A method for optimizing the allocation of distributed resources in a flexible power distribution network

Abstract

The present invention relates to the technical field of power distribution optimization, and discloses a method for optimizing the allocation of distributed resources in an elastic distribution network. The method comprises: collecting historical power supply load data and performing load analysis to obtain power supply resource load prediction results, and constructing an elastic distribution network model. The network model consists of power supply nodes, energy storage nodes, and a distribution network. The network model takes minimizing resource scheduling losses and maximizing the distribution network resource balancing capability as objective functions, and optimizes and solves the constructed elastic distribution network model to obtain an optimal scheduling strategy. The scheduling strategy determines the allocation of power resources between different power supply nodes and energy storage nodes in the distribution network model, and performs elastic distribution network resource scheduling according to the optimal scheduling strategy obtained by solving the solution to achieve power network resource optimization.

Description

Technical Field

[0001] The present invention relates to the technical field of power distribution optimization, and in particular to a method for optimizing the allocation of distributed resources in a flexible power distribution network. Background Art

[0002] With the continuous growth of global energy demand and the enhancement of environmental protection awareness, the power system is undergoing profound changes. Due to its single power structure and fixed power transmission path, the traditional centralized power system is difficult to adapt to the modern society's requirements for high reliability, high efficiency and high flexibility of power supply. At the same time, the rapid development of renewable energy (such as solar energy, wind energy, etc.) and the widespread application of distributed energy resources have made the operating environment of the power system more complex and uncertain. In this context, the elastic distribution network, as an emerging power system architecture, has gradually attracted widespread attention. The elastic distribution network improves the adaptability and risk resistance of the power system through flexible resource scheduling and optimal allocation. Although the elastic distribution network has broad application prospects, with the expansion of network scale and the increase of system complexity, the computational complexity of resource allocation problems has increased significantly, and traditional centralized optimization methods are difficult to operate efficiently in large-scale networks. Summary of the invention

[0003] In view of this, the present invention proposes a method for optimizing the allocation of distributed resources in a flexible power distribution network, which realizes real-time monitoring and dynamic allocation of power resources through intelligent scheduling.

[0004] To achieve the above object, the present invention provides a method for optimizing the allocation of distributed resources in a flexible power distribution network, comprising the following steps:

[0005] S1: Collect historical power supply load data and perform load analysis to obtain power supply resource load forecast results, wherein a long short-term memory network is an implementation method of the load analysis;

[0006] S2: Construct a flexible distribution network model, which consists of power supply nodes, energy storage nodes and distribution network. The network model takes minimizing resource scheduling loss and maximizing distribution network resource balancing ability as objective functions, and multi-dimensional resource scheduling constraints as restriction conditions of the model;

[0007] S3: Optimizing and solving the constructed elastic distribution network model to obtain an optimal scheduling strategy, wherein the scheduling strategy determines the allocation of power resources between different power supply nodes and energy storage nodes in the distribution network model, wherein distributed parallel optimization is an implementation method for optimizing and solving the elastic distribution network model;

[0008] S4: Flexible distribution network resource scheduling is performed according to the optimal scheduling strategy obtained to achieve power network resource optimization.

[0009] As a further improvement method of the present invention:

[0010] Optionally, collecting historical power supply load data and performing load analysis in step S1 includes:

[0011] S11: Collect historical power supply load data, where the data includes a timestamp and a corresponding load value.

[0012] S12: preprocessing the collected data to obtain preprocessed load data, wherein the preprocessing includes normalization and missing value filling;

[0013] S13: Based on the pre-processed load data, load analysis is performed to predict the power supply load. The calculation formula is:

[0014] o t =σ(W o ·[h t-1 ,u t ]+b o )

[0015] in:

[0016] u t Represents the load data at time t;

[0017] h t-1 represents the hidden state at time t-1;

[0018] W o and b o Represent the model weight parameters and bias parameters respectively;

[0019] σ represents the activation function;

[0020] o t Indicates the predicted power supply load.

[0021] Optionally, the network model in step S2 takes minimizing resource scheduling loss and maximizing distribution network resource balancing capability as objective functions, including:

[0022] S21: Construct a function to minimize resource scheduling loss, where the resource scheduling loss includes power transmission loss and energy storage charging and discharging loss, and the calculation formula is:

[0023]

[0024] in:

[0025] P ij represents the power transfer variable from power supply node i to energy storage node j;

[0026] R ijrepresents the line loss rate from power supply node i to energy storage node j;

[0027] E j represents the charging efficiency of energy storage node j;

[0028] E i represents the discharge efficiency of power supply node i;

[0029] L represents the set of distribution network lines;

[0030] S22: Construct a function to maximize the distribution network resource balancing capability evaluation function, which is measured by the load balance between energy storage nodes. The calculation formula is:

[0031]

[0032] in:

[0033] L j represents the load of the jth energy storage node;

[0034] N represents the number of energy storage nodes;

[0035] Represents the average load of all energy storage nodes;

[0036] S23: The constructed resource scheduling loss minimization function and the distribution network resource balancing capacity maximization evaluation function are weighted to obtain a comprehensive objective function, and the calculation formula is:

[0037]

[0038] in:

[0039] α and β represent weight coefficients, which are used to balance the importance of the two objective functions.

[0040] Optionally, the multidimensional resource scheduling constraints in step S2 are restriction conditions of the model, including:

[0041] A21: Construct the power resource supply and demand balance constraint, the calculation formula is:

[0042]

[0043] in:

[0044] I represents the power supply node set;

[0045] J represents the set of energy storage nodes;

[0046] G i,t represents the power generation of power supply node i at time t;

[0047] E j,trepresents the energy storage capacity of energy storage node j at time t;

[0048] L i,j,t represents the line loss from power supply node i to energy storage node j at time t;

[0049] T represents a set of time periods;

[0050] A22: Construct energy storage node state constraints, the calculation formula is:

[0051]

[0052] in:

[0053] E min and E max Represent the minimum and maximum energy states of the energy storage node respectively.

[0054] A23: Construct line current capacity constraint, the calculation formula is:

[0055]

[0056] in:

[0057] S i,j Represents the line current capacity from power supply node i to energy storage node j;

[0058] I i,j,t Represents the line current from power supply node i to energy storage node j at time t.

[0059] Optionally, in step S3, optimizing and solving the constructed elastic distribution network model to obtain an optimal dispatching strategy includes:

[0060] S31: Initialize the load L of each energy storage node in the elastic distribution network model j , the charging and discharging efficiency of energy storage nodes and power supply nodes;

[0061] S32: For each energy storage node, a local objective function is constructed through a set of adjacent nodes and optimized to obtain a local optimal scheduling strategy. The calculation formula of the local objective function is:

[0062]

[0063] in:

[0064] Ω j represents the set of adjacent power supply nodes of energy storage node j;

[0065] S33: All energy storage nodes exchange the load of each energy storage node and the calculated local optimal scheduling strategy information through central upload and distribution, and update the global average load

[0066] S34: Select the best local optimal dispatching strategy calculated by all energy storage nodes, and select the best strategy for global sharing;

[0067] S35: Repeat steps S32 to S34 until the objective function converges or reaches a preset number of iterations, and output the global optimal scheduling strategy.

[0068] Optionally, in step S32, a local objective function is constructed for each energy storage node through a set of adjacent nodes and optimized to obtain a local optimal scheduling strategy, including:

[0069] The objective function is optimized and solved using the adaptive interior point method. The specific process includes:

[0070] S32.1: Randomly initialize the scheduling strategy x0 and initialize the dual variables λ and v;

[0071] in:

[0072] λ represents the Lagrange multiplier of the energy storage node state constraint and the line current capacity constraint;

[0073] v represents the Lagrange multiplier of the power resource supply and demand balance constraint;

[0074] S32.2: Construct the Lagrangian function based on the local objective function. The calculation formula is:

[0075]

[0076] in:

[0077] f(x) represents the local objective function;

[0078] g i (x) represents the i-th inequality constraint, including the energy storage node state constraint and the line current capacity constraint;

[0079] m represents the number of inequality constraints, and its value is 2;

[0080] h j (x) represents the jth equality constraint, which is the power resource supply and demand balance constraint;

[0081] p is the number of equality constraints, which is 1;

[0082] S32.3: The KKT objective function is constructed by introducing slack variables. The calculation formula is:

[0083]

[0084] in:

[0085] Δ represents the gradient operation;

[0086] s i represents the i-th slack variable;

[0087] μ represents the preset relaxation parameter;

[0088] S32.4: Use the gradient quasi-Newton method to iteratively solve the KKT objective function and adaptively update the parameters. The calculation formula is:

[0089]

[0090] in:

[0091] x k represents the scheduling strategy for the kth iteration;

[0092] λ k represents the inequality constraint Lagrange multiplier for the kth iteration;

[0093] v k represents the equality constraint Lagrange multiplier for the kth iteration;

[0094] s k represents the slack variable for the kth iteration;

[0095] Δx, Δλ, Δv, and Δs represent the incremental solutions of the KKT objective function;

[0096] α represents the step size parameter;

[0097] S32.5: Iterate step S32.3 until the maximum number of iterations is reached, then stop the iteration and output the scheduling strategy as the local optimal scheduling strategy.

[0098] Optionally, the step S4 performs flexible distribution network resource scheduling according to the optimal scheduling strategy obtained by solving to achieve power network resource optimization.

[0099] In order to solve the above problem, the present invention provides an electronic device, the electronic device comprising:

[0100] A memory storing at least one instruction;

[0101] Communication interface, enabling electronic equipment to communicate; and

[0102] A processor executes instructions stored in the memory to implement the above-mentioned method for optimizing allocation of distributed resources in a flexible power distribution network.

[0103] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned method for optimizing allocation of distributed resources in a flexible distribution network.

[0104] Compared with the prior art, the present invention proposes a method for optimizing the allocation of distributed resources in a flexible power distribution network, which has the following advantages:

[0105] First, this scheme proposes a bone edge extraction method, which uses a multi-slice spiral CT machine to collect bone slice images at different bone positions, and performs first-order smoothing and sharpening processing and Laplace sharpening processing combined with gradient information on the bone slice images in turn to remove noise information in the bone slice images; combining the pixel values ​​of the neighboring pixels in the image pixels, an adaptive threshold for each pixel is generated to achieve adaptive binarization processing, and a morphological processing method is used to filter isolated pixels in the binary image and perform bone contour refinement processing to obtain a bone area image representing the bone edge.

[0106] At the same time, this scheme uses a multi-directional truncation method to calculate the bone gap and thickness of bones in different positions, selects the two directions that pass through the bone edge the most and the least as the base directions, constructs a function to characterize the bone elastic tensor, and extracts the elastic modulus that characterizes the bone's ability to resist deformation from the elastic tensor, realizes the extraction of bone parameter information based on multi-slice spiral CT images, and uses a deep fracture prediction network model to extract bone position information and bone parameter characteristics, and predict fracture risks for different bone positions.

[0107] (1) This scheme proposes a distribution network resource balancing method. In this method, the elastic distribution network model includes power supply nodes, energy storage nodes and distribution network. The objective function is to minimize resource scheduling losses and maximize the distribution network resource balancing capacity. The introduction of energy storage technology is comprehensively considered, which can more flexibly adjust power resources and improve the resource utilization efficiency and flexibility of the distribution network.

[0108] (2) This solution uses a distributed parallel optimization method to solve the elastic distribution network model, which can fully utilize computing resources, improve solution efficiency, and significantly shorten the solution time while ensuring solution accuracy;

[0109] (3) This solution achieves real-time response and adjustment to dynamically changing power supply loads and conditions, can maintain efficient operation under real-time changing market and environmental conditions, and can continuously optimize resource utilization and power supply quality;

[0110] (4) This solution comprehensively utilizes advanced data analysis technology, flexible distribution network modeling methods, distributed parallel optimization algorithms, and real-time dynamic optimization strategies to achieve more efficient, flexible and intelligent power distribution network management and resource optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0111] Figure 1 A schematic diagram of a flow chart of a method for optimizing allocation of distributed resources in a flexible power distribution network provided by an embodiment of the present invention;

[0112] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0113] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0114] An embodiment of the present application provides a method for optimizing the allocation of distributed resources in an elastic distribution network. The execution subject of the method for optimizing the allocation of distributed resources in an elastic distribution network includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for optimizing the allocation of distributed resources in an elastic distribution network can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0115] Embodiment 1:

[0116] S1: Collect historical power supply load data and perform load analysis to obtain power supply resource load forecast results, wherein the long short-term memory network is an implementation method of the load analysis.

[0117] The step S1 collects historical power supply load data and performs load analysis, including:

[0118] S11: Collect historical power supply load data, where the data includes a timestamp and a corresponding load value.

[0119] S12: preprocessing the collected data to obtain preprocessed load data, wherein the preprocessing includes normalization and missing value filling;

[0120] S13: Based on the pre-processed load data, load analysis is performed to predict the power supply load. The calculation formula is:

[0121] o t =σ(W o ·[h t-1 ,u t ]+bo )

[0122] in:

[0123] u t Represents the load data at time t;

[0124] h t-1 represents the hidden state at time t-1;

[0125] W o and b o Represent the model weight parameters and bias parameters respectively;

[0126] σ represents the activation function;

[0127] o t Indicates the predicted power supply load.

[0128] S2: Construct a flexible distribution network model, which consists of power supply nodes, energy storage nodes and distribution networks. The network model takes minimizing resource scheduling losses and maximizing the distribution network resource balancing capability as objective functions, and multi-dimensional resource scheduling constraints as restriction conditions of the model.

[0129] The network model in step S2 takes minimizing resource scheduling loss and maximizing distribution network resource balancing capability as the objective function, including:

[0130] S21: Construct a function to minimize resource scheduling loss, where the resource scheduling loss includes power transmission loss and energy storage charging and discharging loss, and the calculation formula is:

[0131]

[0132] in:

[0133] P ij represents the power transfer variable from power supply node i to energy storage node j;

[0134] R ij represents the line loss rate from power supply node i to energy storage node j;

[0135] E j represents the charging efficiency of energy storage node j;

[0136] E i represents the discharge efficiency of power supply node i;

[0137] L represents the set of distribution network lines;

[0138] S22: Construct a function to maximize the distribution network resource balancing capability evaluation function, which is measured by the load balance between energy storage nodes. The calculation formula is:

[0139]

[0140] in:

[0141] L j represents the load of the jth energy storage node;

[0142] N represents the number of energy storage nodes;

[0143] Represents the average load of all energy storage nodes;

[0144] S23: The constructed resource scheduling loss minimization function and the distribution network resource balancing capacity maximization evaluation function are weighted to obtain a comprehensive objective function, and the calculation formula is:

[0145]

[0146] in:

[0147] α and β represent weight coefficients, which are used to balance the importance of the two objective functions.

[0148] The multi-dimensional resource scheduling constraints in step S2 are the limiting conditions of the model, including:

[0149] A21: Construct the power resource supply and demand balance constraint, the calculation formula is:

[0150]

[0151] in:

[0152] I represents the power supply node set;

[0153] J represents the set of energy storage nodes;

[0154] G i,t represents the power generation of power supply node i at time t;

[0155] E j,t represents the energy storage capacity of energy storage node j at time t;

[0156] L i,j,t represents the line loss from power supply node i to energy storage node j at time t;

[0157] T represents a set of time periods;

[0158] A22: Construct energy storage node state constraints, the calculation formula is:

[0159]

[0160] in:

[0161] E minand E max Represent the minimum and maximum energy states of the energy storage node respectively.

[0162] A23: Construct line current capacity constraint, the calculation formula is:

[0163]

[0164] in:

[0165] S i,j Represents the line current capacity from power supply node i to energy storage node j;

[0166] I i,j,t Represents the line current from power supply node i to energy storage node j at time t.

[0167] S3: Optimize and solve the constructed elastic distribution network model to obtain the optimal scheduling strategy, which determines the allocation of power resources between different power supply nodes and energy storage nodes in the distribution network model, wherein distributed parallel optimization is an implementation method for optimizing and solving the elastic distribution network model.

[0168] In the step S3, the constructed elastic distribution network model is optimized and solved to obtain the optimal dispatching strategy, including:

[0169] S31: Initialize the load L of each energy storage node in the elastic distribution network model j , the charging and discharging efficiency of energy storage nodes and power supply nodes;

[0170] S32: For each energy storage node, a local objective function is constructed through a set of adjacent nodes and optimized to obtain a local optimal scheduling strategy. The calculation formula of the local objective function is:

[0171]

[0172] in:

[0173] Ω j represents the set of adjacent power supply nodes of energy storage node j;

[0174] S33: All energy storage nodes exchange the load of each energy storage node and the calculated local optimal scheduling strategy information through central upload and distribution, and update the global average load

[0175] S34: Select the best local optimal dispatching strategy calculated by all energy storage nodes, and select the best strategy for global sharing;

[0176] S35: Repeat steps S32 to S34 until the objective function converges or reaches a preset number of iterations, and output the global optimal scheduling strategy.

[0177] In the step S32, a local objective function is constructed for each energy storage node through a set of adjacent nodes and optimized to obtain a local optimal scheduling strategy, including:

[0178] The objective function is optimized and solved using the adaptive interior point method. The specific process includes:

[0179] S32.1: Randomly initialize the scheduling strategy x0 and initialize the dual variables λ and v;

[0180] in:

[0181] λ represents the Lagrange multiplier of the energy storage node state constraint and the line current capacity constraint;

[0182] v represents the Lagrange multiplier of the power resource supply and demand balance constraint;

[0183] S32.2: Construct the Lagrangian function based on the local objective function. The calculation formula is:

[0184]

[0185] in:

[0186] f(x) represents the local objective function;

[0187] g i (x) represents the i-th inequality constraint, including the energy storage node state constraint and the line current capacity constraint;

[0188] m represents the number of inequality constraints, and its value is 2;

[0189] h j (x) represents the jth equality constraint, which is the power resource supply and demand balance constraint;

[0190] p is the number of equality constraints, which is 1;

[0191] S32.3: The KKT objective function is constructed by introducing slack variables. The calculation formula is:

[0192]

[0193] in:

[0194] represents the gradient operation;

[0195] s i represents the i-th slack variable;

[0196] μ represents the preset relaxation parameter;

[0197] S32.4: Use the gradient quasi-Newton method to iteratively solve the KKT objective function and adaptively update the parameters. The calculation formula is:

[0198]

[0199] in:

[0200] x k represents the scheduling strategy for the kth iteration;

[0201] λ k represents the inequality constraint Lagrange multiplier for the kth iteration;

[0202] ν k represents the equality constraint Lagrange multiplier for the kth iteration;

[0203] s k represents the slack variable for the kth iteration;

[0204] Δx, Δλ, Δν, and Δs represent the incremental solutions of the KKT objective function;

[0205] α represents the step size parameter;

[0206] S32.5: Iterate step S32.3 until the maximum number of iterations is reached, then stop the iteration and output the scheduling strategy as the local optimal scheduling strategy.

[0207] S4: Flexible distribution network resource scheduling is performed according to the optimal scheduling strategy obtained to achieve power network resource optimization.

[0208] It should be noted that the serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. And the terms "including", "comprising" or any other variants thereof in this article are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "including a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0209] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0210] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for optimizing the allocation of distributed resources in a flexible power distribution network, characterized in that: The method comprises: S1: Collect historical power supply load data and perform load analysis to obtain power supply resource load forecast results, wherein a long short-term memory network is an implementation method of the load analysis; S2: Construct a flexible distribution network model, which consists of power supply nodes, energy storage nodes and distribution network. The network model takes minimizing resource scheduling loss and maximizing distribution network resource balancing ability as local objective functions, and multi-dimensional resource scheduling constraints as restriction conditions of the model; S3: Optimizing and solving the constructed elastic distribution network model to obtain an optimal scheduling strategy, wherein the scheduling strategy determines the allocation of power resources between different power supply nodes and energy storage nodes in the distribution network model, wherein distributed parallel optimization is an implementation method for optimizing and solving the elastic distribution network model; The constructed elastic distribution network model is optimized and solved to obtain the optimal dispatching strategy, including: S31: Initialize the load of each energy storage node in the elastic distribution network model , the charging and discharging efficiency of energy storage nodes and power supply nodes; S32: constructing a local objective function for each energy storage node through a set of adjacent nodes and optimizing it to obtain a local optimal scheduling strategy; S33: All energy storage nodes exchange the load of each energy storage node and the calculated local optimal scheduling strategy information through central upload and distribution, and update the global average load ; S34: Select the best local optimal dispatching strategy calculated by all energy storage nodes, and select the best strategy for global sharing; S35: Repeat steps S32 to S34 until the objective function converges or reaches a preset number of iterations, and output a global optimal scheduling strategy; The method of constructing a local objective function for each energy storage node through a set of adjacent nodes and optimizing the local optimal scheduling strategy includes: The objective function is optimized and solved using the adaptive interior point method. The specific process includes: S32.1: Random Initialization Scheduling Strategy , and initialize the dual variable and ; in: Lagrange multipliers representing energy storage node state constraints and line current capacity constraints; Lagrange multipliers representing the constraints on the balance between supply and demand of power resources; S32.2: Construct the Lagrangian function based on the local objective function. The calculation formula is: ； in: represents the local objective function; represents the i-th inequality constraint, including energy storage node state constraint and line current capacity constraint; m represents the number of inequality constraints; represents the jth equality constraint, which is the power resource supply and demand balance constraint; p is the number of equality constraints; S32.3: The KKT objective function is constructed by introducing slack variables. The calculation formula is: ； in: represents the gradient operation; represents the i-th slack variable; Represents the preset relaxation parameters; S32.4: Use the gradient quasi-Newton method to iteratively solve the KKT objective function and adaptively update the parameters. The calculation formula is: ； in: represents the scheduling strategy for the kth iteration; represents the inequality constraint Lagrange multiplier for the kth iteration; represents the equality constraint Lagrange multiplier for the kth iteration; represents the slack variable for the kth iteration; , , , They represent the incremental solutions of the KKT objective function respectively; represents the step size parameter; S32.5: Iterate step S32.3 until the maximum number of iterations is reached, then stop the iteration and output the scheduling strategy as the local optimal scheduling strategy; S4: Flexible distribution network resource scheduling is performed according to the optimal scheduling strategy obtained to achieve power network resource optimization.

2. A method for optimizing the allocation of distributed resources in a flexible power distribution network according to claim 1, characterized in that: The step S1 collects historical power supply load data and performs load analysis, including: S11: Collect historical power supply load data, where the data includes a timestamp and a corresponding load value; S12: preprocessing the collected data to obtain preprocessed load data, wherein the preprocessing includes normalization and missing value filling; S13: Load analysis is performed based on the preprocessed load data to predict the power supply load. The calculation formula is: ； in: Represents the load data at time t; represents the hidden state at time t-1; and Represent the model weight parameters and bias parameters respectively; represents the activation function; Indicates the predicted power supply load.

3. The method for optimizing the allocation of distributed resources in a flexible power distribution network according to claim 1, characterized in that: The network model in step S2 takes minimizing resource scheduling loss and maximizing distribution network resource balancing capability as the objective function, including: S21: Construct a function to minimize resource scheduling loss, where the resource scheduling loss includes power transmission loss and energy storage charging and discharging loss, and the calculation formula is: ； in: represents the power transfer variable from power supply node i to energy storage node j; represents the line loss rate from power supply node i to energy storage node j; represents the charging efficiency of energy storage node j; represents the discharge efficiency of power supply node i; Represents a collection of distribution network lines; S22: Construct a function to maximize the distribution network resource balancing capability evaluation function, which is measured by the load balance between energy storage nodes, and the calculation formula is: ； in: represents the load of the jth energy storage node; N represents the number of energy storage nodes; Represents the average load of all energy storage nodes; S23: The constructed resource scheduling loss minimization function and the distribution network resource balancing capacity maximization evaluation function are weighted to obtain a comprehensive objective function, and the calculation formula is: ； in: and Represents the weight coefficient, which is used to balance the importance of the two objective functions.

4. A method for optimizing the allocation of distributed resources in a flexible power distribution network as claimed in claim 3, characterized in that: The multi-dimensional resource scheduling constraints in step S2 are the limiting conditions of the model, including: A21: Construct the power resource supply and demand balance constraint, the calculation formula is: ； in: Represents a collection of power supply nodes; Represents a collection of energy storage nodes; represents the power generation of power supply node i at time t; represents the energy storage capacity of energy storage node j at time t; represents the line loss from power supply node i to energy storage node j at time t; T represents a set of time periods; A22: Construct energy storage node state constraints, the calculation formula is: ； in: and Respectively represent the minimum and maximum energy states of the energy storage node; A23: Construct line current capacity constraint, the calculation formula is: ； in: Represents the line current capacity from power supply node i to energy storage node j; Represents the line current from power supply node i to energy storage node j at time t.

Images