A power distribution network network reconstruction method and system considering dynamic line loss governance

Abstract

The application discloses a power distribution network network reconstruction method considering dynamic line loss governance, which comprises the following steps: S1, acquiring power distribution network data and preprocessing; S2, constructing a line flow distribution prediction model based on deep learning, training based on the preprocessed power distribution network data, and then using the trained line flow distribution prediction model to carry out flow distribution calculation to output line flow distribution prediction results; S3, constructing a power distribution network dynamic line loss evaluation model to obtain the dynamic line loss of each flow direction of the power distribution network flow distribution based on the flow distribution prediction results; S4, taking the minimum equivalent cost of dynamic line loss governance as the target, considering multiple constraint conditions, and constructing a power distribution network network reconstruction optimization model; S5, solving the power distribution network network reconstruction optimization model to obtain all power distribution network reconstruction strategies meeting all constraint conditions, and selecting and outputting the optimal power distribution network reconstruction strategy.

Description

Technical Field

[0001] This invention relates to the field of distribution network reconfiguration technology, and in particular to a distribution network reconfiguration method and system that considers dynamic line loss management. Background Technology

[0002] In recent years, facing the challenges brought about by the growth of electricity load and the increasing complexity of the power grid, distribution network reconfiguration technology has become an important means of optimizing the operation of the distribution system. It can effectively balance the load, eliminate overload, and quickly restore power supply in the event of a fault, thereby greatly improving the overall performance of the distribution system. In particular, distribution network reconfiguration can change the topology of the distribution network by switching the state of tie switches and sectionalizing switches, thereby affecting the power flow distribution and direction in the distribution network, so as to reduce network losses and improve power quality and system reliability.

[0003] Current research on power distribution network reconfiguration mainly focuses on the following aspects:

[0004] 1. Research on distribution network reconfiguration considering distributed generation access: Traditional distribution network structures and operation control methods are difficult to adapt to the access of a large number of distributed generation sources with uncertain characteristics. Based on the impact of the access capacity, location, and power factor of distributed generation sources on network losses, this paper studies a distribution network reconfiguration optimization method for power flow calculation of distribution networks containing distributed generation sources.

[0005] 2. Research on distribution network reconfiguration considering demand response: Based on the impact of time-of-use pricing on user load, and comprehensively considering the capabilities of electric vehicle charging load, user-side energy storage, and large industrial adjustable load to participate in demand response, a distribution network reconfiguration optimization method with the goal of minimizing active power loss is constructed.

[0006] 3. Research on distribution network fault recovery and reconfiguration methods: With the aim of improving the rapid recovery capability and power supply reliability of the distribution network after a fault, this study proposes distribution network reconfiguration methods based on the real-time operating status and topology of the distribution network, and studies power supply recovery strategies for fault areas.

[0007] Research has found that while some scholars have conducted studies on distribution network reconfiguration methods for distributed power sources and new load access, there is still room for improvement: First, traditional distribution network reconfiguration methods are mostly based on static network reconfiguration. However, in the actual operation of the distribution network, the loads of each branch are constantly changing, and the number of times sectionalizing switches and tie switches are opened and closed makes it difficult to achieve accurate dynamic network reconfiguration. Second, when constructing reconfiguration optimization methods, typical economic indicators such as minimizing the number of switching operations and minimizing network operating losses are often used as optimization targets. These methods do not adapt to the development of the power market and do not take into account the adjustable resources on the user side, making it difficult to reflect objective optimization needs.

[0008] Therefore, there is an urgent need to explore a new method and system for reconfiguring distribution networks, so as to meet the dynamic line loss management requirements of distribution networks under the condition of random load changes in each branch at any time, and to ensure the reliable operation of the distribution system and minimize the objective equivalent cost of network reconfiguration. Summary of the Invention

[0009] The technical problem to be solved by the present invention is that the present invention discloses a distribution network reconfiguration method and system that considers dynamic line loss management, so as to solve the problem that the existing distribution network reconfiguration methods cannot meet the dynamic line loss management requirements of the distribution network under the condition of random load changes of each branch at every moment, while ensuring the reliable operation of the distribution system and minimizing the equivalent cost of network reconfiguration.

[0010] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a distribution network reconfiguration method considering dynamic line loss management, which includes the following steps:

[0011] S1: Acquire distribution network data and perform preprocessing;

[0012] S2: Construct a line power flow distribution prediction model based on deep learning, train it based on preprocessed distribution network data, and then use the trained line power flow distribution prediction model to perform power flow distribution calculations and output the line power flow distribution prediction results.

[0013] S3: Construct a dynamic line loss assessment model for the distribution network to obtain the dynamic line loss under each flow direction of the power flow distribution based on the power flow distribution prediction results;

[0014] S4: With the goal of minimizing the equivalent cost of dynamic line loss mitigation, and considering multiple constraints, construct a distribution network reconfiguration optimization model;

[0015] S5: Solve the distribution network reconfiguration optimization model to obtain all distribution network reconfiguration strategies that satisfy all constraints, select and output the optimal distribution network reconfiguration strategy.

[0016] In this invention, the inventors have optimized and designed a novel distribution network reconfiguration method that considers dynamic line loss management. This distribution network reconfiguration method can comprehensively consider key factors such as changes in electric vehicle load and the access of distributed power sources. It can use deep learning optimization algorithms to construct a deep learning-based line power flow distribution prediction model, thereby realizing real-time prediction of power flow for each branch.

[0017] Meanwhile, this invention also constructs a dynamic line loss assessment model for distribution networks to accurately predict dynamic network losses under each flow direction of the power flow distribution network. Furthermore, this invention aims to minimize the equivalent cost of dynamic line loss mitigation, considering the user-side energy storage call-up cost and distribution network reconfiguration cost within the region, while also taking into account the power supply reliability and network flexibility of the distribution network. It constructs a distribution network reconfiguration optimization model and ultimately formulates the optimal distribution network reconfiguration strategy that satisfies all constraints.

[0018] In summary, the power distribution network reconfiguration method considering dynamic line loss management described in this invention can significantly improve the operating efficiency of the power distribution network, reduce power loss and operating costs, and promote energy conservation, emission reduction and technological innovation in the power industry. It has good prospects for promotion and application value.

[0019] Furthermore, in the distribution network reconfiguration method of the present invention, in step S1, the distribution network data includes:

[0020] Distribution network structure, branch parameters, node data, user historical load data, and temperature, humidity, weekday and holiday data consistent with the time scale of user historical load data.

[0021] Furthermore, in the distribution network reconfiguration method of the present invention, the preprocessing in step S1 includes:

[0022] The local outlier detection algorithm is used to identify abnormal load data and missing value data in the user's historical load data, and the moving average method is used for processing.

[0023] Furthermore, in the power distribution network reconfiguration method described in this invention, step S2 specifically includes the following steps:

[0024] S21: Obtain load sequence data based on the preprocessed distribution network data, normalize the load sequence data, and divide the normalized load sequence data into two parts according to the time series, one part as the training set and the other part as the test set.

[0025] S22: Construct a deep learning-based model for predicting power flow distribution on power lines;

[0026] S23: Input the training set into the line power flow distribution prediction model for model training;

[0027] S24: Input the test set into the trained line power flow distribution prediction model to output the line power flow distribution prediction results.

[0028] Furthermore, in the distribution network reconfiguration method of the present invention, in step S22, a branch prediction network is established using a GRU network, and the line power flow distribution prediction model is composed of all branch prediction networks, while sharing the total load characteristics and linking all branch prediction networks together.

[0029] Furthermore, in the distribution network reconfiguration method described in this invention, step S24 specifically includes:

[0030] The test set is input into the trained line power flow distribution prediction model to obtain the load prediction result at time t through inverse normalization. And obtain the load prediction result of branch n at time t. To obtain the power flow of branch n at time t. ;

[0031]

[0032] in, Let n be the power flow of branch n at time t; Let be the voltage of branch n at time t.

[0033] Furthermore, in the distribution network reconfiguration method of the present invention, the constraints in step S4 include: power flow constraints, node voltage constraints, branch capacity constraints, reconfiguration number constraints, and energy storage call-up constraints.

[0034] Furthermore, in the power distribution network reconfiguration method described in this invention, step S5 specifically includes:

[0035] The MISOCP optimization algorithm is applied to solve the distribution network reconfiguration optimization model to obtain all distribution network reconfiguration strategies that satisfy all constraints, and the optimal distribution network reconfiguration strategy is selected and output.

[0036] Accordingly, the present invention also discloses a distribution network reconfiguration system considering dynamic line loss management. This distribution network reconfiguration system can be specifically used to implement the above-described distribution network reconfiguration method of the present invention, and specifically includes:

[0037] The data input and preprocessing module is used to acquire and preprocess data from the power distribution network.

[0038] The branch power flow analysis and prediction module is used to build a deep learning-based line power flow distribution prediction model. It is trained based on preprocessed distribution network data, and then the trained line power flow distribution prediction model is used to perform power flow distribution calculations to output the line power flow distribution prediction results.

[0039] The dynamic line loss calculation and evaluation module is used to construct a dynamic line loss evaluation model for the distribution network, so as to obtain the dynamic line loss under each flow direction of the power flow distribution based on the power flow distribution prediction results.

[0040] The distribution network reconfiguration optimization module is used to construct a distribution network reconfiguration optimization model with the goal of minimizing the equivalent cost of dynamic line loss mitigation and considering multiple constraints.

[0041] The decision implementation and monitoring feedback module is used to solve the distribution network reconfiguration optimization model to obtain all distribution network reconfiguration strategies that satisfy all constraints, and select and output the optimal distribution network reconfiguration strategy.

[0042] Furthermore, in the distribution network reconfiguration system described in this invention, the decision implementation and monitoring feedback module can also monitor the operating status of the distribution network in real time and adjust the distribution network reconfiguration strategy in real time based on the feedback.

[0043] The beneficial effects of this invention are as follows: This invention discloses a distribution network reconfiguration method and system that considers dynamic line loss management, which can effectively solve the technical problems existing in the current distribution network reconfiguration method. It can not only meet the dynamic line loss management requirements of the distribution network under the condition of random load changes of each branch at each moment, but also ensure the reliable operation of the distribution system and minimize the equivalent cost of network reconfiguration. It can significantly improve the operating efficiency of the distribution network, reduce power loss and operating costs, promote energy conservation, emission reduction and technological innovation in the power industry, and has good prospects for promotion and application value. Attached Figure Description

[0044] Figure 1 A schematic diagram illustrating the regional power flow distribution of the distribution network is shown.

[0045] Figure 2 This is a flowchart illustrating the steps of one embodiment of the power distribution network reconfiguration method considering dynamic line loss management described in this invention.

[0046] Figure 3 This is a module design diagram of the power distribution network reconfiguration system considering dynamic line loss management in one embodiment of the present invention.

[0047] Figure 4 This is a flowchart illustrating the steps of implementing the above-described distribution network reconfiguration method using the distribution network reconfiguration system that considers dynamic line loss management as described in this invention. Detailed Implementation

[0048] To explain in detail the technical content, objectives, and effects of the present invention, the following description is provided in conjunction with the embodiments and accompanying drawings.

[0049] Please refer to Figure 1and Figure 2 As shown in this embodiment, the present invention discloses a distribution network reconfiguration method considering dynamic line loss management, which specifically includes the following steps:

[0050] S1: Acquire distribution network data and perform preprocessing;

[0051] S2: Construct a line power flow distribution prediction model based on deep learning, train it based on preprocessed distribution network data, and then use the trained line power flow distribution prediction model to perform power flow distribution calculations and output the line power flow distribution prediction results.

[0052] S3: Construct a dynamic line loss assessment model for the distribution network to obtain the dynamic line loss under each flow direction of the power flow distribution based on the power flow distribution prediction results;

[0053] S4: With the goal of minimizing the equivalent cost of dynamic line loss mitigation, and considering multiple constraints, construct a distribution network reconfiguration optimization model;

[0054] S5: Solve the distribution network reconfiguration optimization model to obtain all distribution network reconfiguration strategies that satisfy all constraints, select and output the optimal distribution network reconfiguration strategy.

[0055] In other words, in this embodiment, the distribution network reconfiguration method disclosed in this invention can first predict the power flow of each branch in the distribution network area in real time through line power flow distribution prediction based on deep learning, so as to build a line power flow distribution prediction model based on deep learning. Then, dynamic network losses are evaluated, and a distribution network dynamic line loss evaluation model is constructed based on the evaluation results. With the goal of minimizing the equivalent cost of dynamic line loss mitigation, multiple constraints are considered to construct a distribution network reconfiguration optimization model. Solving the distribution network reconfiguration optimization model yields the optimal distribution network reconfiguration strategy, thereby achieving accurate dynamic network reconfiguration while ensuring the economic operation of the distribution network system.

[0056] Reference Figure 1 and Figure 2 In this embodiment, in the distribution network reconfiguration method considering dynamic line loss management of the present invention, step S1 specifically includes:

[0057] Input distribution network data for a specific region. This data can be specifically historical distribution network data, which includes the network structure, branch parameters, and node data. Here, G represents the distribution network of that region; G = (V, E) represents the distribution network topology; and V represents the set of branches, V = {v1, v2, ..., v...}. N} means that within the distribution network area G, there are a total of N branches.

[0058] Accordingly, the input distribution network data also includes historical load data for users, which contains the historical hourly load data for all users on each branch, denoted as P. it P it This represents the user load data for the i-th branch at time t. Additionally, in this embodiment, the input distribution network data will also include temperature, humidity, weekday, and holiday data consistent with the time scale of the user's historical load data.

[0059] It should be noted that in step S1 of the present invention, after inputting the above-mentioned distribution network data, the distribution network data will also be preprocessed. For example, in this embodiment, it is necessary to specifically use the local outlier detection algorithm to identify abnormal load data and missing value data in the user's historical load data, and use the moving average method to process it in order to supplement the incomplete or abnormal data.

[0060] Accordingly, in the distribution network reconfiguration method of the present invention, in the above step S2, a line power flow distribution prediction model based on deep learning can be constructed and the line power flow distribution prediction result can be output, which may specifically include the following steps S21-S24.

[0061] S21: Obtain load sequence data based on the preprocessed distribution network data, normalize the load sequence data, and divide the normalized load sequence data into two parts according to the time series, one part as the training set and the other part as the test set.

[0062] For example, in this embodiment, load sequence data can be obtained first from the preprocessed distribution network data; wherein, to achieve dynamic line loss assessment and management of the distribution network, the load of all users on the i-th branch at each moment needs to be predicted. (Using vectors...) The load sequence of the i-th branch is represented by matrix P. The load state in region G can be represented by matrix P. K×N ,like Figure 1 As shown, region G can include K1-K N K represents the length of the load sequence for each branch. Let the preprocessed load data sequence be P = [P1, P2, ... P2]. T ], P t ∈R N×t This represents the total load of all branches at time t, i.e., the characteristic matrix at time t.

[0063] In this embodiment, after obtaining the load sequence data, in step S21, the load sequence data can be processed by max-min normalization, and the load sequence data can be divided into two parts according to the time series using the sliding window method, with 80% as the training set and 20% as the test set.

[0064] S22: Construct a deep learning-based model for predicting power flow distribution on power lines.

[0065] In step S22 of the present invention, according to the definition of deep learning, a branch prediction network can be specifically established using a GRU network. The power flow distribution prediction model of the line is composed of all branch prediction networks, and the total load characteristics are shared, thus linking all branch prediction networks together.

[0066] S23: Input the training set into the power flow distribution prediction model of the line for model training.

[0067] In step S23 of the present invention, a training set can be selected for training, and the total load data P can be... t With load data P of each branch it The data is fused, and the corresponding branch GRU network is forward-trained using the fused data until convergence.

[0068] S24: Input the test set into the trained line power flow distribution prediction model to output the line power flow distribution prediction results.

[0069] In this invention, in this embodiment, see [reference needed]. Figure 1 In the region G shown, there are N branches. In step S24 above, the test set is input into the trained line power flow distribution prediction model to obtain the load prediction result at time t through inverse normalization. Specifically, it can be described as follows:

[0070]

[0071] Accordingly, the load prediction result of branch n at time t can be further obtained. Load prediction result of the nth branch at time t It can be represented as:

[0072]

[0073] Therefore, see Figure 1 Considering that all loads are connected to a 10kV voltage level, the power flow of branch n at time t can be obtained by using the relationship between the branch input power and voltage. ;

[0074]

[0075] in, Let n be the power flow of branch n at time t; Let be the voltage of branch n at time t.

[0076] Based on this, in this invention, Figure 1The power flow in region G of the distribution network shown can be represented as:

[0077]

[0078] Therefore, through S21-S24, a line power flow distribution prediction model based on deep learning can be constructed first, and the line power flow distribution prediction results can be output to make real-time predictions of the power flow of each branch in the distribution network area.

[0079] Similarly, in this distribution network reconfiguration method, in step S3, the dynamic line loss of each flow direction in the distribution network can be obtained by constructing a dynamic line loss assessment model of the distribution network and using the flow distribution prediction results output by the line flow distribution prediction model.

[0080] In this embodiment, based on the power flow calculation results, the loss of the nth branch at time t can be calculated. :

[0081]

[0082] in, This represents the loss of the nth branch at time t. This represents the resistance of the nth branch.

[0083] Therefore, the total line loss in area G of the distribution network is... It can be represented as:

[0084]

[0085] Of course, considering the increased uncertainty in power flow caused by load fluctuations and the changes in distribution network line losses due to local operational inequalities, load distribution can be adjusted by modifying line tie switches or utilizing user-side energy storage, thereby optimizing distribution network line losses. User-side energy storage can be utilized daily, while the number of times line tie switches can be used is limited. Therefore, further optimization is possible, and the power flow in distribution network area G becomes:

[0086]

[0087] in, This is expressed as the change in current in branch n at time t.

[0088] Correspondingly, the total line loss in area G of the distribution network It can be adjusted accordingly:

[0089]

[0090] in, It is represented as the resistance after the structure of the nth branch is adjusted.

[0091] Accordingly, in step S4 of the distribution network reconfiguration method of the present invention, it is necessary to construct a distribution network reconfiguration optimization model with the objective of minimizing the equivalent cost of dynamic line loss mitigation, considering multiple constraints. In other words, the present invention aims to optimize distribution network line losses with the minimum equivalent cost; where minimum equivalent cost refers to the cost level after reducing line losses by taking measures such as switching grid interconnection switches and utilizing user-side energy storage fees. For example, the equivalent cost f is:

[0092]

[0093] Where p represents the power grid price level; C1 represents the cost of switching line interconnection switches; and C2 represents the cost of calling up user-side energy storage.

[0094]

[0095] In the formula, m represents the number of tie switches to be switched; p1 represents the cost of each tie switch switch switch switch switch switch switch switch; P soc p1 indicates the amount of energy stored on the user side that is being used; p2 indicates the energy storage subsidy cost per kilowatt-hour.

[0096] It should be noted that, in this embodiment, to achieve the goal of minimizing the equivalent cost of dynamic line loss mitigation, it is necessary to simultaneously consider power flow constraints, node voltage constraints, branch capacity constraints, reconfiguration frequency constraints, and energy storage dispatch constraints.

[0097] 1. System power flow constraints

[0098] To ensure real-time power balance in the power grid, meaning that user electricity demand equals the sum of grid supply and energy storage supply:

[0099]

[0100] Among them, P grid,nt P soc,nt Let represent the power supplied by the grid and the power supplied by the energy storage in branch n at time t, respectively.

[0101] 2. Node voltage constraints

[0102] Node voltage should not exceed its upper or lower limits:

[0103]

[0104] In the formula, U nmin U nmax U nt These represent the lower limit, upper limit, and node voltage at time t of node n in branch, respectively.

[0105] 3. Branch capacity constraints

[0106] The power flowing through each branch should not exceed the maximum allowable line capacity.

[0107]

[0108] In the formula, P nmax These represent the maximum allowed line capacity for branch n.

[0109] 4. Reconstruction Count Constraint

[0110]

[0111] In the formula, , This indicates the maximum number of times the interconnecting switch can be started and stopped.

[0112] 5. Energy storage dispatch constraints

[0113] The energy storage state of charge does not exceed the rated capacity, and charging and discharging do not occur at the same time.

[0114]

[0115] Among them, SOC t E represents the energy storage capacity at time t; sn Indicates the rated capacity of energy storage; P std With P stc These represent the charge and discharge capacities of the energy storage at time t, respectively.

[0116] Accordingly, in step S5 of the distribution network reconfiguration method of the present invention, combining the dynamic prediction results of power flow distribution and strictly following the rules of distribution network reconfiguration and energy storage dispatch, under all constraints, the MISOCP optimization algorithm can be specifically applied to solve the distribution network reconfiguration optimization model to obtain various distribution network reconfiguration optimization strategies. With the minimum equivalent cost of distribution network reconfiguration optimization as the final objective, the optimal distribution network reconfiguration strategy is determined through comparison. This optimal distribution network reconfiguration strategy may include: energy storage system dispatch strategy, branch tie switch switching strategy, and large grid power supply strategy, etc.

[0117] Meanwhile, in this embodiment, after outputting the optimal distribution network reconfiguration strategy and issuing the corresponding switching operation commands, the operating status of the distribution network can be monitored in real time, and the reconfiguration scheme can be adjusted in real time based on the feedback.

[0118] For ease of understanding and application, the present invention also discloses, as follows: Figure 3 and Figure 4The diagram illustrates a distribution network reconfiguration system that considers dynamic line loss management. This distribution network reconfiguration system can be specifically applied to the distribution network reconfiguration method described above in this invention. It may include: a data input and preprocessing module, a branch power flow analysis and prediction module, a dynamic line loss calculation and evaluation module, a distribution network reconfiguration optimization module, and a decision implementation and monitoring feedback module.

[0119] It should be noted that, in this embodiment, the data input and preprocessing module is used to acquire power distribution network data and perform preprocessing.

[0120] The above-mentioned branch power flow analysis and prediction module is used to construct a line power flow distribution prediction model based on deep learning. It is trained based on preprocessed distribution network data, and then the trained line power flow distribution prediction model is used to perform power flow distribution calculations to output the line power flow distribution prediction results.

[0121] The aforementioned dynamic line loss calculation and evaluation module is used to construct a dynamic line loss evaluation model for the distribution network, so as to obtain the dynamic line loss under each flow direction of the power flow distribution based on the power flow distribution prediction results.

[0122] The aforementioned distribution network reconfiguration optimization module is used to construct a distribution network reconfiguration optimization model with the goal of minimizing the equivalent cost of dynamic line loss mitigation and considering multiple constraints.

[0123] The aforementioned decision implementation and monitoring feedback module is used to solve the distribution network reconfiguration optimization model to obtain all distribution network reconfiguration strategies that satisfy all constraints, select and output the optimal distribution network reconfiguration strategy; at the same time, the decision implementation and monitoring feedback module can also monitor the operation status of the distribution network in real time and adjust the distribution network reconfiguration strategy in real time according to the feedback.

[0124] In summary, the distribution network reconfiguration method and system designed in this invention can effectively solve the technical problems existing in the current distribution network reconfiguration methods. It can meet the dynamic line loss management requirements of the distribution network under the condition of random load changes in each branch at any time, and can also ensure the reliable operation of the distribution system and minimize the equivalent cost of network reconfiguration. It can significantly improve the operating efficiency of the distribution network, reduce power loss and operating costs, and effectively promote energy conservation, emission reduction and technological innovation in the power industry. It has good prospects for promotion and application value.

[0125] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent modifications made based on the content of the present invention specification and drawings, or direct or indirect applications in related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for reconfiguring a distribution network considering dynamic line loss management, characterized in that, Including the following steps: S1: Acquire distribution network data and perform preprocessing; S2: Construct a line power flow distribution prediction model based on deep learning, train it based on preprocessed distribution network data, and then use the trained line power flow distribution prediction model to perform power flow distribution calculations and output the line power flow distribution prediction results. S3: Construct a dynamic line loss assessment model for the distribution network to obtain the dynamic line loss under each flow direction of the power flow distribution based on the power flow distribution prediction results; S4: With the goal of minimizing the equivalent cost of dynamic line loss mitigation, and considering multiple constraints, construct a distribution network reconfiguration optimization model; S5: Solve the distribution network reconfiguration optimization model to obtain all distribution network reconfiguration strategies that satisfy all constraints, select and output the optimal distribution network reconfiguration strategy; Step S2 specifically includes the following steps: S21: Obtain load sequence data based on the preprocessed distribution network data, normalize the load sequence data, and divide the normalized load sequence data into two parts according to the time series, one part as the training set and the other part as the test set. S22: Construct a deep learning-based model for predicting power flow distribution on power lines; S23: Input the training set into the line power flow distribution prediction model for model training; S24: Input the test set into the trained line power flow distribution prediction model to output the line power flow distribution prediction result; In step S22, a branch prediction network is established using the GRU network. The line power flow distribution prediction model is composed of all branch prediction networks, and they share the total load characteristics, thus linking all branch prediction networks together. Step S24 is as follows: The test set is input into the trained line power flow distribution prediction model to obtain the load prediction result at time t through inverse normalization. And obtain the load prediction result of branch n at time t. To obtain the power flow of branch n at time t. ; in, Let n be the power flow of branch n at time t; Let be the voltage of branch n at time t.

2. The distribution network reconfiguration method considering dynamic line loss management according to claim 1, characterized in that, In step S1, the distribution network data includes: Distribution network structure, branch parameters, node data, user historical load data, and temperature, humidity, weekday and holiday data consistent with the time scale of user historical load data.

3. The distribution network reconfiguration method considering dynamic line loss management according to claim 2, characterized in that, In step S1, the preprocessing includes: The local outlier detection algorithm is used to identify abnormal load data and missing value data in the user's historical load data, and the moving average method is used for processing.

4. The distribution network reconfiguration method considering dynamic line loss management according to claim 1, characterized in that, In step S4, the constraints include: power flow constraints, node voltage constraints, branch capacity constraints, reconfiguration times constraints, and energy storage call-up constraints.

5. A distribution network reconfiguration method considering dynamic line loss management according to claim 1, characterized in that, In step S5, specifically: The MISOCP optimization algorithm is applied to solve the distribution network reconfiguration optimization model to obtain all distribution network reconfiguration strategies that satisfy all constraints, and the optimal distribution network reconfiguration strategy is selected and output.

6. A distribution network reconfiguration system considering dynamic line loss management, characterized in that, It is applied to the distribution network reconfiguration method as described in any one of claims 1-5, and includes: The data input and preprocessing module is used to acquire and preprocess data from the power distribution network. The branch power flow analysis and prediction module is used to build a deep learning-based line power flow distribution prediction model. It is trained based on preprocessed distribution network data, and then the trained line power flow distribution prediction model is used to perform power flow distribution calculations to output the line power flow distribution prediction results. The dynamic line loss calculation and evaluation module is used to construct a dynamic line loss evaluation model for the distribution network, so as to obtain the dynamic line loss under each flow direction of the power flow distribution based on the power flow distribution prediction results. The distribution network reconfiguration optimization module is used to construct a distribution network reconfiguration optimization model with the goal of minimizing the equivalent cost of dynamic line loss mitigation and considering multiple constraints. The decision implementation and monitoring feedback module is used to solve the distribution network reconfiguration optimization model to obtain all distribution network reconfiguration strategies that satisfy all constraints, and select and output the optimal distribution network reconfiguration strategy.

7. A distribution network reconfiguration system considering dynamic line loss management according to claim 6, characterized in that, The decision implementation and monitoring feedback module can also monitor the operation status of the distribution network in real time and adjust the distribution network reconfiguration strategy in real time based on the feedback.

Images