , the three-phase voltage phase information obtained for the second line L2 can be recorded as U 2,Aφ , U 2,Bφ , U 2,Cφ, the three-phase current amplitude information obtained for the second line L2 can be recorded as I 2,A ,I 2,B ,I 2,C , the three-phase current phase information obtained for the second line L2 can be recorded as I 2,Aφ ,I 2,Bφ ,I 2,Cφ ; Among them, the data sampling frequency is 500Hz, and the data length of each information saved is 150.

[0061] S302: Integrate the fault data under each fault type to obtain a fault data matrix under the fault type.

[0062] In an optional implementation, the computer device integrates the fault data under each fault type, and obtains a fault data matrix under the fault type as shown in the following formula (1).

[0063] (1)

[0064] In formula (1), A represents the fault data matrix under the fault type; n represents the number of the fault type, where n=1 represents the fault type is A-phase grounding; n=2 represents the fault type is AB-phase short circuit; n=3 represents the fault type is AB-phase short circuit grounding; n=4 represents the fault type is three-phase grounding; n=5 represents the fault type is three-phase grounding short circuit; m represents the number of the fault data matrix under each fault condition. There are 200 fault data matrices under each fault condition, so the value range of m is 1-200.

[0065] In formula (1), the fault data matrix A n,mThe first column is the amplitude information of the A-phase voltage of the first line L1, the second column is the amplitude information of the B-phase voltage of the first line L1, the third column is the amplitude information of the C-phase voltage of the first line L1, the fourth column is the phase information of the A-phase voltage of the first line L1, the fifth column is the phase information of the B-phase voltage of the first line L1, and the sixth column is the phase information of the C-phase voltage of the first line L1; the seventh column is the amplitude information of the A-phase current of the first line L1, the eighth column is the amplitude information of the B-phase current of the first line L1, the ninth column is the amplitude information of the C-phase current of the first line L1, the tenth column is the phase information of the A-phase current of the first line L1, the eleventh column is the phase information of the B-phase current of the first line L1, and the twelfth column is the phase information of the C-phase current of the first line L1; the thirteenth column is the phase information of the The fourteenth column is the amplitude information of the A-phase voltage of the second line L2, the fifteenth column is the amplitude information of the C-phase voltage of the second line L2, the sixteenth column is the phase information of the A-phase voltage of the second line L2, the seventeenth column is the phase information of the B-phase voltage of the second line L2, and the eighteenth column is the phase information of the C-phase voltage of the second line L2; the nineteenth column is the amplitude information of the A-phase current of the second line L2, the twentieth column is the amplitude information of the B-phase current of the second line L2, and the twenty-first column is the amplitude information of the C-phase current of the second line L2; the twenty-second column is the phase information of the A-phase current of the second line L2, the twenty-third column is the phase information of the B-phase current of the second line L2, and the twenty-fourth column is the phase information of the C-phase current of the second line L2.

[0066] S303: normalize the fault data matrix to obtain a normalized fault data matrix.

[0067] In an optional embodiment, the computer device performs a fault data matrix A n,m After normalization, the normalized fault data matrix B can be obtained as shown in the following formula (2): n,m .

[0068] (2)

[0069] In formula (2), B n,m It represents the normalized fault data matrix; A n,m represents the fault data matrix; the physical meanings of n and m can be found in the above description of n and m in formula (1), which will not be repeated here.

[0070] In this implementation, the computer device may use the following formula (3) to calculate the above b n,m,i,j .

[0071] (3)

[0072] In formula (3), b n,m,i,j It represents the normalized fault data value of the fault type numbered n, and i represents the element in the normalized fault data matrix B. n,m The i-th row of b, j means that the element is located in the j-th column of the normalized fault data matrix, that is, b n,m,i,j It represents the normalized fault data matrix B under the fault type numbered n n,m The fault data value of the i-th row and j-th column of a; n,m,i,j Indicates the fault data value under the fault type numbered n; a n,i,max It represents the maximum data in the i-th column of the 200 fault data matrix under the fault type numbered n; a n,i,min It represents the minimum data in the i-th column of the 200 fault data matrix under the fault type numbered n.

[0073] S304 , calling the target feature extraction model to perform feature extraction on the normalized fault data matrix to obtain reference fault data features corresponding to the fault type.

[0074] In an optional implementation, when the computer device calls the target feature extraction model to extract features from the normalized fault data matrix, the following formulas (4) and (5) may be used.

[0075] (4)

[0076] (5)

[0077] In formula (4), c n,i,j It represents the normalized fault data matrix B of the fault type numbered n. n,m The element in the data feature matrix after the element in the i-th row and j-th column passes through the convolution layer; b n,m,i,j represents the fault data value of the i-th row and j-th column of the normalized fault data matrix under the fault type numbered n; K represents the weight of the convolution kernel; b represents the bias of the convolution kernel, and p represents the fault data matrix A. n,m The number of columns.

[0078] Optionally, the computer device may use the above formulas (4) and (5) to calculate each element in the normalized fault data matrix, c i,j Place it in the i-th row and j-th column of the fault data feature matrix, so that the reference fault data feature C corresponding to the fault type numbered n can be obtained n (or called reference fault data feature matrix) is shown in the following formula (6).

[0079] (6)

[0080] S305. When a fault occurs in the distribution network, the target feature extraction model is called to extract features of the current fault data of the distribution network to obtain features of the current fault data.

[0081] In an optional implementation, the computer device calls the target feature extraction model to extract features from the current fault data of the distribution network to obtain the current fault data features, which may include: integrating the current fault data to obtain the current fault data matrix; calling the target feature extraction network to extract features from the current fault data matrix to obtain the current fault data features. The target feature extraction network may be a convolutional neural network after determining the model parameters.

[0082] In this implementation, the computer device integrates the current fault data, and the obtained current fault data matrix can be shown in the following formula (7).

[0083] (7)

[0084] In formula (7), D represents the current fault data matrix; d i,j It represents the element in the i-th row and j-th column, where i=1,2,3,...,150 and j=1,2,3,...,24.

[0085] In this implementation, the computer device calls a target feature extraction network (such as a target convolutional neural network) to extract features from the current fault data matrix, and the obtained current fault data features can be shown in the following formula (8).

[0086] (8)

[0087] In formula (8), F represents the current fault data characteristics; f i,j It represents the element in the i-th row and j-th column, where i=1,2,3,...,150 and j=1,2,3,...,24.

[0088] The computer device can calculate the above f through the following formula (9): i,j .

[0089] (9)

[0090] In formula (9), d i,j It represents the element in the i-th row and j-th column of the current fault data matrix D; K represents the weight of the convolution kernel in the target convolutional neural network; b represents the bias of the convolution kernel in the target convolutional neural network.

[0091] S306. Determine a target reference fault data feature that matches the current fault data feature from a plurality of pre-determined reference fault data features; each reference fault data feature corresponds to a fault type.

[0092] In an optional embodiment, a computer device determines a target reference fault data feature that matches a current fault data feature from a plurality of predetermined reference fault data features, which may include: determining the probability that the current fault data corresponds to each reference fault data feature through a preset probability determination strategy based on the plurality of predetermined reference fault data features and the current fault data feature; and selecting, from multiple probabilities, a reference fault data feature corresponding to a maximum probability as a target reference fault data feature that matches the current fault data feature.

[0093] S307. Obtain a fault type corresponding to the target reference fault data feature as the current fault type of the distribution network.

[0094] In an embodiment of the present application, a computer device can simulate and obtain fault data corresponding to a distribution network under multiple fault types based on a pre-constructed distribution network model; integrate the fault data under each fault type to obtain a fault data matrix under the fault type; normalize the fault data matrix to obtain a normalized fault data matrix; call a target feature extraction model to perform feature extraction on the normalized fault data matrix to obtain a reference fault data feature corresponding to the fault type; in the event of a fault in the distribution network, call the target feature extraction model to perform feature extraction on the current fault data of the distribution network to obtain the current fault data feature; determine a target reference fault data feature that matches the current fault data feature from a plurality of predetermined reference fault data features; each reference fault data feature corresponds to a fault type; obtain the fault type corresponding to the target reference fault data feature as the current fault type of the distribution network. By adopting this method, the computer equipment can obtain reference fault data features corresponding to various fault types by simulating the fault conditions of the distribution network. In the event of a fault in the distribution network, the target reference fault data features that match the current fault data features can be determined by comparing the matching degree between the current fault data features and a plurality of pre-determined reference fault features. Since each reference fault feature corresponds to a fault type, the fault type corresponding to the target reference fault data feature is used as the current fault type of the distribution network, thereby improving the accuracy of the determined distribution network fault type.

[0095] See also Figure 5 , Figure 5 1 is a flow chart of another method for diagnosing a distribution network fault provided in an embodiment of the present application. Figure 3 Compared with the distribution network fault diagnosis method shown in Figure 5The method shown also describes how the computer device specifically determines the target reference fault data feature that matches the current fault data feature from a plurality of predetermined reference fault data features. Figure 5 As shown, the distribution network fault diagnosis method may include but is not limited to the following steps:

[0096] S501. Based on a pre-built distribution network model, simulate and obtain fault data corresponding to various fault types of the distribution network.

[0097] S502: Integrate the fault data under each fault type to obtain a fault data matrix under the fault type.

[0098] S503: normalize the fault data matrix to obtain a normalized fault data matrix.

[0099] S504 , calling the target feature extraction model to perform feature extraction on the normalized fault data matrix to obtain reference fault data features corresponding to the fault type.

[0100] S505. When a fault occurs in the distribution network, the target feature extraction model is called to extract features of current fault data of the distribution network to obtain features of the current fault data.

[0101] In an optional implementation, the relevant descriptions of steps S501 to S505 can refer to the descriptions of the aforementioned steps S301 to S305, and will not be repeated here.

[0102] S506 : Based on the predetermined multiple reference fault data features and the current fault data features, the probability of the current fault data corresponding to each reference fault data feature is determined by a preset probability determination strategy.

[0103] In an optional implementation, the preset probability determination strategy may be as shown in the following formula (10).

[0104] (10)

[0105] In formula (10), p(n) represents the probability of the current fault data corresponding to the nth fault type; f i,j represents the data of the i-th row and j-th column in the previous fault data matrix, where the current fault data matrix is ​​obtained by integrating the current fault data; c n,i,j It represents the data in the i-th row and j-th column of the reference fault data feature corresponding to the n-th fault type; K represents the weight of the convolution kernel in the target convolutional neural network.

[0106] S507 . Select, from the multiple probabilities, a reference fault data feature corresponding to a maximum probability as a target reference fault data feature that matches the current fault data feature.

[0107] S508. Obtain a fault type corresponding to the target reference fault data feature as the current fault type of the distribution network.

[0108] Exemplarily, assuming that the fault type corresponding to the target reference fault data feature is A-phase short circuit, the computer device can determine that the current fault type of the distribution network is A-phase short circuit.

[0109] In the embodiment of the present application, the computer device can determine the probability of the current fault data corresponding to each reference fault data feature through a preset probability determination strategy based on a plurality of pre-determined reference fault data features and the current fault data feature; select the reference fault data feature corresponding to the maximum probability from the multiple probabilities as the target reference fault data feature that matches the current fault data feature; obtain the fault type corresponding to the target reference fault data feature as the current fault type of the distribution network. In this way, by selecting the reference fault feature corresponding to the maximum probability as the target reference fault data feature that matches the current fault data feature, and using the fault type corresponding to the target reference fault data feature as the current fault type of the distribution network, the accuracy of the determined distribution network fault type can be improved.

[0110] In an optional embodiment, after step S303 or step S503, the computer device may further traverse each integer in a preset model parameter range, and input the normalized fault data matrix into a feature extraction model whose model parameters are each integer, to obtain multiple predicted fault data feature matrices; wherein each predicted fault data feature matrix is ​​a predicted fault data feature matrix corresponding to the fault type when the model parameters of the feature extraction model are each integer; determine the loss value between each predicted fault data feature matrix and the normalized fault data matrix; and select the model parameter corresponding to the minimum loss value from the multiple loss values ​​as the model parameter in the target feature extraction model.

[0111] In this implementation, the feature extraction model may be a convolutional neural network; the preset model parameter interval may be a value interval of the weight K of the convolution kernel in the convolutional neural network, and the interval may be 1-1000.

[0112] In this implementation, the computer device may use the following formula (11) to determine the loss value between each predicted fault data feature matrix and the normalized fault data matrix.

[0113] (11)

[0114] In formula (11), J represents the loss function; p represents the fault data matrix A n,m The number of columns; b n,m,i,j It represents the normalized fault data matrix B under the fault type numbered n n,m The fault data value of the i-th row and j-th column of n,i,j It represents the normalized fault data matrix B of the fault type numbered n. n,m The element in the i-th row and j-th column in the data feature matrix after passing through the convolution layer.

[0115] In this embodiment, the computer device can determine that the initial value of K is 1, the incremental compensation is 1, and 1000 loss function values ​​under different weights K are calculated between 1-1000, and based on the 1000 loss function values ​​obtained, the weight with the minimum loss function value is selected as the final parameter of the convolutional neural network.

[0116] See also Figure 6 , Figure 6 is a schematic diagram of a loss function curve provided in an embodiment of the present application. Figure 6 As shown in the figure, the horizontal axis represents the weight K, and the vertical axis represents the loss function value. Figure 6 It can be seen that when the weight value is 1000, the loss function value is the smallest. Therefore, the computer device can determine that the weight value in the target convolutional neural network is 1000.

[0117] By adopting this implementation mode, the computer device can more accurately determine the model parameters in the target feature extraction model, thereby facilitating the extraction of more accurate fault data features based on the target feature extraction model, and further facilitating the improvement of the accuracy of the determined distribution network fault type.

[0118] In order to verify the accuracy of the distribution network fault diagnosis method provided by the embodiment of the present application, the computer device sets the fault to be located on the first line L1 and the second line L2, the initial distance from the left bus M is set to 0.5kM, the incremental compensation is 0.1kM, and the fault data under 10 fault distances are obtained. The fault types are A-phase grounding, AB-phase short circuit, AB-phase short circuit grounding, three-phase grounding, and three-phase grounding short circuit. The fault transition resistance is 5Ω and 10Ω, thus, 20 fault diagnosis data matrices can be obtained for each fault type. Afterwards, the fault diagnosis results obtained by the distribution network fault diagnosis method provided by the embodiment of the present application are shown in Table 1 below.

[0119] Table 1 Fault diagnosis results

[0120]

[0121] It can be seen from Table 1 that the fault diagnosis result obtained by using the distribution network fault diagnosis method provided in the embodiment of the present application has high accuracy.

[0122] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0123] Based on the same inventive concept, the embodiment of the present application also provides a distribution network fault diagnosis device for implementing the distribution network fault diagnosis method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more distribution network fault diagnosis device embodiments provided below can refer to the limitations of the distribution network fault diagnosis method above, and will not be repeated here.

[0124] See also Figure 7 , Figure 7 Schematic diagram of a distribution network fault diagnosis device provided in an embodiment of the present application. Figure 7 As shown, the distribution network fault diagnosis device may include but is not limited to:

[0125] The feature extraction module 701 is used to call the target feature extraction model to extract features from the current fault data of the distribution network to obtain the features of the current fault data when a fault occurs in the distribution network;

[0126] A determination module 702 is used to determine a target reference fault data feature that matches the current fault data feature from a plurality of pre-determined reference fault data features; each reference fault data feature corresponds to a fault type;

[0127] The acquisition module 703 is used to acquire the fault type corresponding to the target reference fault data feature as the current fault type of the distribution network.

[0128] In one embodiment, the device may further include a processing module. The processing module is used to simulate and obtain fault data corresponding to the distribution network under various fault types based on a pre-built distribution network model; the determination module 702 is also used to call the target feature extraction model, and determine the reference fault data feature corresponding to the fault type based on the fault data under each fault type.

[0129] In one embodiment, the distribution network model includes a first line, a second line and a bus; when the processing module is used to simulate and obtain fault data corresponding to the distribution network under multiple fault types based on a pre-constructed distribution network model, it is specifically used to: determine the multiple fault types corresponding to the first line and the second line in the pre-constructed distribution network model; obtain the line information of the first line and the second line respectively, whose distance from the bus is less than or equal to a preset distance threshold under each fault type, and the line information includes voltage amplitude information, voltage phase information, current amplitude information and current phase information; integrate the line information of the first line and the second line respectively under each fault type to obtain the fault data corresponding to each fault type.

[0130] In one embodiment, when the determination module 702 is used to call the target feature extraction model and determine the reference fault data features corresponding to each fault type based on the fault data under the fault type, it is specifically used to: integrate the fault data under each fault type to obtain the fault data matrix under the fault type; normalize the fault data matrix to obtain the normalized fault data matrix; call the target feature extraction model to perform feature extraction on the normalized fault data matrix to obtain the reference fault data features corresponding to the fault type.

[0131] In one embodiment, the processing module is also used to traverse each integer in a preset model parameter range, and input the normalized fault data matrix into a feature extraction model whose model parameters are each integer, to obtain multiple predicted fault data feature matrices; wherein each predicted fault data feature matrix is ​​a predicted fault data feature matrix corresponding to the fault type when the model parameters of the feature extraction model are each integer; the determination module 702 is also used to determine the loss value between each predicted fault data feature matrix and the normalized fault data matrix; the determination module 702 is also used to select the model parameter corresponding to the minimum loss value from multiple loss values ​​as the model parameter in the target feature extraction model.

[0132] In one embodiment, when the determination module 702 is used to determine a target reference fault data feature that matches a current fault data feature from a plurality of predetermined reference fault data features, it is specifically used to: determine the probability of the current fault data corresponding to each reference fault data feature through a preset probability determination strategy based on the plurality of predetermined reference fault data features and the current fault data feature; and select, from multiple probabilities, a reference fault data feature corresponding to a maximum probability as the target reference fault data feature that matches the current fault data feature.

[0133] Each module in the above-mentioned distribution network fault diagnosis device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the terminal device in the form of hardware, or can be stored in the memory in the terminal device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0134] In an exemplary embodiment, the present application provides a computer device, which may be a terminal, and its internal structure diagram may be as follows: Figure 8 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a distribution network fault diagnosis method is implemented.

[0135] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0136] In an exemplary embodiment, the present application provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0137] When a fault occurs in the distribution network, the target feature extraction model is called to extract features from the current fault data of the distribution network to obtain the features of the current fault data;

[0138] Determining a target reference fault data feature that matches the current fault data feature from a plurality of pre-determined reference fault data features; each reference fault data feature corresponds to a fault type;

[0139] The fault type corresponding to the target reference fault data feature is obtained as the current fault type of the distribution network.

[0140] In one embodiment, when the processor executes the computer program, the following steps are also implemented: based on a pre-built distribution network model, the fault data corresponding to the distribution network under various fault types are simulated and obtained; the target feature extraction model is called to determine the reference fault data feature corresponding to the fault type based on the fault data under each fault type.

[0141] In one embodiment, the distribution network model includes a first line, a second line and a bus; when the processor executes a computer program to simulate and obtain fault data corresponding to the distribution network under multiple fault types based on a pre-constructed distribution network model, the following steps are specifically implemented: determining multiple fault types corresponding to the first line and the second line in the pre-constructed distribution network model; obtaining line information of the first line and the second line respectively whose distance from the bus is less than or equal to a preset distance threshold under each fault type, the line information including voltage amplitude information, voltage phase information, current amplitude information and current phase information; integrating the line information of the first line and the second line respectively under each fault type to obtain the fault data corresponding to each fault type.

[0142] In one embodiment, the processor executes a computer program to call the target feature extraction model, and based on the fault data under each fault type, determines the reference fault data feature corresponding to the fault type, specifically implementing the following steps: integrating the fault data under each fault type to obtain a fault data matrix under the fault type; normalizing the fault data matrix to obtain a normalized fault data matrix; calling the target feature extraction model to perform feature extraction on the normalized fault data matrix to obtain a reference fault data feature corresponding to the fault type.

[0143] In one embodiment, when the processor executes the computer program, the following steps are also implemented: traversing each integer in a preset model parameter range, and inputting the normalized fault data matrix into a feature extraction model whose model parameters are each integer, to obtain multiple predicted fault data feature matrices; wherein each predicted fault data feature matrix is ​​a predicted fault data feature matrix corresponding to the fault type when the model parameters of the feature extraction model are each integer; determining the loss value between each predicted fault data feature matrix and the normalized fault data matrix; and selecting, from multiple loss values, the model parameter corresponding to the minimum loss value as the model parameter in the target feature extraction model.

[0144] In one embodiment, when a processor executes a computer program to determine a target reference fault data feature that matches a current fault data feature from a plurality of predetermined reference fault data features, the following steps are specifically implemented: based on the plurality of predetermined reference fault data features and the current fault data feature, the probability that the current fault data corresponds to each reference fault data feature is determined through a preset probability determination strategy; and from multiple probabilities, a reference fault data feature corresponding to a maximum probability is selected as a target reference fault data feature that matches the current fault data feature.

[0145] In an exemplary embodiment, the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the following steps are implemented:

[0146] When a fault occurs in the distribution network, the target feature extraction model is called to extract features from the current fault data of the distribution network to obtain the features of the current fault data;

[0147] Determining a target reference fault data feature that matches the current fault data feature from a plurality of pre-determined reference fault data features; each reference fault data feature corresponds to a fault type;

[0148] The fault type corresponding to the target reference fault data feature is obtained as the current fault type of the distribution network.

[0149] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: based on a pre-built distribution network model, the fault data corresponding to the distribution network under various fault types are simulated and obtained; the target feature extraction model is called to determine the reference fault data feature corresponding to the fault type based on the fault data under each fault type.

[0150] In one embodiment, the distribution network model includes a first line, a second line and a bus; when a computer program is executed by a processor to simulate and obtain fault data corresponding to the distribution network under multiple fault types based on a pre-constructed distribution network model, the following steps are specifically implemented: determining multiple fault types corresponding to the first line and the second line in the pre-constructed distribution network model; obtaining line information of the first line and the second line respectively whose distance from the bus is less than or equal to a preset distance threshold under each fault type, the line information including voltage amplitude information, voltage phase information, current amplitude information and current phase information; integrating the line information of the first line and the second line respectively under each fault type to obtain the fault data corresponding to each fault type.

[0151] In one embodiment, the computer program is executed by a processor to call the target feature extraction model, and based on the fault data under each fault type, when determining the reference fault data feature corresponding to the fault type, the following steps are specifically implemented: integrating the fault data under each fault type to obtain a fault data matrix under the fault type; normalizing the fault data matrix to obtain a normalized fault data matrix; calling the target feature extraction model to perform feature extraction on the normalized fault data matrix to obtain a reference fault data feature corresponding to the fault type.

[0152] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: traversing each integer in a preset model parameter range, and inputting the normalized fault data matrix into a feature extraction model whose model parameters are each integer, to obtain multiple predicted fault data feature matrices; wherein each predicted fault data feature matrix is ​​a predicted fault data feature matrix corresponding to the fault type when the model parameters of the feature extraction model are each integer; determining the loss value between each predicted fault data feature matrix and the normalized fault data matrix; and selecting, from multiple loss values, the model parameter corresponding to the minimum loss value as the model parameter in the target feature extraction model.

[0153] In one embodiment, when a computer program is executed by a processor to determine a target reference fault data feature that matches a current fault data feature from a plurality of predetermined reference fault data features, the following steps are specifically implemented: based on the plurality of predetermined reference fault data features and the current fault data feature, the probability that the current fault data corresponds to each reference fault data feature is determined through a preset probability determination strategy; and from multiple probabilities, a reference fault data feature corresponding to a maximum probability is selected as the target reference fault data feature that matches the current fault data feature.

[0154] In an exemplary embodiment, the present application provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0155] When a fault occurs in the distribution network, the target feature extraction model is called to extract features from the current fault data of the distribution network to obtain the features of the current fault data;

[0156] Determining a target reference fault data feature that matches the current fault data feature from a plurality of pre-determined reference fault data features; each reference fault data feature corresponds to a fault type;

[0157] The fault type corresponding to the target reference fault data feature is obtained as the current fault type of the distribution network.

[0158] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: based on a pre-built distribution network model, the fault data corresponding to the distribution network under various fault types are simulated and obtained; the target feature extraction model is called to determine the reference fault data feature corresponding to the fault type based on the fault data under each fault type.

[0159] In one embodiment, the distribution network model includes a first line, a second line and a bus; when a computer program is executed by a processor to simulate and obtain fault data corresponding to the distribution network under multiple fault types based on a pre-constructed distribution network model, the following steps are specifically implemented: determining multiple fault types corresponding to the first line and the second line in the pre-constructed distribution network model; obtaining line information of the first line and the second line respectively whose distance from the bus is less than or equal to a preset distance threshold under each fault type, the line information including voltage amplitude information, voltage phase information, current amplitude information and current phase information; integrating the line information of the first line and the second line respectively under each fault type to obtain the fault data corresponding to each fault type.

[0160] In one embodiment, the computer program is executed by a processor to call the target feature extraction model, and based on the fault data under each fault type, when determining the reference fault data feature corresponding to the fault type, the following steps are specifically implemented: integrating the fault data under each fault type to obtain a fault data matrix under the fault type; normalizing the fault data matrix to obtain a normalized fault data matrix; calling the target feature extraction model to perform feature extraction on the normalized fault data matrix to obtain a reference fault data feature corresponding to the fault type.

[0161] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: traversing each integer in a preset model parameter range, and inputting the normalized fault data matrix into a feature extraction model whose model parameters are each integer, to obtain multiple predicted fault data feature matrices; wherein each predicted fault data feature matrix is ​​a predicted fault data feature matrix corresponding to the fault type when the model parameters of the feature extraction model are each integer; determining the loss value between each predicted fault data feature matrix and the normalized fault data matrix; and selecting, from multiple loss values, the model parameter corresponding to the minimum loss value as the model parameter in the target feature extraction model.

[0162] In one embodiment, when a computer program is executed by a processor to determine a target reference fault data feature that matches a current fault data feature from a plurality of predetermined reference fault data features, the following steps are specifically implemented: based on the plurality of predetermined reference fault data features and the current fault data feature, the probability that the current fault data corresponds to each reference fault data feature is determined through a preset probability determination strategy; and from multiple probabilities, a reference fault data feature corresponding to a maximum probability is selected as the target reference fault data feature that matches the current fault data feature.

[0163] It should be noted that the data involved in this application (including but not limited to current fault data, reference fault data characteristics, etc.) are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0164] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0165] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0166] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the present application. It should be noted that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A distribution network fault diagnosis method, characterized in that: The method comprises: In the event of a fault in the distribution network, a target feature extraction model is called to extract features from current fault data of the distribution network to obtain features of the current fault data; Determining a target reference fault data feature that matches the current fault data feature from a plurality of pre-determined reference fault data features; each reference fault data feature corresponds to a fault type; A fault type corresponding to the target reference fault data feature is obtained as the current fault type of the distribution network.

2. The method according to claim 1, characterized in that The multiple reference fault data features are determined in the following manner: Based on the pre-built distribution network model, the fault data corresponding to various fault types of the distribution network are simulated; The target feature extraction model is called to determine a reference fault data feature corresponding to each fault type based on the fault data under the fault type.

3. The method according to claim 2, characterized in that The distribution network model includes a first line, a second line and a busbar; based on the pre-built distribution network model, the fault data corresponding to the distribution network under various fault types are simulated, including: Determine multiple fault types corresponding to the first line and the second line in a pre-constructed distribution network model; Acquire line information of the first line and the second line respectively whose distance from the busbar is less than or equal to a preset distance threshold under each of the fault types, wherein the line information includes voltage amplitude information, voltage phase information, current amplitude information, and current phase information; For each of the fault types, the line information of the first line and the second line are respectively integrated to obtain corresponding fault data for each of the fault types.

4. The method according to claim 2, characterized in that: The calling of the target feature extraction model to determine the reference fault data feature corresponding to each fault type based on the fault data under the fault type includes: Integrate the fault data under each fault type to obtain a fault data matrix under the fault type; Normalizing the fault data matrix to obtain a normalized fault data matrix; The target feature extraction model is called to perform feature extraction on the normalized fault data matrix to obtain reference fault data features corresponding to the fault type.

5. The method according to claim 4, characterized in that The method further comprises: Traversing each integer in the preset model parameter interval, and inputting the normalized fault data matrix into a feature extraction model whose model parameter is each of the integers, to obtain a plurality of predicted fault data feature matrices; wherein each of the predicted fault data feature matrices is a predicted fault data feature matrix corresponding to the fault type when the model parameter of the feature extraction model is each of the integers; Determining a loss value between each of the predicted fault data feature matrices and the normalized fault data matrix; From multiple loss values, select the model parameter corresponding to the minimum loss value as the model parameter in the target feature extraction model.

6. The method according to any one of claims 1 to 5, characterized in that: The step of determining a target reference fault data feature that matches the current fault data feature from a plurality of predetermined reference fault data features includes: Based on a plurality of predetermined reference fault data features and the current fault data feature, the probability of the current fault data corresponding to each of the reference fault data features is determined by a preset probability determination strategy; From the multiple probabilities, a reference fault data feature corresponding to a maximum probability is selected as a target reference fault data feature that matches the current fault data feature.

7. A distribution network fault diagnosis device, characterized in that: The device comprises: A feature extraction module is used to call a target feature extraction model to extract features of current fault data of the distribution network to obtain features of the current fault data when a fault occurs in the distribution network; A determination module, used to determine a target reference fault data feature that matches the current fault data feature from a plurality of pre-determined reference fault data features; each reference fault data feature corresponds to a fault type; The acquisition module is used to acquire the fault type corresponding to the target reference fault data feature as the current fault type of the distribution network.

8. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method according to any one of claims 1 to 6 when executing the computer program.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images