A flexible resource cross-taiwan area interaction optimization regulation method and device

Abstract

This invention relates to a method and apparatus for flexible resource cross-regional interactive optimization control, belonging to the field of power system dispatching and operation technology. The method includes the following steps: acquiring multi-source uncertainty data on flexible resources across regions and distribution network operation; constructing a joint probability distribution model of multi-source uncertainties; building a four-level interactive framework for multi-regional flexible resource clusters; establishing a correlation model between cross-regional line power flow and uncertain disturbances, incorporating multiple constraints of the distribution network, and deriving the joint probability density function of cross-regional power flow; constructing a day-ahead stochastic optimization control model and a real-time collaborative look-ahead optimization control model, setting dual-layer optimization objectives and multiple constraints, jointly solving the two optimization control models, and outputting the optimal cross-regional interactive control strategy. Compared with existing technologies, this invention has advantages such as accurately quantifying the impact of distributed power source fluctuations, load changes, and equipment response deviations on cross-regional interaction, and improving the safety and economy of the control strategy.

Description

Technical Field

[0001] This invention relates to the field of power system dispatching and operation technology, and in particular to a flexible resource cross-regional interactive optimization control method and device. Background Technology

[0002] With the large-scale integration of flexible resources such as distributed photovoltaic power, electric vehicles, and user-side energy storage into distribution network areas, the traditional single-area control model is no longer sufficient to meet the needs of optimal resource allocation. Cross-area interaction has become a key means to improve the operating efficiency of distribution networks. By complementing resources between distribution areas, problems such as heavy overload, voltage exceeding limits, and three-phase imbalance can be effectively alleviated.

[0003] Chinese patent CN120638516A discloses a method, device, and electronic equipment for cross-distribution area collaborative control of a distribution network. It acquires sensor data from multiple distribution areas, extracts spatiotemporal features, and constructs a cross-distribution area coupling coefficient matrix. Using a deep reinforcement learning model combined with graph convolutional neural networks and a Transformer model, it generates an initial reconfiguration scheme for the distribution network and determines a target reconfiguration scheme based on local decision-making schemes, thus achieving cross-distribution area collaborative control. However, this method focuses on characterizing the voltage sensitivity between distribution areas through the cross-distribution area coupling coefficient matrix and relies on a deep reinforcement learning model to generate reconfiguration schemes. It fails to consider the inherent multi-source uncertainties of various flexible resources such as distributed photovoltaics, electric vehicles, and energy storage, resulting in insufficient handling of the coupling effects of multi-source uncertainties and poor robustness of the optimization results in a probabilistic sense. Furthermore, its decision-making mechanism prioritizes the global scheme when there is a conflict between the global and local schemes, simply considering emergency events such as voltage exceeding limits, without systematically coordinating multiple physical constraints and uncertainties within the optimization framework. This makes it difficult to guarantee the safety and economy of cross-distribution area interaction under complex uncertain environments.

[0004] In summary, flexible cross-distribution area interaction faces significant challenges from multi-source uncertainties and system constraints: the intermittency of distributed photovoltaic power output, the randomness of electric vehicle charging loads, and the response deviation of energy storage devices make it difficult to accurately predict cross-distribution area power exchange. Simultaneously, physical constraints such as distribution network line transmission capacity, node voltage limits, and three-phase balance requirements further compress the feasible space for resource interaction. Existing cross-distribution area control methods mostly employ deterministic models, failing to fully consider the coupled effects of multi-source uncertainties and the multiple constraints of the distribution network. This can easily lead to conservative or aggressive control strategies, resulting in either resource waste or threats to the safe operation of the distribution network. Furthermore, existing methods lack a unified multi-distribution area interaction framework and an efficient information exchange mechanism, making it difficult to achieve multi-stakeholder collaborative optimization and failing to meet the requirements of new power systems for the accuracy and flexibility of cross-distribution area control. Summary of the Invention

[0005] The purpose of this invention is to solve the problems in existing cross-regional regulation methods, which fail to fully consider the coupling effects of multi-source uncertainties, lack system integration of multiple physical constraints of the distribution network, and improve the multi-entity coordination mechanism, resulting in insufficient exploitation of flexible resource regulation potential and an imbalance between the safety and economy of cross-regional interaction. The invention provides a flexible resource cross-regional interaction optimization regulation method and device that takes into account multi-source uncertainties, which can improve the carrying capacity of the distribution network while achieving synergistic improvement of new energy consumption and system operation efficiency.

[0006] The objective of this invention can be achieved through the following technical solutions: According to a first aspect of the present invention, a flexible resource cross-regional interactive optimization and control method is provided, comprising the following steps: S1. Acquire multi-source uncertainty data of flexible resources across distribution areas and distribution network operation. Classify the uncertainty factors into four categories: source-side fluctuations, load-side fluctuations, equipment response deviations, and grid constraint fluctuations. Construct a multi-source uncertainty joint probability distribution model by classifying, modeling, optimizing, and jointly synthesizing the four types of uncertainty factors. S2, build a four-level interactive framework for flexible resource clusters in multiple distribution areas, clarify the data interaction interface and information interaction mechanism at each level, establish a correlation model between cross-distribution area line power flow and uncertain disturbances based on the AC power flow sensitivity matrix, and incorporate multiple constraints of the distribution network to derive the joint probability density function of cross-distribution area power flow and realize uncertainty transmission mapping; S3. Construct a day-ahead stochastic optimization control model and a real-time collaborative look-ahead optimization control model, set a dual-layer optimization objective and multiple constraints, including uncertainty scenario constraints determined by the joint probability density function of cross-regional power flow, jointly solve the two optimization control models, and output the optimal cross-regional interactive control strategy.

[0007] The multi-source uncertainty data includes: distributed photovoltaic power output fluctuation data, electric vehicle charging load fluctuation data, energy storage device response deviation data, inter-station line transmission capacity constraint data, distribution network voltage amplitude constraint data, three-phase imbalance constraint data, and controllable load adjustment deviation data.

[0008] S1 specifically includes the following steps: S11, acquire cross-regional flexible resources and multi-source uncertainty data of distribution network operation; S12 classifies multi-source uncertainties into four categories: source-side fluctuations, load-side fluctuations, equipment response deviations, and grid constraint fluctuations. These are manifested as random disturbances in cross-regional power exchange. Gaussian mixture models are established for the power disturbances under the influence of the four categories of factors. S13, decompose the four types of deviation vectors corresponding to source-side fluctuations, load-side fluctuations, equipment response deviations, and grid constraint fluctuations into steady-state components and fluctuation components; solve the parameters of the four Gaussian mixture models by the expectation-maximization algorithm, and synthesize the four types of deviation vectors into a unified random variable based on the parameter optimization results, and establish the probability distribution model of the unified random variable, namely the multi-source uncertainty joint probability distribution model.

[0009] The Gaussian mixture model is expressed as follows: ; ; ; ; in, The cross-regional power deviation is caused by source-side fluctuations. It is power deviation The probability density function; This represents the number of sub-models in the Gaussian mixture model corresponding to the source-side fluctuations. In the Gaussian mixture model corresponding to the source-side fluctuations, the first... The weights of each sub-model; The first Gaussian mixture model corresponding to the source-side fluctuations A single Gaussian distribution, in which The mean of this single Gaussian distribution represents the central tendency of the source-side fluctuation power deviation. Let be the covariance matrix of this single Gaussian distribution, representing the degree of dispersion of the power deviation; The power deviation across distribution zones is caused by load-side fluctuations. Power deviation across distribution zones caused by load-side fluctuations The probability density function; This represents the number of sub-models in the Gaussian mixture model corresponding to the load-side fluctuations. In the Gaussian mixture model corresponding to the load-side fluctuations, the first... The weights of each Gaussian sub-model; The first Gaussian mixture model corresponding to the load-side fluctuation A single Gaussian distribution, in which The mean of this single Gaussian distribution represents the central tendency of the load-side fluctuating power deviation. Let be the covariance matrix of this single Gaussian distribution, representing the degree of dispersion of the load-side fluctuation power deviation; This refers to the cross-regional power deviation caused by equipment response deviation. Cross-regional power deviation caused by equipment response deviation The probability density function; The number of sub-models in the Gaussian mixture model corresponding to the device response deviation; In the Gaussian mixture model corresponding to the device response deviation, the first... The weights of each Gaussian sub-model; In the Gaussian mixture model corresponding to the device response deviation, the first... A single Gaussian distribution, in which It is the mean of this single Gaussian distribution, representing the central tendency of the equipment response deviation power. It is the covariance matrix of the single Gaussian distribution, representing the degree of dispersion of the device response deviation power; This refers to the cross-regional power exchange deviation caused by grid constraint fluctuations. Cross-regional power deviation caused by grid constraint fluctuations The probability density function; This represents the number of sub-models in the Gaussian mixture model corresponding to power grid constrained fluctuations. In the Gaussian mixture model corresponding to power grid constrained fluctuations, the first... The weights of each Gaussian sub-model; In the Gaussian mixture model corresponding to power grid constrained fluctuations, the first... A single Gaussian distribution, in which It is the mean of this single Gaussian distribution, representing the central tendency of grid-constrained fluctuating power. It is the covariance matrix of the single Gaussian distribution, representing the degree of dispersion of the power grid constraint fluctuation power.

[0010] The aforementioned multi-source uncertainty joint probability distribution model is expressed as: ; in, The unified random variable resulting from the synthesis of four types of biases The probability density function describes the overall distribution of cross-regional power deviation under comprehensive uncertainty; The cross-regional power deviation is a unified random variable synthesized from four types of deviations: source side, load side, equipment side, and grid constraint. The number of sub-models in the post-multi-source uncertainty joint probability distribution model is equal to the product of the number of Gaussian sub-models for each of the four types of factors, i.e. ; The first multi-source uncertainty joint probability distribution model The weights of each Gaussian sub-model are equal to the product of the weights of the corresponding sub-models for the four categories of factors, i.e. ; The first multi-source uncertainty joint probability distribution model There are single Gaussian distributions, among which... This is the combined mean vector. This is the combined covariance matrix.

[0011] The S2 specifically includes the following steps: S21 establishes a four-level interactive framework consisting of distribution network dispatch center, resource aggregator, transformer area management node, and resource entity. The distribution network dispatch center is responsible for overall strategy formulation, the resource aggregator is responsible for resource aggregation and instruction distribution, the transformer area management node executes fine-grained control, and the resource entity responds to control instructions. S22, Based on the AC power flow sensitivity matrix, a correlation model between power flow and uncertain disturbances on cross-transformer lines is established. Cross-regional trend Represented as: ; in, For cross-regional lines The AC power flow sensitivity matrix quantifies the transmission intensity of multi-source uncertainty disturbances to cross-regional power exchange. The cross-regional power deviation is a unified random variable synthesized from four types of deviations: source side, load side, equipment side, and grid constraints, representing the uncertainty disturbance. S23, based on the linear transformation property of Gaussian distribution, for a unified random variable A linear transformation is performed on the multi-source uncertainty joint probability distribution model to derive the cross-regional power flow. joint probability density function : ; in, The number of sub-models in the multi-source uncertainty joint probability distribution model; The first multi-source uncertainty joint probability distribution model The weights of each Gaussian sub-model; The first step is to perform a linear transformation of the multi-source uncertainty joint probability distribution model. A single Gaussian distribution, For the line The mean vector of the current. For the line The covariance matrix of the current flow; S24, combining distribution network constraints, including line transmission capacity, voltage amplitude, and three-phase imbalance, transforms deterministic constraints into probabilistic constraints, quantifying the pre-set confidence level at which the constraints are satisfied. This confidence level is dynamically selected based on distribution network safety requirements and the intensity of uncertainty disturbances, ranging from 90% to 99%, with a typical value of 95%. Correspondingly, a probability threshold is set, i.e., the probability of exceeding the constraint range is less than a certain value, typically 5% (at a 95% confidence level), 10% (at a 90% confidence level), or 1% (at a 99% confidence level). This achieves a complete mapping from the distribution of uncertainty sources to the feasible control domain across distribution areas, specifically including: Line transmission capacity constraints: ,in, For the line The rated transmission power is required at a confidence level. Down, The probability of exceeding the constraint interval is less than ; Voltage amplitude constraint: based on power flow-voltage sensitivity matrix Derive the node voltage of the distribution area ,constraint ,in, For nodes t voltage, For nodes t The initial voltage, These are the upper and lower limits of the allowable node voltage, and the requirements are as follows: at a confidence level... Below, the probability of voltage exceeding the limit is less than ; Three-phase imbalance constraint: based on cross-regional three-phase power exchange Calculate the three-phase unbalance. ,in, For the line The three-phase active power vector, The lines are respectively The active power of phases A, B, and C in the cross-regional distribution area. To allow for the maximum imbalance, a typical value is ≤2%, which is a safety threshold conforming to national standards and distribution network operation requirements. This threshold is used to constrain the degree of three-phase imbalance during cross-distribution area interactions, and requires a certain confidence level. Below, the probability of exceeding the imbalance limit is less than .

[0012] The S3 specifically includes the following steps: S31. Construct a day-ahead stochastic optimization control model with the objective function of minimizing the comprehensive cost of cross-regional interaction and maximizing the renewable energy consumption rate. Set multiple constraints, including resource regulation capacity constraints within the region, transmission capacity constraints of cross-regional lines, voltage amplitude constraints, three-phase imbalance constraints, and uncertainty scenario constraints. Use the key scenario identification method to solve the day-ahead stochastic optimization control model to obtain the day-ahead cross-regional interaction plan. S32 constructs a real-time collaborative forward-looking optimization and control model. Based on real-time uncertainty disturbance observations, it aims to minimize the tracking deviation of the day-ahead cross-regional interaction plan and achieve the best power fluctuation smoothing effect. A rolling optimization time window is introduced. The window duration is determined according to the real-time control requirements and computing resource allocation, with a typical value of 15 minutes and a step size of 1 to 5 minutes. When the frequency of uncertainty disturbances in the distribution network is high (such as the photovoltaic output fluctuation period < 10 minutes), the step size is 1 minute; when the disturbance is mild, the step size can be 3 to 5 minutes to dynamically adjust the cross-regional power exchange strategy. S33 uses the alternating direction multiplier method and the branch and bound algorithm to solve the two-layer optimization model consisting of the daytime stochastic optimization control model and the real-time collaborative look-ahead optimization control model, and outputs the optimal cross-regional interactive control strategy. The optimal cross-regional interactive control strategy includes flexible resource adjustment instructions for each region, cross-regional power exchange curves, reserve capacity configuration schemes and three-phase balance control strategies. Specifically, the two-layer optimization model is decomposed into a day-ahead master problem and a real-time subproblem by using the alternating direction multiplier method. The master problem is solved by the branch and bound algorithm to solve the mixed integer linear programming problem and output the day-ahead global optimal plan. The subproblem is solved by the fast solution mode of Gurobi under the boundary constraints of the master problem and the real-time fine-tuning strategy is adopted. The two-layer model achieves coordinated convergence through iterative updates of dual variables.

[0013] In the aforementioned daytime stochastic optimization control model, the objective function is expressed as: ; ; in, To improve the comprehensive absorption rate of new energy in multiple districts, This represents the total number of scheduling periods in the previous day. The total number of stations. for Time period The photovoltaic output consumed by the substation for Time period Total photovoltaic output in the area; For electricity purchase costs, , for Peak-valley electricity pricing for different time periods for Taiwan Electricity purchased online during specific time periods; To cover the cost of light curtailment, , The cost of per unit of abandoned light penalty for Time period The photovoltaic output consumed by the substation for Time period Total photovoltaic output in the area; In order to regulate compensation costs, , The unit resource adjustment compensation electricity price, for Taiwan Flexible resource adjustment during different time periods; Multiple constraints include: Resource regulation capacity constraints within the distribution area: ; Transmission capacity constraints for cross-regional lines: ; Voltage amplitude constraint: ; Three-phase unbalance constraint: ; Uncertainty scenario constraints: Based on the joint probability density function of cross-regional power flow derived by S2, typical scenarios with confidence levels of 90%~95% are selected, with a typical value of 95%, to ensure that the constraints are satisfied in each scenario; in, for Taiwan The upper and lower limits of flexible resource output such as time-of-use energy storage / controllable load. , Determined by the physical characteristics of the equipment, for Time-of-day routes Cross-regional power, For the line Rated transmission power, To account for uncertainties, the safety margin factor is determined based on the aging degree of the distribution network lines, their operating years, and historical fault data, with a value range of 0.8 to 0.95. When the operating years of the lines are ≤5 years and there are no fault records, the value is 0.9 to 0.95. When the operating years are >10 years or there are more than 3 overload records, the value is 0.8 to 0.85. for Taiwan Time period node voltage, These are the upper and lower limits of the allowable node voltage, respectively. for Taiwan Three-phase imbalance over time period To allow for the maximum imbalance, a typical value of ≤2% is set. This is a safety threshold that complies with national standards and distribution network operation requirements, and is used to constrain the degree of three-phase imbalance during cross-regional interaction.

[0014] The objective function of the real-time collaborative look-ahead optimization and control model is expressed as: ; ; in, To address the recent discrepancy in the tracking of the cross-regional interaction plan, This represents the number of time periods within the scrolling window. This represents the total number of cross-regional lines. for Time-of-day routes Real-time power adjustment The planned power output for the day is [number]. The standard deviation of cross-regional power exchange. Lines within the scrolling window The average power.

[0015] According to a second aspect of the present invention, a flexible resource cross-regional interactive optimization and control device is provided for implementing the method, comprising: Data acquisition and modeling module: Acquires multi-source uncertainty data of flexible resources across distribution areas and distribution network operation, classifies uncertainty factors into four categories: source-side fluctuations, load-side fluctuations, equipment response deviations, and grid constraint fluctuations, and constructs a multi-source uncertainty joint probability distribution model by classifying, modeling, optimizing parameters, and jointly synthesizing the four types of uncertainty factors. Uncertainty handling module: Build a four-level interactive framework for flexible resource clusters in multiple distribution areas, establish a correlation model between cross-distribution area line power flow and uncertain disturbances based on the AC power flow sensitivity matrix, incorporate multiple constraints of the distribution network, derive the joint probability density function of cross-distribution area power flow, and realize uncertainty transmission mapping; The module for constructing and solving the control model is as follows: It constructs a day-ahead stochastic optimization control model and a real-time collaborative look-ahead optimization control model, sets a two-level optimization objective and multiple constraints, including uncertainty scenario constraints determined by the joint probability density function of cross-regional power flow, jointly solves the two optimization control models, and outputs the optimal cross-regional interactive control strategy. The uncertainty processing module receives the multi-source uncertainty joint probability distribution model output by the data acquisition and modeling module, and the control model construction and solution module receives the cross-regional power flow joint probability density function output by the uncertainty processing module, thereby realizing data interaction between models.

[0016] According to a third aspect of the present invention, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the program to implement the method described thereon.

[0017] According to a fourth aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method described thereon.

[0018] Compared with the prior art, the present invention has the following beneficial effects: (1) This invention constructs a multi-source uncertainty joint probability distribution model and classifies and models various uncertainty factors of flexible resources such as distributed photovoltaics, electric vehicles, and energy storage, thereby systematically and quantitatively characterizing the coupling effect of uncertainty in cross-regional interaction, overcoming the shortcomings of existing methods that do not adequately consider multi-source uncertainty or only perform simple input processing. On this basis, this application further establishes a correlation model between cross-regional line power flow and uncertainty disturbances, realizes the probabilistic transmission mapping of uncertainty in the interactive network, and transforms the joint probability density function derived therefrom into uncertainty scenario constraints in the optimization model. This enables the finally generated cross-regional interaction control strategy to inherently coordinate multiple physical constraints and complex uncertainties such as distribution network line capacity and voltage limits, thereby significantly improving the optimization space of cross-regional resource complementarity and the robustness of the control strategy in stochastic environments under the premise of ensuring the safe operation of the distribution network, effectively alleviating problems such as heavy overload and voltage limit exceedance, and meeting the requirements of new power systems for the accuracy and flexibility of cross-regional control.

[0019] (2) This invention constructs a day-ahead stochastic optimization control model and introduces a key scenario identification method. Under multiple constraints such as resource regulation capacity within the transformer area, transmission capacity of cross-transformer lines, voltage amplitude, three-phase imbalance, and uncertain scenarios, it optimizes the system with the goal of minimizing overall cost and maximizing renewable energy absorption rate. This effectively overcomes the shortcomings of traditional deterministic models in handling multi-source uncertainties, making day-ahead planning more robust in a probabilistic sense, more economical, and with stronger renewable energy absorption capacity. On this basis, a real-time collaborative look-ahead optimization control model is further constructed, based on real-time observations and with the goal of minimizing tracking deviation and smoothing power fluctuations. With the goal of optimization, the rolling optimization window dynamically adjusts the strategy, significantly improving the system's adaptability and operational flexibility to real-time uncertainties. Finally, by jointly solving the two-layer optimization model using the alternating direction multiplier method and the branch and bound algorithm, the decomposition and coordination mechanism efficiently handles mixed integer programming and real-time adjustment problems. While ensuring the coordination and unity of global optimization and local dynamic response, it outputs the optimal control scheme, which includes flexible resource adjustment instructions for each transformer area, power exchange curves, backup configurations, and three-phase balance strategies. Thus, it systematically realizes multi-objective collaborative optimization of cross-transformer interaction in terms of safety constraints, economic dispatch, and uncertainty response. Attached Figure Description

[0020] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a flowchart illustrating the construction process of the multi-source uncertainty joint probability distribution model of the present invention. Figure 3 This is a schematic diagram of the four-level interactive framework for multi-zone flexible resource clustering of the present invention; Figure 4 This is a structural diagram of the device of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. 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 scope of protection of the present invention.

[0022] Unless otherwise defined, the technical or scientific terms used in this application shall have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms “a,” “an,” “an,” “the,” and similar words used in this application do not indicate quantity limitation and may indicate singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units not listed, or may include other steps or units inherent to these processes, methods, products, or devices. The terms “connected,” “linked,” “coupled,” and similar words used in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. “Multiple” used in this application refers to two or more. “And / or” describes the relationship between related objects, indicating that three relationships may exist; for example, “A and / or B” can represent: A alone, A and B simultaneously, and B alone. The character " / " generally indicates that the preceding and following objects are in an "or" relationship. The terms "first," "second," and "third" used in this application are merely to distinguish similar objects and do not represent a specific ordering of the objects.

[0023] Example 1 This embodiment provides a flexible resource cross-regional interactive optimization and control method, such as Figure 1 As shown, it includes the following steps: S1 acquires multi-source uncertainty data on flexible resources across distribution areas and distribution network operation. Through classification modeling, parameter optimization, and joint synthesis of various uncertainty factors, a multi-source uncertainty joint probability distribution model is constructed.

[0024] S1 specifically includes the following steps: S11, acquire cross-regional flexible resources and multi-source uncertainty data of distribution network operation.

[0025] The multi-source uncertainty data includes: distributed photovoltaic power output fluctuation data, electric vehicle charging load fluctuation data, energy storage device response deviation data, inter-station line transmission capacity constraint data, distribution network voltage amplitude constraint data, three-phase imbalance constraint data, and controllable load adjustment deviation data.

[0026] S12 classifies multi-source uncertainties into four categories: source-side fluctuations, load-side fluctuations, equipment response deviations, and grid constraint fluctuations. These are manifested as random disturbances in cross-regional power exchange, and Gaussian mixture models are established for the power disturbances under the influence of the four categories of factors.

[0027] The Gaussian mixture model is expressed as follows: (1) (2) (3) (4) In equation (1), The cross-regional power deviation is caused by source-side fluctuations. It is power deviation The probability density function; This represents the number of sub-models in the Gaussian mixture model corresponding to the source-side fluctuations. The Gaussian mixture model is composed of multiple superimposed Gaussian distributions. It is the number of its sub-models; In the Gaussian mixture model corresponding to the source-side fluctuations, the first... The weight of each sub-model represents the proportion of that sub-model in the mixture distribution, and the sum of the weights of all sub-models is 1. The first Gaussian mixture model corresponding to the source-side fluctuations There are single Gaussian distributions (normal distributions); among them The mean of this single Gaussian distribution represents the central tendency of the source-side fluctuation power deviation; Let be the covariance matrix of this single Gaussian distribution, representing the degree of dispersion of the power deviation. If it is a univariate variable, then it is variance.

[0028] In equation (2), The power deviation across distribution zones is caused by load-side fluctuations. Power deviation across distribution zones caused by load-side fluctuations The probability density function represents Characteristics of the probability distribution of different values; This represents the number of sub-models in the Gaussian mixture model corresponding to the load-side fluctuations. In the Gaussian mixture model corresponding to the load-side fluctuations, the first... The weights of each Gaussian sub-model represent the proportion of that sub-model in the mixture distribution, and the sum of the weights of all sub-models is 1. The first Gaussian mixture model corresponding to the load-side fluctuation There are single Gaussian distributions (normal distributions), among which The mean of this single Gaussian distribution represents the central tendency of the load-side fluctuating power deviation; Let be the covariance matrix of this single Gaussian distribution, representing the degree of dispersion of the load-side fluctuating power deviation, and in the univariate scenario, be the variance.

[0029] Equation (3) describes the impact of the deviation between the actual response of energy storage, controllable load, and other equipment and the control command on cross-regional power exchange, where, This refers to the cross-regional power deviation caused by equipment response deviation. Cross-regional power deviation caused by equipment response deviation The probability density function; This represents the number of sub-models in the Gaussian mixture model corresponding to the device response deviation, the proportion of each sub-model in the mixture distribution, and the sum of all weights is 1. In the Gaussian mixture model corresponding to the device response deviation, the first... The weights of each Gaussian sub-model; In the Gaussian mixture model corresponding to the device response deviation, the first... A single Gaussian distribution, in which It is the mean of this single Gaussian distribution, representing the central tendency of the equipment response deviation power. It is the covariance matrix of the single Gaussian distribution, representing the degree of dispersion of the device response deviation power; In equation (4), This refers to the cross-regional power exchange deviation caused by grid constraint fluctuations. Cross-regional power deviation caused by grid constraint fluctuations The probability density function; This represents the number of sub-models in the Gaussian mixture model corresponding to power grid constrained fluctuations. In the Gaussian mixture model corresponding to power grid constrained fluctuations, the first... The weights of each Gaussian sub-model, the proportion of each sub-model in the mixture distribution, and the sum of all weights is 1; In the Gaussian mixture model corresponding to power grid constrained fluctuations, the first... A single Gaussian distribution, in which It is the mean of this single Gaussian distribution, representing the central tendency of grid-constrained fluctuating power. It is the covariance matrix of the single Gaussian distribution, representing the degree of dispersion of the power grid constraint fluctuation power.

[0030] S13, decompose the four types of deviation vectors corresponding to source-side fluctuations, load-side fluctuations, equipment response deviations, and grid constraint fluctuations into steady-state components and fluctuation components; solve the parameters of the four Gaussian mixture models using the expectation-maximization (EM) algorithm; based on the parameter optimization results, synthesize the four types of deviation vectors into a unified random variable; and establish the probability distribution model of the unified random variable, namely the multi-source uncertainty joint probability distribution model.

[0031] The joint probability distribution model of multi-source uncertainties is expressed as: (5) Equation (5) is the probability density function of the unified power deviation after synthesizing four types of uncertainties (source side, load side, equipment side, and grid constraints), used to integrate multi-dimensional uncertainties into a holistic probability model. Among them, The unified random variable resulting from the synthesis of four types of biases The probability density function describes the overall distribution of cross-regional power deviation under comprehensive uncertainty; The cross-regional power deviation is a unified random variable synthesized from four types of deviations: source side, load side, equipment side, and grid constraint. The number of sub-models in the multi-source uncertainty joint probability distribution model is equal to the product of the number of Gaussian sub-models for each of the four types of factors, i.e. ; The first multi-source uncertainty joint probability distribution model The weights of each Gaussian sub-model are equal to the product of the weights of the corresponding sub-models for the four categories of factors, i.e. ; The first multi-source uncertainty joint probability distribution model There are single Gaussian distributions, among which... This is the combined mean vector. The combined covariance matrix is ​​a block diagonal matrix if the four variables are uncorrelated, meaning that the covariance of each factor corresponds to a diagonal block.

[0032] S2 establishes a four-level interactive framework for flexible resource clusters across multiple distribution areas, clarifies the data interaction interfaces and information interaction mechanisms at each level, establishes a correlation model between cross-distribution area line power flow and uncertain disturbances based on the AC power flow sensitivity matrix, incorporates multiple constraints of the distribution network, derives the joint probability density function of cross-distribution area power flow, and realizes uncertainty transmission mapping.

[0033] like Figure 2 As shown, S2 specifically includes the following steps: S21 establishes a four-level interactive framework consisting of the distribution network dispatch center, resource aggregators, transformer area management nodes, and resource entities. The distribution network dispatch center is responsible for formulating overall strategies, the resource aggregators are responsible for resource aggregation and instruction distribution, the transformer area management nodes execute refined control, and the resource entities respond to control instructions.

[0034] like Figure 3 As shown, the top layer is the distribution network dispatch center, responsible for global optimization decisions and constraint setting; the middle layer is the regional aggregator, responsible for resource aggregation across multiple distribution areas, command decomposition, and information exchange; the lower layer is the distribution area management node, responsible for local resource status collection and control command execution; the terminal is the resource entity, including distributed photovoltaic, energy storage, electric vehicles, etc., which responds to control commands and provides feedback on execution status. Through standardized data interaction interfaces, real-time information sharing between all levels is achieved, ensuring efficient and collaborative interaction across distribution areas.

[0035] S22, Based on the AC power flow sensitivity matrix, a correlation model between power flow and uncertain disturbances on cross-transformer lines is established. Cross-regional trend Represented as: (6) in, For cross-regional lines The AC power flow sensitivity matrix is ​​used to quantify the transmission intensity of multi-source uncertainty disturbances to cross-regional power exchange. The cross-regional power deviation is a unified random variable synthesized from four types of deviations: source side, load side, equipment side, and grid constraint, representing the uncertainty disturbance.

[0036] S23, based on the linear transformation property of Gaussian distribution (if random variable After linear transformation For a uniform random variable The multi-source uncertainty joint probability distribution model (Equation (5)) is linearly transformed to derive the cross-regional power flow. joint probability density function : (7) in, The first step is to perform a linear transformation of the multi-source uncertainty joint probability distribution model. A single Gaussian distribution, For the line The mean vector of the current. For the line The covariance matrix of the power flow fully characterizes the probability distribution of power flow across transformer substations.

[0037] S24, combining distribution network constraints, including line transmission capacity, voltage amplitude, and three-phase imbalance, transforms deterministic constraints into probabilistic constraints, quantifying the pre-set confidence level at which the constraints are satisfied. This confidence level is dynamically selected based on distribution network safety requirements and the intensity of uncertainty disturbances, ranging from 90% to 99%, with a typical value of 95%. Correspondingly, a probability threshold is set, i.e., the probability of exceeding the constraint range is less than a certain value, typically 5% (at a 95% confidence level), 10% (at a 90% confidence level), or 1% (at a 99% confidence level). This achieves a complete mapping from the distribution of uncertainty sources to the feasible control domain across distribution areas, specifically including: Line transmission capacity constraints: ,in, For the line The rated transmission power is required at a confidence level. Down, The probability of exceeding the constraint interval is less than ; Voltage amplitude constraint: based on power flow-voltage sensitivity matrix Derive the node voltage of the distribution area ,constraint ,in, For nodes t voltage, For nodes t The initial voltage, These are the upper and lower limits of the allowable node voltage, and the requirements are as follows: at a confidence level... Below, the probability of voltage exceeding the limit is less than ; Three-phase imbalance constraint: based on cross-regional three-phase power exchange Calculate the three-phase unbalance. ,in, For the line The three-phase active power vector, The lines are respectively The active power of phases A, B, and C in the cross-regional distribution area. To allow for the maximum imbalance, a typical value is ≤2%, which is a safety threshold conforming to national standards and distribution network operation requirements. This threshold is used to constrain the degree of three-phase imbalance during cross-distribution area interactions, and requires a certain confidence level. Below, the probability of exceeding the imbalance limit is less than .

[0038] S3. Construct a day-ahead stochastic optimization control model and a real-time collaborative look-ahead optimization control model, set a dual-layer optimization objective and multiple constraints, including uncertainty scenario constraints determined by the joint probability density function of cross-regional power flow, jointly solve the two optimization control models, and output the optimal cross-regional interactive control strategy.

[0039] S3 specifically includes the following steps: S31. Construct a day-ahead stochastic optimization control model with the objective function of minimizing the comprehensive cost of cross-regional interaction and maximizing the renewable energy absorption rate. Set multiple constraints, including resource regulation capacity constraints within the region, transmission capacity constraints of cross-regional lines, voltage amplitude constraints, three-phase imbalance constraints, and uncertainty scenario constraints. Use the key scenario identification method to solve the day-ahead stochastic optimization control model to obtain the day-ahead cross-regional interaction plan.

[0040] 1. Objective function (8) (9) Equation (8) is an objective function that maximizes the renewable energy absorption rate. Wherein, To improve the comprehensive absorption rate of new energy in multiple districts, This represents the total number of scheduling periods in the previous day. The total number of stations. for Time period The photovoltaic output consumed by the substation for Time period Total photovoltaic output of the transformer substation. Equation (9) is the objective function for minimizing the overall cost, where, For electricity purchase costs, , for Peak-valley electricity pricing for different time periods for Taiwan Electricity purchased online during specific time periods; To cover the cost of light curtailment, , The cost of per unit of abandoned light penalty for Time period The photovoltaic output consumed by the substation for Time period Total photovoltaic output in the area; In order to regulate compensation costs, , The unit resource adjustment compensation electricity price, for Taiwan Flexible resource adjustment during different time periods; 2. Multiple constraints (10) (11) (12) (13) Equation (10) represents the resource regulation capacity constraint within the distribution area, where, for Taiwan The upper and lower limits of flexible resource output such as time-of-use energy storage / controllable load. , Determined by the physical characteristics of the equipment. Equation (11) is the transmission constraint for cross-regional lines. for Time-of-day routes Cross-regional power, For the line Rated transmission power, To account for uncertainties, the safety margin factor is determined based on the aging degree of the distribution network lines, their operating years, and historical fault data, with a value range of 0.8 to 0.95; when the line's operating years are ≤5 years and there are no fault records, the value is 0.9 to 0.95; when the line's operating years are >10 years or there are more than 3 overload records, the value is 0.8 to 0.85. Equation (12) is the voltage amplitude constraint, where, for Taiwan Time period node voltage, These are the upper and lower limits of the allowable node voltage, respectively. Equation (13) is the three-phase unbalance constraint. for Taiwan Three-phase imbalance over time period To allow for the maximum imbalance, ≤2% is a safety threshold that complies with national standards and distribution network operation requirements. It is used to constrain the degree of three-phase imbalance during cross-regional interaction.

[0041] Uncertainty scenario constraints: Based on the joint probability density function of cross-regional power flow derived in S23, typical scenarios with confidence levels of 90% to 95% are selected, with a typical value of 95%, to ensure that the constraints are satisfied in each scenario.

[0042] 3. Solution Method The key scenario identification method is adopted. First, an initial scenario set is generated based on the joint probability distribution of multi-source uncertainties. Key typical scenarios are screened out by K-means clustering, covering high-probability and high-risk scenarios. Second, the multi-objective function is transformed into a weighted single objective, with a typical value of 0.6 for the weight of renewable energy absorption rate and 0.4 for the weight of cost. Finally, the optimization scheme under each key scenario is solved, and the final output is a day-ahead cross-regional interaction plan that takes into account all scenarios, including cross-regional power exchange plans and intra-regional resource adjustment plans for each time period.

[0043] S32 constructs a real-time collaborative forward-looking optimization and control model. Based on real-time uncertainty disturbance observations, it aims to minimize the tracking deviation of the day-ahead cross-regional interaction plan and achieve the best power fluctuation smoothing effect. A rolling optimization time window is introduced. The duration of the rolling optimization time window needs to match the response speed of flexible resources: the response time of energy storage equipment is ≤1 minute, so the minimum window step size is 1 minute; the response time of controllable load is ≤5 minutes, so the maximum step size does not exceed 5 minutes. This ensures that the control command matches the resource response characteristics and dynamically adjusts the cross-regional power exchange strategy.

[0044] 1. Objective function (14) (15) Equation (14) takes minimizing the tracking deviation of the current day plan as the objective function, where, To address the recent discrepancy in the tracking of the cross-regional interaction plan, This represents the number of time periods within the scrolling window (a 15-minute window is divided into 15 time periods with a 1-minute granularity). This represents the total number of cross-regional lines. for Time-of-day routes Real-time power adjustment Let be the planned power for the day. Equation (15) takes the power fluctuation mitigation effect as the optimal objective function, where The standard deviation of cross-regional power exchange. Lines within the scrolling window The smaller the standard deviation of the average power, the smoother the power fluctuation.

[0045] 2. Rolling optimization mechanism The time window adopts a rolling optimization window with a 15-minute look-ahead and a 1-minute step size. The real-time disturbance data is updated every minute to resolve the optimal control strategy for the next 15 minutes. The forward-looking constraint introduces the "N-1" safety verification constraint to forward-lookingly assess the feasibility of power reconfiguration after a single cross-regional line fault within the next 15 minutes, thus avoiding safety risks caused by short-term control strategies. The strategy adjustment is based on the deviation between real-time disturbances and day-ahead plans. It prioritizes smoothing fluctuations through energy storage / controllable load regulation within the distribution area. When the deviation exceeds the threshold, the cross-distribution area power exchange strategy is dynamically adjusted to ensure that the tracking deviation is ≤2% and the power fluctuation rate is ≤3%.

[0046] S33 uses the alternating direction multiplier method (ADMM) and branch-and-bound algorithm to solve the two-layer optimization model consisting of the day-ahead stochastic optimization control model and the real-time collaborative look-ahead optimization control model, and outputs the optimal cross-regional interactive control strategy. The optimal cross-regional interactive control strategy includes flexible resource adjustment instructions for each region, cross-regional power exchange curves, reserve capacity configuration schemes and three-phase balance control strategies.

[0047] 1. Solution framework design Layered decoupling: The two-layer model is decomposed into "day-ahead master problem + real-time sub-problem" using the ADMM algorithm. The master problem solves the day-ahead global optimal plan, while the sub-problem solves the real-time fine-tuning strategy under the boundary constraints of the master problem. The two-layer model achieves coordinated convergence through iterative updates of dual variables.

[0048] Solution tools: The main problem uses the branch and bound algorithm to solve the mixed integer linear programming problem, with a solution accuracy of 1e-6 and an upper limit of 1000 iterations; due to real-time requirements, the subproblems use the fast solution mode of Gurobi, with a single solution time of ≤5 seconds, meeting the minute-level control requirements.

[0049] 2. Output dimension of regulation strategy The basic instructions are flexible resource adjustment instructions for each distribution area, including energy storage charging / discharging power curves, controllable load start / stop time periods / adjustment amounts, and electric vehicle charging power limits; The core curve is the power exchange curve of each cross-regional line, output in 1-minute granularity, including the day-ahead planned curve and the real-time correction curve; The safety configuration is the backup capacity configuration plan for each distribution area, which configures the spinning / cold backup capacity at a 95% confidence level to cope with uncertain disturbances; The power quality strategy is a three-phase balance regulation strategy, which adjusts the output power according to the phase to ensure that the three-phase imbalance is ≤2% and the voltage deviation is ≤±2%. In addition, it includes a verification report, which outputs the constraint satisfaction verification results after the strategy is implemented. The report compares the actual values ​​and limits of line transmission capacity, voltage amplitude, and three-phase imbalance to ensure that the strategy can be implemented.

[0050] This invention fills the gaps in current flexible resource cross-regional interactive regulation by addressing insufficient quantification of multi-source uncertainties and incomplete consideration of distribution network constraints. Traditional regulation methods fail to fully couple the uncertainties and constraints of sources, loads, equipment, and the power grid, resulting in poor adaptability of regulation strategies and an inability to cope with complex and ever-changing operating scenarios. This invention accurately characterizes the joint distribution of multi-source uncertainties using a Gaussian mixture model, establishes a standardized four-level interactive framework across distribution areas, establishes a mapping relationship between uncertainties and regulation responses, and constructs a two-layer optimization model incorporating multiple constraints to solve for the optimal strategy. Under the premise of ensuring the safety constraints of the distribution network, this significantly improves the efficiency of cross-regional resource allocation and the capacity for renewable energy absorption.

[0051] Example 2 This embodiment provides a flexible resource cross-regional interactive optimization and control device, such as... Figure 4 As shown, the method for implementing Embodiment 1 includes: Data Acquisition and Modeling Module 100: Acquires multi-source uncertainty data on flexible resources across distribution areas and the operation of the distribution network. Through classification modeling, parameter optimization and joint synthesis of various uncertainty factors, a joint probability distribution model of multi-source uncertainty is constructed. Uncertainty Handling Module 200: Builds a four-level interactive framework for flexible resource clusters in multiple distribution areas, clarifies the data interaction interfaces and information interaction mechanisms at each level, establishes a correlation model between cross-distribution area line power flow and uncertain disturbances based on the AC power flow sensitivity matrix, incorporates multiple constraints of the distribution network, derives the joint probability density function of cross-distribution area power flow, and realizes uncertainty transmission mapping; The regulation model construction and solution module 300: constructs a day-ahead stochastic optimization regulation model and a real-time collaborative look-ahead optimization regulation model, sets a two-layer optimization objective and multiple constraints, including uncertainty scenario constraints determined based on the joint probability density function of cross-regional power flow, jointly solves the two optimization regulation models, and outputs the optimal cross-regional interactive regulation strategy; The uncertainty processing module 200 receives the multi-source uncertainty joint probability distribution model output by the data acquisition and modeling module 100, and the control model construction and solution module 300 receives the cross-regional power flow joint probability density function output by the uncertainty processing module 200, thereby realizing data interaction between models.

[0052] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the described module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0053] Example 3 The electronic device of this invention includes a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in read-only memory (ROM) or loaded from a storage unit into random access memory (RAM). The RAM may also store various programs and data required for device operation. The CPU, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0054] Multiple components in the device are connected to the I / O interface, including: input units such as keyboards and mice; output units such as various types of displays and speakers; storage units such as disks and optical discs; and communication units such as network interface cards (NICs), modems, and wireless transceivers. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0055] The processing unit executes the various methods and processes described above, such as methods S1 to S3. For example, in some embodiments, methods S1 to S3 may be implemented as computer software programs tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via ROM and / or a communication unit. When the computer program is loaded into RAM and executed by the CPU, one or more steps of methods S1 to S3 described above may be performed. Alternatively, in other embodiments, the CPU may be configured to execute methods S1 to S3 by any other suitable means (e.g., by means of firmware).

[0056] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0057] The program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0058] In the context of this invention, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can include, but are not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0059] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A flexible resource cross-regional interactive optimization and control method, characterized in that, Includes the following steps: S1. Acquire multi-source uncertainty data of flexible resources across distribution areas and distribution network operation. Classify the uncertainty factors into four categories: source-side fluctuations, load-side fluctuations, equipment response deviations, and grid constraint fluctuations. Construct a multi-source uncertainty joint probability distribution model by classifying, modeling, optimizing, and jointly synthesizing the four types of uncertainty factors. S2, build a four-level interactive framework for flexible resource clusters in multiple distribution areas, establish a correlation model between cross-distribution area line power flow and uncertain disturbances based on the AC power flow sensitivity matrix, and incorporate multiple constraints of the distribution network to derive the joint probability density function of cross-distribution area power flow and realize uncertainty transmission mapping; S3. Construct a day-ahead stochastic optimization control model and a real-time collaborative look-ahead optimization control model, set a dual-layer optimization objective and multiple constraints, including uncertainty scenario constraints determined by the joint probability density function of cross-regional power flow, jointly solve the two optimization control models, and output the optimal cross-regional interactive control strategy.

2. The flexible resource cross-regional interactive optimization and control method according to claim 1, characterized in that, The multi-source uncertainty data includes: distributed photovoltaic power output fluctuation data, electric vehicle charging load fluctuation data, energy storage device response deviation data, inter-station line transmission capacity constraint data, distribution network voltage amplitude constraint data, three-phase imbalance constraint data, and controllable load adjustment deviation data.

3. The flexible resource cross-regional interactive optimization and control method according to claim 1, characterized in that, S1 specifically includes the following steps: S11, acquire cross-regional flexible resources and multi-source uncertainty data of distribution network operation; S12 classifies multi-source uncertainties into four categories: source-side fluctuations, load-side fluctuations, equipment response deviations, and grid constraint fluctuations. These are manifested as random disturbances in cross-regional power exchange. Gaussian mixture models are established for the power disturbances under the influence of the four categories of factors. S13, decompose the four types of deviation vectors corresponding to source-side fluctuations, load-side fluctuations, equipment response deviations, and grid constraint fluctuations into steady-state components and fluctuation components; solve the parameters of the four Gaussian mixture models by the expectation-maximization algorithm, and synthesize the four types of deviation vectors into a unified random variable based on the parameter optimization results, and establish the probability distribution model of the unified random variable, namely the multi-source uncertainty joint probability distribution model.

4. The flexible resource cross-regional interactive optimization and control method according to claim 3, characterized in that, The Gaussian mixture model is expressed as follows: ； ； ； ； in, The cross-regional power deviation is caused by source-side fluctuations. It is power deviation The probability density function; This represents the number of sub-models in the Gaussian mixture model corresponding to the source-side fluctuations. In the Gaussian mixture model corresponding to the source-side fluctuations, the first... The weights of each sub-model; The first Gaussian mixture model corresponding to the source-side fluctuations A single Gaussian distribution, in which The mean of this single Gaussian distribution represents the central tendency of the source-side fluctuation power deviation. Let be the covariance matrix of this single Gaussian distribution, representing the degree of dispersion of the power deviation; The power deviation across distribution zones is caused by load-side fluctuations. Power deviation across distribution zones caused by load-side fluctuations The probability density function; This represents the number of sub-models in the Gaussian mixture model corresponding to the load-side fluctuations. In the Gaussian mixture model corresponding to the load-side fluctuations, the first... The weights of each Gaussian sub-model; The first Gaussian mixture model corresponding to the load-side fluctuation A single Gaussian distribution, in which The mean of this single Gaussian distribution represents the central tendency of the load-side fluctuating power deviation. Let be the covariance matrix of this single Gaussian distribution, representing the degree of dispersion of the load-side fluctuation power deviation; This refers to the cross-regional power deviation caused by equipment response deviation. Cross-regional power deviation caused by equipment response deviation The probability density function; The number of sub-models in the Gaussian mixture model corresponding to the device response deviation; In the Gaussian mixture model corresponding to the device response deviation, the first... The weights of each Gaussian sub-model; In the Gaussian mixture model corresponding to the device response deviation, the first... A single Gaussian distribution, in which It is the mean of this single Gaussian distribution, representing the central tendency of the equipment response deviation power. It is the covariance matrix of the single Gaussian distribution, representing the degree of dispersion of the device response deviation power; This refers to the cross-regional power exchange deviation caused by grid constraint fluctuations. Cross-regional power deviation caused by grid constraint fluctuations The probability density function; This represents the number of sub-models in the Gaussian mixture model corresponding to power grid constrained fluctuations. In the Gaussian mixture model corresponding to power grid constrained fluctuations, the first... The weights of each Gaussian sub-model; In the Gaussian mixture model corresponding to power grid constrained fluctuations, the first... A single Gaussian distribution, in which It is the mean of this single Gaussian distribution, representing the central tendency of grid-constrained fluctuating power. It is the covariance matrix of the single Gaussian distribution, representing the degree of dispersion of the power grid constraint fluctuation power.

5. The flexible resource cross-regional interactive optimization and control method according to claim 4, characterized in that, The aforementioned multi-source uncertainty joint probability distribution model is expressed as: ； in, The unified random variable resulting from the synthesis of four types of biases The probability density function describes the overall distribution of cross-regional power deviation under comprehensive uncertainty; The cross-regional power deviation is a unified random variable synthesized from four types of deviations: source side, load side, equipment side, and grid constraint. The number of sub-models in the multi-source uncertainty joint probability distribution model is equal to the product of the number of Gaussian sub-models for each of the four types of factors, i.e. ; The first multi-source uncertainty joint probability distribution model The weights of each Gaussian sub-model are equal to the product of the weights of the corresponding sub-models for the four categories of factors, i.e. ; The first multi-source uncertainty joint probability distribution model There are single Gaussian distributions, among which... This is the combined mean vector. This is the combined covariance matrix.

6. The flexible resource cross-regional interactive optimization and control method according to claim 1, characterized in that, The S2 specifically includes the following steps: S21 establishes a four-level interactive framework consisting of distribution network dispatch center, resource aggregator, transformer area management node, and resource entity. The distribution network dispatch center is responsible for overall strategy formulation, the resource aggregator is responsible for resource aggregation and instruction distribution, the transformer area management node executes fine-grained control, and the resource entity responds to control instructions. S22, Based on the AC power flow sensitivity matrix, a correlation model between power flow and uncertain disturbances on cross-transformer lines is established. Cross-regional trend Represented as: ； in, For cross-regional lines The AC power flow sensitivity matrix quantifies the transmission intensity of multi-source uncertainty disturbances to cross-regional power exchange. The cross-regional power deviation is a unified random variable synthesized from four types of deviations: source side, load side, equipment side, and grid constraints, representing the uncertainty disturbance. S23, based on the linear transformation property of Gaussian distribution, for a unified random variable A linear transformation is performed on the multi-source uncertainty joint probability distribution model to derive the cross-regional power flow. joint probability density function : ； in, The number of sub-models in the multi-source uncertainty joint probability distribution model; The first multi-source uncertainty joint probability distribution model The weights of each Gaussian sub-model; The first step is to perform a linear transformation of the multi-source uncertainty joint probability distribution model. A single Gaussian distribution, For the line The mean vector of the current. For the line The covariance matrix of the current flow; S24, combining distribution network constraints, including line transmission capacity, voltage amplitude, and three-phase imbalance, transforms deterministic constraints into probabilistic constraints, quantifies the pre-set confidence level of constraint satisfaction, and achieves a complete mapping from the distribution of uncertain sources to the feasible control domain across distribution areas. Specifically, it includes: Line transmission capacity constraints: ,in, For the line The rated transmission power is required at a confidence level. Down, The probability of exceeding the constraint interval is less than ; Voltage amplitude constraint: based on power flow-voltage sensitivity matrix Derive the node voltage of the distribution area ,constraint ,in, For nodes t voltage, For nodes t The initial voltage, These are the upper and lower limits of the allowable node voltage, and the requirements are as follows: at a confidence level... Below, the probability of voltage exceeding the limit is less than ; Three-phase imbalance constraint: based on cross-regional three-phase power exchange Calculate the three-phase unbalance. ,in, For the line The three-phase active power vector, The lines are respectively The active power of phases A, B, and C in the cross-regional distribution area. To allow for the maximum imbalance, a certain confidence level is required. Below, the probability of exceeding the imbalance limit is less than .

7. The flexible resource cross-regional interactive optimization and control method according to claim 1, characterized in that, The S3 specifically includes the following steps: S31. Construct a day-ahead stochastic optimization control model with the objective function of minimizing the comprehensive cost of cross-regional interaction and maximizing the renewable energy consumption rate. Set multiple constraints, including resource regulation capacity constraints within the region, transmission capacity constraints of cross-regional lines, voltage amplitude constraints, three-phase imbalance constraints, and uncertainty scenario constraints. Use the key scenario identification method to solve the day-ahead stochastic optimization control model to obtain the day-ahead cross-regional interaction plan. S32, construct a real-time collaborative forward-looking optimization control model. Based on real-time uncertainty disturbance observations, with the goal of minimizing the tracking deviation of the day-ahead cross-regional interaction plan and achieving the best power fluctuation smoothing effect, a rolling optimization time window is introduced. The window duration is determined according to the real-time control requirements and computing resource allocation to dynamically adjust the cross-regional power exchange strategy. S33 uses the alternating direction multiplier method and the branch and bound algorithm to solve the two-layer optimization model consisting of the daytime stochastic optimization control model and the real-time collaborative look-ahead optimization control model, and outputs the optimal cross-regional interactive control strategy. The optimal cross-regional interactive control strategy includes flexible resource adjustment instructions for each region, cross-regional power exchange curves, reserve capacity configuration schemes and three-phase balance control strategies. Specifically, the two-layer optimization model is decomposed into a day-ahead master problem and a real-time subproblem by using the alternating direction multiplier method. The master problem is solved by the branch and bound algorithm to solve the mixed integer linear programming problem and output the day-ahead global optimal plan. The subproblem is solved by the fast solution mode of Gurobi under the boundary constraints of the master problem and the real-time fine-tuning strategy is adopted. The two-layer model achieves coordinated convergence through iterative updates of dual variables.

8. The flexible resource cross-regional interactive optimization and control method according to claim 1, characterized in that, In the aforementioned daytime stochastic optimization control model, the objective function is expressed as: ； ； in, To improve the comprehensive absorption rate of new energy in multiple districts, This represents the total number of scheduling periods in the previous day. The total number of stations. for Time period The photovoltaic output consumed by the substation for Time period Total photovoltaic output in the area; For electricity purchase costs, , for Peak-valley electricity pricing for different time periods for Taiwan Electricity purchased online during specific time periods; To cover the cost of light curtailment, , The cost of per unit of abandoned light penalty for Time period The photovoltaic output consumed by the substation for Time period Total photovoltaic output in the area; In order to regulate compensation costs, , The unit resource adjustment compensation electricity price, for Taiwan Flexible resource adjustment during different time periods; Multiple constraints include: Resource regulation capacity constraints within the distribution area: ; Transmission capacity constraints for cross-regional lines: ; Voltage amplitude constraint: ; Three-phase unbalance constraint: ; Uncertainty scenario constraints: Based on the joint probability density function of cross-regional power flow derived by S2, typical scenarios at a 95% confidence level are selected to ensure that the constraints are satisfied in each scenario; in, for Taiwan The upper and lower limits of flexible resource output such as time-of-use energy storage / controllable load. , Determined by the physical characteristics of the equipment, for Time-of-day routes Cross-regional power, For the line Rated transmission power, To account for the safety margin factor of uncertainty, for Taiwan Time period node voltage, These are the upper and lower limits of the allowable node voltage, respectively. for Taiwan Three-phase imbalance over time period To allow the maximum degree of imbalance.

9. The flexible resource cross-regional interactive optimization and control method according to claim 1, characterized in that, The objective function of the real-time collaborative look-ahead optimization and control model is expressed as: ； ； in, To address the recent discrepancy in the tracking of the cross-regional interaction plan, This represents the number of time periods within the scrolling window. This represents the total number of cross-regional lines. for Time-of-day routes Real-time power adjustment The planned power output for the day is [number]. The standard deviation of cross-regional power exchange. Lines within the scrolling window The average power.

10. A flexible resource cross-regional interactive optimization and control device, characterized in that, For implementing the method as described in any one of claims 1 to 9, comprising: Data acquisition and modeling module: Acquires multi-source uncertainty data of flexible resources across distribution areas and distribution network operation, classifies uncertainty factors into four categories: source-side fluctuations, load-side fluctuations, equipment response deviations, and grid constraint fluctuations, and constructs a multi-source uncertainty joint probability distribution model by classifying, modeling, optimizing parameters, and jointly synthesizing the four types of uncertainty factors. Uncertainty handling module: Build a four-level interactive framework for flexible resource clusters in multiple distribution areas, establish a correlation model between cross-distribution area line power flow and uncertain disturbances based on the AC power flow sensitivity matrix, incorporate multiple constraints of the distribution network, derive the joint probability density function of cross-distribution area power flow, and realize uncertainty transmission mapping; The module for constructing and solving the control model is as follows: It constructs a day-ahead stochastic optimization control model and a real-time collaborative look-ahead optimization control model, sets a two-level optimization objective and multiple constraints, including uncertainty scenario constraints determined by the joint probability density function of cross-regional power flow, jointly solves the two optimization control models, and outputs the optimal cross-regional interactive control strategy. The uncertainty processing module receives the multi-source uncertainty joint probability distribution model output by the data acquisition and modeling module, and the control model construction and solution module receives the cross-regional power flow joint probability density function output by the uncertainty processing module, thereby realizing data interaction between models.

Images