Power distribution area device configuration method, device, equipment, storage medium and program product

Abstract

The application relates to a power distribution area device configuration method, device, equipment, storage medium and program product. The method comprises the following steps: acquiring a plurality of hybrid decision vectors of a power distribution area; the hybrid decision vector comprises a transformer capacity level, a static reactive power generator capacity and an access position, an active power filter capacity and an access position, a charging pile technology level and a charging facility scale coefficient; constraint parameters of a worst-case scenario are determined according to an uncertainty radius and a source-load uncertainty variable, and the plurality of hybrid decision vectors are subjected to optimization processing by using a pre-determined multi-objective optimization model, a deterministic constraint condition and the constraint parameters of the worst-case scenario to obtain an optimal solution set; the multi-objective optimization model takes the maximum investment return rate, the minimum power quality comprehensive index, the minimum load rate deviation and the minimum total investment as targets; and a preset balancing algorithm is used to screen a best configuration scheme from the optimal solution set. The method takes into account the economy and robustness, and is beneficial to the operation of the power distribution area.

Description

Technical Field

[0001] This application relates to the field of power distribution network planning technology, and in particular to a configuration method, device, equipment, storage medium and program product for a power distribution substation. Background Technology

[0002] As the "dual carbon" goals are being implemented more deeply, the operating environment of distribution transformer substations, as key terminal media connecting the user side and the distribution network, is undergoing profound changes. On the one hand, the large-scale integration of distributed photovoltaic power generation has led to high penetration and strong fluctuations in the substations; on the other hand, the rapid growth of electric vehicle charging load has brought significant randomness and impact. The superposition of these dual uncertainties of source and load has resulted in highly complex and variable operating conditions for the substations, placing higher demands on the configuration and coordinated operation of the equipment within the substations.

[0003] Currently, traditional distribution transformer substation configuration methods typically rely on deterministic load forecasting and photovoltaic (PV) output forecasting, treating future operating scenarios as deterministic values ​​for planning. This fails to adequately consider uncertainties such as the intermittency and volatility of distributed PV output and the randomness of charging loads. To address these uncertainties, existing research has proposed methods such as stochastic optimization, robust optimization, and interval optimization. However, stochastic optimization requires accurately obtaining the probability density function of uncertain variables, which is often difficult to achieve in practical engineering. Traditional robust optimization models based on worst-case scenarios, easily leading to overly conservative configuration schemes and high investment costs. Interval optimization requires pre-determining the fluctuation range of variables, which is highly subjective and difficult to adapt to dynamic changes in actual operation. Furthermore, existing substation configuration methods often plan transformers, reactive power compensation devices, harmonic mitigation devices, and charging facilities independently, lacking a system-level collaborative optimization mechanism. This results in functional overlap or insufficient coordination between different devices, further restricting the economic efficiency and reliability of substation operation.

[0004] Therefore, there is an urgent need for a decision-making method for distribution radio station configuration that balances economy and robustness. Summary of the Invention

[0005] Therefore, it is necessary to provide a configuration method, apparatus, device, storage medium, and program product for a distribution radio station that can balance economy and robustness in response to the above-mentioned technical problems.

[0006] In a first aspect, this application provides a method for configuring a distribution radio station device, including:

[0007] Obtain multiple hybrid decision vectors for the distribution transformer area; the hybrid decision vectors include transformer capacity level, static var generator capacity and access location, active power filter capacity and access location, charging pile technology level and charging facility scale coefficient;

[0008] The constraint parameters for the worst-case scenario are determined based on the uncertainty radius and source load uncertainty variables. Then, multiple mixed decision vectors are optimized using a pre-determined multi-objective optimization model, deterministic constraints, and the constraint parameters for the worst-case scenario to obtain the optimal solution set. The multi-objective optimization model aims to maximize the return on investment, minimize the comprehensive power quality index, minimize the load factor deviation, and minimize the total investment.

[0009] The optimal configuration scheme is selected from the set of optimal solutions using a pre-defined balancing algorithm.

[0010] In one embodiment, the constraint parameters for the worst-case scenario are determined based on the uncertainty radius and the source load uncertainty variables. Then, multiple hybrid decision vectors are optimized using a pre-determined multi-objective optimization model, deterministic constraints, and the constraint parameters for the worst-case scenario to obtain an optimal solution set, including:

[0011] Obtain the set of source and load uncertainties and the uncertainty radius of the distribution transformer area. Determine the constraint parameters for the worst-case scenario based on the set of uncertainty radius and the set of source and load uncertainties. The set of source and load uncertainties includes load, photovoltaic and charging simultaneity rate. The worst-case scenario includes peak worst-case scenario and valley worst-case scenario.

[0012] Technical and economic parameters are determined based on the constraints of the worst-case scenario; among them, technical parameters include voltage deviation, harmonic parameters, and load rate, while economic parameters include total investment, annual operating cost, and charging revenue.

[0013] Technical and economic parameters are input into a multi-objective optimization model for calculation, resulting in a combination of objective values ​​for multiple hybrid decision vectors.

[0014] The non-dominated solution for the corresponding round is determined by calculating the combination of target values ​​using a pre-defined crossover algorithm.

[0015] Adjust each mixed decision vector and return to the step of obtaining the set of source-load uncertain variables and uncertainty radius of the distribution radio station until the maximum number of iterations is reached. Combine the non-dominated solutions of all iterations to obtain the optimal solution set.

[0016] In one embodiment, the multi-objective optimization model includes a return on investment sub-model, a comprehensive power quality index sub-model, a load factor deviation sub-model, and a total investment sub-model; the objective value combination includes a first objective value, a second objective value, a third objective value, and a fourth objective value of the mixed decision vector. Technical and economic parameters are input into the multi-objective optimization model for calculation to obtain multiple objective value combinations of the mixed decision vector, including:

[0017] For each hybrid decision vector, a simplified rate of return is determined based on economic parameters. The simplified rate of return is then input into the rate of return sub-model for calculation to obtain the first objective value of the hybrid decision vector.

[0018] The voltage deviation and harmonic parameters are input into the power quality comprehensive index sub-model for calculation, and the second objective value of the hybrid decision vector is obtained.

[0019] The load factor is input into the load factor deviation sub-model for calculation to obtain the third objective value of the hybrid decision vector;

[0020] The total investment is input into the total investment sub-model for calculation, resulting in the fourth objective value of the hybrid decision vector.

[0021] In one embodiment, a preset crossover algorithm is used to calculate the combination of target values ​​to determine the non-dominated solution for the corresponding round, including:

[0022] The combination of target values ​​is sorted by non-dominated order to obtain multiple non-dominated levels;

[0023] Calculate the crowding degree of the combination of objective values ​​for each non-dominated level, and preserve the boundary solutions and sparse region solutions;

[0024] The parent is determined based on the non-dominance level and crowding degree, and multiple offspring individuals of the parent are determined based on a preset crossover algorithm;

[0025] Based on the crowding level, the parent and offspring individuals are selected in descending order to obtain the non-dominated solution for the corresponding round.

[0026] In one embodiment, a preset balancing algorithm is used to select the optimal configuration scheme from the optimal solution set, including:

[0027] Normalize each optimal solution in the optimal solution set to obtain multiple normalized optimal solutions;

[0028] For each normalized optimal solution, calculate the Euclidean distance between multiple objective values ​​in the normalized optimal solution, and determine the optimal configuration scheme based on the Euclidean distance.

[0029] In one embodiment, a preset balancing algorithm is used to select the optimal configuration scheme from the optimal solution set, including:

[0030] Normalize each optimal solution in the optimal solution set to obtain multiple normalized optimal solutions;

[0031] For each normalized optimal solution, the multiple objective values ​​in the normalized optimal solution are weighted and summed to obtain the comprehensive score of each optimal solution;

[0032] The optimal solution with the lowest overall score is determined as the best configuration scheme.

[0033] Secondly, this application also provides a configuration device for a distribution radio station, comprising:

[0034] The acquisition module is used to acquire multiple hybrid decision vectors of the distribution transformer area; the hybrid decision vectors include transformer capacity level, static var generator capacity and access location, active power filter capacity and access location, charging pile technology level and charging facility scale coefficient;

[0035] The determination module is used to determine the constraint parameters of the worst-case scenario based on the uncertainty radius and source load uncertainty variables, and to optimize multiple mixed decision vectors using a pre-determined multi-objective optimization model, deterministic constraints, and the constraint parameters of the worst-case scenario to obtain the optimal solution set; wherein, the multi-objective optimization model aims to maximize the return on investment, minimize the comprehensive power quality index, minimize the load factor deviation, and minimize the total investment.

[0036] The filtering module is used to select the best configuration scheme from the set of optimal solutions using a preset balancing algorithm.

[0037] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0038] Obtain multiple hybrid decision vectors for the distribution transformer area; the hybrid decision vectors include transformer capacity level, static var generator capacity and access location, active power filter capacity and access location, charging pile technology level and charging facility scale coefficient;

[0039] The constraint parameters for the worst-case scenario are determined based on the uncertainty radius and source load uncertainty variables. Then, multiple mixed decision vectors are optimized using a pre-determined multi-objective optimization model, deterministic constraints, and the constraint parameters for the worst-case scenario to obtain the optimal solution set. The multi-objective optimization model aims to maximize the return on investment, minimize the comprehensive power quality index, minimize the load factor deviation, and minimize the total investment.

[0040] The optimal configuration scheme is selected from the set of optimal solutions using a pre-defined balancing algorithm.

[0041] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0042] Obtain multiple hybrid decision vectors for the distribution transformer area; the hybrid decision vectors include transformer capacity level, static var generator capacity and access location, active power filter capacity and access location, charging pile technology level and charging facility scale coefficient;

[0043] The constraint parameters for the worst-case scenario are determined based on the uncertainty radius and source load uncertainty variables. Then, multiple mixed decision vectors are optimized using a pre-determined multi-objective optimization model, deterministic constraints, and the constraint parameters for the worst-case scenario to obtain the optimal solution set. The multi-objective optimization model aims to maximize the return on investment, minimize the comprehensive power quality index, minimize the load factor deviation, and minimize the total investment.

[0044] The optimal configuration scheme is selected from the set of optimal solutions using a pre-defined balancing algorithm.

[0045] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0046] Obtain multiple hybrid decision vectors for the distribution transformer area; the hybrid decision vectors include transformer capacity level, static var generator capacity and access location, active power filter capacity and access location, charging pile technology level and charging facility scale coefficient;

[0047] The constraint parameters for the worst-case scenario are determined based on the uncertainty radius and source load uncertainty variables. Then, multiple mixed decision vectors are optimized using a pre-determined multi-objective optimization model, deterministic constraints, and the constraint parameters for the worst-case scenario to obtain the optimal solution set. The multi-objective optimization model aims to maximize the return on investment, minimize the comprehensive power quality index, minimize the load factor deviation, and minimize the total investment.

[0048] The optimal configuration scheme is selected from the set of optimal solutions using a pre-defined balancing algorithm.

[0049] The aforementioned configuration method, device, equipment, storage medium, and program product for the distribution transformer substation acquire multiple hybrid decision vectors for the substation. These hybrid decision vectors include transformer capacity level, static var generator capacity and connection location, active power filter capacity and connection location, charging pile technology level, and charging facility scale coefficient. Based on the uncertainty radius and source-load uncertainty variables, the constraint parameters for the worst-case scenario are determined. Then, using a pre-determined multi-objective optimization model, deterministic constraints, and the constraint parameters for the worst-case scenario, the multiple hybrid decision vectors are optimized to obtain the optimal solution set. The multi-objective optimization model aims to maximize return on investment, minimize comprehensive power quality indicators, minimize load factor deviation, and minimize total investment. Finally, a pre-defined balancing algorithm is used to select the optimal configuration scheme from the optimal solution set. By setting an uncertainty radius, the robustness of the distribution substation equipment configuration scheme can be enhanced. Based on the uncertainty radius, the worst-case peak scenario that may lead to overload and voltage drop and the worst-case valley scenario that may lead to light load can be explicitly constructed, reducing the computational load of the configuration process. Finally, by combining a multi-objective optimization model to optimize the mixed decision variables, it is possible to perform synergistic optimization from multiple dimensions such as economy, power quality, operating efficiency and total investment, thereby improving the economic efficiency of the distribution substation equipment configuration and contributing to the stable operation of the distribution substation. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a structural block diagram of a computer device for configuring a distribution radio station area device in one embodiment;

[0052] Figure 2 This is a flowchart illustrating a method for configuring a distribution radio area device in one embodiment;

[0053] Figure 3 This is a flowchart illustrating the process of optimizing multiple mixed decision vectors using a pre-determined multi-objective optimization model, deterministic constraints, and worst-case scenario constraint parameters in one embodiment.

[0054] Figure 4 This is a schematic diagram illustrating the process of inputting technical and economic parameters into a multi-objective optimization model to determine a combination of objective values ​​for multiple hybrid decision vectors in one embodiment.

[0055] Figure 5This is a flowchart illustrating the process of calculating the combination of target values ​​using a preset crossover algorithm in one embodiment to determine the non-dominated solution for the corresponding round;

[0056] Figure 6 This is a flowchart illustrating the process of selecting the optimal configuration scheme from the optimal solution set using a preset balancing algorithm in one embodiment.

[0057] Figure 7 This is a flowchart illustrating the process of selecting the optimal configuration scheme from the optimal solution set using a preset balancing algorithm in another embodiment.

[0058] Figure 8 This is a structural block diagram of the configuration device of the distribution radio station area device in one embodiment. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0060] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0061] With the deepening implementation of the "dual carbon" goals, the operating environment of distribution transformer substations, as key terminal media connecting the user side and the distribution network, is undergoing profound changes. On the one hand, the large-scale integration of distributed photovoltaic (PV) power generation has led to high penetration and strong fluctuations in these substations; on the other hand, the rapid growth of electric vehicle charging loads has brought significant randomness and impact. The superposition of these dual uncertainties in source and load results in highly complex and variable operating conditions for the substations, placing higher demands on the configuration and coordinated operation of equipment within the substations. Currently, traditional methods for configuring distribution transformer substation equipment typically rely on deterministic load forecasting and PV output forecasting, treating future operating scenarios as deterministic values ​​for planning, failing to fully consider uncertainties such as the intermittency and volatility of distributed PV output and the randomness of charging loads. To address these uncertainties, existing research has proposed methods such as stochastic optimization, robust optimization, and interval optimization. Among these, stochastic optimization requires accurate acquisition of the probability density function of uncertain variables, which is often difficult to achieve in engineering practice; traditional robust optimization is based on worst-case scenario modeling, which can easily lead to overly conservative configuration schemes and high investment costs; interval optimization requires pre-determining the fluctuation range of variables, which is highly subjective and difficult to adapt to dynamic changes in actual operation. Furthermore, existing distribution transformer substation configuration methods often involve independent planning for transformers, reactive power compensation devices, harmonic mitigation devices, and charging facilities, lacking a system-level collaborative optimization mechanism. This leads to functional overlap or insufficient coordination between different devices, further restricting the economy and reliability of distribution transformer substation operation. Therefore, there is an urgent need for a distribution transformer substation configuration decision-making method that balances economy and robustness.

[0062] In view of the above-mentioned technical problems, the present application provides a configuration method for a distribution radio station device that can balance economy and robustness. The following embodiments will specifically illustrate the configuration method of the distribution radio station device.

[0063] The configuration method for the distribution radio station device provided in this application embodiment can be applied to, for example, Figure 1 The computer device shown may be a server, and its internal structure diagram may be as follows. Figure 1As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores hybrid decision vectors, such as transformer capacity levels, static var generator capacity and connection location, active power filter capacity and connection location, charging pile technology levels, and charging facility scale coefficients. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a configuration method for a distribution substation device.

[0064] Those skilled in the art will understand that Figure 1 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0065] In one exemplary embodiment, such as Figure 2 As shown, a configuration method for a distribution radio station device is provided. This embodiment illustrates the method by applying it to a computer device. In this embodiment, the method includes:

[0066] S201, obtain multiple hybrid decision vectors for the distribution radio area.

[0067] The hybrid decision vector includes transformer capacity level, static var generator capacity and access location, active power filter capacity and access location, charging pile technology level, and charging facility scale coefficient.

[0068] In the embodiments of this application, the computer device can retrieve pre-stored multi-dimensional decision parameters from a database and combine the multi-dimensional decision parameters with different values ​​to obtain a hybrid decision vector. Alternatively, decision parameters can be directly collected from the device in the distribution radio area in real time and combined to obtain a hybrid decision vector.

[0069] In some embodiments, the hybrid decision vector can be represented by the following relation (1):

[0070] (1);

[0071] In the formula, Transformer capacity rating; For SVG access location; For SVG size; Location of APF access; For APF capacity; , , These are the technical levels for slow charging, fast charging, and ultra-fast charging, respectively. This represents the scale coefficient of charging facilities.

[0072] The value of the transformer capacity level can be expressed by the following relationship (2):

[0073] (2);

[0074] In the formula, It is the transformer grade number; The rated capacity of the transformer is obtained by mapping the number, and the unit is kVA.

[0075] The values ​​of the static var generator (SVG) connection location and the active power filter (APF) connection location can be expressed by the following relationship (3):

[0076] (3);

[0077] In the formula, and These represent the access locations of SVG and APF, respectively; values ​​0, 1, 2, and 3 represent no installation, low-voltage side, first end of the transformer area, and charging pile cluster, respectively.

[0078] The values ​​of the static var generator capacity and the active power filter capacity can be expressed by the following relationship (4):

[0079] (4);

[0080] In the formula, This indicates the rated reactive power capacity of the SVG, in kVar. This indicates the rated compensation current of the APF, in amperes (A).

[0081] The values ​​of the charging pile technical grade and the charging facility scale coefficient can be expressed by the following relationship (5):

[0082] (5);

[0083] In the formula, , , The technical level of the three types of charging piles; This is a scaling factor for the size of charging facilities.

[0084] S202. Based on the uncertainty radius and the source load uncertainty variables, determine the constraint parameters of the worst scenario, and use the pre-determined multi-objective optimization model, deterministic constraints and the constraint parameters of the worst scenario to optimize multiple mixed decision vectors to obtain the optimal solution set.

[0085] The multi-objective optimization model aims to maximize return on investment, minimize comprehensive power quality indicators, minimize load factor deviation, and minimize total investment. The worst-case scenarios include the worst-case scenario for peak load and the worst-case scenario for valley load. Uncertain variables related to source and load can include the predicted values ​​of load, photovoltaic (PV) output, and charging simultaneity rate; the uncertainty radius can be the maximum relative deviation allowed from the predicted value. For example, if the predicted value for conventional load is Pbasepred = 200 kW, and α = 0.2, it indicates that the conventional load may fluctuate within 20% above and below the predicted value, where Pbase ∈ [200 × (1 − 0.2), 200 × (1 + 0.2)], Pbase ∈ [160, 240] kW. A larger α indicates a wider tolerance range for prediction errors, resulting in a more conservative model; a smaller α indicates a higher expectation of accuracy in the prediction, leading to a more aggressive configuration. Specifically, when α = 0, it means the predicted value is completely trusted, and uncertainty is not considered; in this case, the model degenerates into deterministic optimization. When α = 0.1, it means that the load, PV output, and charging simultaneity rate are all allowed to vary within 10% above and below the predicted value. When α = 0.2, it indicates that a fluctuation of 20% is allowed, which is a more conservative and robust configuration. When α approaches 1, it indicates that a large prediction deviation is allowed, and the configuration scheme will be very conservative, which may lead to an over-conservative equipment capacity and increased investment costs. The optimal solution set consists of multiple configuration schemes that satisfy different indicators, including multiple optimal solutions. Each optimal solution is essentially a mixture of decision variables, including multiple normalized decision parameters.

[0086] In the embodiments of this application, the computer device can set the uncertainty radius and the predicted values ​​of the source load uncertain variables based on the scenario requirements, and obtain the constraint parameters of the worst-case scenario based on the predicted values ​​of the uncertainty radius and the source load uncertain variables. Finally, multiple mixed decision vectors and the constraint parameters of the worst-case scenario are input into a multi-objective optimization model for optimization processing. During the optimization process, deterministic constraints must be met until the maximum number of iterations is reached, and the optimal solution set is obtained.

[0087] In some embodiments, the multi-objective optimization model can be represented by the following relation (6):

[0088] (6);

[0089] In the formula, The search space consists of all allowed configuration schemes; The objective function vector; to These correspond to the inverse of the rate of return on investment, the comprehensive power quality index, the economic operation deviation, and the total investment, respectively. For the first Constraint functions.

[0090] S203 uses a preset balancing algorithm to select the best configuration scheme from the optimal solution set.

[0091] The preset balancing algorithm can be either the weighted summation method or the least Euclidean distance method.

[0092] In embodiments of this application, a computer device can calculate the decision parameters contained in each optimal solution in the optimal solution set using a weighted summation method, and compare the calculation results to determine the optimal configuration scheme. In some embodiments, the decision parameters contained in each optimal solution in the optimal solution set can also be calculated using the least Euclidean distance method, and the calculation results can be compared to determine the optimal configuration scheme. In some embodiments, the computer device can first calculate the decision parameters contained in each optimal solution in the optimal solution set using a weighted summation method, and compare the calculation results to determine the optimal configuration scheme; then, it can calculate the decision parameters contained in each optimal solution in the optimal solution set using the least Euclidean distance method to obtain a second calculation result; and then sum the two calculation results for the same optimal solution, determining the optimal solution with the largest sum as the optimal configuration scheme.

[0093] The above-mentioned method for configuring distribution transformer substations involves obtaining multiple hybrid decision vectors for the substation. These hybrid decision vectors include transformer capacity level, static var generator capacity and connection location, active power filter capacity and connection location, charging pile technology level, and charging facility scale coefficient. Based on these hybrid decision vectors, the constraint parameters for the worst-case scenario are determined. Then, using a pre-determined multi-objective optimization model, deterministic constraints, and the constraint parameters for the worst-case scenario, the multiple hybrid decision vectors are optimized to obtain an optimal solution set. The multi-objective optimization model aims to maximize return on investment, minimize comprehensive power quality indicators, minimize load factor deviation, and minimize total investment. Finally, a pre-defined balancing algorithm is used to select the optimal configuration scheme from the optimal solution set. By setting an uncertainty radius, the robustness of the distribution substation equipment configuration scheme can be enhanced. Based on the uncertainty radius, the worst-case peak scenario that may lead to overload and voltage drop and the worst-case valley scenario that may lead to light load can be explicitly constructed, reducing the computational load of the configuration process. Finally, by combining a multi-objective optimization model to optimize the mixed decision variables, it is possible to perform synergistic optimization from multiple dimensions such as economy, power quality, operating efficiency and total investment, thereby improving the economic efficiency of the distribution substation equipment configuration and contributing to the stable operation of the distribution substation.

[0094] In one exemplary embodiment, such as Figure 3 As shown, the constraint parameters for the worst-case scenario are determined based on the uncertainty radius and the source load uncertainty variables. Then, using a pre-determined multi-objective optimization model, deterministic constraints, and the constraint parameters for the worst-case scenario, multiple mixed decision vectors are optimized to obtain the optimal solution set, including:

[0095] S301, obtain the set of uncertain variables of the source load and the uncertainty radius of the distribution radio area, and determine the constraint parameters of the worst scenario based on the set of uncertainty radius and the set of uncertain variables of the source load.

[0096] The source-load uncertainties include load, photovoltaic (PV) capacity, and charging simultaneity rate. The set of source-load uncertainties can include different predicted values ​​of multiple sets of source-load uncertainties. The uncertainty radius can be any value in [0,1]. The worst-case scenarios include peak worst-case scenario and valley worst-case scenario. The peak scenario is used to verify issues such as overload, voltage drop, and N-1 safety. The higher the load, the lower the PV capacity, and the higher the charging simultaneity rate, the more likely overload and low voltage will occur. The valley scenario is used to verify whether the transformer is in an excessively low load range for a long period of time. The lower the normal load, the higher the PV capacity, and the lower the charging simultaneity rate, the lower the net load of the transformer area.

[0097] In some embodiments, the set of source load uncertainties can be represented by the following relation (7):

[0098] (7);

[0099] In the formula, For an uncertain set of source charges; , , These represent actual conventional load, photovoltaic output, and charging simultaneity rate, respectively; with superscripts. The variable is the predicted value; The radius represents the relative uncertainty.

[0100] The range of values ​​for the normal load can be expressed by the following relationship (8):

[0101] (8);

[0102] In the formula, This represents the actual value of the normal load. These are the values ​​for the standard load forecast. This represents the allowable relative fluctuation range. This formula gives the upper and lower bounds of the normal load under the IGDT model, with the highest load value used for peak scenarios and the lowest value used for valley scenarios.

[0103] The range of photovoltaic output values ​​can be expressed by the following relationship (9):

[0104] (9);

[0105] In the formula, To contribute to the actual development of photovoltaics; Contribute to photovoltaic forecasting; Let be the radius of uncertainty in photovoltaic (PV) forecasts. This formula gives the fluctuation range of PV output; the lower limit of PV output will exacerbate peak loads, while the upper limit will intensify the risk of light loads.

[0106] The charging simultaneity rate can be expressed by the following equation (10):

[0107] (10);

[0108] In the formula, This represents the actual charging simultaneity rate. This is a predicted value for the simultaneous charging rate. Let be the radius of the simultaneous rate fluctuation. This formula is used to describe the range of influence of the randomness of electric vehicle charging behavior on the actual charging load.

[0109] In the embodiments of this application, the constraint parameters for the worst-case peak scenario may include the conventional load, photovoltaic output, charging simultaneity rate, and worst net load under the worst-case peak scenario. The constraint parameters for the worst-case valley scenario may include the conventional load, photovoltaic output, charging simultaneity rate, and worst net load under the worst-case valley scenario. After obtaining the set of uncertain variables and the set of uncertain radii of the source load, the predicted values ​​in the set of uncertain variables of the source load can be multiplied by any value in the set of uncertain radii of the source load to obtain the constraint parameters for the worst-case scenario.

[0110] In some embodiments, the normal load under the worst-case peak scenario can be represented by the following relationship (11):

[0111] (11);

[0112] In the formula, This represents the normal load under the most severe peak scenario; This is the load forecast value; This is the upward adjustment ratio.

[0113] The photovoltaic output under the worst-case peak scenario can be expressed by the following relationship (12):

[0114] (12);

[0115] In the formula, Provide photovoltaic power during peak scenarios; This represents the predicted photovoltaic (PV) output. The formula uses the lower limit of PV output because the lower the PV output, the higher the net load, and the more unfavorable the operation of the distribution area.

[0116] The charging simultaneity rate under the worst peak scenario can be expressed by the following relationship (13):

[0117] (13);

[0118] In the formula, The charging simultaneous rate under peak scenarios; This is the predicted simultaneous charging rate. This formula takes the upper limit of the simultaneous charging rate to reflect the load impact caused by a large number of vehicles charging at the same time.

[0119] The peak worst-case net load under the peak worst-case scenario can be expressed by the following relationship (14):

[0120] (14);

[0121] In the formula, For configuration The corresponding peak worst net load; This refers to the total installed capacity of the charging piles; This indicates that the net normal load is not negative, meaning that when the photovoltaic output exceeds the normal load, the net load is not negative, thus avoiding the misapplication of the reverse power flow simplification model to peak load scenarios. This formula combines the load increase, photovoltaic decrease, and charging simultaneity rate increase into the most unfavorable peak load, used for overload, voltage, and safety constraint verification.

[0122] In some embodiments, the normal load under the worst-case scenario of the valley can be represented by the following relationship (15):

[0123] (15);

[0124] In the formula, This represents the normal load under the worst-case scenario of the valley value.

[0125] The photovoltaic output under the worst-case scenario of valley power can be expressed by the following relationship (16):

[0126] (16);

[0127] In the formula, This is for photovoltaic (PV) output during off-peak hours. The formula takes the upper limit of PV output because the higher the PV output, the lower the equivalent net load of the distribution area, and the more obvious the light load problem becomes.

[0128] The charging simultaneity rate under the worst-case scenario can be expressed by the following equation (17):

[0129] (17);

[0130] In the formula, The charging simultaneous rate during off-peak hours.

[0131] The worst-case scenario of the trough net load can be represented by the following relationship (18):

[0132] (18);

[0133] In the formula, For configuration The corresponding trough value represents the worst-case net load; the meanings of the other variables are the same as in the peak load formula. Among them, , This refers to the total installed capacity of the charging piles; For the first The standard number of charging piles; For the first Rated power of a single charging pile; This indicates slow charging, fast charging, or ultra-fast charging.

[0134] S302, determine the technical and economic parameters based on the constraint parameters of the worst-case scenario.

[0135] The technical parameters include voltage deviation, harmonic parameters, and load rate, while the economic parameters include total investment, annual operating cost, and charging revenue.

[0136] In the embodiments of this application, after the computer device has calculated the constraint parameters for the worst-case scenario, it can substitute the constraint parameters into the relationship between the technical parameters and economic parameters to calculate the technical parameters and economic parameters.

[0137] In some embodiments, the voltage deviation can be expressed by the following relationships (19)-(24):

[0138] (19);

[0139] In the formula, For line voltage drop; , These are the active and reactive power of the line, respectively. , These are the line resistance and reactance, respectively. This is the rated voltage. This formula is an engineering approximation of the voltage drop in a power distribution line, illustrating that active power passes through the resistance and reactive power passes through the reactance, both causing voltage deviation.

[0140] Furthermore, the deterministic code provides a basis for feeder resistance. Feeder Reactor Line length Simplified model of power factor components:

[0141] (20);

[0142] In the formula, Peak load; This refers to the line length; and These reflect the active and reactive power components, respectively; the other variables are the same as before. This formula substitutes the line impedance, line length, and power factor into the voltage drop model for deterministic feeder voltage drop estimation.

[0143] The robust code further employs a normalized approximation to facilitate rapid search optimization:

[0144] (twenty one);

[0145] In the formula, Voltage deviation caused by transformer impedance; This is the transformer short-circuit impedance or equivalent impedance voltage drop coefficient; This represents the ratio of peak load to capacity. This formula uses per-unit values ​​to approximate the transformer voltage drop; the closer the load is to the upper limit of capacity, the greater the transformer voltage drop.

[0146] (twenty two);

[0147] In the formula, This is the amount of compensation for voltage deviation by the SVG; For SVG reactive power capacity; This is the position efficiency coefficient; This is a compensation strength calibration factor. This formula indicates that the larger the SVG capacity and the more effective the installation location, the stronger the voltage support capability. The function is used to limit the compensation ratio to no more than 1.

[0148] (twenty three);

[0149] In the formula, For comprehensive voltage deviation; For transformer voltage drop; For line voltage drop; This represents the voltage improvement resulting from reactive power compensation. The formula combines the voltage drop factor with the SVG compensation factor, using them as inputs for voltage deviation constraints and power quality targets.

[0150] (twenty four);

[0151] In the formula, Let be the SVG location efficiency vector, whose elements correspond sequentially to no installation, low-voltage side, transformer substation head, and charging pile cluster. This formula uses empirical coefficients to describe the differences in voltage support efficiency at different access locations. These correspond to no installation, low-voltage side, first end of the transformer area, and charging pile cluster, respectively. The location coefficient reflects the different voltage support effects of SVG of the same capacity at different access locations.

[0152] In some embodiments, harmonics can be represented by the following relationships (25)-(29):

[0153] (25);

[0154] In the formula, The total harmonic distortion of the current; For the first RMS value of subharmonic current; This is the effective value of the fundamental current. This formula is the strict definition of THD, representing the degree of distortion of all higher harmonic currents relative to the fundamental current.

[0155] In the code model, to facilitate rapid evaluation during the planning phase, the harmonic levels of different types of charging piles are represented by grade coefficients. This is expressed as a total harmonic injection, multiplied by a baseline quantity:

[0156] (26);

[0157] In the formula, Injecting indices into the equivalent harmonics before treatment; For the first Number of charging piles; This is the scale factor; This represents the harmonic coefficient corresponding to the technical level of this type of charging pile. This formula summarizes the harmonic contributions of different charging pile types and technical levels into a harmonic index that can be calculated during the planning stage.

[0158] (27);

[0159] In the formula, The equivalent compensation rate for APF; For APF capacity; This represents the APF location efficiency coefficient. This formula indicates that the APF capacity and the access location jointly determine the harmonic mitigation capability. The function limits the upper limit of the compensation rate.

[0160] (28);

[0161] In the formula, Let be the APF location efficiency vector, with elements corresponding to no installation, low-voltage side, transformer substation head, and charging pile cluster, respectively. This formula reflects the impact of different installation locations on harmonic mitigation effectiveness.

[0162] (29);

[0163] In the formula, For configuration Equivalent harmonic index after treatment; Injecting initial harmonics; Let APF be the compensation rate. This formula shows that after APF compensation, the residual harmonics decrease as the compensation rate increases, and through... The formula ensures that the indicators are non-negative. It reflects two trade-offs: first, high-level charging piles have lower harmonics but higher investment; second, the better the APF capacity and access location, the lower the THD, but the higher the investment.

[0164] In some embodiments, the peak load rate can be expressed by the following relationship (30):

[0165] (30);

[0166] In the formula, This refers to the transformer load rate under the worst-case scenario at peak power. The rated capacity of the selected transformer; This is the power factor. This formula converts kW load to kVA capacity utilization, and is used to determine whether the transformer is overloaded during peak hours and whether it is close to the economic operating range.

[0167] The valley load factor can be expressed by the following equation (31):

[0168] (31);

[0169] In the formula, This represents the transformer load rate under the worst-case scenario of the valley value. This represents the valley net load. This formula is used to determine whether the transformer is below the economic operating limit under low load conditions, avoiding over-capacity configuration that leads to prolonged under-loading. The closer the load rate is to the optimal economic operating load rate... The more balanced the transformer loss and capacity utilization, the better.

[0170] In some embodiments, the total investment can be expressed by the following relationships (32)-(34):

[0171] (32);

[0172] In the formula, Total investment; , , , These are investments in transformers, SVG, APF, and charging piles, respectively. This is an optional feeder investment. Among them, , This represents the total investment in charging piles; For the first Baseline cost of similar charging piles; This refers to the technical level of this type of charging pile; This indicates the increased cost resulting from upgrading the level.

[0173] (33);

[0174] In the formula, Invest in SVG; Install the instruction function; Cost per unit capacity; This represents the location cost coefficient. The formula indicates that investment is only incurred when the SVG is installed, and that the investment is determined by capacity, unit price, and installation location.

[0175] (34);

[0176] In the formula, Investing in APF; Install the instruction function; Cost per unit capacity; This represents the location cost coefficient. and This represents the construction / connection cost coefficient for different installation locations. A value of 0 indicates no installation, with corresponding costs and losses of zero. This formula establishes the relationship between APF capacity, location, and investment; the cost is zero when no installation is performed.

[0177] In some embodiments, the annual operating cost can be expressed by the following relationship (35):

[0178] (35);

[0179] In the formula, Annual operating costs; Transformer loss rate; To predict peak load; The electricity purchase price; , The running efficiency of SVG and APF respectively; This represents the annual peak operating hours. This formula converts transformer losses and the losses of the treatment equipment itself into annual operating costs. This means that the annual calculation is based on 6 hours of peak operation per day. The electricity purchase price.

[0180] The charging revenue can be expressed by the following equation (36):

[0181] (36);

[0182] In the formula, Annual charging revenue; Electricity price for charging services; For the first Power of similar charging piles; The utilization rate is used to estimate charging service revenue based on annual operating hours, equipment size, and utilization rate.

[0183]

[0184] In the formula, Electricity price for charging services; The electricity purchase price is used as the basis for calculating the annual revenue. This formula represents a 20% increase in the charging service fee based on the electricity purchase price. To optimize the utilization of slow charging, fast charging, and ultra-fast charging, the code takes... .

[0185] S303 inputs technical parameters, economic parameters, and multiple mixed decision vectors into a multi-objective optimization model for calculation, and obtains a combination of objective values ​​for multiple mixed decision vectors.

[0186] In the embodiments of this application, the computer device can first calculate a simplified rate of return on investment through economic parameters, and then input the simplified rate of return on investment, technical parameters, economic parameters and multiple mixed decision vectors into a multi-objective optimization model for calculation. At the same time, the deterministic constraints in the multi-objective optimization model must be satisfied to obtain a combination of target values ​​of multiple mixed decision vectors.

[0187] In some embodiments, the deterministic constraints of the multi-objective optimization model can be expressed by the following relations (37)-(42):

[0188] (37);

[0189] In the formula, This refers to the amount of voltage deviation constraint violation. This refers to the allowable voltage deviation limit. When the voltage deviation exceeds the limit, If the value is 0, it indicates that the constraint is satisfied; otherwise, it is 0.

[0190] (38);

[0191] In the formula, This refers to the amount of harmonic constraint violation. This is the limit for harmonic distortion rate. This formula is used to determine whether the harmonic distortion meets the power quality requirements after APF (Area-of-Fluid Power) treatment.

[0192] (39);

[0193] In the formula, This refers to the amount of overload constraint violation. This represents the maximum permissible load rate of the transformer. When the peak worst-case load rate exceeds the upper limit, this formula gives the degree of overload.

[0194] (40);

[0195] In the formula, This is the amount of light-load constraint violation; This represents the minimum economic load rate of the transformer. When the worst-case load rate is below the lower limit, this formula indicates the degree of underload.

[0196] (41);

[0197] In the formula, , These are the SVG and APF capacity constraint violation amounts, respectively. and These represent the maximum capacity of the equipment. This formula ensures that the capacity of the selected treatment equipment does not exceed the upper limit that the project can procure or allow access to.

[0198] (42);

[0199] In the formula, The amount of violation of the N-1 safety margin constraint; This indicates an additional 10% safety margin based on the worst-case peak load. The amount of violation is constrained by economic returns; This represents the minimum required rate of return on investment. The formula simultaneously constrains safety redundancy and economic feasibility, ensuring that the solution is both safe and has basic investment value.

[0200] S304. Calculate the combination of target values ​​using a preset crossover algorithm to determine the non-dominated solution for the corresponding round.

[0201] The preset crossover algorithm can be the NSGA-II solution algorithm.

[0202] In the embodiments of this application, the computer device can perform non-dominated sorting of target value combinations based on a preset crossover algorithm to obtain multiple non-dominated levels; calculate the crowding degree of target value combinations at each non-dominated level, and retain boundary solutions and sparse region solutions; determine the parent based on the non-dominated level and crowding degree, and determine multiple offspring individuals of the parent based on the preset crossover algorithm; select the parent and offspring individuals in descending order based on the crowding degree to obtain the non-dominated solution for the corresponding round.

[0203] S305, adjust each mixed decision vector, and return to the step of obtaining the set of source-load uncertain variables and uncertainty radius of the distribution station area, until the iteration round reaches the maximum round, combine the non-dominated solutions of all rounds to obtain the optimal solution set.

[0204] In the embodiments of this application, after the computer completes the calculation of the hybrid decision vector for the current round, it can extract new values ​​from a preset range of values ​​to adjust the hybrid decision vector, obtain a set of source load uncertain variables with new predicted values ​​and a new uncertainty radius, continue to calculate the constraint parameters of the new worst-case scenario, and substitute the new hybrid decision vector of the constraint parameters of the new worst-case scenario into the multi-objective optimization model for calculation to obtain the non-dominated solution of the new round; repeat this step until the iteration round reaches the maximum round, and combine the non-dominated solutions of all rounds to obtain the optimal solution set.

[0205] By setting an uncertainty radius, the robustness of the distribution substation equipment configuration scheme can be enhanced. Based on the uncertainty radius, the worst-case peak scenario that may lead to overload and voltage drop and the worst-case valley scenario that may lead to light load can be explicitly constructed, reducing the computational load of the configuration process. Finally, by combining a multi-objective optimization model to optimize the mixed decision variables, it is possible to perform synergistic optimization from multiple dimensions such as economy, power quality, operating efficiency and total investment, thereby improving the economic efficiency of the distribution substation equipment configuration and contributing to the stable operation of the distribution substation.

[0206] In one exemplary embodiment, such as Figure 4 As shown, the multi-objective optimization model includes a return on investment sub-model, a comprehensive power quality index sub-model, a load factor deviation sub-model, and a total investment sub-model. The objective value combinations include the first, second, third, and fourth objective values ​​of the mixed decision vector. Technical and economic parameters are input into the multi-objective optimization model for calculation, resulting in multiple combinations of objective values ​​for the mixed decision vector, including:

[0207] S401, for each hybrid decision vector, determine the simplified rate of return on investment based on economic parameters, input the simplified rate of return on investment into the rate of return on investment sub-model for calculation, and obtain the first target value of the hybrid decision vector.

[0208] In the embodiments of this application, economic parameters are substituted into the simplified investment return relationship for calculation to obtain the simplified investment return, which can be expressed by the following relationships (43)-(44):

[0209] (43);

[0210] In the formula, For a strict internal rate of return; The project's lifespan; The variable represents the year; the remaining variables are investment, income, and operating costs. This formula is the definition of the internal rate of return with a net present value of zero. If a strict financial approach is used, this equation should be used to solve the problem. .

[0211] Simplified ROI is used in the code and patent embodiments:

[0212] (44);

[0213] In the formula, To simplify the return on investment; Annual net income; Let be the total investment. This formula uses the ratio of annual net income to total investment to approximate the economic return, facilitating rapid calculation in optimization.

[0214] In some embodiments, the rate of return sub-model can be represented by the following relation (45):

[0215] (45);

[0216] In the formula, This is the first objective function; For configuration The rate of return on investment.

[0217] S402, input the voltage deviation and harmonic parameters into the power quality comprehensive index sub-model for calculation, and obtain the second objective value of the hybrid decision vector.

[0218] In the embodiments of this application, the power quality comprehensive index sub-model can be represented by the following relationship (46):

[0219] (46);

[0220] In the formula, For comprehensive power quality objectives; For voltage deviation; This is a harmonic index.

[0221] S403, the load rate is input into the load rate deviation sub-model for calculation, and the third objective value of the hybrid decision vector is obtained.

[0222] In the embodiments of this application, the load rate deviation sub-model can be represented by the following relationship (47):

[0223] (47);

[0224] In the formula, This is due to deviations in economic performance; This represents the peak and worst-case load rate. This formula represents the optimal economic load rate. It brings the transformer load rate as close as possible to the economic operating point, reducing the risks of idle capacity and overload.

[0225] S404: Input the total investment into the total investment sub-model for calculation to obtain the fourth objective value of the mixed decision vector.

[0226] In the embodiments of this application, the total investment sub-model can be represented by the following relation (48):

[0227] (48);

[0228] In the formula, To achieve the target investment scale; For total investment, the denominator is... This formula is used to convert units to millions of yuan. It controls construction costs and prevents excessive equipment configuration in order to improve technical specifications.

[0229] By using a multi-objective optimization model to optimize mixed decision variables, it is possible to achieve synergistic optimization from multiple dimensions such as economy, power quality, operating efficiency, and total investment, thereby improving the economic efficiency of distribution substation equipment configuration and contributing to the stable operation of the distribution substation.

[0230] In one exemplary embodiment, such as Figure 5 As shown, a preset crossover algorithm is used to calculate the combination of target values ​​to determine the non-dominated solutions for the corresponding rounds, including:

[0231] S501, perform non-dominated sorting on the combination of target values ​​to obtain multiple non-dominated levels.

[0232] In the embodiments of this application, after calculating the target value combination of each hybrid decision vector, the computer device treats each hybrid decision vector as a configuration scheme, compares any two configuration schemes, and obtains the non-dominated level; that is, if the scheme... The plan is no worse than the solution in all objectives. And at least one objective is strictly superior to Then it is called Dominate And it is expressed by the following relation (49):

[0233] (49);

[0234] In some embodiments, after all mixed decision vectors have performed the above operations, the non-dominated levels are sorted and the frontier is obtained by filtering.

[0235] S502 calculates the crowding degree of the combination of objective values ​​for each non-dominated level, preserving the boundary solutions and sparse region solutions.

[0236] In the embodiments of this application, the degree of congestion can be represented by the following relationships (50)-(51):

[0237] (50);

[0238] In the formula, For the first The solution is at the th solution. The crowding component on each target; and The objective value of the adjacent solution; , These are the maximum and minimum values ​​of the target.

[0239] (51);

[0240] In the formula, For the first The total crowding distance of each solution.

[0241] S503 determines the parent based on non-dominance level and crowding degree, and determines multiple offspring individuals of the parent based on a preset crossover algorithm.

[0242] In the embodiments of this application, for any two configuration schemes, if the non-dominance levels of the two configuration schemes are different, the configuration scheme with the smaller non-dominance level is extracted; if the non-dominance levels of the two configuration schemes are the same, the configuration scheme with the greater crowding is extracted. The above operations are performed on any two mixed decision vectors, and the extracted configuration scheme (mixed decision variable) is used as the parent. Finally, the parent is processed by simulating binary crossover and polynomial mutation to generate multiple offspring individuals.

[0243] S504: Select the parent and offspring individuals in descending order based on the crowding level to obtain the non-dominated solution for the corresponding round.

[0244] In the embodiments of this application, the target value corresponding to multiple offspring individuals is calculated, and the parent and offspring individuals are merged. The merged individuals are then sorted in a non-dominated order, and individuals are selected sequentially starting from the best frontier until a preset number is filled, thus obtaining the non-dominated solution for the corresponding round. In some embodiments, if the last frontier exceeds the limit, a preset number is selected in descending order of crowding to obtain the non-dominated solution for the corresponding round.

[0245] In one exemplary embodiment, such as Figure 6 As shown, a preset balancing algorithm is used to select the optimal configuration scheme from the set of optimal solutions, including:

[0246] S601, normalize each optimal solution in the optimal solution set to obtain multiple normalized optimal solutions.

[0247] In the embodiments of this application, the normalization process can be represented by the following relation (52):

[0248] (52);

[0249] In the formula, For the first The Pareto solution is at the th . Normalized values ​​on each target; To prevent small positive numbers with a denominator of zero.

[0250] S602, for each normalized optimal solution, calculate the Euclidean distance between multiple objective values ​​in the normalized optimal solution, and determine the optimal configuration scheme based on the Euclidean distance.

[0251] In the embodiments of this application, for each normalized optimal solution, the comprehensive Euclidean distance between multiple objective values ​​of each optimal solution is calculated. That is, the Euclidean distances between different objective values ​​in the same optimal solution are summed to obtain the Euclidean distance of each optimal solution. The Euclidean distances of all optimal solutions are compared, and the optimal solution with the smallest Euclidean distance is taken as the best configuration scheme.

[0252] In one exemplary embodiment, such as Figure 7 As shown, a preset balancing algorithm is used to select the optimal configuration scheme from the set of optimal solutions, including:

[0253] S701, normalize each optimal solution in the optimal solution set to obtain multiple normalized optimal solutions.

[0254] In the embodiments of this application, the normalization process is the same as the principle of step S601, and will not be described in detail here.

[0255] S702, for each normalized optimal solution, the multiple objective values ​​in the normalized optimal solution are weighted and summed to obtain the comprehensive score of each optimal solution.

[0256] In the embodiments of this application, for each normalized optimal solution, a comprehensive score of multiple objective values ​​of each optimal solution is calculated, that is, a weighted sum of multiple objective values ​​in the same optimal solution is performed to obtain a comprehensive score of each optimal solution.

[0257] In some embodiments, the weighted summation can be expressed by the following relation (53):

[0258] (53);

[0259] In the formula, For the first The weighted comprehensive score of each option; For the first Each target weight can be adaptively set according to project preferences.

[0260] S703 determines the optimal solution with the lowest overall score as the best configuration scheme.

[0261] In the embodiments of this application, the comprehensive scores of all optimal solutions are compared, and the optimal solution with the smallest comprehensive score is taken as the best configuration scheme.

[0262] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0263] Based on the same inventive concept, this application also provides a configuration device for a distribution radio station device to implement the configuration method of the distribution radio station device described above. The solution provided by this device is similar to the implementation solution described in the above method. Therefore, the specific limitations of one or more configuration device embodiments of distribution radio station devices provided below can be found in the limitations of the configuration method of distribution radio station devices above, and will not be repeated here.

[0264] In one exemplary embodiment, such as Figure 8 As shown, a configuration device for a distribution radio area device is provided, comprising: an acquisition module 81, a determination module 82, and a filtering module 83, wherein:

[0265] The acquisition module 81 is used to acquire multiple hybrid decision vectors of the distribution transformer area; the hybrid decision vectors include transformer capacity level, static var generator capacity and access location, active power filter capacity and access location, charging pile technology level and charging facility scale coefficient;

[0266] The determination module 82 is used to determine the constraint parameters of the worst-case scenario based on the uncertainty radius and source load uncertainty variables, and to optimize multiple mixed decision vectors using a pre-determined multi-objective optimization model, deterministic constraints, and the constraint parameters of the worst-case scenario to obtain the optimal solution set; wherein, the multi-objective optimization model aims to maximize the return on investment, minimize the comprehensive power quality index, minimize the load rate deviation, and minimize the total investment.

[0267] The filtering module 83 is used to filter the best configuration scheme from the set of optimal solutions using a preset balancing algorithm.

[0268] In an exemplary embodiment, the determining module 82 described above is specifically used for:

[0269] Obtain the set of source and load uncertainties and the uncertainty radius of the distribution transformer area. Determine the constraint parameters for the worst-case scenario based on the set of uncertainty radius and the set of source and load uncertainties. The set of source and load uncertainties includes load, photovoltaic and charging simultaneity rate. The worst-case scenario includes peak worst-case scenario and valley worst-case scenario.

[0270] Technical and economic parameters are determined based on the constraints of the worst-case scenario; among them, technical parameters include voltage deviation, harmonic parameters, and load rate, while economic parameters include total investment, annual operating cost, and charging revenue.

[0271] Technical and economic parameters are input into a multi-objective optimization model for calculation, resulting in a combination of objective values ​​for multiple hybrid decision vectors.

[0272] The non-dominated solution for the corresponding round is determined by calculating the combination of target values ​​using a pre-defined crossover algorithm.

[0273] Adjust each mixed decision vector and return to the step of obtaining the set of source-load uncertain variables and uncertainty radius of the distribution radio station until the maximum number of iterations is reached. Combine the non-dominated solutions of all iterations to obtain the optimal solution set.

[0274] In an exemplary embodiment, the determining module 82 described above is specifically used for:

[0275] For each hybrid decision vector, a simplified rate of return is determined based on economic parameters. The simplified rate of return is then input into the rate of return sub-model for calculation to obtain the first objective value of the hybrid decision vector.

[0276] The voltage deviation and harmonic parameters are input into the power quality comprehensive index sub-model for calculation, and the second objective value of the hybrid decision vector is obtained.

[0277] The load factor is input into the load factor deviation sub-model for calculation to obtain the third objective value of the hybrid decision vector;

[0278] The total investment is input into the total investment sub-model for calculation, resulting in the fourth objective value of the hybrid decision vector.

[0279] In an exemplary embodiment, the determining module 82 described above is specifically used for:

[0280] The combination of target values ​​is sorted by non-dominated order to obtain multiple non-dominated levels;

[0281] Calculate the crowding degree of the combination of objective values ​​for each non-dominated level, and preserve the boundary solutions and sparse region solutions;

[0282] The parent is determined based on the non-dominance level and crowding degree, and multiple offspring individuals of the parent are determined based on a preset crossover algorithm;

[0283] Based on the crowding level, the parent and offspring individuals are selected in descending order to obtain the non-dominated solution for the corresponding round.

[0284] In an exemplary embodiment, the filtering module 83 is specifically used for:

[0285] Normalize each optimal solution in the optimal solution set to obtain multiple normalized optimal solutions;

[0286] For each normalized optimal solution, calculate the Euclidean distance between multiple objective values ​​in the normalized optimal solution, and determine the optimal configuration scheme based on the Euclidean distance.

[0287] In an exemplary embodiment, the filtering module 83 is specifically used for:

[0288] Normalize each optimal solution in the optimal solution set to obtain multiple normalized optimal solutions;

[0289] For each normalized optimal solution, the multiple objective values ​​in the normalized optimal solution are weighted and summed to obtain the comprehensive score of each optimal solution;

[0290] The optimal solution with the lowest overall score is determined as the best configuration scheme.

[0291] Each module in the configuration device of the aforementioned distribution radio station can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0292] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program that, when executed by the processor, implements the steps in the above-described method embodiments.

[0293] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0294] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0295] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0296] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0297] The above embodiments are merely illustrative of several implementation methods of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for configuring a distribution radio area device, characterized in that, The method includes: Multiple hybrid decision vectors for the distribution transformer area are obtained; the hybrid decision vectors include transformer capacity level, static var generator capacity and access location, active power filter capacity and access location, charging pile technology level and charging facility scale coefficient; The constraint parameters for the worst-case scenario are determined based on the uncertainty radius and source load uncertainty variables. Then, multiple hybrid decision vectors are optimized using a pre-determined multi-objective optimization model, deterministic constraints, and the constraint parameters of the worst-case scenario to obtain the optimal solution set. The multi-objective optimization model aims to maximize the return on investment, minimize the comprehensive power quality index, minimize the load factor deviation, and minimize the total investment. The optimal configuration scheme is selected from the set of optimal solutions using a preset balancing algorithm.

2. The method according to claim 1, characterized in that, The process involves determining the constraint parameters for the worst-case scenario based on the uncertainty radius and source load uncertainty variables, and then using a pre-determined multi-objective optimization model, deterministic constraints, and the constraint parameters for the worst-case scenario to optimize multiple hybrid decision vectors, resulting in an optimal solution set, including: Obtain the set of source-load uncertain variables and the uncertainty radius of the distribution transformer area, and determine the constraint parameters of the worst-case scenario based on the set of uncertainty radius and the set of source-load uncertain variables; the set of source-load uncertain variables includes load, photovoltaic and charging simultaneity rate; the worst-case scenario includes peak worst-case scenario and valley worst-case scenario; Technical and economic parameters are determined based on the constraints of the worst-case scenario; wherein the technical parameters include voltage deviation, harmonic parameters, and load rate, and the economic parameters include total investment, annual operating cost, and charging revenue; The technical parameters and economic parameters are input into the multi-objective optimization model for calculation to obtain a combination of objective values ​​for multiple hybrid decision vectors. The target value combination is calculated using a preset crossover algorithm to determine the non-dominated solution for the corresponding round; The mixed decision vectors are adjusted, and the process returns to the step of obtaining the set of source-load uncertain variables and the uncertainty radius of the distribution station, until the iteration reaches the maximum number of iterations. The non-dominated solutions of all iterations are combined to obtain the optimal solution set.

3. The method according to claim 2, characterized in that, The multi-objective optimization model includes a return on investment sub-model, a comprehensive power quality index sub-model, a load factor deviation sub-model, and a total investment sub-model; the objective value combination includes a first objective value, a second objective value, a third objective value, and a fourth objective value of the hybrid decision vector. The technical parameters and economic parameters are input into the multi-objective optimization model for calculation to obtain multiple objective value combinations of the hybrid decision vector, including: For each of the hybrid decision vectors, a simplified rate of return is determined based on the economic parameters, and the simplified rate of return is input into the rate of return sub-model for calculation to obtain the first target value of the hybrid decision vector. The voltage deviation and the harmonic parameters are input into the power quality comprehensive index sub-model for calculation to obtain the second objective value of the hybrid decision vector; The load rate is input into the load rate deviation sub-model for calculation to obtain the third objective value of the hybrid decision vector; The total investment is input into the total investment sub-model for calculation to obtain the fourth objective value of the hybrid decision vector.

4. The method according to claim 2, characterized in that, The step of using a preset crossover algorithm to calculate the combination of target values ​​and determine the non-dominated solution for the corresponding round includes: The target value combinations are sorted using non-dominated ranking to obtain multiple non-dominated levels; Calculate the crowding degree of the target value combination for each of the aforementioned non-dominated levels, preserving the boundary solutions and sparse region solutions; The parent is determined based on the non-dominance level and the crowding degree, and multiple offspring individuals of the parent are determined based on a preset crossover algorithm; Based on the crowding degree, the parent and offspring individuals are selected in descending order to obtain the non-dominated solution for the corresponding round.

5. The method according to any one of claims 1-4, characterized in that, The step of using a preset balancing algorithm to select the optimal configuration scheme from the optimal solution set includes: Normalize each optimal solution in the optimal solution set to obtain multiple normalized optimal solutions; For each normalized optimal solution, calculate the Euclidean distance between multiple objective values ​​in the normalized optimal solution, and determine the optimal configuration scheme based on the Euclidean distance.

6. The method according to any one of claims 1-4, characterized in that, The step of using a preset balancing algorithm to select the optimal configuration scheme from the optimal solution set includes: Normalize each optimal solution in the optimal solution set to obtain multiple normalized optimal solutions; For each normalized optimal solution, a weighted sum of multiple objective values ​​in the normalized optimal solution is performed to obtain a comprehensive score for each optimal solution. The optimal solution with the lowest overall score is determined as the best configuration scheme.

7. A configuration device for a distribution radio station area device, characterized in that, The device includes: The acquisition module is used to acquire multiple hybrid decision vectors of the distribution transformer area; the hybrid decision vectors include transformer capacity level, static var generator capacity and access location, active power filter capacity and access location, charging pile technology level and charging facility scale coefficient; The determination module is used to determine the constraint parameters of the worst-case scenario based on the uncertainty radius and source load uncertainty variables, and to optimize multiple hybrid decision vectors using a pre-determined multi-objective optimization model, deterministic constraints, and the constraint parameters of the worst-case scenario to obtain the optimal solution set; wherein, the multi-objective optimization model aims to maximize the return on investment, minimize the comprehensive power quality index, minimize the load factor deviation, and minimize the total investment. The filtering module is used to select the best configuration scheme from the set of optimal solutions using a preset balancing algorithm.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

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

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

Images