Power distribution network architecture evolution planning method, system, equipment and medium

Abstract

The invention discloses a power distribution network architecture evolution planning method, system, equipment and medium, and the method comprises the steps: obtaining power distribution network basic data and social power demand driving factors, and obtaining a power distribution network basic data set through data preprocessing; obtaining a typical development scene set through uncertainty modeling and multi-scene generation and reduction; in combination with system operation constraints, site selection and sizing of the multi-port flexible interconnection device are optimized, and a preliminary site selection and sizing scheme is obtained; a multi-stage investment decision model is constructed and solved, and a preliminary staged construction path is obtained; calculating a power supply reliability index through a fault mode analysis method, and evaluating economy in combination with a cost model to obtain a quantitative evaluation result; and performing global sensitivity analysis through a Morse algorithm, and adjusting an optimization strategy to obtain a final power distribution network architecture evolution planning scheme. The flexibility and the reliability of the power distribution network are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of power distribution network planning technology, and in particular to a method, system, equipment and medium for planning the evolution of power distribution network architecture. Background Technology

[0002] With the large-scale integration of distributed renewable energy, energy storage systems, and electric vehicle charging facilities, traditional distribution networks face severe challenges. Existing distribution network planning methods suffer from the following main technical deficiencies: Traditional distribution network planning methods, based on static load growth forecasting, are ill-suited to the dynamic integration of distributed resources and bidirectional power flow characteristics. Existing planning methods typically employ load forecasting at a single time point, neglecting the spatiotemporal distribution characteristics of intermittent power sources such as photovoltaics and wind power. This leads to overload or low equipment utilization issues in actual operation of the planned solutions.

[0003] Existing flexible interconnection technologies mainly focus on two-port or three-port devices, which have limited adaptability to complex distribution network topologies. Traditional UPQC and DVR devices have relatively simple functions and cannot simultaneously meet the requirements for power regulation, power quality management, and fault isolation among multiple feeders. Although existing soft-switching devices have a certain degree of flexible regulation capability, they lack systematic solutions for multi-port coordinated control.

[0004] In terms of distribution network evolution planning, existing methods mainly focus on equipment capacity expansion and line upgrades, lacking a systematic consideration of the gradual evolution of the network structure. Existing planning methods often adopt one-time investment decisions, which are difficult to adapt to the realities of rapid technological development and load uncertainty, resulting in low investment efficiency and the risk of technology route lock-in. Summary of the Invention

[0005] In view of the aforementioned existing problems, the present invention is proposed.

[0006] Therefore, this invention provides a distribution network architecture evolution planning method, system, equipment, and medium to address the problems that existing distribution network planning methods are unable to adapt to the dynamic changes in operating status and uncertainties in multiple scenarios brought about by distributed resource access, and lack the systematic planning capability for multi-terminal collaborative control of flexible interconnection devices and the progressive evolution of the network structure.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for planning the evolution of a power distribution network architecture, comprising: Acquire basic data of the power distribution network and driving factors of social electricity demand, and obtain the basic dataset of the power distribution network through data preprocessing; Based on the basic dataset, a set of typical development scenarios is obtained through uncertainty modeling and multi-scenario generation and reduction; Based on a set of typical development scenarios and combined with system operation constraints, the site selection and capacity determination of multi-port flexible interconnection devices are optimized to obtain a preliminary site selection and capacity determination scheme. Based on the preliminary site selection and capacity determination plan, a multi-stage investment decision model was constructed and solved to obtain a preliminary phased construction path; Based on the initial phased construction path, the power supply reliability index is calculated by the failure mode analysis method and the economic efficiency is evaluated by combining the cost model to obtain quantitative evaluation results. Based on the quantitative evaluation results, global sensitivity analysis is performed using the Morris algorithm, and optimization strategies are adjusted to obtain the final power distribution network architecture evolution plan.

[0008] As a preferred embodiment of the distribution network architecture evolution planning method described in this invention, the method for obtaining the basic distribution network dataset includes: Based on the basic data of the distribution network and the driving factors of social electricity demand, key model parameters related to the probability distribution of photovoltaic output, load growth trend and energy storage operation characteristics are identified, and a set of key model parameters is formed. Based on the key model parameter set, quality verification and outlier processing are performed on the corresponding data fields in the multi-source original datasets. Missing values ​​and erroneous records are identified and eliminated. In accordance with the input format required by the model parameters, the data from different sources are normalized in terms of dimensions and aligned in terms of time scale to form a consistent intermediate dataset. Based on a consistent intermediate dataset, a multi-dimensional structured database integrating the power grid physical architecture, equipment operating characteristics, load change patterns, and external development driving factors was established. Based on a multi-dimensional structured database, all processed structured data and calibration parameters are integrated to obtain the basic dataset of the power distribution network.

[0009] As a preferred embodiment of the power distribution network architecture evolution planning method described in this invention, the typical development scenario set obtained includes: Based on the basic dataset of the power distribution network, a joint output model of photovoltaic, energy storage and electric vehicle charging power is constructed; Based on the joint output model, combined with the probability distribution characteristics of photovoltaic output and the temporal evolution law of load growth, an initial random sample set is generated, and a stratified sampling method is used to generate an initial scenario set that conforms to the actual distribution characteristics. Based on the initial scene set, the difference metric between each scene is calculated. Using the scene distance criterion with optimal transmission cost as the core, the similarity between scenes is quantified, redundant scene paths are identified and merged, and a filtered candidate scene set is generated. The candidate scenario set is clustered and divided, and the most representative typical scenarios in each cluster are retained to obtain the typical development scenario set.

[0010] As a preferred embodiment of the distribution network architecture evolution planning method described in this invention, the preliminary site selection and capacity determination scheme includes: Based on the distribution network topology and load distribution characteristics, combined with network connectivity, critical load density and potential fault risk level, a set of candidate installation locations for multi-port flexible interconnection devices is determined. Based on the candidate location set, a device operation model is constructed that includes active and reactive power exchange capabilities and apparent power capacity constraints. The charging and discharging dynamic characteristics of the energy storage system are integrated to form a device-energy storage collaborative operation model. Based on the device-energy storage collaborative operation model and combined with the distribution network operation constraints, an optimization model is constructed with the goal of minimizing the overall cost, taking into account investment cost, network loss cost and reliability improvement benefits; The fault transfer capability model of the device to the power loss load is introduced into the optimization model to evaluate the load recovery capability of the multi-port flexible interconnection device under the condition of line fault and equipment fault, and incorporate it as a hard constraint into the optimization objective to generate an optimization model with integrated reliability constraints. The fault transfer capability model of the device to the power loss load is introduced into the optimization model to evaluate the load recovery capability of the multi-port flexible interconnection device under line faults and equipment faults. This model is then incorporated as a hard constraint in the solution to obtain a preliminary site selection and capacity determination scheme.

[0011] As a preferred embodiment of the power distribution network architecture evolution planning method described in this invention, the step of obtaining the preliminary phased construction path includes: Based on a set of typical development scenarios, a multi-stage progressive planning strategy is adopted to divide the power grid development into multiple planning stages, determine the time span and boundary conditions of each stage, and form a multi-stage planning framework. Based on the planning framework, and combined with the load level and network status at each stage, a distribution network operation model is established, and system operation constraints are constructed. Based on the preliminary site selection and capacity determination scheme, the commissioning time series variables of the equipment at different stages are constructed, and time-series coupling constraints are applied to ensure that the upper limit of the single-stage installation capacity and the cumulative construction capacity do not exceed the rated capacity. By integrating investment costs, network loss costs, and reliability benefits at each stage, an objective function is constructed with the goal of minimizing the overall cost throughout the entire lifecycle. By combining the dynamic evolution constraint of the device's fault recovery capability as a hard constraint with the objective function, a multi-stage mixed integer optimization problem is formed and solved, resulting in a preliminary phased construction path.

[0012] The beneficial effects of this preferred technical solution are that by constructing a full life cycle optimization model that integrates multi-stage progressive planning, time-series coupling constraints, and dynamic evolution of fault recovery capabilities, the synergistic optimization of the power distribution network construction path in terms of technical feasibility, economic efficiency, and power supply reliability is achieved, effectively avoiding resource waste and technology lock-in risks caused by premature or delayed investment.

[0013] As a preferred embodiment of the power distribution network architecture evolution planning method described in this invention, the step of obtaining the quantitative evaluation result includes: Based on the load growth trend and distributed power supply access timing characteristics represented by typical development scenario sets, the evaluation boundary conditions and operating parameters for each planning stage are determined. Based on the determined evaluation boundary conditions and operating parameters, combined with the access location, operating status and distribution network topology of the multi-port flexible interconnection device at each stage, the power loss load nodes within the device's power supply control range after line faults and equipment faults are identified. Based on the power loss load node, the maximum power value that can be transferred to the load at each stage under the support of the equipment is quantified using the load recovery capability model after the fault. The load reduction and power restoration duration during the fault are calculated to obtain the reliability improvement index at each stage. Based on reliability improvement indicators, by comparing the load reduction and power outage duration in scenarios with and without device support, the load reduction and power outage duration reduction at each stage are calculated. Combining the unit power loss cost and the user power outage loss function, the economic loss of power outage is calculated and a monetized representation is formed. Based on monetization, a net benefit assessment model for the entire life cycle is constructed by integrating the investment cost of the integrated equipment, operation and maintenance costs, network loss savings and reliability improvement benefits, to support dynamic decision optimization and obtain quantitative assessment results.

[0014] As a preferred embodiment of the distribution network architecture evolution planning method described in this invention, the step of obtaining the final distribution network architecture evolution planning scheme includes: Based on the initial phased construction path, perturbations are applied to key uncertainty parameters, including load growth rate, unit investment cost of equipment, and output fluctuation range of distributed power sources, forming multiple sets of differentiated parameter scenarios. Based on multiple sets of differentiated parameter scenarios, the phased construction scheme is re-solved for each set of scenarios, the construction sequence of each candidate node under different scenarios is recorded, and its construction frequency and capacity distribution are statistically analyzed to identify the set of key nodes with high construction probability. Based on the construction sequence of each candidate node under different scenarios, the degree of deviation of the construction sequence of each candidate node with parameter disturbance is evaluated, and the stability level of construction timing is quantified. Based on the set of key nodes, and by combining the key node identification results with the stability index of construction timing, multi-scenario analysis and robust optimization algorithms are used to select the device access location and capacity sequence, thus forming the final distribution network architecture evolution planning scheme.

[0015] The beneficial effects of this preferred technical solution are that, by taking the initial construction path as a benchmark and combining multi-scenario disturbance simulation with construction frequency and time sequence stability analysis, multi-scenario analysis and robust optimization algorithms are used to identify key access nodes and capacity sequences with high probability and low fluctuations. This significantly improves the adaptability of the final planning scheme in multiple uncertain environments and the robustness of investment decisions, and effectively reduces planning risks caused by prediction deviations.

[0016] Secondly, the present invention provides a power distribution network architecture evolution planning system, comprising: The basic data integration module is used to acquire basic data of the distribution network and driving factors of social electricity demand. Through data preprocessing, the basic data set of the distribution network is obtained. The typical scenario construction module is used to obtain a set of typical development scenarios based on a basic dataset through uncertainty modeling and multi-scenario generation and reduction. The preliminary planning and decision-making module is used to optimize the site selection and capacity determination of multi-port flexible interconnection devices based on typical development scenario sets and system operation constraints, and to obtain a preliminary site selection and capacity determination scheme. The multi-stage investment decision module is used to construct and solve a multi-stage investment decision model based on the preliminary site selection and capacity determination plan, so as to obtain the preliminary phased construction path; The life-cycle benefit assessment module is used to calculate power supply reliability indicators based on the initial phased construction path, and to evaluate the economic efficiency by combining the failure mode analysis method with the cost model, so as to obtain quantitative assessment results. The global sensitivity feedback module is used to perform global sensitivity analysis based on the quantitative evaluation results using the Morris algorithm, adjust the optimization strategy, and obtain the final distribution network architecture evolution plan.

[0017] Thirdly, the present invention provides an electronic device, comprising: Memory, used to store programs; A processor is configured to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the power distribution network architecture evolution planning method.

[0018] Fourthly, the present invention provides a computer-readable storage medium, comprising: when the program is executed by a processor, the steps of implementing the power distribution network architecture evolution planning method.

[0019] The beneficial effects of this invention are as follows: By constructing a multi-source data fusion and modeling module, this invention integrates basic data of the power distribution network and driving factors of social electricity demand, and performs preprocessing to achieve systematic organization of planning input data and accurate characterization of uncertainty driving factors, providing a reliable data foundation for subsequent scenario generation; through the uncertainty modeling and scenario generation module, based on load growth trends and distributed power source access timing characteristics, it generates and reduces typical development scenario sets, achieving representative coverage of multiple possible future development paths and effectively characterizing multiple uncertainties in long-term planning; through the device location and capacity optimization module and the multi-stage investment decision module, combined with system operation constraints and full life cycle cost targets, it constructs and solves a multi-stage mixed integer optimization model with timing coupling constraints, realizing the transformation from static planning... The evolution towards a dynamic and gradual construction path significantly improves the economic efficiency and technological adaptability of investment timing. Through the full life-cycle benefit assessment module, which integrates failure mode analysis and cost-benefit models, the monetary value of improved power supply reliability and reduced power outage losses is quantified, achieving a unified assessment of technical performance and economic value, and enhancing the decision support capability of the planning scheme. Through the robustness optimization and scheme generation module, based on the initial phased construction path, multi-scenario disturbance analysis and quantitative methods for construction frequency and timing stability are introduced. Multi-scenario analysis and robust optimization algorithms are used to screen access schemes with high construction probability and stable timing, achieving strong robustness and low investment risk in the final planning results under complex and uncertain environments, and comprehensively improving the scientificity, foresight, and feasibility of the distribution network architecture evolution planning. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a basic flowchart of a power distribution network architecture evolution planning method provided in one embodiment of the present invention. Detailed Implementation

[0021] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0022] Example 1, referring to Figure 1As an embodiment of the present invention, a method for planning the evolution of a power distribution network architecture is provided, comprising: S100: Obtain basic data of the power distribution network and driving factors of social electricity demand, and obtain the basic dataset of the power distribution network through data preprocessing; S200: Based on the basic dataset, a set of typical development scenarios is obtained through uncertainty modeling and multi-scenario generation and reduction; S300: Based on a set of typical development scenarios and combined with system operation constraints, optimize the location and capacity of multi-port flexible interconnection devices to obtain a preliminary location and capacity scheme; S400: Based on the preliminary site selection and capacity determination scheme, a multi-stage investment decision model is constructed and solved to obtain the preliminary phased construction path; S500: Based on the initial phased construction path, the power supply reliability index is calculated by the failure mode analysis method and the economic efficiency is evaluated by combining the cost model to obtain the quantitative evaluation results. S600: Based on the quantitative evaluation results, global sensitivity analysis is performed using the Morris algorithm, and optimization strategies are adjusted to obtain the final power distribution network architecture evolution plan.

[0023] It should be noted that existing distribution network planning methods face multiple severe challenges during operation. Traditional planning methods, largely based on static load forecasting and deterministic scenarios, struggle to adapt to the strong randomness and spatiotemporal fluctuations of intermittent power sources such as distributed photovoltaic and wind power, leading to problems like localized overloads, voltage exceeding limits, or low equipment utilization during actual operation. With the large-scale integration of new loads and power sources such as electric vehicles and energy storage systems, distribution network power flow has shifted from unidirectional to bidirectional, making the existing radial structure and protection configurations ill-suited to handle complex power interactions and fault propagation risks. Existing planning often employs a one-time investment decision-making model, lacking dynamic response capabilities to the uncertainties of technological evolution and load development, easily resulting in premature investment leading to capital tied up or delayed investment causing grid bottlenecks. Furthermore, traditional methods emphasize line expansion and substation capacity enhancement, neglecting the coordinated optimization of new regulation methods such as flexible interconnection devices, making it difficult to simultaneously address multiple objectives such as improved power supply reliability, enhanced power quality, and rapid fault recovery. The overall planning scheme exhibits poor adaptability and weak robustness in complex and ever-changing operating environments, failing to support the high-quality development needs of modern distribution networks.

[0024] Therefore, addressing the aforementioned issues that existing distribution network planning methods struggle to adapt to the dynamic changes in operational status and uncertainties across multiple scenarios brought about by distributed resource access, and lack systematic planning capabilities for multi-terminal collaborative control of flexible interconnection devices and the progressive evolution of the network structure, this paper addresses these problems through steps S100-S600. By integrating multi-source data-driven approaches, multi-stage progressive planning, full life-cycle benefit assessment, and multi-scenario robust optimization algorithms, it achieves a synergistic improvement in the dynamic adaptability, economic rationality, and construction robustness of distribution network architecture evolution schemes under uncertain environments. This significantly enhances the scientific rigor, foresight, and feasibility of distribution network planning in the context of high-proportion distributed energy access.

[0025] Example 2, this is an embodiment of the present invention, which provides a distribution network architecture evolution planning method based on the previous embodiment, including: In this embodiment of the application, the distribution network basic dataset obtained in step S100 includes: Based on the basic data of the distribution network and the driving factors of social electricity demand, key model parameters related to the probability distribution of photovoltaic output, load growth trend and energy storage operation characteristics are identified, and a set of key model parameters is formed. Based on the key model parameter set, quality verification and outlier processing are performed on the corresponding data fields in the multi-source original datasets. Missing values ​​and erroneous records are identified and eliminated. In accordance with the input format required by the model parameters, the data from different sources are normalized in terms of dimensions and aligned in terms of time scale to form a consistent intermediate dataset. Based on a consistent intermediate dataset, a multi-dimensional structured database integrating the power grid physical architecture, equipment operating characteristics, load change patterns, and external development driving factors was established. Based on a multi-dimensional structured database, all processed structured data and calibration parameters are integrated to obtain the basic dataset of the power distribution network.

[0026] In this embodiment, based on basic data of the distribution network and driving factors of social electricity demand, key model parameters related to the probability distribution of photovoltaic power output, load growth trend, and energy storage operation characteristics are identified, forming a set of key model parameters; wherein, the randomness of photovoltaic power output is modeled using a beta distribution, and its probability density function is expressed as: (1) in, Let be the probability density function. This is the per-unit value for photovoltaic power output. , For Beta distribution parameters, This is the Gamma function.

[0027] The distribution parameter was obtained by fitting historical photovoltaic power output data and is used to characterize the power output fluctuation characteristics under different weather conditions; the load growth trend is predicted based on factors such as economic development, population growth, and the improvement of electrification level, and its expression is shown below: (2) in, For the load demand in year t, Based on the baseline annual load, , , These are the GDP growth rate, population growth rate, and electrification rate growth rate, respectively. This is a random disturbance term, reflecting the uncertainty of the prediction.

[0028] The operating characteristics of an energy storage system are described by the state-of-charge dynamic equation as follows: (3) In the formula, Let t be the state of charge of the stored energy. , These represent charge and discharge efficiencies, , These are the charging and discharging powers, For energy storage capacity, The time step is used to characterize the energy change pattern during the charging and discharging process of energy storage.

[0029] Based on the key model parameter set, quality verification and outlier processing are performed on the corresponding data fields in the multi-source original datasets. Missing values ​​and erroneous records are identified and eliminated. In accordance with the input format required by the model parameters, the data from different sources are normalized in terms of dimensions and aligned in terms of time scale to form a consistent intermediate dataset. Based on a consistent intermediate dataset, a multi-dimensional structured database integrating the power grid physical architecture, equipment operating characteristics, load variation patterns, and external development drivers was established. The spatial distribution of the load was modeled using a gravity model, as shown below: (4) In the formula, For the load attraction from region i to region j, , The load densities for regions i and j are respectively. Let k be the distance between regions, and k be a proportionality constant.

[0030] In this embodiment of the application, the typical development scenario set obtained in step S200 includes: Based on the basic dataset of the power distribution network, a joint output model of photovoltaic, energy storage and electric vehicle charging power is constructed; Based on the joint output model, combined with the probability distribution characteristics of photovoltaic output and the temporal evolution law of load growth, an initial random sample set is generated, and a stratified sampling method is used to generate an initial scenario set that conforms to the actual distribution characteristics. Based on the initial scene set, the difference metric between each scene is calculated. Using the scene distance criterion with optimal transmission cost as the core, the similarity between scenes is quantified, redundant scene paths are identified and merged, and a filtered candidate scene set is generated. The candidate scenario set is clustered and divided, and the most representative typical scenarios in each cluster are retained to obtain the typical development scenario set.

[0031] In this embodiment of the application, a joint output model of photovoltaic, energy storage and electric vehicle charging power is constructed based on the basic dataset of the distribution network. The total output is expressed as follows: (5) This model comprehensively considers the collaborative and competitive relationships among three types of distributed resources, supporting the generation of multiple scenarios.

[0032] Based on the joint output model, combined with the probability distribution characteristics of photovoltaic output (Formula (1)) and the temporal evolution law of load growth (Formula (2)), an initial random sample set is generated, and a stratified sampling method is used to generate an initial scenario set that conforms to the actual distribution characteristics. Based on the initial scene set, the difference measure between each scene is calculated. The scene distance criterion with optimal transmission cost as the core, namely Kantorovich distance, is used to quantify the similarity between scenes. Its definition is shown in formula (6): (6) In the formula, For probability measure and The distance between Kantorovich, For the set of coupling measures, Let be the cost function.

[0033] This distance criterion is combined with the cost function It measures the cost of transitions between scenes, identifies and merges redundant scene paths, and generates a filtered set of candidate scenes. The candidate scene set is clustered, and the most representative typical scenes in each cluster are retained to obtain the typical development scene set. The initial scene set is generated using the Latin hypercube sampling method, mathematically expressed as follows: (7) In the formula, Let be a vector of random variables in scene s. It is the inverse function of the cumulative distribution function. It is a uniformly distributed sample in the interval [0,1].

[0034] In this embodiment of the application, the generation of the scenario set in step S200 includes constructing a joint output model of photovoltaic, energy storage and electric vehicle charging power based on the distribution network basic dataset, generating an initial random scenario set by combining the load growth time series law and Latin hypercube sampling, then using the Kantorovich distance criterion to quantify the similarity between scenarios and merging redundant paths to form a candidate scenario set, and retaining representative scenarios of each cluster through cluster analysis to obtain a typical development scenario set covering multiple uncertainties.

[0035] In an optional implementation, the generation of the scenario set in step S200 can also be achieved by constructing a joint output model of photovoltaic, energy storage and electric vehicle charging power based on the distribution network basic dataset, generating a large number of initial random scenarios using the Monte Carlo method in combination with the load growth time series law, and then using clustering algorithms such as K-means to group the scenarios and select the centroid of each cluster as a representative scenario, finally forming a typical development scenario set covering the main uncertainty features.

[0036] In an optional implementation, the scenario set generated in step S200 can also be generated by training a conditional generative adversarial network (cGAN) using multi-source data such as historical photovoltaic power output, load data and electric vehicle charging behavior, so that the network learns the spatiotemporal correlation and joint probability distribution of distributed resources and loads, and then inputs random noise and scenario condition variables, so that the trained generator can directly output a typical development scenario set that conforms to the actual evolution law and has statistical representativeness.

[0037] In this embodiment of the application, step S300 yields a preliminary site selection and capacity determination scheme, including: Based on the distribution network topology and load distribution characteristics, combined with network connectivity, critical load density and potential fault risk level, a set of candidate installation locations for multi-port flexible interconnection devices is determined. Based on the candidate location set, a device operation model is constructed that includes active and reactive power exchange capabilities and apparent power capacity constraints. The charging and discharging dynamic characteristics of the energy storage system are integrated to form a device-energy storage collaborative operation model. Based on the device-energy storage collaborative operation model and combined with the distribution network operation constraints, an optimization model is constructed with the goal of minimizing the overall cost, taking into account investment cost, network loss cost and reliability improvement benefits; The fault transfer capability model of the device to the power loss load is introduced into the optimization model to evaluate the load recovery capability of the multi-port flexible interconnection device under the condition of line fault and equipment fault, and incorporate it as a hard constraint into the optimization objective to generate an optimization model with integrated reliability constraints. The fault transfer capability model of the device to the power loss load is introduced into the optimization model to evaluate the load recovery capability of the multi-port flexible interconnection device under line faults and equipment faults. This model is then incorporated as a hard constraint in the solution to obtain a preliminary site selection and capacity determination scheme.

[0038] In this embodiment of the application, a device operation model is constructed based on the candidate location set, which includes active and reactive power exchange capabilities and apparent power capacity constraints; wherein, the active and reactive power transmission characteristics of the multi-port flexible interconnection device are described by formulas (8) and (9), respectively: (8) (9) In the formula, Let be the active power from port i to port j. , Let i and j be the voltage amplitudes at ports i and j, respectively. , These are the real and imaginary parts of the admittance matrix, respectively. , The voltage phase angles at ports i and j are respectively. Let be the reactive power from port i to port j.

[0039] Its capacity constraint is determined by the apparent power limit, as shown in formula (10): (10) In the formula, The apparent power capacity limit for ports i to j.

[0040] The charging and discharging dynamic characteristics of the integrated energy storage system (Formula (3)) form a device-energy storage collaborative operation model; The fault transfer capability model of the device to the power outage load is introduced into the optimization model to evaluate the load recovery capability of the multi-port flexible interconnection device under the condition of line fault and equipment fault, and is included as a hard constraint in the optimization objective to generate an optimization model with integrated reliability constraints; wherein, the load transfer capability of the node is calculated by formula (11): (11) In the formula, Node i can transfer load. To meet load demand, Let be the transfer coefficient from flexible interconnect device j to node i.

[0041] In this embodiment of the application, step S400 obtains a preliminary phased construction path, including: Based on a set of typical development scenarios, a multi-stage progressive planning strategy is adopted to divide the power grid development into multiple planning stages, determine the time span and boundary conditions of each stage, and form a multi-stage planning framework. Based on the planning framework, and combined with the load level and network status at each stage, a distribution network operation model is established, and system operation constraints are constructed. Based on the preliminary site selection and capacity determination scheme, the commissioning time series variables of the equipment at different stages are constructed, and time-series coupling constraints are applied to ensure that the upper limit of the single-stage installation capacity and the cumulative construction capacity do not exceed the rated capacity. By integrating investment costs, network loss costs, and reliability benefits at each stage, an objective function is constructed with the goal of minimizing the overall cost throughout the entire life cycle. The objective function includes equipment investment expenditure, power loss costs, and power outage loss savings. By combining the dynamic evolution constraint of the device's fault recovery capability as a hard constraint with the objective function, a multi-stage mixed integer optimization problem is formed and solved, resulting in a preliminary phased construction path.

[0042] In this embodiment, based on the planning framework and combined with the load level and network status at each stage, a distribution network operation model is established, and system operation constraints are constructed; wherein, the network power flow is described by a branch power flow model, and the active and reactive power balance equations of the nodes are shown in formulas (12) and (13): (12) (13) In the formula, , Let E represent the active and reactive power injections at node i, respectively, and E be the set of branches. , These are the real and imaginary parts of the branch impedance, respectively. This represents the branch current.

[0043] The voltage amplitude variation relationship is described by formula (14): (14) And it meets the voltage safety boundary constraints: (15) In the formula, , These are the lower and upper limits of the node voltage, respectively.

[0044] In this embodiment, the investment cost, network loss cost, and reliability benefits at each stage are integrated to construct an objective function that aims to minimize the overall cost throughout the entire lifecycle. Its general form is shown in formula (16): (16) In the formula, Z represents the present value of the total investment cost, T represents the planning period, and r represents the discount rate. Let be the investment cost in year t. For operating costs, For network loss costs, To compensate for the cost of reliability.

[0045] Among them investment costs This includes equipment such as flexible interconnection devices, lines, and transformers, and its expression is shown in formula (17): (17) In the formula, These are flexible interconnection devices, lines, and transformer sets. These represent the unit capacity investment costs of the corresponding equipment. These refer to equipment capacity or length, respectively. These are variables for route investment decisions.

[0046] In this embodiment of the application, the multi-stage investment decision-making algorithm in step S400 includes constructing a multi-stage mixed integer optimization model with the goal of minimizing the comprehensive cost throughout the entire life cycle, combining the preliminary site selection and capacity determination scheme with the load level, network status and system operation constraints of each stage, using an improved dynamic programming algorithm to perform reverse recursive solution, and using the value function recursive formula and state transition equation to optimize the investment decision stage by stage, and finally obtain a technically feasible and economically optimal preliminary phased construction path.

[0047] In an optional implementation, the multi-stage investment decision-making algorithm in step S400 can also decouple the planning problem into two stages: the first stage determines irreversible investment decisions such as the location, capacity, and construction sequence of the multi-port flexible interconnection device; the second stage simulates the operating status under each scenario based on a set of typical development scenarios and adjusts adjustable variables such as power allocation and load transfer; constructs a stochastic programming model with the goal of minimizing the expected operating cost; and uses Benders decomposition or L-shaped algorithm for iterative solution, ultimately obtaining a phased construction scheme that balances economic efficiency and adaptability to uncertain environments.

[0048] In an optional implementation, the multi-stage investment decision-making algorithm in step S400 can also be modeled as a Markov decision process, with the grid state (such as load level, equipment capacity, network topology) as the state space, the equipment investment decision (site selection, capacity determination, construction sequence) as the action space, and the comprehensive cost saving and power supply reliability improvement throughout the entire life cycle as the reward function. It is trained in a simulation environment composed of typical development scenarios using the Deep Q Network (DQN) or Proximal Policy Optimization (PPO) algorithm, and learns the optimal strategy through interaction with the environment, and finally outputs a highly robust preliminary phased construction path.

[0049] In this embodiment, the solution process employs an improved dynamic programming algorithm to handle multi-stage decision-making and uncertainty propagation problems, defining the first... Phase in state The value function under To minimize the sum of current cost and expected future cost, the recursive relationship is shown in formula (18): (18) In the formula, State of stage t The value function, As decision variables, For immediate costs, As the discount factor, Let be the mathematical expectation.

[0050] The state transition is described by a nonlinear function, the expression of which is shown in equation (19): (19) In the formula, This is the state transition function. It is a random perturbation.

[0051] In this embodiment of the application, obtaining the quantitative evaluation result in step S500 includes: Based on the load growth trend and distributed power supply access timing characteristics represented by typical development scenario sets, the evaluation boundary conditions and operating parameters for each planning stage are determined. Based on the determined evaluation boundary conditions and operating parameters, combined with the access location, operating status and distribution network topology of the multi-port flexible interconnection device at each stage, the power loss load nodes within the device's power supply control range after line faults and equipment faults are identified. Based on the power loss load node, the maximum power value that can be transferred to the load at each stage under the support of the equipment is quantified using the load recovery capability model after the fault. The load reduction and power restoration duration during the fault are calculated to obtain the reliability improvement index at each stage. Based on reliability improvement indicators, by comparing the load reduction and power outage duration in scenarios with and without device support, the load reduction and power outage duration reduction at each stage are calculated. Combining the unit power loss cost and the user power outage loss function, the economic loss of power outage is calculated and a monetized representation is formed. Based on monetization, a net benefit assessment model for the entire life cycle is constructed by integrating the investment cost of the integrated equipment, operation and maintenance costs, network loss savings and reliability improvement benefits, to support dynamic decision optimization and obtain quantitative assessment results.

[0052] In this embodiment of the application, the reliability assessment in step S500 includes determining the operating conditions of each planning stage based on a typical development scenario set, identifying recoverable power-loss load nodes after a fault by combining the access location and fault transfer capability model of the multi-port flexible interconnection device, calculating the SAIFI and SAIDI indices using the fault mode analysis method, quantifying the load reduction and power outage time reduction at each stage, and converting the reliability improvement into monetary benefits by combining the unit power loss cost, and finally incorporating it into the full life cycle net benefit assessment model to achieve dynamic quantitative assessment of the reliability of the planning scheme.

[0053] In an optional implementation, the reliability assessment in step S500 can also be carried out by generating long-term load and distributed power output samples based on typical development scenario sets, combining a stochastic process model of equipment failure rate and repair time, simulating the evolution of the distribution network's operating status at each stage in chronological order, statistically considering the load reduction, outage duration, and expected power shortage (EENS) under the fault transfer capability of multi-port flexible interconnection devices, and then calculating reliability indicators and converting them into economic losses, thereby achieving a refined dynamic assessment of the reliability benefits of the planning scheme throughout its entire life cycle.

[0054] In an optional implementation, the reliability assessment in step S500 can also be carried out by modeling the distribution network as a complex network consisting of nodes (substations, load points) and edges (line and device connections), extracting the topological features (such as degree centrality, betweenness, and connectivity) of each planning stage, and combining the connection relationship of multi-port flexible interconnection devices to analyze the structural robustness and load recovery path capability of the network after the failure of critical edges or nodes. The network resilience index is used to quantify the functional degradation and recovery performance under faults, and it is incorporated into the full life cycle assessment as an alternative representation of reliability improvement, so as to achieve a comprehensive evaluation of the resilience and recovery capability of the planning scheme.

[0055] In this embodiment, based on the reliability improvement index, the load reduction and power outage duration are compared with and without device support to calculate the load reduction and power outage duration reduction at each stage. Combined with the unit power loss cost and the user power outage loss function, the economic loss of the power outage is calculated to form a monetized representation. Among them, the reliability index is calculated using the failure mode analysis method, and SAIFI and SAIDI are defined by formulas (20) and (21), respectively: (20) (twenty one) In the formula, This is an average power supply reliability indicator. This is an indicator of the average power outage duration. Let i be the number of users at node i. For failure rate, This represents the average power outage time.

[0056] In this embodiment of the application, step S600, which yields the final distribution network architecture evolution plan, includes: Based on the initial phased construction path, perturbations are applied to key uncertainty parameters, including load growth rate, unit investment cost of equipment, and output fluctuation range of distributed power sources, forming multiple sets of differentiated parameter scenarios. Based on multiple sets of differentiated parameter scenarios, the phased construction scheme is re-solved for each set of scenarios, the construction sequence of each candidate node under different scenarios is recorded, and its construction frequency and capacity distribution are statistically analyzed to identify the set of key nodes with high construction probability. Based on the construction sequence of each candidate node under different scenarios, the degree of deviation of the construction sequence of each candidate node with parameter disturbance is evaluated, and the stability level of construction timing is quantified. Based on the set of key nodes, and by combining the key node identification results with the stability index of construction timing, multi-scenario analysis and robust optimization algorithms are used to select the device access location and capacity sequence, thus forming the final distribution network architecture evolution planning scheme.

[0057] In this embodiment of the application, the perturbation of key uncertainty parameters in step S600 includes determining the fluctuation range of key parameters such as load growth rate, equipment investment cost and distributed power output based on a set of typical development scenarios, constructing an ellipsoidal uncertainty set centered on the mean, and solving it using a robust optimization framework to ensure that the planning scheme still meets technical constraints and has optimal performance under the most unfavorable parameter perturbation, thereby generating a highly robust final distribution network architecture evolution planning scheme.

[0058] In an optional implementation, the perturbation of key uncertainty parameters in step S600 can also be achieved by determining the probability distribution of uncertain parameters such as load growth and distributed power output based on distribution network basic data and statistical characteristics of typical development scenarios. An optimization model is constructed with the goal of minimizing the expected value of investment cost, network loss cost and reliability penalty. Key constraints such as power flow exceeding limits and voltage deviation are set as chance constraints that are valid under a preset high probability level. After converting them into deterministic equivalent forms using convex approximation or scenario method, they are solved to obtain a preliminary phased construction path that takes into account both economic efficiency and risk tolerance.

[0059] In an optional implementation, the perturbation of key uncertainty parameters in step S600 can also be achieved by defining the maximum tolerable deviation range of key parameters such as load growth rate and equipment investment cost relative to the predicted value in the absence of an accurate probability distribution, constructing an optimization model with the goal of minimizing the maximum regret value or maximizing the robustness level, evaluating the performance degradation of the planning scheme under extreme deviations without relying on scenario probabilities, and solving to obtain the final distribution network architecture evolution planning scheme with strong adaptability to parameter missetting.

[0060] In this embodiment, the initial phased construction path is used as a benchmark, and perturbations are applied to key uncertainty parameters, including load growth rate, unit investment cost of equipment, and output fluctuation range of distributed power sources, forming multiple sets of differentiated parameter scenarios; wherein, the parameter uncertainty is described by an ellipsoidal uncertainty set, as shown in formula (22): (twenty two) In the formula, The mean of the uncertain parameter, Let covariance matrix be the variance matrix. This represents the confidence level parameter.

[0061] A robust optimization framework is used for processing, and its general form is as follows: (twenty three) In the formula, As decision variables, Let U be an uncertain parameter and U be an uncertain set. The objective function is denoted as .

[0062] In this embodiment, the global sensitivity analysis uses the Morris method, and its sensitivity mean and standard deviation are calculated by formulas (24) and (25), respectively: (twenty four) (25) In the formula, Let be the mean sensitivity of parameter i. Here, r represents the sensitivity standard deviation, and r is the sample size. For the perturbation step size, The objective function is denoted as .

[0063] In this embodiment, the present invention starts with data collection and preprocessing, and sequentially goes through key steps such as scenario generation and analysis, load forecasting modeling, distributed power source modeling, multi-port device location and capacity determination, and multi-stage investment decision optimization to form a preliminary construction path; the feasibility of the scheme is verified by power flow calculation and constraint verification, and reliability assessment, economic analysis and sensitivity analysis are carried out to optimize and adjust the scheme; if convergence is not achieved, the previous steps are returned for iterative optimization until the performance requirements are met, and the final planning scheme is output. The effectiveness of the scheme in actual operation is verified by simulation, and finally, a long-term evolution plan for the distribution network that takes into account technical feasibility, economic rationality and system resilience improvement is achieved.

[0064] Example 3, referring to Tables 1-3, is an embodiment of the present invention. This embodiment provides a method for planning the evolution of a power distribution network architecture. In order to verify the beneficial effects of the present invention, specific implementation methods and implementation effects are used for scientific demonstration.

[0065] The specific details of this embodiment are as follows: 1. Experimental Environment This study uses a simulation environment built on the MATLAB / Simulink platform and selects a mountainous power distribution network as a case study for verification. This power distribution network includes four 10kV feeders with a total load capacity of approximately 45MW, and a planning period of 2024-2033 (10 years). The simulation parameters are set as follows: discount rate 8%, planning period 10 years, 1000 scenarios (reduced to 50 typical scenarios). The unit capacity investment cost of the multi-port flexible interconnection device is 3000 yuan / kVA, and the cost of upgrading traditional equipment is 60% of the original investment.

[0066] 2. Experimental Procedure Step 1: Building the basic dataset Based on basic data from the power distribution network and driving factors of social electricity demand, key model parameters related to the probability distribution of photovoltaic power output, load growth trends, and energy storage operation characteristics are identified, forming a set of key model parameters. Quality verification and outlier processing are performed on the corresponding data fields in the multi-source raw datasets. Furthermore, according to the input format required by the model parameters, the data from different sources are normalized in terms of dimensions and aligned in terms of time scale to form a consistent intermediate dataset.

[0067] Step 2: Generate a set of typical development scenarios By utilizing Latin hypercube sampling combined with the Kantorovich distance criterion, and based on the joint output model of photovoltaic, energy storage, and electric vehicle charging power, as well as the temporal pattern of load growth, an initial set of random scenarios was generated. By calculating the similarity between scenarios and merging redundant paths, 50 of the most representative typical scenario sets were finally selected.

[0068] Step 3: Establish a collaborative operation model between multi-port flexible interconnection devices and energy storage. A device operation model is constructed that incorporates active and reactive power exchange capabilities and apparent power capacity constraints. This model integrates the charging and discharging dynamic characteristics of the energy storage system to form a device-energy storage collaborative operation model. Furthermore, considering distribution network operation constraints, a multi-stage investment decision optimization model is developed with the goal of minimizing the overall lifecycle cost.

[0069] Step 4: Solve the multi-stage optimization problem An improved dynamic programming algorithm is used to solve a multi-stage mixed-integer optimization problem in reverse order, determining the investment decisions and operational strategies for each stage. A fault recovery capability model is incorporated into the optimization objective to ensure the scheme possesses practical fault recovery capabilities.

[0070] Step 5: Reliability Assessment and Quantification Based on the load growth trend and distributed power supply access timing characteristics represented by typical development scenario sets, the SAIFI and SAIDI indices are calculated using the Failure Mode and Effects Analysis (FMEA) method. The reliability improvement is converted into monetization benefits by combining the unit power loss cost and incorporated into the full life cycle net benefit assessment model.

[0071] Step Six: Adaptive Analysis and Robust Optimization Key uncertainty parameters (such as load growth rate, unit investment cost of equipment, and output fluctuation range of distributed power sources) are perturbed to the initial phased construction path. The phased construction scheme is re-solved for each set of scenarios, the set of key nodes with high construction probability is identified, and the degree of deviation of the construction sequence of each candidate node with parameter perturbation is evaluated. Finally, multi-scenario analysis and robust optimization techniques are used to select the optimal configuration scheme.

[0072] 3. Experimental Results As shown in Table 1, the comparison results of the technical and economic indicators of different planning schemes show that the present invention is superior to traditional planning and flexible transformation schemes in terms of investment cost, operating efficiency, power supply reliability and comprehensive benefits.

[0073] Table 1 Comparison of Technical and Economic Indicators of Different Planning Schemes

[0074] As shown in Table 2, the comparison of investment adaptability of the three schemes under different development scenarios shows that the invention has stronger robustness and planning stability when facing multiple uncertainties.

[0075] Table 2 Adaptability Analysis under Different Scenarios (Investment Increase %)

[0076] As shown in Table 3, the configuration scheme of the multi-port flexible interconnection device planned in this invention clarifies the installation location, capacity scale and feeder connection relationship of the device, reflecting the feasibility and rationality of the phased construction path.

[0077] Table 3 Configuration Scheme for Multi-Port Flexible Interconnection Device

[0078] Simulation results show that the proposed distribution network architecture evolution planning method based on multi-port flexible interconnection devices has significant advantages in terms of technical and economic efficiency. It can effectively reduce investment costs, improve system reliability, and exhibit good adaptability and robustness under different development scenarios, providing a scientific and systematic solution for the sustainable development of distribution networks under the background of high proportion of distributed energy access.

[0079] Example 4 is an embodiment of the present invention. This embodiment differs from the first embodiment in that it provides a power distribution network architecture evolution planning system.

[0080] It should be noted that the technical solution of the distribution network architecture evolution planning system and the technical solution of the aforementioned distribution network architecture evolution planning method belong to the same concept. For details not described in detail in the technical solution of the distribution network architecture evolution planning system in this embodiment, please refer to the description of the technical solution of the aforementioned distribution network architecture evolution planning method.

[0081] This embodiment of a power distribution network architecture evolution planning system includes: The basic data integration module is used to acquire basic data of the distribution network and driving factors of social electricity demand. Through data preprocessing, the basic data set of the distribution network is obtained. The typical scenario construction module is used to obtain a set of typical development scenarios based on a basic dataset through uncertainty modeling and multi-scenario generation and reduction. The preliminary planning and decision-making module is used to optimize the site selection and capacity determination of multi-port flexible interconnection devices based on typical development scenario sets and system operation constraints, and to obtain a preliminary site selection and capacity determination scheme. The multi-stage investment decision module is used to construct and solve a multi-stage investment decision model based on the preliminary site selection and capacity determination plan, so as to obtain the preliminary phased construction path; The life-cycle benefit assessment module is used to calculate power supply reliability indicators based on the initial phased construction path, and to evaluate the economic efficiency by combining the failure mode analysis method with the cost model, so as to obtain quantitative assessment results. The global sensitivity feedback module is used to perform global sensitivity analysis based on the quantitative evaluation results using the Morris algorithm, adjust the optimization strategy, and obtain the final distribution network architecture evolution plan.

[0082] This embodiment also provides an electronic device applicable to a power distribution network architecture evolution planning method, including: The system includes a memory and a processor. The memory stores computer-executable instructions, and the processor executes these instructions to implement a power distribution network architecture evolution planning method as proposed in the above embodiments.

[0083] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, it implements a power distribution network architecture evolution planning method as proposed in the above embodiments.

[0084] The storage medium proposed in this embodiment and the method for implementing a power distribution network architecture evolution planning proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0085] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0086] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for planning the evolution of a power distribution network architecture, characterized in that, include: Acquire basic data of the power distribution network and driving factors of social electricity demand, and obtain the basic dataset of the power distribution network through data preprocessing; Based on the basic dataset, a set of typical development scenarios is obtained through uncertainty modeling and multi-scenario generation and reduction; Based on a set of typical development scenarios and combined with system operation constraints, the site selection and capacity determination of multi-port flexible interconnection devices are optimized to obtain a preliminary site selection and capacity determination scheme. Based on the preliminary site selection and capacity determination plan, a multi-stage investment decision model was constructed and solved to obtain a preliminary phased construction path; Based on the initial phased construction path, the power supply reliability index is calculated by the failure mode analysis method and the economic efficiency is evaluated by combining the cost model to obtain quantitative evaluation results. Based on the quantitative evaluation results, global sensitivity analysis is performed using the Morris algorithm, and optimization strategies are adjusted to obtain the final power distribution network architecture evolution plan.

2. The distribution network architecture evolution planning method as described in claim 1, characterized in that: The obtained distribution network basic dataset includes: Based on the basic data of the distribution network and the driving factors of social electricity demand, key model parameters related to the probability distribution of photovoltaic output, load growth trend and energy storage operation characteristics are identified, and a set of key model parameters is formed. Based on the key model parameter set, quality verification and outlier processing are performed on the corresponding data fields in the multi-source original datasets. Missing values ​​and erroneous records are identified and eliminated. In accordance with the input format required by the model parameters, the data from different sources are normalized in terms of dimensions and aligned in terms of time scale to form a consistent intermediate dataset. Based on a consistent intermediate dataset, a multi-dimensional structured database integrating the power grid physical architecture, equipment operating characteristics, load change patterns, and external development driving factors was established. Based on a multi-dimensional structured database, all processed structured data and calibration parameters are integrated to obtain the basic dataset of the power distribution network.

3. The distribution network architecture evolution planning method as described in claim 1 or 2, characterized in that: The obtained typical development scenario set includes: Based on the basic dataset of the power distribution network, a joint output model of photovoltaic, energy storage and electric vehicle charging power is constructed; Based on the joint output model, combined with the probability distribution characteristics of photovoltaic output and the temporal evolution law of load growth, an initial random sample set is generated, and a stratified sampling method is used to generate an initial scenario set that conforms to the actual distribution characteristics. Based on the initial scene set, the difference metric between each scene is calculated. Using the scene distance criterion with optimal transmission cost as the core, the similarity between scenes is quantified, redundant scene paths are identified and merged, and a filtered candidate scene set is generated. The candidate scenario set is clustered and divided, and the most representative typical scenarios in each cluster are retained to obtain the typical development scenario set.

4. The distribution network architecture evolution planning method as described in claim 3, characterized in that: The preliminary site selection and capacity determination scheme obtained includes: Based on the distribution network topology and load distribution characteristics, combined with network connectivity, critical load density and potential fault risk level, a set of candidate installation locations for multi-port flexible interconnection devices is determined. Based on the candidate location set, a device operation model is constructed that includes active and reactive power exchange capabilities and apparent power capacity constraints. The charging and discharging dynamic characteristics of the energy storage system are integrated to form a device-energy storage collaborative operation model. Based on the device-energy storage collaborative operation model and combined with the distribution network operation constraints, an optimization model is constructed with the goal of minimizing the overall cost, taking into account investment cost, network loss cost and reliability improvement benefits; The fault transfer capability model of the device to the power loss load is introduced into the optimization model to evaluate the load recovery capability of the multi-port flexible interconnection device under the condition of line fault and equipment fault, and incorporate it as a hard constraint into the optimization objective to generate an optimization model with integrated reliability constraints. The fault transfer capability model of the device to the power loss load is introduced into the optimization model to evaluate the load recovery capability of the multi-port flexible interconnection device under line faults and equipment faults. This model is then incorporated as a hard constraint in the solution to obtain a preliminary site selection and capacity determination scheme.

5. The distribution network architecture evolution planning method as described in claim 4, characterized in that: The preliminary phased construction path obtained includes: Based on a set of typical development scenarios, a multi-stage progressive planning strategy is adopted to divide the power grid development into multiple planning stages, determine the time span and boundary conditions of each stage, and form a multi-stage planning framework. Based on the planning framework, and combined with the load level and network status at each stage, a distribution network operation model is established, and system operation constraints are constructed. Based on the preliminary site selection and capacity determination scheme, the commissioning time series variables of the equipment at different stages are constructed, and time-series coupling constraints are applied to ensure that the upper limit of the single-stage installation capacity and the cumulative construction capacity do not exceed the rated capacity. By integrating investment costs, network loss costs, and reliability benefits at each stage, an objective function is constructed with the goal of minimizing the overall cost throughout the entire lifecycle. By combining the dynamic evolution constraint of the device's fault recovery capability as a hard constraint with the objective function, a multi-stage mixed integer optimization problem is formed and solved, resulting in a preliminary phased construction path.

6. The distribution network architecture evolution planning method as described in claim 5, characterized in that: The obtained quantitative evaluation results include: Based on the load growth trend and distributed power supply access timing characteristics represented by typical development scenario sets, the evaluation boundary conditions and operating parameters for each planning stage are determined. Based on the determined evaluation boundary conditions and operating parameters, combined with the access location, operating status and distribution network topology of the multi-port flexible interconnection device at each stage, the power loss load nodes within the device's power supply control range after line faults and equipment faults are identified. Based on the power loss load node, the maximum power value that can be transferred to the load at each stage under the support of the equipment is quantified using the load recovery capability model after the fault. The load reduction and power restoration duration during the fault are calculated to obtain the reliability improvement index at each stage. Based on reliability improvement indicators, by comparing the load reduction and power outage duration in scenarios with and without device support, the load reduction and power outage duration reduction at each stage are calculated. Combining the unit power loss cost and the user power outage loss function, the economic loss of power outage is calculated and a monetized representation is formed. Based on monetization, a net benefit assessment model for the entire life cycle is constructed by integrating the investment cost of the integrated equipment, operation and maintenance costs, network loss savings and reliability improvement benefits, to support dynamic decision optimization and obtain quantitative assessment results.

7. The distribution network architecture evolution planning method as described in claim 6, characterized in that: The final distribution network architecture evolution plan includes: Based on the initial phased construction path, perturbations are applied to key uncertainty parameters, including load growth rate, unit investment cost of equipment, and output fluctuation range of distributed power sources, forming multiple sets of differentiated parameter scenarios. Based on multiple sets of differentiated parameter scenarios, the phased construction scheme is re-solved for each set of scenarios, the construction sequence of each candidate node under different scenarios is recorded, and its construction frequency and capacity distribution are statistically analyzed to identify the set of key nodes with high construction probability. Based on the construction sequence of each candidate node under different scenarios, the degree of deviation of the construction sequence of each candidate node with parameter disturbance is evaluated, and the stability level of construction timing is quantified. Based on the set of key nodes, and by combining the key node identification results with the stability index of construction timing, multi-scenario analysis and robust optimization algorithms are used to select the device access location and capacity sequence, thus forming the final distribution network architecture evolution planning scheme.

8. A power distribution network architecture evolution planning system, using the method described in any one of claims 1-7, characterized in that, include: The basic data integration module is used to acquire basic data of the distribution network and driving factors of social electricity demand. Through data preprocessing, the basic data set of the distribution network is obtained. The typical scenario construction module is used to obtain a set of typical development scenarios based on a basic dataset through uncertainty modeling and multi-scenario generation and reduction. The preliminary planning and decision-making module is used to optimize the site selection and capacity determination of multi-port flexible interconnection devices based on typical development scenario sets and system operation constraints, and to obtain a preliminary site selection and capacity determination scheme. The multi-stage investment decision module is used to construct and solve a multi-stage investment decision model based on the preliminary site selection and capacity determination plan, so as to obtain the preliminary phased construction path; The life-cycle benefit assessment module is used to calculate power supply reliability indicators based on the initial phased construction path, and to evaluate the economic efficiency by combining the failure mode analysis method with the cost model, so as to obtain quantitative assessment results. The global sensitivity feedback module is used to perform global sensitivity analysis based on the quantitative evaluation results using the Morris algorithm, adjust the optimization strategy, and obtain the final distribution network architecture evolution plan.

9. An electronic device, characterized in that, include: Memory, used to store programs; A processor for loading the program to perform the steps of the method as claimed in any one of claims 1-7.

10. A computer-readable storage medium storing a program, characterized in that, When the program is executed by a processor, it implements the steps of the method as described in any one of claims 1-7.

Images