Parameter optimization method and system for flexible interconnected phase shifting device

Abstract

The present invention discloses a parameter optimization method and system for a flexible interconnected phase shifter, which is applied to the technical field of flexible interconnected phase shifters. The method includes obtaining power data of the current medium-voltage distribution network; constructing a multi-objective two-layer optimization model with interconnected network feeder power flow balance and network loss as optimization objectives; obtaining the target results output by the multi-objective two-layer optimization model, wherein both the upper optimization model and the lower optimization model are designed to process randomly generated initial parameters in sequence through a probabilistic proxy model and a TPE strategy; and processing the target results based on the Pareto frontier solution set to obtain parameter optimization decision results. The parameter optimization method and system for a flexible interconnected phase shifter provided by the present invention propose a multi-objective optimization objective function that considers improving medium-voltage distribution network feeder power flow balance and network loss. Based on the Bayesian optimization algorithm combined with multi-objective optimization, a Bayesian nested two-layer optimization solution algorithm is constructed that considers the parameters of the flexible phase shifter and is oriented to installation location optimization.

Description

Technical Field

[0001] The present invention relates to the technical field of flexible interconnected phase shifters, and in particular to a parameter optimization method and system for a flexible interconnected phase shifter. Background Art

[0002] The distribution network's status and role as a platform for carrying electric energy have been further strengthened. Characteristics such as active generation and source-load interaction are becoming increasingly prominent on the distribution network side. The rapid development of distributed generation (DG) and electric vehicles (EVs) has resulted in distribution networks exhibiting high volatility in both source and load. High DG penetration and source-load fluctuations in localized areas have led to or exacerbated a series of problems in distribution networks, including difficulty accommodating DG, voltage overshooting, and reduced operational efficiency.

[0003] Traditional medium and low voltage AC distribution networks, with their "closed-loop design and open-loop operation," are no longer able to adapt to the evolving demands of distribution network sources and loads. Flexible interconnected phase-shifting devices, based on power electronic devices, enable flexible interconnection of distribution networks, effectively improving the ability to regulate node voltages and branch power. With minimal investment and on-site modifications, they can fully unleash the power supply capacity of existing distribution networks, efficiently utilize redundant capacity, and achieve power sharing, precise power flow control, reduced line losses, and improved power quality across distribution networks in different areas and substations.

[0004] However, the existing parameter control method for flexible interconnected phase shifters analyzes all parameters one by one under the condition of a single optimization objective. The algorithm is inefficient and has poor dynamic adaptability, which seriously affects the rational control of medium and low voltage AC distribution networks. Summary of the Invention

[0005] The present invention provides a parameter optimization method and system for a flexible interconnected phase-shifting device. Based on dual optimization objectives, a two-layer model is designed. The upper and lower models optimize and feedback each other, and parameters are rationally selected for subsequent power flow calculations. This solves the problem of low algorithm efficiency caused by calculating all parameters, and is beneficial to the rational regulation of medium and low voltage AC power distribution networks that incorporate the flexible interconnected phase-shifting device.

[0006] In order to solve the above technical problems, an embodiment of the present invention provides a parameter optimization method for a flexible interconnected phase shifting device, comprising:

[0007] Obtain current power data of the medium voltage distribution network;

[0008] Based on the power data, with interconnected network feeder power flow balance and network loss as optimization objectives, a multi-objective two-level optimization model with mutual feedback between an upper-level optimization model and a lower-level optimization model is constructed, wherein the upper-level optimization model is configured to use position parameters and control parameters of the flexible interconnected phase shifting device as decision variables, and the lower-level optimization model is configured to optimize the control parameters based on the selected position parameters of the flexible interconnected phase shifting device;

[0009] Obtaining a target result output by the multi-objective two-level optimization model, wherein both the upper-level optimization model and the lower-level optimization model are designed to sequentially process randomly generated initial parameters using a probabilistic proxy model and a TPE strategy, and perform power flow optimization calculations based on the processed results, wherein the initial parameters include position parameters and control parameters of the flexible interconnected phase shifting device;

[0010] The target result is processed based on the Pareto frontier solution set to obtain a parameter optimization decision result of the flexible interconnected phase shifting device in the current medium-voltage distribution network.

[0011] As one of the preferred solutions, the power data at least includes balancing node information data, active node information data, reactive node information data and network element information data.

[0012] As one of the preferred solutions, both the upper-layer optimization model and the lower-layer optimization model are used to construct the probabilistic proxy model using a Gaussian process of Bayesian optimization to process the randomly generated initial parameters.

[0013] As one preferred solution, both the upper-layer optimization model and the lower-layer optimization model are used to obtain an objective function corresponding to the current network state through power flow calculation, and the objective function is used as the target result, wherein the power flow calculation process includes:

[0014] Under the current randomly generated initial parameter conditions, the topological structure information, component parameter information, power generation information and load parameter information of the current medium-voltage distribution network are processed based on the selected power flow calculation method to obtain the steady-state operating status parameters of the current medium-voltage distribution network.

[0015] As one of the preferred solutions, both the upper-layer optimization model and the lower-layer optimization model are used to construct a sampling function using the TPE strategy, and at least a maximum improved observation value is obtained based on the sampling function.

[0016] Another embodiment of the present invention provides a parameter optimization system for a flexible interconnected phase shifting device, comprising:

[0017] The acquisition module is used to obtain the power data of the current medium voltage distribution network;

[0018] a construction module for constructing, based on the power data and with interconnected network feeder power flow balance and network loss as optimization objectives, a multi-objective two-level optimization model with mutual feedback between an upper-level optimization model and a lower-level optimization model, wherein the upper-level optimization model is configured to use position parameters and control parameters of the flexible interconnected phase shifting device as decision variables, and the lower-level optimization model is configured to optimize the control parameters using the position parameters of the selected flexible interconnected phase shifting device;

[0019] an output module for obtaining a target result output by the multi-objective two-layer optimization model, wherein both the upper-layer optimization model and the lower-layer optimization model are designed to sequentially process randomly generated initial parameters using a probabilistic proxy model and a TPE strategy, and perform power flow optimization calculations based on the processed results, wherein the initial parameters include position parameters and control parameters of the flexible interconnected phase shifting device;

[0020] A decision module is used to process the target result based on the Pareto frontier solution set to obtain a parameter optimization decision result of the flexible interconnected phase shifting device in the current medium-voltage distribution network.

[0021] As one of the preferred solutions, the power data at least includes balancing node information data, active node information data, reactive node information data and network element information data.

[0022] As one of the preferred solutions, both the upper-layer optimization model and the lower-layer optimization model are used to construct the probabilistic proxy model using a Gaussian process of Bayesian optimization to process the randomly generated initial parameters.

[0023] As one preferred solution, both the upper-layer optimization model and the lower-layer optimization model are used to obtain an objective function corresponding to the current network state through power flow calculation, and the objective function is used as the target result, wherein the power flow calculation process includes:

[0024] Under the current randomly generated initial parameter conditions, the topological structure information, component parameter information, power generation information and load parameter information of the current medium-voltage distribution network are processed based on the selected power flow calculation method to obtain the steady-state operating status parameters of the current medium-voltage distribution network.

[0025] As one of the preferred solutions, both the upper-layer optimization model and the lower-layer optimization model are used to construct a sampling function using the TPE strategy, and at least a maximum improved observation value is obtained based on the sampling function.

[0026] Compared with the prior art, the embodiments of the present invention have the following advantages:

[0027] This invention proposes a multi-objective optimization objective function that considers improving feeder power flow balance and network losses in a medium-voltage distribution network. Based on a Bayesian optimization algorithm and combined with multi-objective optimization, a Bayesian nested two-level optimization solution algorithm is constructed that considers the parameters of flexible phase shifters and optimizes their installation locations. This eliminates the need for power flow calculations for all parameters. Instead, this embodiment employs a simplified proxy model to predict feeder balance and network losses under different control parameters, thereby more quickly finding the optimal control parameter set without requiring tedious power flow calculations each time. Furthermore, this embodiment employs a Tree-structured Parzen Estimator (TPE) strategy to collect observations and control parameters that maximize feeder balance and network losses, helping to more quickly find the optimal solution. Power flow calculations are then performed to verify the effectiveness of this parameter set. The flexible phase shifter interconnected distribution network, derived through the optimization algorithm, enables the joint consumption of distributed photovoltaic power on two feeders, improves active power balance on the feeder source side, and significantly mitigates the negative impact of load fluctuations from distributed photovoltaic power generation and electric vehicles on the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 1 is a flow chart of a parameter optimization method for a flexible interconnected phase shifting device in one embodiment of the present invention;

[0029] Figure 2 is a flow chart of an algorithm for optimizing the configuration of flexible interconnected assembly in one embodiment of the present invention (upper-level optimization model);

[0030] Figure 3 is a flow chart of an algorithm for optimizing the configuration of flexible interconnected assembly in one embodiment of the present invention (lower-layer optimization model);

[0031] Figure 4 This is an example of one embodiment of the present invention - a 10kV distribution network electrical topology diagram;

[0032] Figure 5 This is an example of one embodiment of the present invention - a typical daily active load curve;

[0033] Figure 6 This is an example of one embodiment of the present invention - a typical daily photovoltaic power generation curve;

[0034] Figure 7 This is an example of one embodiment of the present invention - a multi-objective two-level optimization Pareto solution;

[0035] Figure 8 This is an example of one embodiment of the present invention - a 10kV distribution network after flexible interconnection;

[0036] Figure 9 This is an example of one embodiment of the present invention - a ratio optimization curve of a flexible interconnection device;

[0037] Figure 10 This is an example of one embodiment of the present invention - a phase shift angle optimization curve of a flexible interconnecting device;

[0038] Figure 11 This is an example of one embodiment of the present invention - a curve showing the variation of feeder power supply power after interconnection;

[0039] Figure 12 This is an example of one embodiment of the present invention - a curve of network loss change after interconnection;

[0040] Figure 13 It is a structural block diagram of a parameter optimization system of a flexible interconnected phase shifting device in one embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0042] In the description of this application, the terms "first," "second," "third," etc. are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first," "second," "third," etc. may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0043] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or an indirect connection through an intermediate medium, or it can be a communication between the two components. The terms "vertical", "horizontal", "left", "right", "up", "down" and similar expressions used herein are for illustrative purposes only, and do not indicate or imply that the device or component referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0044] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in this application have the same meanings as those commonly understood by those skilled in the art. The terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. Those skilled in the art will understand the specific meanings of the above terms in this application in specific circumstances.

[0045] An embodiment of the present invention provides a parameter optimization method for a flexible interconnected phase shifting device. For details, see Figure 1 , Figure 1 FIG. 1 is a flow chart of a parameter optimization method for a flexible interconnected phase shifting device according to one embodiment of the present invention, which includes steps S1 to S4:

[0046] S1. Obtain the current power data of the medium voltage distribution network;

[0047] S2. Based on the power data, and with interconnected network feeder power flow balance and network loss as optimization objectives, construct a multi-objective, two-level optimization model with mutual feedback between an upper-level optimization model and a lower-level optimization model, wherein the upper-level optimization model is configured to use position parameters and control parameters of the flexible interconnected phase shifting device as decision variables, and the lower-level optimization model is configured to optimize the control parameters based on the position parameters of the selected flexible interconnected phase shifting device;

[0048] S3. Obtaining a target result output by the multi-objective two-level optimization model, wherein both the upper-level optimization model and the lower-level optimization model are designed to sequentially process randomly generated initial parameters using a probabilistic proxy model and a TPE strategy, and perform power flow optimization calculations based on the processed results, wherein the initial parameters include position parameters and control parameters of the flexible interconnected phase shifter;

[0049] S4. Process the target result based on the Pareto frontier solution set to obtain a parameter optimization decision result of the flexible interconnected phase shifting device in the current medium-voltage distribution network.

[0050] A flexible interconnected phase shifter is a device used to control the phase difference between voltage and current in an AC circuit. It can change the relative time relationship between current and voltage in the circuit, thereby adjusting the ratio of active power to reactive power to meet load demand and stabilize the power system. In an embodiment of the present invention, a multi-objective two-layer optimization model is used for processing. In a medium-voltage active power distribution network, in order to give full play to the role of the flexible interconnection device in power flow regulation in the interconnected network, the interconnection network feeder power balance and network loss are used as optimization objectives, and the interconnection position and device parameters of the flexible interconnection phase shifter are used as decision variables. A multi-objective two-layer optimization model for the site selection of the flexible interconnection device is constructed. The two-layer optimization model for the site selection of the flexible interconnection device is as follows:

[0051]

[0052] In the two-layer optimization model, in the upper-layer optimization, the node locations of the flexible interconnecting devices and the equipment control parameters are both decision variables, but in the lower-layer optimization, the node locations of the flexible interconnecting devices are parameters and the equipment control parameters are decision variables. The two-layer optimization model is explained in detail below.

[0053] Optionally, for the upper optimization model, it includes the following steps, please combine with the participation Figure 2 , Figure 2 The following is a flow chart of an algorithm for optimizing the configuration of flexible interconnected assembly in one embodiment of the present invention (upper optimization model):

[0054] Step 1: Read the current power data of the medium-voltage distribution network. This primarily includes node information, such as balancing nodes, active nodes, and reactive nodes. Active nodes are nodes to which active power flows from both sides, and reactive nodes are nodes to which reactive power flows from both sides. This data also includes network component information (such as series lines, transformers, phase shifters, and shunt compensators), their connection relationships and parameters, and the connection relationships and parameters (active and reactive) between generation and loads. This information forms the basis for subsequent steps.

[0055] Step 2: Set the upper and lower layer variable information, including setting the upper layer variable information, i.e., the installation start and end nodes and their upper and lower limits in the flexible interconnection device decision variables, and setting the lower layer variable information, i.e., the transformation ratio and phase shift angle information.

[0056] The control parameter constraints of the flexible phase shifter are as follows:

[0057]

[0058] Among them: tk min , tk max They are the upper and lower limits of the transformation ratio, usually 0.9 and 1.1, respectively. min and Angle max They are the upper and lower limits of the phase shift angle respectively.

[0059] Step 3: Solve the current medium-voltage distribution network objective function and the optimization solution of its control parameters, which specifically includes calling the lower-level optimization module, passing the installation location parameters, and solving the distribution network objective function and the optimization solution of its control parameters. and The expression is as follows:

[0060]

[0061] It should be noted that P loss and P feeder It is a function of x and y, but there is no explicit functional expression for it. This model uses a numerical solution rather than an analytical solution. x refers to the position vector formed by the two connected nodes, and y refers to the vector formed by tk and angle.

[0062] in, To obtain the network loss, To calculate the active power balance degree on the source side of the medium voltage interconnection network. loss is the active power loss of the ith branch element in the network, in MW; n is the total number of branch elements in the network; The active power distributed on the source side of interconnected feeder 1, in MW; is the active power distributed on the source side of interconnected feeder 2, unit, MW; x is a two-dimensional integer vector, representing the installation location of the interconnected equipment, x(1) is the head node vector, x(2) is the end node vector; y is a two-dimensional real number variable, representing the control parameter of the interconnected equipment, y(1) is the flexible phase shifter ratio vector, y(2) is the flexible phase shifter angle vector.

[0063] In an embodiment of the present invention, optionally, the medium voltage distribution network is constrained as follows:

[0064]

[0065] Among them, P ij is the branch ij power, V i is the voltage at node i, V j is the voltage at node j, G ij is the real part of the admittance of branch ij, B ij is the imaginary part of the admittance of branch ij, θ ij is the voltage phase angle difference between nodes i and j.

[0066] Voltage Constraints:

[0067]

[0068] Among them, V i is the voltage at node i, V mini , V maxi are the lower and upper limits of the voltage constraint at node i, respectively.

[0069] Branch power constraints:

[0070]

[0071] Among them, P ij is the branch ij power, P minij , P maxij are the lower and upper limits of the power constraints of branch ij respectively.

[0072] Step 4: Construct the upper-level optimization observation space and probabilistic proxy function. In this step, the previously obtained data is used to build a model that can predict the value of the objective function at different node positions. This allows for quick predictions of the objective function without performing actual simulations.

[0073] Specifically, in the embodiment of the present invention, the Gaussian Process (GP) commonly used in Bayesian optimization is used as the probability proxy model. For the input X=[x1,x2,..,x n ] T , the target function output y=[y1,y2,..,y n ] T , through the joint probability distribution to satisfy the following formula:

[0074]

[0075] in: is the mean function, 、 and is the covariance function, is to predict new input, yes The corresponding output value.

[0076] Step 5: Use the proxy model to predict the objective function value and uncertainty of the new parameter point. This embodiment of the present invention uses joint probabilistic distributed computing to derive the mean and variance of the predicted value, which is then used to determine the next sampling point. After obtaining the sample, the parameters of the Gaussian process model are updated based on the new observation space.

[0077] Step 6: Use the Tree-Structure Parzen Estimator (TPE) sampling strategy to construct a sampling function and collect observations that maximize the objective function improvement, along with information about their installation locations. TPE is an optimization strategy that uses previous optimization results to guide subsequent search directions. In this step, TPE collects probabilistic information about the optimization target to determine the optimal search direction. Guided by the TPE strategy, the node locations most likely to improve the objective function are identified and the model is updated.

[0078] The TPE sampling strategy used in the model is based on a certain threshold y*, which divides the objective function value into two categories. The better part is classified as Y L , the corresponding parameter set is recorded as X L ; The other type is recorded as Y R and X R The probability density functions (see the following formula) are constructed separately, and new sampling points are generated using the estimator to achieve effective exploration of unknown areas and rational use of better areas, thus achieving a good balance in the process of finding the optimal solution.

[0079]

[0080]

[0081]

[0082] Where l(x) is in the region X L The probability density function in the region X is g(x). R The probability density function in x i is the sample point, K(x,x i ) is the kernel function. N L and N R Corresponding to X L and X R The number of samples.

[0083] Step 7: Call the lower-level optimization module again, pass the installation location parameters, and solve the distribution network objective function and its control parameter optimization solution.

[0084] Step 8: Determine whether the termination conditions are met. That is, after updating the node positions, the control parameters need to be adjusted again to obtain a new optimized solution for the objective function. By adjusting the control parameters and updating the node positions multiple times, one or more optimization target solutions can be obtained. These solutions may correspond to different network configurations and performance. After obtaining the optimization target solution, the equipment needs to be installed according to the actual situation. This may involve adding new equipment to the space or adjusting the location and parameters of existing equipment. The equipment installation process continues until certain termination conditions are met, such as achieving the expected network performance or installing a sufficient number of devices.

[0085] Preferably, the termination condition is usually: reaching a preset number of iterations or the improvement of the objective function value is less than a given threshold. If the condition is met, go to step 11 of this layer, otherwise go to step 9 of this layer.

[0086] Step 9: Add new installation location information and its corresponding objective function value in the observation space.

[0087] Step 10: Update the parameters of the upper-layer optimization probability proxy model and proceed to step 5 of the current-layer optimization.

[0088] Step 11: Calculate the Pareto frontier solution set for the objective function and make an optimal solution decision. Choose the optimal solution set among multiple optimization objective solutions. This typically requires considering multiple factors, such as cost, performance, and reliability.

[0089] Optionally, for the lower layer optimization model, it includes the following steps, please combine with the participation Figure 3 , Figure 3 The following is a flow chart of an algorithm for optimizing the configuration of flexible interconnected assembly in one embodiment of the present invention (lower-level optimization model):

[0090] Step 1: Set the interconnection device installation node information according to the variable information transferred by the upper layer optimization.

[0091] Step 2: Randomly generate a phase shift device transformation ratio and phase shift angle parameter set.

[0092] Step 3: Connect the flexible interconnection device to the distribution network, obtain the network status through power flow calculation, and calculate the objective function (see the above-mentioned dual-objective function expression for details). In this embodiment of the present invention, the topology structure, component parameters, and power generation and load parameters of the distribution network after connecting the flexible interconnection device are given, and the steady-state operating state parameters of the distribution network can be calculated by conventional power flow calculation methods. Conventional power flow calculation algorithms include but are not limited to the Gauss-Seidel method (GS method), the Newton-Raphson method (NR method), the rapid decoupling method (PQ decomposition method), etc.

[0093] Step 4: Construct the lower-level optimization observation space (including objective function and control parameters) and probabilistic proxy function.

[0094] Step 5: Use the surrogate model to predict the objective function value and uncertainty of the new parameter point.

[0095] Step 6: Use the TPE sampling strategy to construct a sampling function to collect the maximum improvement observation value of the objective function and its installation location node information.

[0096] Step 7: Calculate the power flow and obtain the network status, and calculate the objective function (see the above-mentioned flexible phase shift device control parameter constraint formula for details).

[0097] Step 8: Determine whether the termination condition is met.

[0098] The termination condition is usually: reaching a preset number of iterations or the objective function value improvement is less than a given threshold. If the condition is met, go to step 11 of the current layer; otherwise, go to step 9 of the current layer.

[0099] Step 9: Add new transformation ratio and phase shift angle information and their corresponding objective function values in the observation space.

[0100] Step 10: Update the parameters of the upper-layer optimization probability proxy model and proceed to step 5 of the current-layer optimization.

[0101] Step 11: Obtain the Pareto frontier solution set of the objective function and make an optimization solution decision.

[0102] The above describes the upper-layer and lower-layer optimization models, respectively. In this two-layer optimization model, the upper and lower layers focus on different aspects of the design of the FPSID installation node locations and device control parameters. The upper layer determines the FPSID installation node locations, a global decision subject to various practical constraints, such as network structure, device capacity, and safe operating conditions. The lower layer focuses on optimizing device control parameters, such as phase shift angle and transmission power, given the FPSID installation node locations. These control parameters directly impact the FPSID's power flow regulation capability and network losses. The results of the lower layer optimization serve as feedback for the upper layer optimization. Through continuous iteration and optimization, a closed-loop feedback mechanism is formed between the upper and lower layers, continuously improving the efficiency and accuracy of the entire optimization process. This balances global strategies and local performance, improving optimization efficiency, and enhancing the practicality of the results.

[0103] To further illustrate the beneficial effects of the embodiments of the present invention, a specific example is described in detail below.

[0104] The following describes the network structure and features proposed by the present invention in detail using a 10kV distribution network in a certain area as an example: Figure 4 As shown in Figure 1, the topology of the 10kV distribution network is as follows: Feeder 1 (starting with the number 1 in the figure) and Feeder 2 (starting with the number 1 in the figure) are interconnected through phase shifters. Feeder 1 and Feeder 2 each contain 5 loads and 1 photovoltaic power station. Among them, node 11 in feeder 1 is the 10kV busbar of the substation, and load nodes 12 to 17 and 100 are photovoltaic nodes; node 21 in feeder 2 is the 10kV busbar of the substation, and load nodes 22 to 27 and 200 are photovoltaic nodes. The power supply load level, characteristics, and access node status within the grid are detailed in Table 1, and the characteristic curves of residential and commercial loads are detailed in Table 1. Figure 5 .

[0105] Table 1: Power supply load access, load level and load characteristics

[0106]

[0107] The photovoltaic access situation, installed capacity and typical daily active power characteristics of the grid are detailed in Table 2 and Figure 6 .

[0108] Table 2: Photovoltaic access and power generation levels

[0109]

[0110] Typical daily electricity load characteristics and photovoltaic power generation curve data show that during the daytime photovoltaic peak period (12:00-14:00), because the power supply load of feeder 1 is mainly residential, the midday power load is relatively low, less than 1MW, and the photovoltaic power generation level is close to 10MW. Feeder 1 has a large active power flow feedback, which is at a heavy load level. Feeder 2, on the other hand, mainly supplies administrative office power and has a heavy load (0.8-1.0MW) and a small photovoltaic installed capacity (1.0MW). Feeder 2 has a large active power flow, which is at a heavy load state. During the nighttime photovoltaic off-time period, feeder 1 has a large charging load demand and a heavy load, while feeder 2 has a light power supply load.

[0111] (2) Flexible interconnection optimization results and analysis

[0112] Combining network topology information, source-load composition, and typical daily operating characteristics, the Pareto front solution set is obtained based on the Bayesian nested two-level optimization solution for the two objectives of active power deviation and network loss of feeder 1 and feeder 2 (see Figure 7 , Figure 7 This is an example of one embodiment of the present invention - a multi-objective dual-level optimization Pareto solution. Using the approximate ideal solution sorting method to evaluate the Pareto front solution set, it is concluded that a flexible interconnection device is installed between node 14 and node 25 (see Figure 8 , Figure 8This is an example of one embodiment of the present invention - a 10kV distribution network after flexible interconnection).

[0113] Typical daily optimization results of the distribution network after flexible interconnection (see Figures 9-12 ,in Figure 9 This is an example of one embodiment of the present invention - the ratio optimization curve of the flexible interconnection device. Figure 10 This is an example of one embodiment of the present invention - the phase shift angle optimization curve of the flexible interconnection device. Figure 11 This is an example of one embodiment of the present invention - the feeder power supply power variation curve after interconnection, Figure 12 The network loss curve after interconnection (an example of one embodiment of the present invention) shows that by controlling the transformation ratio (0.983-1.014) and phase shift angle parameters (4.2-5.6 degrees) of the flexible phase shifter, the active power deviation between feeders 1 and 2 is reduced to between 0.02 and 0.084, and the network loss is reduced to between 0.02 and 0.23. The results of this example demonstrate that the flexible phase shifter interconnected distribution network, derived through the optimization algorithm, can jointly accommodate distributed photovoltaic power generation on both feeders, improving active power balance on the source side of the feeders and significantly alleviating the negative impact of load fluctuations from distributed photovoltaic power generation and electric vehicles on the grid.

[0114] Another embodiment of the present invention provides a parameter optimization system for a flexible interconnected phase shifting device. Figure 13 , Figure 13 FIG. 1 is a block diagram of a parameter optimization system for a flexible interconnected phase shifting device according to one embodiment of the present invention, which includes:

[0115] Acquisition module A is used to obtain the current power data of the medium voltage distribution network;

[0116] a construction module B, configured to construct, based on the power data and with interconnected network feeder power flow balance and network loss as optimization objectives, a multi-objective two-level optimization model with mutual feedback between an upper-level optimization model and a lower-level optimization model, wherein the upper-level optimization model is configured to use position parameters and control parameters of the flexible interconnected phase shifting device as decision variables, and the lower-level optimization model is configured to optimize the control parameters using the position parameters of the selected flexible interconnected phase shifting device;

[0117] an output module C for obtaining a target result output by the multi-objective two-level optimization model, wherein both the upper-level optimization model and the lower-level optimization model are designed to sequentially process randomly generated initial parameters using a probabilistic proxy model and a TPE strategy, and perform power flow optimization calculations based on the processed results, wherein the initial parameters include position parameters and control parameters of the flexible interconnected phase shifting device;

[0118] The decision module D is used to process the target result based on the Pareto frontier solution set to obtain a parameter optimization decision result of the flexible interconnected phase shifting device in the current medium-voltage distribution network.

[0119] Furthermore, in the above embodiment, the power data at least includes balancing node information data, active node information data, reactive node information data and network element information data.

[0120] Furthermore, in the above embodiment, both the upper-layer optimization model and the lower-layer optimization model are used to construct the probabilistic proxy model using a Gaussian process of Bayesian optimization to process the randomly generated initial parameters.

[0121] Furthermore, in the above embodiment, both the upper-layer optimization model and the lower-layer optimization model are used to obtain an objective function corresponding to the current network state through power flow calculation, and the objective function is used as the target result, wherein the power flow calculation process includes:

[0122] Under the current randomly generated initial parameter conditions, the topological structure information, component parameter information, power generation information and load parameter information of the current medium-voltage distribution network are processed based on the selected power flow calculation method to obtain the steady-state operating status parameters of the current medium-voltage distribution network.

[0123] Furthermore, in the above embodiment, both the upper-layer optimization model and the lower-layer optimization model are used to construct a sampling function using the TPE strategy, and at least a maximum improved observation value is obtained based on the sampling function.

[0124] The parameter optimization method and system for the flexible interconnected phase shifting device provided in the embodiments of the present invention have the following beneficial effects:

[0125] This invention proposes a multi-objective optimization objective function that considers improving feeder power flow balance and network losses in a medium-voltage distribution network. Based on a Bayesian optimization algorithm and combined with multi-objective optimization, a Bayesian nested two-level optimization solution algorithm is constructed that considers the parameters of flexible phase shifters and optimizes their installation locations. This eliminates the need for power flow calculations for all parameters. Instead, this embodiment employs a simplified proxy model to predict feeder balance and network losses under different control parameters, thereby more quickly finding the optimal control parameter set without requiring tedious power flow calculations each time. Furthermore, this embodiment employs a Tree-structured Parzen Estimator (TPE) strategy to collect observations and control parameters that maximize feeder balance and network losses, helping to more quickly find the optimal solution. Power flow calculations are then performed to verify the effectiveness of this parameter set. The flexible phase shifter interconnected distribution network, derived through the optimization algorithm, enables the joint consumption of distributed photovoltaic power on two feeders, improves active power balance on the feeder source side, and significantly mitigates the negative impact of load fluctuations from distributed photovoltaic power generation and electric vehicles on the power grid.

[0126] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A parameter optimization method for a flexible interconnected phase shifting device, characterized in that: include: Obtain current power data of the medium voltage distribution network; Based on the power data, with interconnected network feeder power flow balance and network loss as optimization objectives, a multi-objective two-level optimization model with mutual feedback between an upper-level optimization model and a lower-level optimization model is constructed, wherein the upper-level optimization model is configured to use position parameters and control parameters of the flexible interconnected phase shifting device as decision variables, and the lower-level optimization model is configured to optimize the control parameters based on the selected position parameters of the flexible interconnected phase shifting device; Obtaining a target result output by the multi-objective two-level optimization model, wherein both the upper-level optimization model and the lower-level optimization model are designed to sequentially process randomly generated initial parameters using a probabilistic proxy model and a TPE strategy, and perform power flow optimization calculations based on the processed results, wherein the initial parameters include position parameters and control parameters of the flexible interconnected phase shifting device; The target result is processed based on the Pareto frontier solution set to obtain a parameter optimization decision result of the flexible interconnected phase shifting device in the current medium-voltage distribution network.

2. The parameter optimization method of the flexible interconnect phase shifting device according to claim 1, characterized in that: The power data at least includes balancing node information data, active node information data, reactive node information data and network element information data.

3. The parameter optimization method of the flexible interconnected phase shifting device according to claim 1, characterized in that: The upper layer optimization model and the lower layer optimization model are both used to construct the probabilistic proxy model using a Gaussian process of Bayesian optimization to process the randomly generated initial parameters.

4. The parameter optimization method of the flexible interconnected phase shifting device according to claim 1, wherein: The upper-layer optimization model and the lower-layer optimization model are both used to obtain an objective function corresponding to the current network state through power flow calculation, and the objective function is used as the target result, wherein the power flow calculation process includes: Under the current randomly generated initial parameter conditions, the topological structure information, component parameter information, power generation information and load parameter information of the current medium-voltage distribution network are processed based on the selected power flow calculation method to obtain the steady-state operating status parameters of the current medium-voltage distribution network.

5. The parameter optimization method of the flexible interconnected phase shifting device according to claim 1, characterized in that: The upper-layer optimization model and the lower-layer optimization model are both used to construct a sampling function using the TPE strategy, and to obtain at least a maximum improved observation value based on the sampling function.

6. A parameter optimization system for a flexible interconnected phase shifting device, characterized in that: include: The acquisition module is used to obtain the power data of the current medium voltage distribution network; a construction module for constructing, based on the power data and with interconnected network feeder power flow balance and network loss as optimization objectives, a multi-objective two-level optimization model with mutual feedback between an upper-level optimization model and a lower-level optimization model, wherein the upper-level optimization model is configured to use position parameters and control parameters of the flexible interconnected phase shifting device as decision variables, and the lower-level optimization model is configured to optimize the control parameters using the position parameters of the selected flexible interconnected phase shifting device; an output module for obtaining a target result output by the multi-objective two-layer optimization model, wherein both the upper-layer optimization model and the lower-layer optimization model are designed to sequentially process randomly generated initial parameters using a probabilistic proxy model and a TPE strategy, and perform power flow optimization calculations based on the processed results, wherein the initial parameters include position parameters and control parameters of the flexible interconnected phase shifting device; A decision module is used to process the target result based on the Pareto frontier solution set to obtain a parameter optimization decision result of the flexible interconnected phase shifting device in the current medium-voltage distribution network.

7. The parameter optimization system for the flexible interconnected phase shifting device according to claim 6, characterized in that: The power data at least includes balancing node information data, active node information data, reactive node information data and network element information data.

8. The parameter optimization system for the flexible interconnected phase shifting device according to claim 6, wherein: The upper layer optimization model and the lower layer optimization model are both used to construct the probabilistic proxy model using a Gaussian process of Bayesian optimization to process the randomly generated initial parameters.

9. The parameter optimization system for the flexible interconnected phase shifting device according to claim 6, wherein: The upper-layer optimization model and the lower-layer optimization model are both used to obtain an objective function corresponding to the current network state through power flow calculation, and the objective function is used as the target result, wherein the power flow calculation process includes: Under the current randomly generated initial parameter conditions, the topological structure information, component parameter information, power generation information and load parameter information of the current medium-voltage distribution network are processed based on the selected power flow calculation method to obtain the steady-state operating status parameters of the current medium-voltage distribution network.

10. The parameter optimization system for the flexible interconnected phase shifting device according to claim 6, wherein: The upper-layer optimization model and the lower-layer optimization model are both used to construct a sampling function using the TPE strategy, and to obtain at least a maximum improved observation value based on the sampling function.

Images