A distribution network flexibility resource regulation method and device considering network constraints and heterogeneity, a terminal device, and a storage medium

Abstract

The application discloses a kind of considering network constraint and flexibility resource regulation method, device and heterogeneity of distribution network, terminal equipment and storage medium belong to flexible resource aggregation technical field, the method includes: first, distribution network aggregation node, energy storage, dispatchable load, internal combustion engine group parameter and dispatching duration are collected, with the maximum node power adjustable range as target, construct objective function and flexibility resource operation constraint set;Solving the objective function obtains the upper and lower limit of each node optimal power, generates node time period decoupling rectangular feasible region.Minkowski summation is obtained to the total node feasible region, after neural network safety evaluation, to unsafe boundary iteration contraction, obtain safe final total feasible region;Accordingly, complete feeder outlet global regulation and distributed energy cluster local regulation.By the application, the problem that the existing technology is prone to cause grid overrun in actual dispatch execution process can be solved.

Description

Technical Field

[0001] This invention relates to the technical field of flexible resource aggregation technology, and in particular to a method, apparatus, terminal equipment, and storage medium for flexible resource regulation of distribution networks that takes into account network constraints and heterogeneity. Background Technology

[0002] With the construction of new power systems, a large number of heterogeneous and flexible resources such as energy storage, temperature-controlled loads, and distributed photovoltaics have emerged in the distribution network. If these resources can be effectively integrated to participate in the dispatch of the upper-level power grid, the flexibility of the power system can be improved. However, due to their heterogeneity and geographically dispersed characteristics, they face many technical challenges that restrict the realization of their value and affect the safe operation of the distribution network.

[0003] Existing technologies mainly employ a simple parameter overlay method to integrate heterogeneous flexible resources to meet the dispatching requirements of the upper-level power grid. However, simple parameter overlay does not fully consider the relevant constraints of various flexible resources, which can easily lead to grid over-limit problems during actual dispatching and execution, thereby causing safety hazards. Summary of the Invention

[0004] This invention provides a method, apparatus, terminal equipment, and storage medium for distribution network flexibility resource regulation that takes into account network constraints and heterogeneity. The method can solve the problem of power grid over-limits that are easily caused in the actual scheduling process of existing technologies.

[0005] An embodiment of the present invention provides a method for flexible resource regulation of distribution networks that considers network constraints and heterogeneity, comprising: Obtain the number of nodes participating in aggregation in the distribution network, energy storage system parameters, dispatchable load parameters, internal combustion engine unit parameters, and planned dispatch duration; Based on the energy storage system parameters, demand response parameters, internal combustion engine parameters, and planned scheduling duration, and with the goal of maximizing the node power range of each node, a corresponding objective function and set of constraints are constructed; wherein, the set of constraints includes: energy storage constraints, controllable load constraints, and internal combustion engine constraints. Under the set of constraints, the objective function is solved to obtain the optimal upper limit vector and the optimal lower limit vector of the node power of each node in each time period when the function value of the objective function is maximized. Based on the optimal upper bound vector and optimal lower bound vector of the node power of each node in each time period, generate the time period decoupling rectangular feasible region for each node. The initial aggregated feasible region at the feeder outlet is obtained by performing Minkowski summation on the time-decoupled rectangular feasible region of all nodes. For the boundary power points in the initial aggregated feasible region, the power information corresponding to the boundary power points is input into the trained neural network for safety evaluation. If there are boundary power points that are unsafe according to the safety evaluation, the initial aggregated feasible region is subjected to gradient iterative shrinkage to obtain the final total feasible region that meets the safety threshold. The feeder outlet is regulated based on the final total feasible region; the distributed energy cluster is regulated based on the time period of the node decoupled rectangular feasible region.

[0006] Furthermore, the controllable load constraints include shiftable load constraints, transferable load constraints, and interruptible load constraints. The energy storage constraints are constructed using the following formula: ; ; ; in, for The state of charge of the stored energy at the end of the time period; for The state of charge of the stored energy at the end of the time period; Energy storage coefficient; For energy storage charging efficiency; The discharge efficiency of energy storage; This refers to the rated capacity of the energy storage. yes Time-limited energy storage charging and discharging power; Indicates energy storage and discharge; Indicates charging; This represents the lower limit of energy storage capacity. Represents the upper limit of energy storage capacity; represent The upper limit of charging power for time-limited energy storage; represent The upper limit of discharge power for time-limited energy storage; For time step; The first in the total scheduling period Each time period; The transferable load constraint condition is constructed using the following formula: ; ; ; in, The schedulable time period for loads that can be shifted; The planned scheduling duration; The allowable start time for dispatching of loads that can be moved; Allowable termination time for dispatchable loads; for The start status of the time period, when set to 1 indicates The load is started during the designated time period; As an intermediate variable, 0 indicates The load is not in operation during the period, represented by a value of 1. The load for that period is already in operation; This refers to the electrical power consumption of a transferable load when it is in operation. That is The original planned power consumption of the load that can be shifted during the time period; This represents the deviation of the actual power consumption of the transferable load during time period t from the originally planned power. The transferable load constraint is constructed using the following formula: ; ; in, for The maximum permissible upward adjustment of load during a given period relative to its original electricity consumption plan; for The maximum permissible reduction in the power of the load that can be transferred during a given period relative to its original power consumption plan; Allowable start time for dispatching transferable loads; Allowable termination time for dispatchable load; For transferable loads in Actual power consumption after adjustment during the time period; The interruptible load constraint is constructed using the following formula: ; in, for The reduction limit for interruptible loads during certain periods is based on the original electricity consumption plan. for The upward adjustment limit for interruptible loads during certain periods is based on the original electricity consumption plan; interruptible load Actual power during the time period; The internal combustion engine constraint conditions are constructed using the following formula: ; ; in, This refers to the minimum permissible output power of an internal combustion engine; This refers to the maximum permissible output power of the internal combustion engine; For internal combustion engines Output power during the time period; For internal combustion engines Output power during the time period; Limits on the rate of increase of internal combustion engine units; Limits the rate of decline for internal combustion engine units.

[0007] Furthermore, the objective function is constructed in the following manner: ; ; in, For nodes exist Lower limit of aggregation power over a given time period; For nodes exist Maximum aggregation power for a given time period; This is the lower power limit vector for each node in each time period; This is the upper limit vector of power for each node in each time period; This represents the power of the node at different time periods; for Weighting coefficients for time-of-day flexibility value; The feasible region is a set of constraints including energy storage constraints, controllable load constraints, and internal combustion engine constraints. This represents the total number of scheduling periods; Decouple the node into a rectangular feasible region for the time period.

[0008] Furthermore, in the case of boundary power points deemed unsafe by the safety assessment, gradient iterative shrinkage is performed on the initial aggregated feasible region, including: If there are boundary power points that are deemed unsafe in the safety assessment, repeat the following gradient iterative shrinkage operation until the final total feasible region that meets the safety threshold is obtained: Get the current penalty factor; the initial penalty factor is a preset value. The current sensitivity factor is calculated based on the safe probability output of the current target boundary power point and the power of all sampling points in the preliminary aggregated feasible region. Calculate the updated boundary power point based on the current target boundary power point, the current penalty factor, and the current sensitivity factor; The updated power information of the boundary power points is input into the trained neural network for security assessment. If the security assessment is safe, the final total feasible region is generated based on the updated boundary power points; If the security assessment is unsafe, a linear growth strategy is adopted to adjust the current penalty factor with a fixed step size to obtain the updated current penalty factor.

[0009] Further, the training of the neural network includes: Acquire several power point training samples that include power information and labeled safety status tags; wherein, the safety status tags are obtained through power flow equation verification. If the power flow equation verification passes, the tag is set as the first identifier; otherwise, the tag is set as the second identifier. Each power point training sample is input into the neural network to be trained one by one, and iterative training is performed until the loss function converges, thus obtaining the trained neural network. In each training session, a predicted safety probability value is generated based on the power information in the power point training samples; the predicted safety probability value is compared with the labeled safety status label, and the loss function value is calculated based on the comparison result.

[0010] An embodiment of the present invention also provides a distribution network flexibility resource regulation device that considers network constraints and heterogeneity, including: a data acquisition module, a node-level model construction module, a solution module, a time-period decoupling feasible region generation module, a feeder-level preliminary aggregation module, a security verification module, and a regulation module; The data acquisition module is used to acquire the number of nodes participating in aggregation in the distribution network, energy storage system parameters, dispatchable load parameters, internal combustion engine unit parameters, and planned scheduling duration; The node-level model construction module is used to construct the corresponding objective function and constraint set based on the energy storage system parameters, demand response parameters, internal combustion engine parameters, and planned scheduling duration, with the goal of maximizing the node power range of each node; wherein, the constraint set includes: energy storage constraints, controllable load constraints, and internal combustion engine constraints. The solution module is used to solve the objective function under the set of constraints, and obtain the optimal upper limit vector and the optimal lower limit vector of the node power of each node in each time period when the function value of the objective function is maximized. The time-period decoupling feasible region generation module is used to generate the time-period decoupling rectangular feasible region for each node based on the optimal upper limit vector and the optimal lower limit vector corresponding to the node power of each node in each time period. The feeder-level preliminary aggregation module is used to perform Minkowski summation on the time-decoupled rectangular feasible regions of all nodes to obtain the preliminary aggregated feasible region at the feeder outlet. The safety verification module is used to input the power information corresponding to the boundary power points in the preliminary aggregated feasible region into the trained neural network for safety evaluation. If there are boundary power points whose safety evaluation results are unsafe, the preliminary aggregated feasible region is subjected to gradient iterative shrinkage to obtain the final total feasible region that meets the safety threshold. The control module is used to control the feeder outlet based on the final total feasible region; and to control the distributed energy cluster based on the time period decoupling rectangular feasible region of the node.

[0011] Furthermore, the controllable load constraints include shiftable load constraints, transferable load constraints, and interruptible load constraints. The node-level model construction module includes: an energy storage constraint construction unit, a transferable load constraint construction unit, a transferable load constraint construction unit, an interruptible load constraint construction unit, and an internal combustion engine constraint construction unit. The energy storage constraint construction unit is used to construct the energy storage constraint using the following formula: ; ; ; in, for The state of charge of the stored energy at the end of the time period; for The state of charge of the stored energy at the end of the time period; Energy storage coefficient; For energy storage charging efficiency; The discharge efficiency of energy storage; This refers to the rated capacity of the energy storage. yes Time-limited energy storage charging and discharging power; Indicates energy storage and discharge; Indicates charging; This represents the lower limit of energy storage capacity. Represents the upper limit of energy storage capacity; This represents the upper limit of the charging power of energy storage; Represents the upper limit of the discharge power of energy storage; For time step; The first in the total scheduling period Each time period; The transferable load constraint construction unit is used to construct the transferable load constraint using the following formula: ; ; ; in, The schedulable time period for loads that can be shifted; The planned scheduling duration; The allowable start time for dispatching of loads that can be moved; Allowable termination time for dispatchable loads; for The start status of the time period, when set to 1 indicates The load is started during the designated time period; As an intermediate variable, 0 indicates The load is not in operation during the period, represented by a value of 1. The load for that period is already in operation; This refers to the electrical power consumption of a transferable load when it is in operation. That is The original planned power consumption of the load that can be shifted during the time period; This represents the deviation of the actual power consumption of the transferable load during time period t from the originally planned power. The transferable load constraint construction unit is used to construct the transferable load constraint using the following formula: ; ; in, for The maximum permissible upward adjustment of load during a given period relative to its original electricity consumption plan; for The maximum permissible reduction in the power of the load that can be transferred during a given period relative to its original power consumption plan; Allowable start time for dispatching transferable loads; Allowable termination time for dispatchable load; For transferable loads in Actual power consumption after adjustment during the time period; The interruptible load constraint construction unit is used to construct the interruptible load constraint using the following formula: ; in, The limit for the reduction of interruptible load during time period t, based on the original electricity consumption plan; The upper limit for interruptible load during time period t is based on the original electricity consumption plan. The actual power of the interruptible load during time period t; The internal combustion engine constraint construction unit is used to construct the internal combustion engine constraint conditions using the following formula: ; ; in, This refers to the minimum permissible output power of an internal combustion engine; This refers to the maximum permissible output power of the internal combustion engine; For internal combustion engines Output power during the time period; For internal combustion engines Output power during the time period; Limits on the rate of increase of internal combustion engine units; Limits the rate of decline for internal combustion engine units.

[0012] Furthermore, the node-level model construction module also includes: an objective function construction unit; The objective function construction unit is used to construct the objective function using the following formula: ; ; in, For nodes exist Lower limit of aggregation power over a given time period; For nodes exist Maximum aggregation power for a given time period; This is the lower power limit vector for each node in each time period; This is the upper limit vector of power for each node in each time period; This represents the power of the node at different time periods; for Weighting coefficients for time-of-day flexibility value; The feasible region is a set of constraints including energy storage constraints, controllable load constraints, and internal combustion engine constraints. This represents the total number of scheduling periods; Decouple the node into a rectangular feasible region for the time period.

[0013] This application also provides a terminal device, including: One or more processors; A memory, coupled to the processor, for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement a resource regulation method for distribution networks that takes into account network constraints and heterogeneity, as described in the above embodiments of the invention.

[0014] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a resource regulation method for distribution network flexibility that considers network constraints and heterogeneity as described in the above embodiments.

[0015] The following benefits can be obtained by implementing the present invention: This invention discloses a multi-layer aggregation method for distribution network flexibility resources that considers network constraints and heterogeneity. The core of this method is to first obtain relevant parameters of the distribution network and various flexibility resources, then construct an objective function and various resource constraints with the goal of maximizing the node power range of each node. Solving for the maximum inscribed rectangle of each node yields the initial total feasible region at the feeder outlet using Minkowski summation. A trained neural network is then used to evaluate safety and iteratively shrink the region using gradients to obtain the final total feasible region that meets the safety threshold. Finally, based on this feasible region and the optimal power vector of each node, the feeder outlet and distributed energy clusters are regulated respectively. This method, by constructing a set of constraints including various resources and combining it with a neural network safety evaluation iteration mechanism, can effectively avoid grid over-limit problems in actual dispatching, eliminate safety hazards to ensure the safe and stable operation of the distribution network, and facilitate the efficient integration of heterogeneous flexibility resources such as energy storage, temperature-controlled loads, and internal combustion engines, fully leveraging their value and enabling them to participate in the dispatching of the upper-level power grid. Attached Figure Description

[0016] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating a method for adjusting the flexibility of distribution network resources that takes into account network constraints and heterogeneity, provided in an embodiment of this application.

[0018] Figure 2 This is a schematic diagram of the structure of a distribution network flexibility resource regulation device that takes into account network constraints and heterogeneity, according to an embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0021] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0022] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0023] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0024] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0025] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0026] See Figure 1To address the issue of grid over-limit problems easily arising during actual dispatching in existing technologies, an embodiment of the present invention provides a distribution network flexibility resource regulation method considering network constraints and heterogeneity, comprising: S1. Obtain the number of nodes participating in aggregation in the distribution network, energy storage system parameters, dispatchable load parameters, internal combustion engine unit parameters, and planned dispatch duration; Specifically, energy storage system parameters include energy storage charging efficiency, energy storage discharging efficiency, energy storage rated capacity, energy storage lower limit capacity, energy storage upper limit capacity, energy storage charging power upper limit, and energy storage discharging power upper limit; dispatchable load parameters include dispatchable time periods for shiftable loads, original planned power consumption of shiftable loads, maximum allowable upward adjustment power of shiftable loads relative to their original power consumption plan, maximum allowable downward adjustment power of shiftable loads relative to their original power consumption plan, downward adjustment limit of interruptible loads based on the original power consumption plan, and upward adjustment limit of interruptible loads based on the original power consumption plan; internal combustion engine unit parameters include minimum allowable output power of internal combustion engines, maximum allowable output power of internal combustion engines, upward rate limit of internal combustion engine units, and downward rate limit of internal combustion engine units.

[0027] S2. Based on the energy storage system parameters, demand response parameters, internal combustion engine parameters, and planned scheduling duration, construct the corresponding objective function and constraint set with the goal of maximizing the node power range of each node; wherein, the constraint set includes: energy storage constraints, controllable load constraints, and internal combustion engine constraints. Preferably, the controllable load constraints include shiftable load constraints, transferable load constraints, and interruptible load constraints; The energy storage constraints are constructed using the following formula: ; ; ; in, for The state of charge of the stored energy at the end of the time period; for The state of charge of the stored energy at the end of the time period; Energy storage coefficient; For energy storage charging efficiency; The discharge efficiency of energy storage; This refers to the rated capacity of the energy storage. yes Time-limited energy storage charging and discharging power; Indicates energy storage and discharge; Indicates charging; This represents the lower limit of energy storage capacity. Represents the upper limit of energy storage capacity; represent The upper limit of charging power for time-limited energy storage; represent The upper limit of discharge power for time-limited energy storage; For time step; The first in the total scheduling period Each time period; The transferable load constraint condition is constructed using the following formula: ; ; ; in, The schedulable time period for loads that can be shifted; The planned scheduling duration; The allowable start time for dispatching of loads that can be moved; Allowable termination time for dispatchable loads; for The start status of the time period, when set to 1 indicates The load is started during the designated time period; As an intermediate variable, 0 indicates The load is not in operation during the period, represented by a value of 1. The load for that period is already in operation; This refers to the electrical power consumption of a transferable load when it is in operation. That is The original planned power consumption of the load that can be shifted during the time period; This represents the deviation of the actual power consumption of the transferable load during time period t from the originally planned power. The transferable load constraint is constructed using the following formula: ; ; in, for The maximum permissible upward adjustment of load during a given period relative to its original electricity consumption plan; for The maximum permissible reduction in the power of the load that can be transferred during a given period relative to its original power consumption plan; Allowable start time for dispatching transferable loads; Allowable termination time for dispatchable load; For transferable loads in Actual power consumption after adjustment during the time period; The interruptible load constraint is constructed using the following formula: ; in, for The reduction limit for interruptible loads during certain periods is based on the original electricity consumption plan. for The upward adjustment limit for interruptible loads during certain periods is based on the original electricity consumption plan; interruptible load Actual power during the time period; The internal combustion engine constraint conditions are constructed using the following formula: ; ; in, This refers to the minimum permissible output power of an internal combustion engine; This refers to the maximum permissible output power of the internal combustion engine; For internal combustion engines Output power during the time period; For internal combustion engines Output power during the time period; Limits on the rate of increase of internal combustion engine units; Limits the rate of decline for internal combustion engine units.

[0028] Specifically, flexible resources are divided into energy storage, controllable loads, and internal combustion engine units, with the following constraints: The energy storage constraints are constructed using the following formula: ; ; ; Among them, the first item This is a dynamic model for updating the energy storage capacity over time periods, used to describe the change in the energy storage's state of charge with charging and discharging power. for The state of charge of the stored energy at the end of the time period; for The state of charge of the stored energy at the end of the time period; For energy storage charging efficiency; The discharge efficiency of energy storage; Indicates energy storage and discharge; Indicates charging; Energy storage coefficient: when Take charging efficiency The power consumption increases; when Discharge efficiency conversion The battery level decreased; This refers to the rated capacity of the energy storage. yes Bidirectional charging and discharging power during the same period; For time step; The first in the total scheduling period Each time period.

[0029] Second item This is a battery charge safety constraint used to limit the real-time state of charge of the energy storage device to a reasonable and safe range, effectively preventing overcharging and over-discharging, and avoiding damage to the energy storage battery. This represents the lower limit of energy storage capacity. This represents the upper limit of energy storage capacity.

[0030] Third item This is a real-time power operation constraint for energy storage, used to limit the amplitude of energy storage charging and discharging power to not exceed the rated limit. represent The upper limit of charging power for time-limited energy storage; represent The upper limit of discharge power for time-limited energy storage. The constraint range spans the negative charging range, the zero-power shutdown range, and the positive discharging range, and allows bidirectional charging and discharging operation. Only the maximum charging and discharging power is limited, and normal charging and discharging behavior is not restricted.

[0031] The entire set of energy storage constraints comprehensively and rigorously describes the full-condition operation characteristics of energy storage charging, discharging, and resting from three dimensions: dynamic change law of energy quantity, safe range of energy quantity, and real-time charging and discharging power limit. It not only considers the physical dynamic process of energy storage charging and discharging efficiency loss, but also ensures battery safety through upper and lower limits of state of charge, and limits the charging and discharging power to not exceed the rated operating capacity of the equipment through bidirectional upper and lower power limits, forming a complete closed-loop energy storage optimization operation constraint system.

[0032] Next, the controllable load constraints include shiftable load constraints, transferable load constraints, and interruptible load constraints; The transferable load constraint condition is constructed using the following formula: ; ; ; Among them, the first item To constrain the dispatchable time range of the transferable load, ensuring that the load can operate completely within the dispatch window. Duration. The schedulable time period for loads that can be shifted; The required scheduling duration for the plan to be completed; The allowable start time for dispatching of loads that can be moved; The allowable termination time for dispatching of loads that can be moved.

[0033] Second item As a logical constraint for the running state, it is stated that once the load is started, it must run continuously for a full period of time. Duration, the total runtime of the entire scheduling cycle is strictly equal to It can only shift the running time, not increase or decrease the power consumption. And... for The start status of the time period, when set to 1 indicates The load is started during a specific time period, and this is the only start-up time in the entire scheduling cycle; As an intermediate variable, 0 indicates The load is not in operation during the period, represented by a value of 1. The load for this period is already in operation.

[0034] Third item This is the formula for calculating power deviation. This refers to the electrical power consumption of a transferable load when it is in operation. That is The original planned power consumption of the load that can be shifted during the time period; This is the deviation of the actual power consumption of the shiftable load in time period t from the original planned power. A positive value indicates that more power is used than originally planned, and a negative value indicates that less power is used than originally planned. It is used to characterize the power regulation contribution of the load shift to the power grid.

[0035] The complete set of load shifting constraints fully describes the flexible adjustment characteristics of load shifting, which only shifts the electricity consumption period, does not change the total electricity consumption, and operates continuously for a fixed duration, from three dimensions: adjustable time range, start-stop operation rules, and actual power deviation characteristics. It not only limits the time window range of load scheduling, but also standardizes the operation logic of load starting only once, running continuously, and for a fixed duration. At the same time, it quantifies the power adjustment deviation of the load after shifting relative to the original plan, forming a closed-loop complete load shifting flexibility resource optimization scheduling constraint system.

[0036] Next, the transferable load constraint is constructed using the following formula: ; ; Among them, the first item This is a constraint on the real-time power adjustment range of transferable loads in a single time period, used to limit the load power adjustment amplitude in each time period from exceeding the equipment's allowable adjustment limit. for The maximum permissible upward adjustment of load during a given period relative to its original electricity consumption plan; for The maximum allowable reduction in power of the load that can be transferred during a given period relative to its original power consumption plan.

[0037] Second item To maintain the total power conservation constraint during the transferable load dispatch cycle, the cumulative power deviation is 0 throughout the entire adjustable dispatch cycle; this means that the transferable load can only transfer and shift electrical loads between different time periods, and the overall total electricity consumption remains consistent with the original plan, without increasing electricity consumption or reducing total power consumption. Allowable start time for dispatching transferable loads; Allowable termination time for dispatchable load; For transferable loads in Actual power consumption after adjustment during the time period.

[0038] The entire set of transferable load constraints fully describes the flexible adjustment characteristics of transferable loads between time periods while maintaining the overall total electricity consumption, from two dimensions: the upper limit of single-time period adjustment capacity and the conservation of total electricity consumption throughout the entire cycle. It ensures that the power adjustment in each time period does not exceed the allowable range and that the load transfer does not change the overall total electricity consumption, forming a closed-loop and standardized transferable load optimization scheduling constraint system.

[0039] Next, the interruptible load constraint is constructed using the following formula: ; in, for The interruptible load for a given period is the lower limit based on the original electricity consumption plan. The value is negative, representing the interruptible and reduced electricity load. for The upward adjustment limit for interruptible load during a given period, based on the original electricity consumption plan, is a positive value, indicating that the electricity load can be increased. interruptible load Actual power consumption during a given time period. Constraints only limit the adjustable range of power consumption within a single time period, and do not set full-cycle power conservation constraints, allowing direct interruption to reduce load and decrease overall total power consumption.

[0040] Next, the internal combustion engine constraints are constructed using the following formula: ; ; Among them, the first item To constrain the steady-state operating power of the internal combustion engine, the engine only has forward power generation capability, with the power always greater than or equal to 0. There is no reverse charging or grid power absorption operating mode. This constraint limits the engine's power generation capacity to within its rated safe operating range. This refers to the minimum permissible output power of an internal combustion engine; This represents the maximum permissible output power of the internal combustion engine.

[0041] Second item This is a power ramp-up constraint for internal combustion engines in adjacent time periods, used to prevent excessive power surges from impacting the power grid and ensuring smooth and stable power regulation of the unit. For internal combustion engines Output power during the time period; For internal combustion engines Output power during the time period; Limits on the rate of increase of internal combustion engine units; Limits the rate of decline for internal combustion engine units.

[0042] The entire set of internal combustion engine constraints fully describes the unidirectional power generation and stable and adjustable operation characteristics of internal combustion engines from two dimensions: the safe range of steady-state power generation and the limit of dynamic power change rate. It not only limits the power generation of the unit to not exceed the rated operating capacity, but also limits the rate of power increase and decrease to avoid grid impact, forming a closed-loop complete internal combustion engine optimization scheduling constraint system that conforms to the actual physical operating characteristics of the unit.

[0043] This invention constructs operational constraints for various flexible resources, including energy storage, controllable loads, and internal combustion engines, achieving comprehensive safety management from both equipment safety and grid operation safety perspectives. These constraints work together to form a closed loop, ensuring the operational safety of each type of flexible resource and extending equipment lifespan, while also enabling coordinated and orderly scheduling of various resources. This avoids grid safety risks caused by scheduling conflicts between different resources, providing a solid safety guarantee for the integration of flexible resources into the grid and their efficient and stable operation. Simultaneously, it considers the rationality and economy of scheduling, ensuring that the regulatory capacity of flexible resources is fully utilized without affecting the overall safety and stability of the grid.

[0044] Preferably, the objective function is constructed in the following manner: ; ; in, For nodes exist Lower limit of aggregation power over a given time period; For nodes exist Maximum aggregation power for a given time period; This is the lower power limit vector for each node in each time period; This is the upper limit vector of power for each node in each time period; This represents the power of the node at different time periods; for Weighting coefficients for time-of-day flexibility value; The feasible region is a set of constraints including energy storage constraints, controllable load constraints, and internal combustion engine constraints. This represents the total number of scheduling periods; Decouple the node into a rectangular feasible region for the time period.

[0045] Specifically, The objective function targets all flexibility resources connected to the same bus node k, optimizing by maximizing the adjustable range of weighted aggregated power across different time periods. For nodes exist Lower limit of aggregation power over a given period For nodes exist The upper limit of the aggregation power for a time period, and the difference between the two is the single-time node flexibility adjustment capacity; This is the lower power limit vector for each node in each time period; This is the upper limit vector of power for each node in each time period; for The weighting coefficients for time-period flexibility value are used to reflect the degree of power grid demand for flexibility adjustment capabilities in different time periods, and are preset known parameters; This represents the total number of scheduling periods.

[0046] Subsequently, the feasible region of the decoupled rectangle for the node time period is defined as follows: ; It is composed of all heterogeneous flexibility resource physical safety operation constraints, including the energy storage constraints, controllable load constraints, and internal combustion engine constraints mentioned above. It is an irregular feasible domain with complex form and strong coupling across multiple time periods. The rectangular feasible region for time-separation of nodes represents the flexibility capacity of nodes to be completely decoupled along the time axis; containment relationship. This means that the rectangular feasible region obtained by the solution falls completely inside the original feasible region, strictly meets all equipment safety operation constraints, and will not exceed the safety limits of energy storage charging and discharging, load regulation, and internal combustion engine power generation. The actual operating power of the node in each time period is constrained. The operating power of the representative node at each time period must be within the aggregate upper and lower limit range obtained by the solution.

[0047] This invention solves for the maximum inscribed rectangular feasible region of the original complex coupled feasible region by maximizing the weighted flexibility adjustable range of nodes in each time period. This transforms the original strongly coupled and irregular node flexibility feasible region in multiple time periods into a regular rectangular power feasible region in which each time period is independent and completely decoupled. Under the premise of strictly ensuring the safe operation of various flexible resource equipment and the stable operation of the power grid, this invention eliminates the strong coupling relationship between multiple time periods and realizes the simplified aggregation and efficient optimized scheduling of heterogeneous flexible resources of nodes.

[0048] S3. Under the set of constraints, solve the objective function to obtain the optimal upper limit vector and the optimal lower limit vector of the node power of each node in each time period when the function value of the objective function is maximized. S4. Based on the optimal upper limit vector and optimal lower limit vector of the node power of each node in each time period, generate the time period decoupling rectangle feasible region for each node. Specifically, under the constraints of energy storage, controllable load, and internal combustion engine, the objective function for maximizing node flexibility is optimized. Under the condition that the objective function reaches its maximum value and the node can provide the optimal flexibility adjustment capability, the optimal lower power limit vector and the optimal upper power limit vector for each node in each time period are calculated. Then, the optimal upper and lower limit vectors obtained by solving are substituted into the feasible region construction formula to generate the time-period decoupling rectangular feasible region corresponding to each bus node. Under the premise of strictly ensuring the safe operation of various flexibility resources, the inner rectangle approximation of the original complex coupling feasible region of the node is completed, eliminating the strong coupling relationship between multiple time periods.

[0049] S5. Perform Minkowski summation on the time-decoupled rectangular feasible regions of all nodes to obtain the preliminary aggregated feasible region at the feeder outlet. Specifically, since the approximate nodes of the distribution network have been processed into corresponding time-period decoupled rectangular feasible regions, the total feasible region of the feeder outlet can be obtained by performing Minkowski summation on these time-period decoupled rectangular feasible regions.

[0050] ; in, The overall preliminary aggregation feasible region at the feeder outlet; Let be the set of feasible rectangular regions for decoupling Minkowski and operators over time intervals, whose sum is equivalent to the direct summation of boundary vectors; This is the total regulated power vector of the distribution network at the feeder; This represents the total number of nodes or sub-regions participating in aggregation within the distribution network.

[0051] This invention, after processing the time-decoupling rectangular feasible regions of all nodes in the distribution network, performs overall aggregation calculations on the independent time-decoupling rectangular feasible regions of each node using the Minkowski summation method, directly obtaining the overall aggregated feasible region of the entire network at the feeder outlet. This method relies on the simplified feasible regions of each node, which have already undergone time-decoupling and are independent of each other, eliminating the need to process the original complex multi-time-period strongly coupled constraints. The overall adjustable power range of the feeder can be quickly solved by directly summing the upper and lower power limits of each node. Under the premise of strictly adhering to all safe operation constraints of the flexibility resources of each node and ensuring the safety of equipment and grid operation, it significantly reduces the computational complexity of aggregating the flexibility resources of the entire network, achieving efficient and rapid aggregation and evaluation from node-level flexibility resources to feeder system-level flexibility capacity, greatly improving the computational efficiency of overall distribution network flexibility analysis and optimized scheduling.

[0052] S6. For the boundary power points in the preliminary aggregated feasible region, the power information corresponding to the boundary power points is input into the trained neural network for safety evaluation. If there are boundary power points whose safety evaluation results are unsafe, the preliminary aggregated feasible region is subjected to gradient iterative shrinkage to obtain the final total feasible region that meets the safety threshold. Preferably, in the case of boundary power points that are deemed unsafe in the safety assessment, gradient iterative shrinkage is performed on the initial aggregated feasible region, including: If there are boundary power points that are deemed unsafe in the safety assessment, repeat the following gradient iterative shrinkage operation until the final total feasible region that meets the safety threshold is obtained: Get the current penalty factor; the initial penalty factor is a preset value. The current sensitivity factor is calculated based on the safe probability output of the current target boundary power point and the power of all sampling points in the preliminary aggregated feasible region. Calculate the updated boundary power point based on the current target boundary power point, the current penalty factor, and the current sensitivity factor; The updated power information of the boundary power points is input into the trained neural network for security assessment. If the security assessment is safe, the final total feasible region is generated based on the updated boundary power points; If the security assessment is unsafe, a linear growth strategy is adopted to adjust the current penalty factor with a fixed step size to obtain the updated current penalty factor.

[0053] Preferably, training the neural network includes: Acquire several power point training samples that include power information and labeled safety status tags; wherein, the safety status tags are obtained through power flow equation verification. If the power flow equation verification passes, the tag is set as the first identifier; otherwise, the tag is set as the second identifier. Each power point training sample is input into the neural network to be trained one by one, and iterative training is performed until the loss function converges, thus obtaining the trained neural network. In each training session, a predicted safety probability value is generated based on the power information in the power point training samples; the predicted safety probability value is compared with the labeled safety status label, and the loss function value is calculated based on the comparison result.

[0054] Specifically, the training of the neural network is first based on the nonlinear mapping equation of the power flow in the distribution network. Define the physical security domain S of the distribution network: ; The nonlinear mapping equation of the power flow in the distribution network DistFlow It consists of a complete set of branch power flow recursive equations, the specific expression of which is: = ; in, For the physical security domain of the distribution network; The nonlinear mapping equation of power flow in distribution networks describes the relationship between power distribution and network state. This refers to the voltage amplitude at each node and its allowable lower and upper operating limits; The voltage amplitude at each node; This refers to the branch current vector and its allowable current carrying capacity limit; To contribute to the efforts of each node; branch road Flow to Node The active and reactive power; For nodes The active and reactive power demand of the load at the location; For nodes The active and reactive power output of the flexible resources accessed at the location; branch road The resistance and reactance parameters; For nodes and nodes The voltage amplitude; branch road The current amplitude.

[0055] According to the upper and lower limits of voltage Branch current carrying limit Safety verification: If the power combination meets all node voltage constraints and all branch current constraints, the power flow verification passes, the power grid operation is safe, and the sample is marked as a positive sample and labeled as real. If the power combination experiences voltage over-limit or branch overload issues, the power flow verification will fail, and the grid operation will be unsafe. In this case, the sample will be labeled as a negative sample, and the true label will be removed. Simultaneously, the node power allocation ratio and network topology features (i.e., power information) of each sample are extracted as input features for the neural network. Ultimately, this forms a set of input features. Authentic safety label The complete training dataset.

[0056] Next, we build a data-driven security proxy neural network, with the following expression: ; in, Used to represent the weights of a neural network, Both represent the bias of the neural network and are empirical coefficients obtained from training the neural network. This represents the Sigmoid activation function; Represents an exponential function; The vector of parameters to be learned in a data-driven neural network; For the first The input features of each training sample include node power allocation ratios and network topology features; This is the predicted probability of safe operation of the power grid for the current sample output by the model.

[0057] Further construct the binary classification cross-entropy loss function The formula is: ; in, This represents the total number of training samples.

[0058] Neural network training to minimize the loss function To achieve this goal, the backpropagation algorithm is used to iteratively update the internal parameter vector of the neural network. (weight) Bias In each round of training, all training samples are input sequentially. Predicting security probability through network formula calculation model Substitute the values ​​into the loss function formula to calculate the overall prediction error for this round; then, based on the backpropagation gradient of the error, adjust and update the weights and biases, and gradually reduce the value of the loss function; repeat the iterative training until the loss function converges and stabilizes, and the accuracy of the model's safety classification prediction meets the preset requirements. At this point, the neural network training is complete, and a neural network with optimal parameters that can accurately judge the safety of power grid operation is obtained.

[0059] After obtaining the trained neural network, the initial aggregated feasible region is subjected to safety boundary correction and iterative shrinking. The specific implementation steps are as follows: First, set the initial penalty factor. The initial value is a preset fixed parameter, which serves as the benchmark value for subsequent iterative shrinkage processes and is used to measure the strength of each round of boundary shrinkage.

[0060] The penalty factor is updated using a linear growth strategy, with adjustments made in fixed steps during each iteration. The update formula is as follows: ; in, For the first The penalty factor for the next iteration; The step size is fixed, typically ranging from 0.01 to 0.05. By using a linearly increasing penalty factor, the boundary contraction strength can be gradually enhanced, ensuring effective correction of unsafe boundary points.

[0061] If the boundary power points in the feasible region are initially aggregated If the security probability is lower than the preset confidence threshold, the current feasible region is determined to be "unsafe," and a sensitivity-guided boundary iterative shrinkage procedure needs to be executed. The specific correction formula is as follows: ; ; in, For safety indicators The sensitivity factor of the total power over a time period is defined as the gradient of the safe probability output with respect to the total power. When it is the nth iteration Maximum aggregation power for a given time period; For the first The penalty factor in each iteration increases with each iteration when the safety standard is not met, thereby strengthening the contraction force. Output the probability that the sample is a safe point for the model; Preliminary feasible domain spatial projection The sum of the power of all sampling points within the time period.

[0062] By leveraging the synergistic effect of the penalty factor and the sensitivity factor, insecure boundary power points are iteratively corrected and shrunk. The updated boundary power points are then input into the trained neural network for security assessment. If the assessment is secure, the final total feasible region is generated based on the updated boundary. If it is still insecure, the penalty factor is increased according to the rules, and the iterative shrinkage process is repeated until the final total feasible region that meets the security threshold is obtained.

[0063] This invention replaces the complex and time-consuming original power flow safety verification with a data-driven neural network proxy model trained based on DistFlow power flow physical safety rules, achieving rapid and accurate assessment of grid safety under power operation conditions. Furthermore, by combining a collaborative gradient iterative contraction strategy using sensitivity factors and linearly increasing penalty factors, the unsafe boundary power points of the initially aggregated feasible region are adaptively corrected, contracted, and re-verified round by round, continuously optimizing the feasible region boundary range until all boundaries meet the grid voltage and branch current safety operation threshold requirements. Ultimately, while fully preserving the adjustability of distribution network flexibility resources, the invention completely eliminates the safety hazards of voltage exceeding limits and branch overload caused by flexibility resource aggregation, obtaining the final total feasible region of feeders that fully complies with the physical safety operation constraints of the distribution network. This achieves a synergistic balance between efficient aggregation analysis of distribution network flexibility resources and safe and stable grid operation.

[0064] S7. Regulate the feeder outlet based on the final total feasible region; regulate the distributed energy cluster based on the decoupled rectangular feasible region of the node's time period.

[0065] Specifically, the final total feasible region obtained after neural network security assessment and iterative shrinkage using sensitivity and penalty factors is the overall adjustable power range of the feeder outlet that satisfies the safety constraints of all grid voltages or branch currents. By conducting feeder outlet regulation based on this final total feasible region, the upper and lower boundaries of the overall injected or absorbed power at the feeder outlet can be precisely constrained under the premise of safe and stable grid operation. This enables safe and efficient overall regulation of the global power of the entire feeder, avoiding safety issues such as branch overload and voltage exceeding limits in the feeder.

[0066] The feasible rectangular domain of decoupling for each node is the node-level flexible power adjustable range that completes the decoupling of strong coupling across multiple time periods and is independently regularized. It corresponds to all distributed energy clusters connected to the node, such as energy storage, controllable loads, and internal combustion engines. Based on the independent feasible rectangular domain of decoupling for each node, regulation can be carried out to finely constrain and schedule the active power output and regulation power of the distributed energy cluster within the node by time period and node. This fully taps the regulation potential of the distributed flexible resources within the node, while strictly adhering to the node's own safe operation constraints, thus achieving orderly, safe, and efficient localized collaborative regulation of the distributed energy cluster.

[0067] Based on the final total feasible region of the feeder level after safety pruning, this invention can complete the safe overall control of the global power output of the entire feeder, ensuring the stable operation of the overall power grid of the feeder; based on the rectangular feasible region after decoupling of each node time period, it can perform time-segmented and refined local scheduling of the distributed energy clusters connected within the node, fully releasing the distributed flexibility resource adjustment capability, and realizing hierarchical coordination and safe and efficient operation of the global macro-control of the feeder and the micro-control of the distributed energy of the node.

[0068] See Figure 2 This invention provides a flexible resource regulation device for distribution networks that considers network constraints and heterogeneity, comprising: a data acquisition module, a node-level model construction module, a solution module, a time-period decoupling feasible domain generation module, a feeder-level preliminary aggregation module, a security verification module, and a regulation module. The data acquisition module is used to acquire the number of nodes participating in aggregation in the distribution network, energy storage system parameters, dispatchable load parameters, internal combustion engine unit parameters, and planned scheduling duration; The node-level model construction module is used to construct the corresponding objective function and constraint set based on the energy storage system parameters, demand response parameters, internal combustion engine parameters, and planned scheduling duration, with the goal of maximizing the node power range of each node; wherein, the constraint set includes: energy storage constraints, controllable load constraints, and internal combustion engine constraints. The solution module is used to solve the objective function under the set of constraints, and obtain the optimal upper limit vector and the optimal lower limit vector of the node power of each node in each time period when the function value of the objective function is maximized. The time-period decoupling feasible region generation module is used to generate the time-period decoupling rectangular feasible region for each node based on the optimal upper limit vector and the optimal lower limit vector corresponding to the node power of each node in each time period. The feeder-level preliminary aggregation module is used to perform Minkowski summation on the time-decoupled rectangular feasible regions of all nodes to obtain the preliminary aggregated feasible region at the feeder outlet. The safety verification module is used to input the power information corresponding to the boundary power points in the preliminary aggregated feasible region into the trained neural network for safety evaluation. If there are boundary power points whose safety evaluation results are unsafe, the preliminary aggregated feasible region is subjected to gradient iterative shrinkage to obtain the final total feasible region that meets the safety threshold. The control module is used to control the feeder outlet based on the final total feasible region; and to control the distributed energy cluster based on the time period decoupling rectangular feasible region of the node.

[0069] Preferably, the controllable load constraints include shiftable load constraints, transferable load constraints, and interruptible load constraints; The node-level model construction module includes: an energy storage constraint construction unit, a transferable load constraint construction unit, a transferable load constraint construction unit, an interruptible load constraint construction unit, and an internal combustion engine constraint construction unit. The energy storage constraint construction unit is used to construct the energy storage constraint using the following formula: ; ; ; in, for The state of charge of the stored energy at the end of the time period; for The state of charge of the stored energy at the end of the time period; Energy storage coefficient; For energy storage charging efficiency; The discharge efficiency of energy storage; This refers to the rated capacity of the energy storage. yes Time-limited energy storage charging and discharging power; Indicates energy storage and discharge; Indicates charging; This represents the lower limit of energy storage capacity. Represents the upper limit of energy storage capacity; This represents the upper limit of the charging power of energy storage; Represents the upper limit of the discharge power of energy storage; For time step; The first in the total scheduling period Each time period; The transferable load constraint construction unit is used to construct the transferable load constraint using the following formula: ; ; ; in, The schedulable time period for loads that can be shifted; The planned scheduling duration; The allowable start time for dispatching of loads that can be moved; Allowable termination time for dispatchable loads; for The start status of the time period, when set to 1 indicates The load is started during the designated time period; As an intermediate variable, 0 indicates The load is not in operation during the period, represented by a value of 1. The load for that period is already in operation; This refers to the electrical power consumption of a transferable load when it is in operation. That is The original planned power consumption of the load that can be shifted during the time period; This represents the deviation of the actual power consumption of the transferable load during time period t from the originally planned power. The transferable load constraint construction unit is used to construct the transferable load constraint using the following formula: ; ; in, for The maximum permissible upward adjustment of load during a given period relative to its original electricity consumption plan; for The maximum permissible reduction in the power of the load that can be transferred during a given period relative to its original power consumption plan; Allowable start time for dispatching transferable loads; Allowable termination time for dispatchable load; For transferable loads in Actual power consumption after adjustment during the time period; The interruptible load constraint construction unit is used to construct the interruptible load constraint using the following formula: ; in, The limit for the reduction of interruptible load during time period t, based on the original electricity consumption plan; The upper limit for interruptible load during time period t is based on the original electricity consumption plan. The actual power of the interruptible load during time period t; The internal combustion engine constraint construction unit is used to construct the internal combustion engine constraint conditions using the following formula: ; ; in, This refers to the minimum permissible output power of an internal combustion engine; This refers to the maximum permissible output power of the internal combustion engine; For internal combustion engines Output power during the time period; For internal combustion engines Output power during the time period; Limits on the rate of increase of internal combustion engine units; Limits the rate of decline for internal combustion engine units.

[0070] Preferably, the node-level model construction module further includes: an objective function construction unit; The objective function construction unit is used to construct the objective function using the following formula: ; ; in, For nodes exist Lower limit of aggregation power over a given time period; For nodes exist Maximum aggregation power for a given time period; This is the lower power limit vector for each node in each time period; This is the upper limit vector of power for each node in each time period; This represents the power of the node at different time periods; for Weighting coefficients for time-of-day flexibility value; The feasible region is a set of constraints including energy storage constraints, controllable load constraints, and internal combustion engine constraints. This represents the total number of scheduling periods; Decouple the node into a rectangular feasible region for the time period.

[0071] It is understood that the above-described device embodiments correspond to the method embodiments of the present invention, and can realize any of the above-described method embodiments of the present invention to provide a distribution network flexibility resource regulation method that considers network constraints and heterogeneity.

[0072] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can specifically be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0073] Accordingly, one embodiment of this application also provides a terminal device, including: One or more processors; A memory, coupled to the processor, for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement a resource regulation method for distribution network flexibility that takes into account network constraints and heterogeneity, as described above.

[0074] The processor is used to control the overall operation of the terminal device to complete all or part of the steps of the above-mentioned method for flexible resource regulation of distribution networks that takes into account network constraints and heterogeneity.

[0075] Memory is used to store various types of data to support the operation of the terminal device. This data may include, for example, instructions for any application or method operating on the terminal device, as well as application-related data. The memory can be implemented using any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0076] In an exemplary embodiment, the terminal device may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform a distribution network flexibility resource regulation method considering network constraints and heterogeneity as described in any of the foregoing embodiments, and to achieve the same technical effects as the methods described above.

[0077] Accordingly, the present invention provides a computer-readable storage medium including a computer program, which, when executed by a processor, implements the steps of a distribution network flexibility resource regulation method considering network constraints and heterogeneity as described in any of the above embodiments. For example, the computer-readable storage medium may be the aforementioned memory including the computer program, which may be executed by a processor of a terminal device to complete the distribution network flexibility resource regulation method considering network constraints and heterogeneity as described in any of the above embodiments, and achieve the same technical effects as the aforementioned method.

[0078] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for flexible resource regulation of distribution networks considering network constraints and heterogeneity, characterized in that, include: Obtain the number of nodes participating in aggregation in the distribution network, energy storage system parameters, dispatchable load parameters, internal combustion engine unit parameters, and planned dispatch duration; Based on the energy storage system parameters, demand response parameters, internal combustion engine parameters, and planned scheduling duration, and with the goal of maximizing the node power range of each node, a corresponding objective function and set of constraints are constructed; wherein, the set of constraints includes: energy storage constraints, controllable load constraints, and internal combustion engine constraints. Under the set of constraints, the objective function is solved to obtain the optimal upper limit vector and the optimal lower limit vector of the node power of each node in each time period when the function value of the objective function is maximized. Based on the optimal upper bound vector and optimal lower bound vector of the node power of each node in each time period, generate the time period decoupling rectangular feasible region for each node. The initial aggregated feasible region at the feeder outlet is obtained by performing Minkowski summation on the time-decoupled rectangular feasible region of all nodes. For the boundary power points in the initial aggregated feasible region, the power information corresponding to the boundary power points is input into the trained neural network for safety evaluation. If there are boundary power points that are unsafe according to the safety evaluation, the initial aggregated feasible region is subjected to gradient iterative shrinkage to obtain the final total feasible region that meets the safety threshold. The feeder outlet is regulated based on the final total feasible region; the distributed energy cluster is regulated based on the time period of the node decoupled rectangular feasible region.

2. The distribution network flexibility resource regulation method considering network constraints and heterogeneity as described in claim 1, characterized in that, The controllable load constraints include shiftable load constraints, transferable load constraints, and interruptible load constraints. The energy storage constraints are constructed using the following formula: ； ； ； in, for The state of charge of the stored energy at the end of the time period; for The state of charge of the stored energy at the end of the time period; Energy storage coefficient; For energy storage charging efficiency; The discharge efficiency of energy storage; This refers to the rated capacity of the energy storage. yes Time-limited energy storage charging and discharging power; Indicates energy storage and discharge; Indicates charging; This represents the lower limit of energy storage capacity. Represents the upper limit of energy storage capacity; represent The upper limit of charging power for time-limited energy storage; represent The upper limit of discharge power for time-limited energy storage; For time step; The first in the total scheduling period Each time period; The transferable load constraint condition is constructed using the following formula: ； ； ； in, The schedulable time period for loads that can be shifted; The planned scheduling duration; The allowable start time for dispatching of loads that can be moved; Allowable termination time for dispatchable loads; for The start status of the time period, when set to 1 indicates The load is started during the designated time period; As an intermediate variable, 0 indicates The load is not in operation during the period, represented by a value of 1. The load for that period is already in operation; This refers to the electrical power consumption of a transferable load when it is in operation. That is The original planned power consumption of the load that can be shifted during the time period; This represents the deviation of the actual power consumption of the transferable load during time period t from the originally planned power. The transferable load constraint is constructed using the following formula: ； ； in, for The maximum permissible upward adjustment of load during a given period relative to its original electricity consumption plan; for The maximum permissible reduction in the power of the load that can be transferred during a given period relative to its original power consumption plan; Allowable start time for dispatching transferable loads; Allowable termination time for dispatchable load; For transferable loads in Actual power consumption after adjustment during the time period; The interruptible load constraint is constructed using the following formula: ； in, for The reduction limit for interruptible loads during certain periods is based on the original electricity consumption plan. for The upward adjustment limit for interruptible loads during certain periods is based on the original electricity consumption plan; interruptible load Actual power during the time period; The internal combustion engine constraint conditions are constructed using the following formula: ； ； in, This refers to the minimum permissible output power of an internal combustion engine; This refers to the maximum permissible output power of the internal combustion engine; For internal combustion engines Output power during the time period; For internal combustion engines Output power during the time period; Limits on the rate of increase of internal combustion engine units; Limits the rate of decline for internal combustion engine units.

3. The distribution network flexibility resource regulation method considering network constraints and heterogeneity as described in claim 2, characterized in that, The objective function is constructed in the following way: ； ； in, For nodes exist Lower limit of aggregation power over a given time period; For nodes exist Maximum aggregation power for a given time period; This is the lower power limit vector for each node in each time period; This is the upper limit vector of power for each node in each time period; This represents the power of the node at different time periods; for Weighting coefficients for time-of-day flexibility value; The feasible region is a set of constraints including energy storage constraints, controllable load constraints, and internal combustion engine constraints. This represents the total number of scheduling periods; Decouple the node into a rectangular feasible region for the time period.

4. The distribution network flexibility resource regulation method considering network constraints and heterogeneity as described in claim 3, characterized in that, In the presence of boundary power points deemed unsafe by the safety assessment, gradient iterative shrinkage is performed on the initial aggregated feasible region, including: If there are boundary power points that are deemed unsafe in the safety assessment, repeat the following gradient iterative shrinkage operation until the final total feasible region that meets the safety threshold is obtained: Get the current penalty factor; the initial penalty factor is a preset value. The current sensitivity factor is calculated based on the safe probability output of the current target boundary power point and the power of all sampling points in the preliminary aggregated feasible region. Calculate the updated boundary power point based on the current target boundary power point, the current penalty factor, and the current sensitivity factor; The updated power information of the boundary power points is input into the trained neural network for security assessment. If the security assessment is safe, the final total feasible region is generated based on the updated boundary power points; If the security assessment is unsafe, a linear growth strategy is adopted to adjust the current penalty factor with a fixed step size to obtain the updated current penalty factor.

5. A method for flexible resource regulation of distribution networks considering network constraints and heterogeneity as described in claim 4, characterized in that, Training a neural network includes: Acquire several power point training samples that include power information and labeled safety status tags; wherein, the safety status tags are obtained through power flow equation verification. If the power flow equation verification passes, the tag is set as the first identifier; otherwise, the tag is set as the second identifier. Each power point training sample is input into the neural network to be trained one by one, and iterative training is performed until the loss function converges, thus obtaining the trained neural network. In each training session, a predicted safety probability value is generated based on the power information in the power point training samples; the predicted safety probability value is compared with the labeled safety status label, and the loss function value is calculated based on the comparison result.

6. A distribution network flexibility resource regulation device considering network constraints and heterogeneity, characterized in that, include: The system includes a data acquisition module, a node-level model building module, a solution module, a time-period decoupling feasible region generation module, a feeder-level preliminary aggregation module, a safety verification module, and a control module. The data acquisition module is used to acquire the number of nodes participating in aggregation in the distribution network, energy storage system parameters, dispatchable load parameters, internal combustion engine unit parameters, and planned scheduling duration; The node-level model construction module is used to construct the corresponding objective function and constraint set based on the energy storage system parameters, demand response parameters, internal combustion engine parameters, and planned scheduling duration, with the goal of maximizing the node power range of each node; wherein, the constraint set includes: energy storage constraints, controllable load constraints, and internal combustion engine constraints. The solution module is used to solve the objective function under the set of constraints, and obtain the optimal upper limit vector and the optimal lower limit vector of the node power of each node in each time period when the function value of the objective function is maximized. The time-period decoupling feasible region generation module is used to generate the time-period decoupling rectangular feasible region for each node based on the optimal upper limit vector and the optimal lower limit vector corresponding to the node power of each node in each time period. The feeder-level preliminary aggregation module is used to perform Minkowski summation on the time-decoupled rectangular feasible regions of all nodes to obtain the preliminary aggregated feasible region at the feeder outlet. The safety verification module is used to input the power information corresponding to the boundary power points in the preliminary aggregated feasible region into the trained neural network for safety evaluation. If there are boundary power points whose safety evaluation results are unsafe, the preliminary aggregated feasible region is subjected to gradient iterative shrinkage to obtain the final total feasible region that meets the safety threshold. The control module is used to control the feeder outlet based on the final total feasible region; and to control the distributed energy cluster based on the time period decoupling rectangular feasible region of the node.

7. A distribution network flexibility resource regulation device considering network constraints and heterogeneity as described in claim 6, characterized in that, The controllable load constraints include shiftable load constraints, transferable load constraints, and interruptible load constraints. The node-level model construction module includes: an energy storage constraint construction unit, a transferable load constraint construction unit, a transferable load constraint construction unit, an interruptible load constraint construction unit, and an internal combustion engine constraint construction unit. The energy storage constraint construction unit is used to construct the energy storage constraint using the following formula: ； ； ； in, for The state of charge of the stored energy at the end of the time period; for The state of charge of the stored energy at the end of the time period; Energy storage coefficient; For energy storage charging efficiency; The discharge efficiency of energy storage; This refers to the rated capacity of the energy storage. yes Time-limited energy storage charging and discharging power; Indicates energy storage and discharge; Indicates charging; This represents the lower limit of energy storage capacity. Represents the upper limit of energy storage capacity; This represents the upper limit of the charging power of energy storage; Represents the upper limit of the discharge power of energy storage; For time step; The first in the total scheduling period Each time period; The transferable load constraint construction unit is used to construct the transferable load constraint using the following formula: ； ； ； in, The schedulable time period for loads that can be shifted; The planned scheduling duration; The allowable start time for dispatching of loads that can be moved; Allowable termination time for dispatchable loads; for The start status of the time period, when set to 1 indicates The load is started during the designated time period; As an intermediate variable, 0 indicates The load is not in operation during the period, represented by a value of 1. The load for that period is already in operation; This refers to the electrical power consumption of a transferable load when it is in operation. That is The original planned power consumption of the load that can be shifted during the time period; This represents the deviation of the actual power consumption of the transferable load during time period t from the originally planned power. The transferable load constraint construction unit is used to construct the transferable load constraint using the following formula: ； ； in, for The maximum permissible upward adjustment of load during a given period relative to its original electricity consumption plan; for The maximum permissible reduction in the power of the load that can be transferred during a given period relative to its original power consumption plan; Allowable start time for dispatching transferable loads; Allowable termination time for dispatchable load; For transferable loads in Actual power consumption after adjustment during the time period; The interruptible load constraint construction unit is used to construct the interruptible load constraint using the following formula: ； in, The limit for the reduction of interruptible load during time period t, based on the original electricity consumption plan; The upper limit for interruptible load during time period t is based on the original electricity consumption plan. The actual power of the interruptible load during time period t; The internal combustion engine constraint construction unit is used to construct the internal combustion engine constraint conditions using the following formula: ； ； in, This refers to the minimum permissible output power of an internal combustion engine; This refers to the maximum permissible output power of the internal combustion engine; For internal combustion engines Output power during the time period; For internal combustion engines Output power during the time period; Limits on the rate of increase of internal combustion engine units; Limits the rate of decline for internal combustion engine units.

8. A distribution network flexibility resource regulation device considering network constraints and heterogeneity as described in claim 7, characterized in that, The node-level model construction module further includes: an objective function construction unit; The objective function construction unit is used to construct the objective function using the following formula: ； ； in, For nodes exist Lower limit of aggregation power over a given time period; For nodes exist Maximum aggregation power for a given time period; This is the lower power limit vector for each node in each time period; This is the upper limit vector of power for each node in each time period; This represents the power of the node at different time periods; for Weighting coefficients for time-of-day flexibility value; The feasible region is a set of constraints including energy storage constraints, controllable load constraints, and internal combustion engine constraints. This represents the total number of scheduling periods; Decouple the node into a rectangular feasible region for the time period.

9. A terminal device, characterized in that, include: One or more processors; A memory, coupled to the processor, for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement a resource regulation method for distribution networks that takes into account network constraints and heterogeneity as described in any one of claims 1-5.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements a resource regulation method for distribution networks that considers network constraints and heterogeneity, as described in any one of claims 1-5.

Images