Power distribution line topology identification method and device, equipment and storage medium

Abstract

The application discloses a power distribution line topology identification method, device and equipment and a storage medium. The power distribution line topology identification method specifically comprises the following steps: obtaining an injection point and an identification point to be identified in a power distribution line, and collecting topology verification data to be identified corresponding to the injection point and the identification point; obtaining a connection probability corresponding to the identification point based on the topology verification data to be identified and a pre-trained topology relationship identification network, wherein the topology relationship identification network is obtained by training a neural network model based on sample topology verification data and expected label data corresponding to the sample topology verification data, and the expected label data comprises data for indicating a connection relationship between each identification point and the injection point; and in the case that the connection probability corresponding to the identification point is greater than or equal to a preset first probability threshold, the connection relationship between the injection point and the identification point is determined as an existing connection relationship. The accurate identification of the line topology relationship is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power distribution network, and particularly relates to a power distribution line topology identification method, device and equipment and a storage medium. BACKGROUND

[0002] It is necessary for power enterprises to pay attention to the fine management of power distribution network. The connection relationship of power distribution network is an important part of fine management. Inaccurate station-line relationship, line-transformer relationship and house-transformer relationship may cause the consequences of power outage notification not in place and abnormal line loss management. Therefore, in the related art, the connection relationship of power distribution network is often verified by various technical means.

[0003] At present, the main technical means in this field are manual verification, data method and signal method. Manual verification depends on the production experience of operating personnel, and it is difficult to deal with complex cable lines, multi-branch lines erected on the same pole and double power distribution stations. The data method requires stable operation of the line flow, but with high proportion of distributed photovoltaic, small hydropower, dynamic interaction of energy storage facilities and electric vehicles, the new type of power distribution network will present a complex form of source-load interaction and multi-directional flow. Therefore, the algorithm driven by data alone has systematic errors. The signal method refers to injecting a characteristic signal into the line by means of equipment, and verifying the topology relationship by the principle that the characteristic signal can be present at any position on the same line. The signal method can perfectly avoid the accuracy problem caused by imperfect power data, and ensures the correctness of topology verification by using the intuitive physical connection method.

[0004] However, since the characteristic signal is propagated in the line with power frequency voltage and current, it is easy to be disturbed by harmonics on the line, affecting the identification effect. SUMMARY

[0005] The present application provides a power distribution line topology identification method, device, equipment and storage medium, to solve the problem that in the topology relationship verification process of power distribution line, the characteristic signal is easy to be disturbed by harmonics on the line, thereby affecting the topology identification effect of power distribution line. The beneficial effects of effectively reducing the disturbance of abnormal harmonics on the line to the pulse signal, improving the identification accuracy and speed, and improving the line loss management level of power supply company are achieved.

[0006] According to an aspect of the present application, a power distribution line topology identification method is provided, which comprises:

[0007] Obtaining an injection point and an identification point to be identified in the power distribution line, and collecting topology verification data to be identified corresponding to the injection point and the identification point;

[0008] obtain a connection probability corresponding to the identification point based on the to-be-identified topology verification data and a pre-trained topology relationship identification network, wherein the topology relationship identification network is obtained by training a neural network model based on sample topology verification data and expected label data corresponding to the sample topology verification data, and the expected label data includes data for indicating a connection relationship between each identification point and the injection point;

[0009] In a case where the connection probability corresponding to the identification point is greater than or equal to a preset first probability threshold, the connection relationship between the injection point and the identification point is determined as an existing connection relationship.

[0010] According to another aspect of the present application, there is provided a topology identification device for a power distribution line, which comprises:

[0011] a data acquisition module configured to acquire an injection point and an identification point to be identified in the power distribution line, and collect to-be-identified topology verification data corresponding to the injection point and the identification point;

[0012] obtain a connection probability corresponding to the identification point based on the to-be-identified topology verification data and a pre-trained topology relationship identification network, wherein the topology relationship identification network is obtained by training a neural network model based on sample topology verification data and expected label data corresponding to the sample topology verification data, and the expected label data includes data for indicating a connection relationship between each identification point and the injection point;

[0013] In a case where the connection probability corresponding to the identification point is greater than or equal to a preset first probability threshold, the connection relationship between the injection point and the identification point is determined as an existing connection relationship.

[0014] According to another aspect of the present application, there is provided an electronic device, which comprises:

[0015] at least one processor; and

[0016] a memory in communication connection with the at least one processor; wherein

[0017] the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the topology identification method for the power distribution line according to any one of the embodiments of the present application.

[0018] According to another aspect of the present application, there is provided a computer readable storage medium storing computer instructions for enabling a processor to implement the topology identification method for the power distribution line according to any one of the embodiments of the present application when executed by the processor.

[0019] The technical scheme of the embodiment of the application first acquires an injection point and a recognition point to be identified in a power distribution line, collects topology verification data to be identified corresponding to the injection point and the recognition point, and provides a data source for identifying the connection relationship of the injection point and the recognition point. Then, based on the topology verification data to be identified and a pre-trained topology relationship identification network, a connection probability corresponding to the recognition point is obtained, wherein the topology relationship identification network is obtained by training a neural network model based on sample topology verification data and expected label data corresponding to the sample topology verification data, and the expected label data includes data for indicating the connection relationship between each recognition point and the injection point, thereby providing a judgment basis for determining the connection relationship between the injection point and the recognition point. Finally, in a case where the connection probability corresponding to the recognition point is greater than or equal to a preset first probability threshold, the connection relationship between the injection point and the recognition point is determined as an existing connection relationship, thereby improving the accuracy and efficiency of topology relationship identification. The problem that feature signals are easily disturbed by harmonics on the line in the topology identification process of the power distribution line, thereby affecting the topology identification effect of the power distribution line, is solved. The beneficial effects of effectively reducing the disturbance of abnormal harmonics on the line on the pulse signal, improving the identification accuracy and speed, and further improving the line loss management level of the power supply company are achieved.

[0020] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the application, nor is it used to limit the scope of the application. Other features of the application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0022] Figure 1 is a flow chart of a topology identification method of a power distribution line according to the first embodiment of the application;

[0023] Figure 2 is a flow chart of a topology identification method of a power distribution line according to the first embodiment of the application;

[0024] Figure 3 is a flow chart of a topology identification method of a power distribution line according to the second embodiment of the application;

[0025] Figure 4 is a flow chart of a topology identification method of a power distribution line according to the third embodiment of the application;

[0026] Figure 5is a structural schematic diagram of a power distribution line topology identification device provided according to an embodiment four of the present application;

[0027] Figure 6 is a structural schematic diagram of an electronic device implementing a power distribution line topology identification method of an embodiment of the present application. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts should fall within the protection scope of the present application.

[0029] It should be noted that the terms "first", "second" and "target" and the like in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or a chronological sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but can include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0030] Embodiment one

[0031] Figure 1 A flowchart of a power distribution line topology identification method is provided for the embodiment one of the present application. The present embodiment can be applicable to the case of identifying the topology connection relationship of a power distribution line. The method can be executed by a power distribution line topology identification device. The power distribution line topology identification device can be realized in the form of hardware and / or software, and can be configured in an electronic device. As shown in the figure, the method comprises: Figure 1

[0032] S110, acquiring an injection point and an identification point to be identified in the power distribution line, and collecting topology verification data to be identified corresponding to the injection point and the identification point.

[0033] ​In this embodiment, the injection point can be any electrical connection position on the distribution line, such as a 10kV conductor, a circuit breaker, a sectional switch, a distribution transformer high and low voltage side, a 0.4kV conductor, and a user meter, etc. Optionally, the injection point can be on the low-voltage line. The identification point can be the switch cabinet of the 10kV line of the substation, the switch cabinet of the first end of any branch line on the 10kV line, and the low-voltage bus of the distribution transformer, etc. It can be understood that the injection point and the identification point can be one or more, and the identification point is electrically connected to the injection point. The electrical circuit can be at least one closed state. The to-be-identified topology verification data can be corresponding to the injection point and the identification point, and the topology vector data used to identify the topology connection relationship between the injection point and the identification point.

[0034] Optionally, a preset injection signal at the injection point is obtained, an injection signal is injected at the injection point, a feature value of a plurality of target feature parameters associated with the injection point and the identification point is collected, a to-be-identified topology vector is constructed based on the feature values of the plurality of target feature parameters, and to-be-identified topology verification data corresponding to the injection point and the identification point is constructed based on the to-be-identified topology vector.

[0035] Optionally, the to-be-identified topology vector is taken as the to-be-identified topology verification data corresponding to the injection point and the identification point; or a plurality of to-be-identified topology vectors constructed by injecting the injection signal at the injection point multiple times are obtained, and the to-be-identified topology verification data corresponding to the injection point and the identification point is constructed based on the plurality of to-be-identified topology vectors.

[0036] In this embodiment, the injection signal can be a pulse current signal injected at the injection point. The target feature parameter can be a parameter associated with the injection point and the identification point, such as a pulse width, a peak value, and other electrical characteristic parameters collected by the injection point and the identification point.

[0037] For example, a 10kV conductor on a low-voltage line can be taken as an injection point, and a switch cabinet of a 10kV line of a substation can be taken as an identification point. To identify the topology connection relationship between the 10kV conductor and the switch cabinet of the 10kV line of the substation, a pulse current signal can be injected at the 10kV conductor, a feature value of a plurality of electrical characteristic parameters associated with the 10kV conductor and the switch cabinet of the 10kV line of the substation is collected, and then a to-be-identified topology vector can be constructed based on the feature values of the plurality of electrical characteristic parameters collected.

[0038] In this embodiment, to increase the number of samples of electrical characteristic parameters, a pulse current signal can be injected at the 10kV conductor multiple times, and then a to-be-identified topology vector can be constructed based on the feature values of a plurality of electrical characteristic parameters collected by the multiple times of injecting the pulse current signal. This setting has the advantage that it can avoid the situation that the connection relationship is not accurately identified due to deviation of the target feature data collected in a single collection process.

[0039] Optionally, the target characteristic parameters include signal parameters of the injection signal at the injection point, signal parameters of the target identification signal corresponding to the injection signal at the identification point, interval parameters from the injection point to the identification point, and electrical parameters of the identification point.

[0040] Optionally, by injecting an injection signal at the injection point, characteristic values ​​of multiple target characteristic parameters associated with the injection point and the identification point are collected, including: determining the characteristic values ​​of the target signal parameters of the injection signal, wherein the signal parameters include pulse width, peak value and frequency; injecting the injection signal at the injection point, collecting the target identification signal corresponding to the injection signal at the identification point corresponding to the injection point, and determining the characteristic values ​​of the target signal parameters of the target identification signal; obtaining the characteristic values ​​of the target interval parameters from the injection point to the identification point and the characteristic values ​​of the target electrical parameters of the identification point, wherein the target interval signal includes at least electrical distance and impedance value.

[0041] For example, the topology verification data M to be identified can be represented as follows:

[0042]

[0043] The topology vector L to be identified may be a row vector in the topology verification data M to be identified, as shown below:

[0044] L=(X n , Y n1 ,…,Y nm , D n , C n1 ,…,C nm )

[0045] Among them, X n Y is the characteristic value of the target signal parameter corresponding to the injection signal when the injection signal is injected at the injection point for the nth time, nm is the characteristic value of the target signal parameter corresponding to the target recognition signal collected at the mth recognition point when the injection signal is injected at the injection point for the nth time, D n The characteristic value of the target interval parameter between the injection point and the identification point when the injection signal is injected at the injection point for the nth time, C nm is the characteristic value of the target electrical parameter of the mth identification point when the injection signal is injected at the injection point for the nth time.

[0046] Optionally, the target signal parameters corresponding to the injection signal may be multiple, such as the pulse width, peak value and frequency corresponding to the injection signal. Assuming that there are i target signal parameters corresponding to the injection signal, then, It can be expressed as:

[0047]

[0048] Among them, X1i Xi is the characteristic value of the i-th electrical characteristic parameter of the injection signal obtained at the first injection of the injection signal at the injection point, ni Xi is the characteristic value of the i-th electrical characteristic parameter of the injection signal obtained at the first injection of the injection signal at the injection point,

[0049] Optionally, n can be 3-5 times, and the interval time of each injection of the pulse current signal can be 1 min.

[0050] In the embodiment of the present application, optionally, the signal parameters of the target identification signal can correspond to the injection signal and contain the same or corresponding electrical signal parameters.

[0051] Similarly, the target signal parameters corresponding to the target identification signal can be multiple, such as pulse width, peak value, and frequency corresponding to the target identification signal, assuming that the target signal parameters corresponding to the target identification signal are j, at this time, which can be expressed as:

[0052]

[0053] wherein Ymj is the characteristic value of the j-th target electrical parameter of the m-th identification point obtained at the first injection at the injection point, 1j wherein Ymj is the characteristic value of the j-th target electrical parameter of the m-th identification point obtained at the first injection at the injection point, ni Ymnj is the characteristic value of the j-th target electrical parameter of the m-th identification point of the target identification signal obtained at the n-th injection of the injection signal at the injection point.

[0054] Optionally, the target interval parameters from the injection point to the identification point can be multiple, such as electrical distance and impedance value from the injection point to the identification point, assuming that the target interval parameters from the injection point to the identification point are g, at this time, which can be expressed as:

[0055]

[0056] wherein Dmg is the characteristic value of the g-th target interval parameter from the injection point to the identification point obtained at the first injection of the injection signal at the injection point, 1g wherein Dmg is the characteristic value of the g-th target interval parameter from the injection point to the identification point obtained at the first injection of the injection signal at the injection point, ni Dmn is the characteristic value of the g-th target interval parameter from the injection point to the identification point obtained at the n-th injection of the injection signal at the injection point.

[0057] Optionally, the target electrical parameters corresponding to the identification point can be multiple, such as the transformation ratio and accuracy of the mutual inductor of the identification point, assuming that the target electrical parameters corresponding to the identification point are h, at this time, which can be expressed as:

[0058]

[0059] Cm,h 1h Cm,h nh Cm,h

[0060] S120, based on the to-be-identified topology verification data and the pre-trained topology relationship identification network, obtain the connection probability corresponding to the identification point, wherein the topology relationship identification network is obtained by training the neural network model based on the sample topology verification data and the expected label data corresponding to the sample topology verification data, and the expected label data includes data for indicating the connection relationship between each identification point and the injection point.

[0061] In this embodiment, the pre-trained topology relationship identification network can be a neural network model pre-trained for identifying topology relationship, for example, can be an LSTM model, a GRU model, etc. The connection probability corresponding to the identification point can be the probability of the connection relationship between the identification point and the injection point. The sample topology verification data can be the topology verification data for training the neural network model, and the expected label data can be obtained in advance according to the actual situation, including the label data for indicating the connection relationship between each identification point and the injection point.

[0062] Specifically, based on the to-be-identified topology verification data constructed based on the feature values of the collected multiple electrical characteristic parameters, and the pre-trained neural network model for identifying topology relationship, the connection probability corresponding to the identification point can be obtained. The neural network model for identifying topology relationship can be obtained by training the neural network model based on the topology verification data for training the neural network model and the expected label data corresponding to the sample topology verification data. The expected label data can include data for indicating the connection relationship between each identification point and the injection point.

[0063] S130, in the case where the connection probability corresponding to the identification point is greater than or equal to a preset first probability threshold, determining that the connection relationship between the injection point and the identification point exists.

[0064] In this embodiment, the first probability threshold can be a lower limit value representing the probability of the connection relationship between the identification point and the injection point, for example, can be 0.8. In the case where the connection probability corresponding to the identification point is greater than or equal to the first probability threshold, the connection relationship between the injection point and the identification point can be determined to exist.

[0065] Optionally, in the case that there are multiple identification points connected with the injection point, the connection relationship between the identification point with the largest connection probability and the injection point is determined as the main connection relationship, and the connection relationship between the remaining identification points connected with the injection point and the injection point is determined as the secondary connection relationship.

[0066] In this embodiment, the main connection relationship can be a connection relationship indicating that the identification point and the injection point are directly powered in the same power flow direction. The secondary connection relationship can be a closed electrical loop formed by the identification point and the injection point, but the power flow directions of the two are inconsistent, and the injection point is not directly powered by the identification point.

[0067] Specifically, in the case that there are multiple identification points connected with the injection point, the connection relationship between the identification point with the largest connection probability and the injection point can be determined as the main connection relationship in the same power flow direction, and the injection point is directly powered by the identification point. The connection relationship between the remaining identification points connected with the injection point and the injection point is determined as the secondary connection relationship in the inconsistent power flow direction, and the injection point is not directly powered by the identification point.

[0068] Optionally, in the case that the connection probability corresponding to the identification point is less than the preset second probability threshold, the operation of inputting the to-be-identified topology verification data corresponding to the injection point and the identification point into the pre-trained topology relationship identification network to obtain the connection probability corresponding to the identification point is performed based on a preset collection number threshold.

[0069] In this embodiment, the second probability threshold can be the maximum value in the connection probability results corresponding to all identification points in the case that it is determined that the to-be-identified topology verification data needs to be re-collected, which can be 0.7, for example. In the case that the connection probability corresponding to all identification points is less than 0.7, it can be determined that all identification points do not have a connection relationship with the injection point, and the topology verification data needs to be re-collected. The preset collection number threshold can be the maximum number of times of re-collecting the topology verification data, which can be 5 times, for example.

[0070] As Figure 2As shown, when the operation of repeatedly inputting the to-be-identified topology verification data into the pre-trained topology relationship identification network to obtain the connection probability corresponding to the identification point is performed, if the number of repeated operations exceeds the preset collection number threshold, it is still not determined that the identification point and the injection point have a connection relationship, and whether the electrical loop connection relationship of the identification point and the injection point meets the requirements can be re-evaluated. If the electrical loop connection relationship of the identification point and the injection point meets the requirements, the number of times of collecting topology verification data can be cleared, and the operations of collecting the to-be-identified topology verification data corresponding to the injection point and the identification point and S120-S130 can be performed again. If the evaluation result is that the electrical loop connection relationship of the identification point and the injection point does not meet the requirements, the positions of the injection point and the identification point can be replaced, and the operations of S110-S130 can be performed again.

[0071] The technical scheme of the embodiment, by inputting the to-be-identified topology verification data corresponding to the injection point and the identification point into the pre-trained topology relationship identification network, obtaining the connection probability corresponding to the identification point, and then determining the connection relationship of the power distribution line based on the connection probability corresponding to the identification point, improves the accuracy and efficiency of topology relationship identification. The problem that the feature signal is easily disturbed by harmonics on the line in the topology identification process of the power distribution line is solved. The beneficial effects of effectively reducing the disturbance of abnormal harmonics on the line to the pulse signal and improving the identification accuracy and speed are achieved.

[0072] Embodiment Two

[0073] Figure 3 A flowchart of a topology identification method of a power distribution line provided for the second embodiment of the application, the embodiment is based on the above-mentioned embodiments, and specifically describes a method of performing data scaling processing on to-be-identified topology verification data to obtain and use the scaled identification topology verification data. The specific implementation can be referred to the description of the embodiment. Among them, the same or similar technical features as the foregoing embodiments will not be described again. As shown in the figure, the method comprises: Figure 2

[0074] S310, obtaining an injection point and an identification point to be identified in a power distribution line, and collecting to-be-identified topology verification data corresponding to the injection point and the identification point.

[0075] S320, for each feature value in the to-be-identified topology verification data, obtaining a plurality of feature values corresponding to a target feature parameter corresponding to the feature value as reference values.

[0076] In the embodiment, the plurality of feature values corresponding to the target feature parameter can be a plurality of feature values corresponding to columns of pulse width, peak value and frequency signal parameters in the to-be-identified topology verification data matrix constructed in the foregoing embodiments when the injection signal is multiple times.

[0077] ​Assuming that the target signal parameters corresponding to the injection signal have i, may be expressed as:

[0078]

[0079] At this time, the characteristic value X 11 The reference value of the characteristic value X 11 ~X n1 corresponding to the same target signal parameter is the characteristic value X 11 of the pulse width of the injection signal obtained when the injection signal is injected for the first time. 11 The reference value corresponding to the characteristic value X i is the characteristic value of the pulse width of the injection signal obtained when the injection signal is injected for the first time to the nth time.

[0080] S330, scaling each characteristic value based on the average and standard deviation of the plurality of reference values to obtain a target value corresponding to the characteristic value. The calculation method of the target value is shown in the following formula:

[0081]

[0082] wherein, is the target value obtained after scaling each characteristic value, μ i is the average of the plurality of reference values in the foregoing embodiments, and σ i is the standard deviation of the plurality of reference values in the foregoing embodiments.

[0083] S340, updating each characteristic value in the to-be-identified topology verification data to a target value corresponding to the characteristic value to obtain target identification topology verification data.

[0084] S350, inputting the target identification topology verification data into the pre-trained topology relationship identification network to obtain a connection probability corresponding to the identification point.

[0085] S360, in the case where the connection probability corresponding to the identification point is greater than or equal to a preset first probability threshold, determining that the connection relationship between the injection point and the identification point exists.

[0086] The technical scheme of the embodiment, by inputting the target value obtained after scaling the to-be-identified topology verification data corresponding to the injection point and the identification point into the pre-trained topology relationship identification network, obtaining a connection probability corresponding to the identification point, and then determining the connection relationship of the power distribution line based on the connection probability corresponding to the identification point. It can effectively avoid the influence of noise data in multiple injections on the identification result, further improve the accuracy of topology relationship identification. It realizes the beneficial effects of effectively reducing the interference of abnormal harmonics on the pulse signal on the line, improving the identification accuracy and speed.

[0087] Embodiment Three

[0088] Figure 4 A flowchart of a power distribution line topology identification method provided for Embodiment Three of the present application is provided, and this embodiment is based on the above-mentioned embodiments and specifically describes a training method for identifying the topology relationship network. The specific implementation can be referred to the description of this embodiment. Among them, the same or similar technical features as the foregoing embodiments will not be described here again. As shown in the following table, the method comprises: Figure 4

[0089] S410, obtaining sample topology verification data and determining expected label data corresponding to the sample topology verification data.

[0090] Specifically, a plurality of sample topology verification data is obtained. Like the topology verification data to be identified, each sample topology verification data can also include the pulse width, peak value, and frequency of the injected signal, the pulse width, peak value, and frequency of the plurality of identification signals, the electrical distance and impedance value from the plurality of injection points to the identification points, and the electrical parameters of the plurality of identification points, etc. corresponding to the target feature parameters of the injection points and identification points as samples.

[0091] As mentioned above, the expected label data corresponding to the sample topology verification data can be a sequence composed of data indicating the connection relationship between each identification point and the injection point corresponding to the sample topology verification data. Exemplarily, assuming that there is 1 injection point, and the identification points corresponding to the injection point are m, at this time, the expected label data is a label sequence composed of correct connection relationship data between the injection point and each identification point, and the sequence contains m elements. Specifically, the element value of each element in the expected label data can be "1" or "0", for example, when the identification point and the injection point are the main connection relationship, the corresponding element value in the expected label data is 1, otherwise it is "0".

[0092] S420, constructing a training set and a test set based on the sample topology verification data and the expected label data corresponding to the sample topology verification data.

[0093] In this embodiment, the training set and the test set can be constructed by selecting the obtained sample topology verification data, a part of which is used as the training set and the remaining part is used as the test set. Optionally, the training set can include a set of 300-500 topology verification data matrices as shown below.

[0094] Optionally, similar to the data scaling processing of the topology verification data of the training set and the test set in the foregoing embodiments, the constructed topology verification data of the training set and the test set can be subjected to data scaling processing.

[0095] ​Specifically, for each feature value in the training set and test set topology verification data, a plurality of feature values corresponding to the target feature parameter corresponding to the feature value are obtained as reference values. Then, each feature value can be scaled based on the average and standard deviation of the plurality of reference values to obtain a target value corresponding to the feature value. The calculation method of the target value can be shown in the following formula:

[0096]

[0097] wherein, is the target value obtained by scaling each feature value, μ i is the average of the plurality of reference values in the foregoing embodiments, σ i is the standard deviation of the plurality of reference values in the foregoing embodiments. Finally, each feature value in the training set and test set topology verification data is updated to the target value corresponding to the feature value to obtain the target training set and test set topology verification data.

[0098] S430, based on the sample topology verification data and the expected label data in the training set, the pre-established neural network model is trained to obtain a preliminary recognition network.

[0099] wherein, the neural network model includes a long short-term memory neural network (long short-term memory neural network) or a gated recurrent unit.

[0100] Specifically, the neural network model with preset model parameters is trained based on the training data set to obtain a preliminary recognition network. The model parameters include batch_size, activation function, number of neural network layers, gradient descent algorithm, optimization algorithm, learning rate, etc.

[0101] Exemplarily, the sample topology verification data in the training set can be input into the neural network model with preset model parameters to obtain model output data; based on the preset loss function, the model output data and the expected label data, the model loss is determined, and the model parameters of the neural network model are adjusted based on the model loss to obtain the preliminary recognition network.

[0102] It should be noted that in different topological relationship scenarios, the topological relationship to be identified can be different, and the corresponding injection points and identification points are also different. Exemplarily, in a power distribution network line identification scenario, the connection relationship between a 10kV line and a substation or the connection relationship between the 10kV line and a main transformer in the substation is determined. At this time, the injection point is a conductor point at any position of the 10kV conductor of the 10kV line, and the identification point is a 10kV bus of the main transformer in the substation. The number of identification points is determined based on the number of main transformers in the substation. In a power distribution network line transformer identification scenario, the connection relationship between a distribution transformer and a 10kV line is determined. At this time, the injection point is the low-voltage bus of the distribution transformer. The identification point is the first end of the 10kV line in the substation, and the number of identification points is determined based on the number of 10kV outgoing lines of the substation. In a power distribution network user transformer identification scenario, the connection relationship between a low-voltage user and a distribution transformer is determined. The injection point is the meter bus. The identification point is the low-voltage bus of the distribution transformer, and the number of identification points is determined based on the number of adjacent distribution transformers.

[0103] In the embodiment of the application, the topological relationship identification network can be trained for each scenario to achieve accurate identification of the topological relationship in different scenarios.

[0104] S440, test the preliminary identification network based on the sample topological verification data and the expected label data in the test set, and in the case where the output accuracy of the preliminary identification network reaches a preset accuracy, the preliminary identification network is used as the topological relationship identification network.

[0105] In this embodiment, the output accuracy of the preliminary identification network can be the accuracy obtained by comparing the expected label data and the test output result of the preliminary identification network. The preset accuracy can be a preset accuracy for determining whether the preliminary identification network can be used as a topological relationship identification network. For example, the preset accuracy can be 90%.

[0106] S450, obtaining the injection point and the identification point to be identified in the power distribution line, and collecting the topological verification data to be identified corresponding to the injection point and the identification point.

[0107] S460, obtaining the connection probability corresponding to the identification point based on the topological verification data to be identified and the pre-trained topological relationship identification network.

[0108] S470, in the case where the connection probability corresponding to the identification point is greater than or equal to a preset first probability threshold, the connection relationship between the injection point and the identification point is determined as an existing connection relationship.

[0109] The technical scheme of the embodiment, by constructing the training set and the test set of the topological verification data, the training set and the test set data are scaled and processed, and the topological relationship identification network is trained and tested, and the topological relationship identification network meeting the preset condition of output accuracy is obtained. The accuracy of the topological relationship identification is improved. The beneficial effects of effectively reducing the interference of abnormal harmonics on the pulse signal on the line, improving the identification accuracy and speed are realized.

[0110] Embodiment four

[0111] Figure 5 A structural schematic diagram of a topological identification device of a power distribution line provided by the embodiment three of the application is shown in the figure. Figure 5 As shown, the device comprises a data acquisition module 510, a topological identification module 520 and a connection relationship determination module 530.

[0112] The data acquisition module is configured to acquire an injection point and an identification point to be identified in the power distribution line, and collect topological verification data to be identified corresponding to the injection point and the identification point. The topological identification module is configured to obtain a connection probability corresponding to the identification point based on the topological verification data to be identified and a pre-trained topological relationship identification network. The topological relationship identification network is obtained by training a neural network model based on sample topological verification data and expected label data corresponding to the sample topological verification data. The expected label data comprises data for indicating the connection relationship between each identification point and the injection point. The connection relationship determination module is configured to determine the connection relationship between the injection point and the identification point as an existing connection relationship in the case that the connection probability corresponding to the identification point is greater than or equal to a preset first probability threshold.

[0113] The technical scheme of the embodiment of the application first acquires an injection point and a recognition point to be identified in a power distribution line, collects topology verification data to be identified corresponding to the injection point and the recognition point, and provides a data source for identifying the connection relationship of the injection point and the recognition point. Then, based on the topology verification data to be identified and a pre-trained topology relationship identification network, a connection probability corresponding to the recognition point is obtained, wherein the topology relationship identification network is obtained by training a neural network model based on sample topology verification data and expected label data corresponding to the sample topology verification data, and the expected label data includes data for indicating the connection relationship between each recognition point and the injection point, thereby providing a judgment basis for determining the connection relationship between the injection point and the recognition point. Finally, in a case where the connection probability corresponding to the recognition point is greater than or equal to a preset first probability threshold, the connection relationship between the injection point and the recognition point is determined as an existing connection relationship, thereby improving the accuracy and efficiency of topology relationship identification. The problem that feature signals are easily disturbed by harmonics on the line in the topology identification process of the power distribution line, thereby affecting the topology identification effect of the power distribution line, is solved. The beneficial effects of effectively reducing the disturbance of abnormal harmonics on the line on the pulse signal, improving the identification accuracy and speed, and further improving the line loss management level of the power supply company are achieved.

[0114] On the basis of the above technical scheme, further, the data acquisition module 510 can include a topology vector to be identified acquisition unit, which is specifically used for:

[0115] acquiring a preset injection signal at the injection point, collecting feature values of a plurality of target feature parameters associated with the injection point and the recognition point by injecting the injection signal at the injection point, and constructing a topology vector to be identified based on the feature values of the plurality of target feature parameters;

[0116] acquiring a plurality of topology vectors to be identified constructed by injecting the injection signal at the injection point multiple times, and constructing topology verification data to be identified corresponding to the injection point and the recognition point based on the plurality of topology vectors to be identified.

[0117] On the basis of the above technical scheme, further, the topology vector to be identified acquisition unit is further used for:

[0118] The target feature parameters include a signal parameter of the injection signal of the injection point, a signal parameter of a target recognition signal corresponding to the injection signal at the recognition point, an interval parameter from the injection point to the recognition point, and an electrical parameter of the recognition point;

[0119] The feature values of the target feature parameters associated with the injection point and the identification point are collected by injecting an injection signal at the injection point, including: determining the feature values of the target signal parameters of the injection signal, wherein the signal parameters include pulse width, peak value and frequency; injecting the injection signal at the injection point, collecting the target identification signal corresponding to the injection signal at the identification point corresponding to the injection point, and determining the feature values of the target signal parameters of the target identification signal; obtaining the feature values of the target interval parameters from the injection point to the identification point and the feature values of the target electrical parameters of the identification point, wherein the target interval signal at least includes electrical distance and impedance value.

[0120] On the basis of the above technical solutions, further, the topology identification module 520 can include a data scaling processing unit, and the data scaling processing unit is specifically used for:

[0121] performing data scaling processing on the to-be-identified topology verification data to obtain target identification topology verification data;

[0122] inputting the target identification topology verification data into the pre-trained topology relationship identification network to obtain the connection probability corresponding to the identification point.

[0123] On the basis of the above technical solutions, further, the data scaling processing unit is specifically used for:

[0124] for each feature value in the to-be-identified topology verification data, obtaining a plurality of feature values corresponding to the target feature parameter corresponding to the feature value as reference values;

[0125] scaling processing each feature value based on the average value and the standard deviation of the plurality of reference values to obtain a target value corresponding to the feature value;

[0126] updating each feature value in the to-be-identified topology verification data to the target value corresponding to the feature value to obtain the target identification topology verification data.

[0127] On the basis of the above technical solutions, further, the connection relationship determination module 530 can include a connection relationship classification unit, and the to-be-connected relationship classification unit is specifically used for:

[0128] in the case where there are a plurality of identification points having a connection relationship with the injection point, determining the connection relationship between the identification point with the largest connection probability and the injection point as the main connection relationship, and determining the connection relationship between the remaining identification points having a connection relationship with the injection point and the injection point as the secondary connection relationship.

[0129] On the basis of the above technical solutions, further, the connection relationship determination module 530 can include a connection relationship classification unit, and the to-be-connected relationship classification unit is specifically used for:

[0130] In a case where the connection probability corresponding to the identification point is less than a preset second probability threshold, the topology verification data corresponding to the injection point and the identification point is re-collected based on a preset collection number threshold, and the operation of inputting the topology verification data to be identified into the topology relationship identification network is performed again to obtain the connection probability corresponding to the identification point.

[0131] On the basis of the above technical solution, further, the power distribution line topology identification device can comprise a topology relationship identification network training unit, and the topology relationship identification network training unit is specifically used for:

[0132] Obtaining sample topology verification data, and determining expected label data corresponding to the sample topology verification data;

[0133] Based on the sample topology verification data and the expected label data corresponding to the sample topology verification data, a training set and a test set are constructed;

[0134] Based on the sample topology verification data and the expected label data in the training set, a neural network model is trained to obtain a preliminary identification network, wherein the neural network model comprises a long short-term memory neural network or a gated recurrent unit;

[0135] Based on the sample topology verification data and the expected label data in the test set, the preliminary identification network is tested, and in a case where the output accuracy of the preliminary identification network reaches a preset accuracy, the preliminary identification network is taken as the topology relationship identification network.

[0136] The power distribution line topology identification device provided by the embodiment of the present application can execute the power distribution line topology identification method provided by any embodiment of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0137] Embodiment five

[0138] Figure 6 A structural schematic diagram of an electronic device 10 that can be used to implement embodiments of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular telephones, smart phones, wearable devices (e.g., headsets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit the implementations of the present application described and / or claimed in this document.

[0139] As Figure 6As shown, the electronic device 10 includes at least one processor 11, and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., communicatively connected to the at least one processor 11, where the memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer programs stored in the read-only memory (ROM) 12 or loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the RAM 13. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0140] Various components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc., an output unit 17, such as various types of displays, a speaker, etc., a storage unit 18, such as a magnetic disk, an optical disk, etc., and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network, such as the Internet, and / or various telecommunication networks.

[0141] The processor 11 can be various general and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs various methods and processes described above, such as the topology identification method of the power distribution line.

[0142] In some embodiments, the topology identification method of the power distribution line can be implemented as a computer program tangibly embodied in a computer readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the topology identification method of the power distribution line described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to perform the topology identification method of the power distribution line by any other appropriate means, such as by means of firmware.

[0143] The various embodiments of the systems and techniques described above can be implemented in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a load programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0144] Computer programs used to implement the processes of the application can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer program, when executed, can cause instructions defined in the flow charts and / or block diagrams to be implemented. The computer program can be executed entirely on a machine, partially on a machine, partially on a machine as a standalone software package and partially on a remote machine or entirely on a remote machine or server.

[0145] In the context of the present application, a computer-readable storage medium can be a tangible medium that can contain or store computer programs for use by or in connection with an instruction execution system, apparatus, or device. Computer-readable storage media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium will include one or more lines of electrical connections, portable computer disks, hard disk drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fibers, portable compact disc read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0146] To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0147] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0148] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. A server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system, to solve the defects of large management difficulty and weak business scalability in traditional physical host and VPS service.

[0149] It should be understood that the various forms of flow shown above can be re-ordered, added to, or deleted from without departing from the scope of the present disclosure. For example, the steps recited in the present disclosure can be executed in parallel, executed in sequence, or executed in a different order, as long as the desired results of the present disclosure are achieved, and the present disclosure is not limited herein.

[0150] The specific embodiments described above are not intended to be limiting, and persons skilled in the art will appreciate that various modifications, combinations, sub-combinations and alternatives can be made to the specific embodiments without departing from the spirit and principles of the disclosure. Accordingly, the disclosure is not limited to the specific embodiments described above, but only by the scope of the appended claims.

Claims

1. A method for topology identification of a power distribution line, characterized in that, The method comprises the following steps: obtaining an injection point and a recognition point to be identified in a power distribution line, collecting topology verification data to be identified corresponding to the injection point and the recognition point; obtaining a connection probability corresponding to the recognition point based on the topology verification data to be identified and a pre-trained topology relationship identification network, wherein the topology relationship identification network is obtained by training a neural network model based on sample topology verification data and expected label data corresponding to the sample topology verification data, and the expected label data includes data for indicating a connection relationship between each recognition point and the injection point; in a case where the connection probability corresponding to the recognition point is greater than or equal to a preset first probability threshold, determining that there is a connection relationship between the injection point and the recognition point.

2. The method of claim 1, wherein, The collection of the topology verification data to be identified corresponding to the injection point and the recognition point comprises: obtaining a preset injection signal at the injection point, injecting the injection signal at the injection point, collecting characteristic values of a plurality of target characteristic parameters associated with the injection point and the recognition point, and constructing a topology vector to be identified based on the characteristic values of the plurality of target characteristic parameters; obtaining a plurality of topology vectors to be identified constructed by injecting the injection signal at the injection point multiple times, and constructing topology verification data to be identified corresponding to the injection point and the recognition point based on the plurality of topology vectors to be identified.

3. The method of claim 2, wherein, The target characteristic parameters include signal parameters of the injection signal at the injection point, signal parameters of a target recognition signal corresponding to the injection signal at the recognition point, interval parameters from the injection point to the recognition point, and electrical parameters of the recognition point; The collection of the characteristic values of the plurality of target characteristic parameters associated with the injection point and the recognition point by injecting the injection signal at the injection point comprises: determining characteristic values of target signal parameters of the injection signal, wherein the signal parameters include pulse width, peak value and frequency; injecting the injection signal at the injection point, collecting a target recognition signal corresponding to the injection signal at the recognition point corresponding to the injection point, and determining characteristic values of the target signal parameters of the target recognition signal; obtaining characteristic values of target interval parameters from the injection point to the recognition point and characteristic values of target electrical parameters of the recognition point, wherein the target interval parameters at least include electrical distance and impedance value.

4. The method of claim 2, wherein, The obtaining of the connection probability corresponding to the recognition point based on the topology verification data to be identified and the pre-trained topology relationship identification network comprises: performing data scaling processing on the topology verification data to be identified to obtain target identification topology verification data; inputting the target identification topology verification data into the pre-trained topology relationship identification network to obtain the connection probability corresponding to the recognition point.

5. The method of claim 4, wherein, The data scaling processing on the topology verification data to be identified to obtain the target identification topology verification data comprises: for each characteristic value in the topology verification data to be identified, obtaining a plurality of characteristic values corresponding to the target characteristic parameters corresponding to the characteristic value as reference values; scaling each of the feature values based on a mean value and a standard deviation of a plurality of the reference values, to obtain a target value corresponding to the feature value; updating each of the feature values in the to-be-identified topology verification data to the target value corresponding to the feature value, to obtain target to-be-identified topology verification data.

6. The method of claim 1, wherein, Further comprising: In the case where there are a plurality of the identified points having a connection relationship with the injection point, the connection relationship between the identified point with the largest connection probability and the injection point is determined as the main connection relationship, and the connection relationship between the injection point and the remaining identified points having a connection relationship with the injection point is determined as the secondary connection relationship.

7. The method of claim 1, wherein, Further comprising: In the case where the connection probability corresponding to the identified point is less than a preset second probability threshold, reacquiring the to-be-identified topology verification data corresponding to the injection point and the identified point based on a preset acquisition frequency threshold, and returning to the operation of inputting the to-be-identified topology verification data into the pre-trained topology relationship identification network to obtain the connection probability corresponding to the identified point.

8. The method of claim 1, wherein, Before the connection probability corresponding to the identified point is obtained based on the to-be-identified topology verification data and the pre-trained topology relationship identification network, further comprising: acquiring sample topology verification data and determining expected label data corresponding to the sample topology verification data; constructing a training set and a test set based on the sample topology verification data and the expected label data corresponding to the sample topology verification data; training a pre-established neural network model based on the sample topology verification data and the expected label data in the training set, to obtain a preliminary identification network, wherein the neural network model includes a long short-term memory neural network or a gated recurrent unit; testing the preliminary identification network based on the sample topology verification data and the expected label data in the test set, and taking the preliminary identification network as the topology relationship identification network in the case where the output accuracy of the preliminary identification network reaches a preset accuracy.

9. A topology identification device for a power distribution line, characterized by, Comprise: a data acquisition module configured to acquire an injection point and an identified point to be identified in a power distribution line, and acquire to-be-identified topology verification data corresponding to the injection point and the identified point; a topology identification module configured to obtain a connection probability corresponding to the identified point based on the to-be-identified topology verification data and a pre-trained topology relationship identification network, wherein the topology relationship identification network is obtained by training a neural network model based on sample topology verification data and expected label data corresponding to the sample topology verification data, and the expected label data includes data for indicating a connection relationship between each of the identified points and the injection point; a connection relationship determination module configured to determine that a connection relationship between the injection point and the identified point exists in the case where the connection probability corresponding to the identified point is greater than or equal to a preset first probability threshold.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions for causing the processor to implement the topology identification method for a power distribution line according to any one of claims 1-8 when executed.

Images