Low-inertia ac-dc system network type energy storage robust planning method and device

Abstract

The application provides a low-inertia AC-DC system network-type energy storage robust planning method and equipment. According to the inertia response of synchronous machines and network-type energy storage, the primary frequency response of synchronous machines and the fast frequency response of network-type energy storage in the low-inertia AC-DC system, a frequency change rate constraint and a frequency minimum point constraint are constructed. A network-type energy storage candidate node set is constructed. According to the new energy output parameters, a wind power and photovoltaic output uncertainty set is constructed. According to the frequency change rate constraint, the frequency minimum point constraint, the wind power and photovoltaic output uncertainty set and the network-type energy storage candidate node set, a network-type energy storage two-stage robust planning model is established. The installation node, rated power, energy capacity, day-ahead operation plan and real-time rescheduling strategy of the network-type energy storage are solved and output by using a nested column and constraint generation algorithm, and a network-type energy storage site selection, capacity determination, day-ahead operation plan and uncertain scene rescheduling scheme are formed.

Description

Technical Field

[0001] This application relates to the field of power system planning and optimized operation technology, and in particular to a robust planning method and equipment for grid-type energy storage in low-inertia AC / DC systems. Background Technology

[0002] As the proportion of new energy power generation such as wind power and photovoltaic power, as well as high-voltage direct current transmission systems, integrated into the power system continues to increase, the number of power sources and transmission equipment connected via power electronic interfaces is gradually increasing, leading to a corresponding decrease in the number of traditional synchronous machines in operation and their share of power generation. Since power electronic interface equipment typically does not possess the same rotational inertia as synchronous machines, the synchronous inertia and conventional primary frequency response capability obtainable by the system after active power disturbances decrease, easily resulting in problems such as excessive frequency change rate and the frequency minimum point falling below the allowable value.

[0003] The rate of frequency change is primarily affected by disturbance power and system inertia, while the lowest frequency point is influenced by system inertia, frequency response capacity, and response speed. Therefore, energy storage planning based solely on system power balance, reserve capacity, or total inertia constraints is insufficient to simultaneously reflect the safety requirements during both the initial disturbance phase and the frequency decline phase. Grid-based energy storage can provide virtual inertia and rapid frequency response through converter control, but its installation location, rated power, energy capacity, and operating status need to be coordinated with the physical inertia of the synchronous motor, the primary frequency response of the synchronous motor, and the system operating mode.

[0004] Furthermore, inertia sources in low-inertia AC / DC systems are typically unevenly distributed, with varying degrees of electrical coupling between different nodes and existing inertia sources. Even if the total system inertia meets the preset requirements, the local inertia support capacity reflected by different nodes may still show significant differences. If all nodes are directly used as candidate nodes for grid-type energy storage, it will not only be difficult to reflect the impact of inertia spatial distribution on energy storage site selection, but it will also increase the number of integer variables and the solution scale in the planning model.

[0005] The actual output of wind and solar power is still uncertain. Based on the grid-based energy storage configuration and day-ahead operation plan derived from the predicted output, adverse deviations in renewable energy output may prevent the fulfillment of power balance and frequency security requirements. Therefore, it is necessary to consider both day-ahead operation decisions and real-time rescheduling decisions under uncertain renewable energy scenarios during the energy storage planning stage.

[0006] In a two-stage robust programming model, the real-time charging and discharging states of grid-type energy storage are typically represented by binary variables. When the second-stage real-time rescheduling problem includes these binary variables, its inner-layer minimization problem becomes a mixed-integer programming problem, which is difficult to directly transform into a single-layer worst-case scenario search problem using strong duality theory. Consequently, conventional column and constraint generation methods are difficult to apply directly.

[0007] Therefore, a robust planning method for grid-based energy storage is needed to coordinate the handling of frequency change rate, minimum frequency point, inertia spatial distribution, and real-time operating status of grid-based energy storage while considering the uncertainty of wind and solar power output, and to form a site selection, capacity determination, day-ahead operation plan, and rescheduling scheme for uncertain scenarios for grid-based energy storage. Summary of the Invention

[0008] This application provides a robust planning method and device for grid-based energy storage in low-inertia AC / DC systems. It solves the problems of existing grid-based energy storage planning methods, which are difficult to consider both the rate of frequency change and the safety requirements of the lowest frequency point, are difficult to reflect the spatial distribution of inertia support capacity in the network, and are difficult to adapt to the uncertainty of wind power and photovoltaic output. It achieves coordinated processing of the rate of frequency change, the lowest frequency point, the spatial distribution of inertia, and the real-time operating status of grid-based energy storage, and forms a site selection, capacity determination, day-ahead operation plan, and rescheduling scheme for uncertain scenarios for grid-based energy storage.

[0009] To achieve the above objectives, in a first aspect, embodiments of this application provide a robust planning method for grid-based energy storage in low-inertia AC / DC systems, the method comprising: Obtain network topology parameters, system operating parameters, frequency safety parameters, new energy output parameters, and grid-type energy storage planning parameters for low-inertia AC / DC systems; Based on the inertia response of synchronous machines and grid-type energy storage in low-inertia AC / DC systems, the primary frequency response of synchronous machines and the fast frequency response of grid-type energy storage, frequency change rate constraints and frequency minimum point constraints are constructed. Based on the synchronous machine and the electrical coupling relationship between the synchronous machine and the access node, the computing node inertia of the access node is determined; the computing node inertia is sorted to identify nodes with weak inertia and to construct a set of candidate nodes for grid-type energy storage. Construct an uncertain set of wind power and photovoltaic power output based on the aforementioned new energy output parameters; Based on the frequency change rate constraint, the frequency minimum point constraint, the uncertain set of wind power and photovoltaic output, and the candidate node set of grid-type energy storage, a two-stage robust planning model for grid-type energy storage is established. The nested column and constraint generation algorithm is used to solve the two-stage robust programming model of the grid-type energy storage, and the solution results are obtained. The installation nodes, rated power, energy capacity, day-ahead operation plan and real-time rescheduling strategy of the grid-type energy storage are output.

[0010] In one possible implementation, the system operating parameters are line parameters, node load parameters, synchronous machine operating parameters, and DC transmission power parameters; the frequency safety parameters are the system rated frequency, equivalent active power disturbance, maximum allowable frequency change rate, maximum allowable frequency deviation, synchronous machine primary frequency complete response time, and grid-type energy storage fast frequency complete response time. The new energy output parameters are the predicted output of wind power and photovoltaic power, the down-biased ratio of the maximum allowable output, and the output uncertainty budget; the grid-type energy storage planning parameters are the candidate installation nodes, the upper and lower limits of power capacity, the upper and lower limits of energy capacity, the charging and discharging efficiency, the range of inertia time constant values, the fast frequency response capability, and the investment and maintenance cost parameters.

[0011] In one possible implementation, the step of constructing frequency change rate constraints and frequency minimum point constraints based on the inertia response of the synchronous machine and the grid-type energy storage in the low-inertia AC / DC system, the primary frequency response of the synchronous machine, and the fast frequency response of the grid-type energy storage includes: Ignoring load damping and secondary frequency regulation during the transient frequency response process in the low-inertia AC / DC system, a dynamic model of system frequency deviation is constructed based on the primary frequency response of the synchronous machine, the fast frequency response of the grid-type energy storage, and the equivalent active power deficit. The total system inertia is determined based on the physical inertia response of the synchronous machine in operation and the virtual inertia response of the grid-type energy storage. Based on the dynamic model of the system frequency deviation, the total inertia requirement of the system is determined by combining the total inertia of the system and the disturbance power, and the frequency change rate constraint is constructed. A segmented response model is constructed based on the primary frequency response of the synchronous machine and the fast frequency response of the grid-type energy storage. Based on the piecewise response model, the minimum frequency constraint is constructed.

[0012] In one possible implementation, the step of determining the computing node inertia of the access node based on the synchronizing machine and the electrical coupling relationship between the synchronizing machine and the access node; sorting the computing node inertia to identify nodes with weak inertia and constructing a set of candidate nodes for grid-type energy storage includes: Before planning a grid-type energy storage system, the synchronous machine in operation is identified as an existing inertial source; the inertia of the computational node is determined based on the node disturbance power and the initial frequency change rate of the node at the access node. The inertia of the computing nodes are arranged in ascending order. A first preset number of nodes whose inertia values ​​are less than a preset inertia threshold are identified as weak inertia nodes. The set of candidate nodes for grid-type energy storage is constructed based on the weak inertia nodes.

[0013] In one possible implementation, the two-stage robust planning model for grid-based energy storage includes an objective function, a first-stage planning and day-ahead operation constraints, and a second-stage real-time rescheduling constraints. The first-stage planning and day-ahead operation constraints are determined based on the grid-based energy storage installation status, power capacity, energy capacity, and number of installations, and are used to determine the installation location, rated power, energy capacity, and day-ahead operation plan of the grid-based energy storage. The second-stage real-time rescheduling constraints are determined based on the real-time charging and discharging power and real-time charging and discharging status of the grid-based energy storage, and are used to determine the real-time rescheduling strategy in scenarios where wind power and photovoltaic output are uncertain.

[0014] In one possible implementation, the first phase planning and day-ahead operation constraints also include day-ahead start-stop constraints of synchronous machines, day-ahead output constraints of synchronous machines, ramping constraints, upper and lower reserve constraints, primary frequency regulation reserve constraints, node power balance constraints, DC power flow constraints, line transmission capacity constraints, day-ahead charge and discharge mutual exclusion constraints of grid-type energy storage, day-ahead energy balance constraints, state of charge constraints, and new energy utilization constraints. The second-stage real-time rescheduling constraints also include real-time rescheduling constraints for synchronous machines, real-time utilization constraints for new energy sources, load reduction constraints, real-time power balance constraints for nodes, DC power flow constraints, line transmission capacity constraints, frequency change rate constraints, and frequency minimum point constraints.

[0015] In one possible implementation, the method of using a nested column and constraint generation algorithm to solve the two-stage robust programming model of the grid-type energy storage, obtaining the solution results, and outputting the installation nodes, rated power, energy capacity, day-ahead operation plan, and real-time rescheduling strategy of the grid-type energy storage includes: Establish an outer master problem; the outer master problem is used to determine the planning variables, day-ahead scheduling variables, and auxiliary variables for the maximum real-time rescheduling cost of grid-type energy storage. Initialize the set of worst-case scenarios in history and solve the outer master problem to obtain the grid-type energy storage planning solution, the day-ahead scheduling solution, and the maximum real-time rescheduling cost threshold. Establish a worst-case scenario screening problem to determine candidate worst-case scenarios from the uncertain set of wind power and photovoltaic output; When the objective function value of the worst-case scenario selection problem is greater than the preset function threshold, the candidate worst-case scenario is substituted into the original real-time rescheduling model to re-optimize the grid-type energy storage charging and discharging state, charging and discharging power and fast frequency response power to obtain the rescheduling verification cost. Based on the rescheduling verification cost, valid violation scenarios are determined, and the installation node, rated power, energy capacity, day-ahead operating plan, and real-time rescheduling strategy are output.

[0016] In one possible implementation, determining valid violation scenarios based on the rescheduling verification cost and outputting the installation node, rated power, energy capacity, day-ahead operating plan, and real-time rescheduling strategy includes: If the rescheduling verification cost is greater than a preset cost threshold, the corresponding worst-case candidate scenario is determined as the valid violation scenario. If the rescheduling verification cost is less than or equal to a preset cost threshold, the corresponding candidate worst-case scenario is added to the scenario exclusion set of a typical day, the worst-case scenario screening problem is solved again, and the installation node, rated power, energy capacity, day-ahead operation plan and real-time rescheduling strategy are output.

[0017] In one possible implementation, the resolving of the worst-case scenario selection problem and the output of the installation node, rated power, energy capacity, day-ahead operating plan, and real-time rescheduling strategy include: From the multiple valid violation scenarios, a second preset number of target valid violation scenarios are selected in descending order of the rescheduling verification cost; The outer master problem is to incorporate the uncertain variables of new energy output, the real-time rescheduling variable replicas, the real-time operation constraints, the frequency security constraints, and the real-time rescheduling cost constraints corresponding to the scenario in which the objective is effectively violated. Repeatedly solve the adjusted outer master problem and the worst-case scenario screening problem to determine a new rescheduling verification cost until the new rescheduling verification cost is less than or equal to the preset cost threshold, or until the number of iterations is greater than or equal to the preset number of iterations. Output the installation node, rated power, energy capacity, day-ahead operation plan and real-time rescheduling strategy.

[0018] Secondly, embodiments of this application provide an electronic device, including: a memory storing a computer program thereon; and a processor for executing the computer program in the memory to implement the steps of the method described in the first aspect.

[0019] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: (1) By simultaneously constructing frequency change rate constraints and frequency minimum point constraints, the physical inertia of the synchronous machine, the virtual inertia of the grid-type energy storage, the primary frequency response of the synchronous machine and the fast frequency response of the grid-type energy storage are coordinated, so that the planning results can simultaneously consider the frequency safety requirements of the initial stage of disturbance and the frequency minimum point stage within the range of the given disturbance model and the uncertainty set of new energy output. (2) By calculating the node inertia, the electrical coupling relationship between different nodes and existing inertia sources can be characterized, and weak inertia nodes can be screened accordingly. This can reflect the spatial differences in inertia support capacity in grid-type energy storage planning, while narrowing the range of candidate nodes for grid-type energy storage and reducing the number of site selection variables. (3) Limit the cumulative degree of downward deviation of new energy output by the uncertain budget of wind power and photovoltaic output, and adopt a two-stage robust planning to coordinate day-ahead operation decision and real-time rescheduling decision under uncertain scenarios, so that the site selection, capacity setting and operation plan of grid-type energy storage can adapt to the changes in new energy output within the constructed uncertain set; (4) By transforming the analytical form of the minimum frequency constraint into a time discretized scan constraint, the bilinear, quadratic and fractional terms generated at the minimum frequency moment are avoided in the mixed integer robust programming model, so that the minimum frequency safety requirement can be embedded in the two-stage robust programming model. (5) By using the real-time charging and discharging state of fixed grid-type energy storage, the inner-layer problem in the worst scenario screening is transformed into a continuous linear programming problem that can undergo strong dual transformation; then the original real-time rescheduling model that allows re-optimization of charging and discharging state is used to verify candidate scenarios, avoiding the direct addition of scenarios generated under fixed state but no longer violating the current cost threshold after flexible rescheduling to the outer-layer main problem. (6) By iteratively adding valid violation scenarios confirmed by real-time flexible rescheduling, it is not necessary to exhaust all scenarios in the uncertainty set of wind power and photovoltaic output. The planning variables, day-ahead operation variables and historical scenario rescheduling variables of grid-type energy storage can be updated, thereby realizing the two-stage robust planning solution of grid-type energy storage including the second-stage binary operation state. (7) By outputting the grid-type energy storage installation nodes, rated power, energy capacity, day-ahead operation plan and real-time rescheduling strategy under corresponding uncertain scenarios, it can provide a basis for the planning and operation scheme of grid-type energy storage for low-inertia AC / DC systems. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments of this application or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A flowchart illustrating a robust planning method for grid-type energy storage in low-inertia AC / DC systems, provided in this application embodiment; Figure 2 This is a block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0023] In the description of the embodiments of this application, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the embodiments of this application and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application. The terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, the terms "installed," "connected," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; 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 connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the embodiments of this application according to the specific circumstances.

[0024] Figure 1 A flowchart illustrating a robust planning method for grid-based energy storage in a low-inertia AC / DC system, provided as an embodiment of this application. Figure 1 As shown, the method may include the following steps.

[0025] S101. Obtain the network topology parameters, system operation parameters, frequency safety parameters, new energy output parameters, and grid-type energy storage planning parameters of the low-inertia AC / DC system.

[0026] S102. Based on the inertia response of the synchronous machine and the grid-type energy storage in the low-inertia AC / DC system, the primary frequency response of the synchronous machine and the fast frequency response of the grid-type energy storage, construct the frequency change rate constraint and the minimum frequency point constraint.

[0027] S103. Based on the synchronous machine and the electrical coupling relationship between the synchronous machine and the access node, determine the computational node inertia of the access node; sort the computational node inertia, identify weak inertia nodes, and construct a set of candidate nodes for grid-type energy storage.

[0028] S104. Construct an uncertain set of wind power and photovoltaic power output based on the new energy output parameters.

[0029] S105. Based on the frequency change rate constraint, the frequency minimum point constraint, the uncertain set of wind power and photovoltaic output, and the candidate node set of grid-type energy storage, establish a two-stage robust planning model for grid-type energy storage.

[0030] S106. The nested column and constraint generation algorithm is used to solve the two-stage robust programming model of the grid-type energy storage, and the solution results are obtained. The installation nodes, rated power, energy capacity, day-ahead operation plan and real-time rescheduling strategy of the grid-type energy storage are output.

[0031] According to the above technical solution, frequency change rate constraints and frequency minimum point constraints can be constructed simultaneously, coordinating the physical inertia of the synchronous machine, the virtual inertia of the grid-type energy storage, the primary frequency response of the synchronous machine, and the rapid frequency response of the grid-type energy storage. This allows the planning results to simultaneously consider the frequency safety requirements of the initial stage of the disturbance and the frequency minimum point stage within the given disturbance model and the uncertainty set of new energy output.

[0032] In one possible implementation, the system operating parameters include line parameters, node load parameters, synchronous machine operating parameters, and DC transmission power parameters; the frequency safety parameters include the system rated frequency, equivalent active power disturbance, maximum allowable frequency change rate, maximum allowable frequency deviation, synchronous machine primary frequency complete response time, and grid-type energy storage fast frequency complete response time; the renewable energy output parameters include predicted wind and solar power output, maximum allowable output under-bias ratio, and output uncertainty budget; the grid-type energy storage planning parameters include candidate installation nodes, upper and lower limits of power capacity, upper and lower limits of energy capacity, charge and discharge efficiency, inertia time constant range, fast frequency response capability, and investment and maintenance cost parameters. For example, the network topology parameters can include the type of access nodes and branches, connection relationships, and network topology configuration.

[0033] In one possible implementation, S102 may include: ignoring load damping and secondary frequency regulation during the transient frequency response process in the low-inertia AC / DC system, constructing a dynamic model of system frequency deviation based on the primary frequency response of the synchronous machine, the fast frequency response of the grid-type energy storage, and the equivalent active power deficit; determining the total system inertia based on the physical inertia response of the synchronous machine in operation and the virtual inertia response of the grid-type energy storage; determining the total system inertia requirement based on the dynamic model of system frequency deviation, combined with the total system inertia and disturbance power, and constructing the frequency change rate constraint; constructing a segmented response model based on the primary frequency response of the synchronous machine and the fast frequency response of the grid-type energy storage; and constructing the minimum frequency point constraint based on the segmented response model.

[0034] For example, (1) construct the frequency change rate constraint. The dynamic model of the system's frequency deviation can be expressed as: ;in, This represents the total inertia of the system. This refers to the system's frequency deviation. and These are the quantities of synchronous generator units and grid-type energy storage, respectively. and Synchronous generator units i Grid-type energy storage n The actual power provided for primary frequency response and fast frequency response; This represents the disturbance to the system's equivalent active power.

[0035] The total inertia of the system consists of the physical inertia of the operating synchronous machine (online synchronous machine) and the virtual inertia of the installed grid-type energy storage, which can be expressed as: ;in, and Synchronizer i The inertial time constant and rated power; Synchronizer i The running state variables; For grid-type energy storage n The inertia time constant; For grid-type energy storage n Rated power; For grid-type energy storage n Installation status variables.

[0036] This frequency change rate constraint can be expressed as: ;in, The maximum rate of change of frequency is preset. Equivalently, the total inertia of the system can satisfy: .

[0037] (2) Constructing the minimum frequency constraint. A piecewise response model is used to represent the process of establishing the primary frequency response of the synchronous machine and the fast frequency response of the grid-type energy storage: , , , ; in, and Synchronous generator units i First-order frequency response capacity and grid-type energy storage n Fast frequency response capacity; and These are the primary frequency response power and the fast frequency response power, respectively. and These are the complete response times for the first frequency and the complete response times for the fast frequency, respectively. .

[0038] When the lowest frequency point is located and When the system frequency deviation is between these values, it can be expressed as: By setting the rate of change of frequency deviation to 0, the moment of minimum frequency can be determined. : Based on the maximum permissible frequency deviation Construct the minimum frequency constraint: .

[0039] Substituting the time of the lowest frequency into the system frequency deviation expression, we get: Embed the minimum frequency constraint into a mixed-integer robust programming model, in and Construct a set of scan times with the lowest frequency between them: , ;in, This is the set of scan times at the lowest frequency points; For the first q Each scan moment; Q This represents the total number of scan times. Let be the scan step size. The following linear scan constraint is applied at each scan time: The condition for the lowest continuous-time frequency is discretized using the linear scanning constraint, and the frequency response is verified after the planning solution is obtained.

[0040] In one possible implementation, S103 may include: before planning a grid-type energy storage system, identifying the operating synchronous machine as an existing inertial source; determining the inertia of the computing node based on the node disturbance power and the initial frequency change rate of the node; arranging the computing node inertia in ascending order, identifying a first preset number of nodes whose computing node inertia values ​​are less than a preset inertia threshold as weak inertia nodes, and constructing a set of candidate nodes for the grid-type energy storage system based on the weak inertia nodes.

[0041] For example, before grid-based energy storage planning, the inertia of this computing node can be expressed as: ;in, For nodes m inertia; For nodes m Active power disturbance; The system's rated frequency; For nodes m Frequency deviation; For nodes m The rate of change of frequency. The calculated nodal inertia can also be expressed based on the electrical coupling relationship between the node and existing inertia sources as: ;in, Inertia source h The inertial time constant; For the set of system inertia sources; For nodes m With inertia source h The voltage correlation coefficient between them; Inertia source h With nodes m The equivalent susceptance between them.

[0042] During the candidate node selection phase, only existing synchronous machines before the installation of the grid-type energy storage were included in the existing inertia source set. The computational node inertia of each node was sorted in ascending order, and the node with the lowest computational node inertia was selected. k A set of weak inertia nodes was identified, and a network-type energy storage candidate node set was constructed based on these weak inertia nodes. .

[0043] In one possible implementation, constructing the uncertain set of wind and solar power output in S104 above may include the following steps. For a typical day r scheduling period t The actual available power output of wind power and photovoltaic power are expressed as follows: , ;in, and Wind power and solar power respectively during the time period t The actual available output; and Wind power and solar power respectively during the time period t The predicted output; and These are the downward offset ratios of the maximum allowable output of wind power and solar power, respectively. and These are the output deviations for wind power and photovoltaic power, respectively.

[0044] The output deviation variable between wind power and photovoltaic power satisfies: , , , ;in, and These are the output uncertainty budgets for wind power and solar power, used to limit the total number of periods or the cumulative degree of down-biasing of wind power and solar power output during a typical day.

[0045] In one possible implementation, the two-stage robust planning model for grid-based energy storage includes an objective function, a first-stage planning and day-ahead operation constraints, and a second-stage real-time rescheduling constraints. The first-stage planning and day-ahead operation constraints are determined based on the grid-based energy storage's installation status, power capacity, energy capacity, and number of installations, and are used to determine the installation location, rated power, energy capacity, and day-ahead operation plan. The second-stage real-time rescheduling constraints are determined based on the grid-based energy storage's real-time charge / discharge power and real-time charge / discharge status, and are used to determine the real-time rescheduling strategy under uncertain wind and solar power output scenarios. For example, the two-stage robust planning model for grid-based energy storage may include two stages: the first stage determines the grid-based energy storage's installation location, rated power, energy capacity, and day-ahead operation plan, and the second stage determines the real-time rescheduling strategy under uncertain wind and solar power output scenarios.

[0046] In one possible implementation, the first-stage planning and day-ahead operation constraints also include day-ahead start / stop constraints for synchronous machines, day-ahead output constraints for synchronous machines, ramping constraints, upper and lower reserve constraints, primary frequency regulation reserve constraints, node power balance constraints, DC power flow constraints, line transmission capacity constraints, day-ahead charge / discharge mutual exclusion constraints for grid-type energy storage, day-ahead energy balance constraints, state of charge constraints, and renewable energy utilization constraints; the second-stage real-time rescheduling constraints also include real-time rescheduling constraints for synchronous machines, real-time utilization constraints for renewable energy, load reduction constraints, real-time node power balance constraints, DC power flow constraints, line transmission capacity constraints, frequency change rate constraints, and frequency minimum point constraints.

[0047] For example, the objective function of this two-stage robust planning model for grid-type energy storage can be expressed as: ;in, The daily investment cost for grid-type energy storage; Maintenance costs for grid-type energy storage; R A typical set of days; For typical days r The weights; For typical days r Daily operating costs; Uncertain variables in contributing to new energy sources; For typical days r The output of wind and solar power is uncertain; For real-time rescheduling variables; For a given planning, day-ahead decision, and uncertain scenarios, the real-time rescheduling feasible region is defined. x Determine the capacity variable for energy storage site selection; y For day-ahead scheduling variables; For real-time rescheduling costs; To reduce the penalty costs for renewable energy curtailment and load reduction during the real-time rescheduling phase.

[0048] The installation status, rated power, energy capacity, and number of grid-type energy storage devices meet the following requirements: , , ;in, For grid-type energy storage n The configuration variables, and ; and These are grid-type energy storage n Rated power and rated capacity; and These are grid-type energy storage n Upper and lower limits of rated power; and These are grid-type energy storage n Upper and lower limits of rated capacity; The upper limit for the number of grid-type energy storage configurations.

[0049] During the real-time rescheduling process of the second stage of the two-stage robust planning model for grid-based energy storage, the real-time charging and discharging power and real-time charging and discharging state of the grid-based energy storage satisfy the following: , , ;in, and These are grid-type energy storage n During the real-time rescheduling phase t The charging and discharging power; and These are grid-type energy storage n During the real-time rescheduling phase t The charging and discharging state variables; This represents the installation status of the grid-type energy storage system as determined in the first phase.

[0050] The fast frequency response power provided by grid-based energy storage and the energy balance of grid-based energy storage should meet the following requirements: , , , ;in, For grid-type energy storage n During the real-time rescheduling phase t Fast frequency response power; For grid-type energy storage n During the real-time rescheduling phase t Energy loss when providing fast frequency response power; The complete frequency response time of a synchronous machine is used to characterize the longest support time for grid-type energy storage to continuously provide a fast frequency response until the upper limit of the interval may appear at the lowest frequency point. and These are the charging and discharging efficiencies of grid-type energy storage; and These represent the real-time energy storage capacity of the grid-type energy storage system for the current and previous time periods, respectively. and These represent the lower and upper limits of the allowable energy storage capacity for grid-type energy storage, respectively.

[0051] Furthermore, during the second phase of real-time rescheduling, the start / stop status of the synchronous machine, the upper and lower reserve capacities of the synchronous machine, and the primary frequency regulation reserve capacity of the synchronous machine determined in the first phase can be fixed. The real-time charging / discharging status, charging / discharging power, and fast frequency response power of grid-type energy storage can be re-optimized based on uncertain renewable energy output scenarios. Load shedding is prohibited in normal day-ahead forecasting scenarios, and load shedding is only used as an emergency real-time rescheduling measure under adverse renewable energy scenarios.

[0052] In one possible implementation, S106 may include: establishing an outer master problem; the outer master problem is used to determine the planning variables, day-ahead scheduling variables, and auxiliary variables of the maximum real-time rescheduling cost for grid-based energy storage; initializing the set of worst-case scenarios in history and solving the outer master problem to obtain the planning solution for grid-based energy storage, the day-ahead scheduling solution, and the maximum real-time rescheduling cost threshold; establishing a worst-case scenario screening problem to determine candidate worst-case scenarios from the set of uncertain wind and solar power output; if the objective function value of the worst-case scenario screening problem is greater than a preset function threshold, substituting the candidate worst-case scenario into the original real-time rescheduling model, re-optimizing the charging and discharging state, charging and discharging power, and fast frequency response power of the grid-based energy storage to obtain the rescheduling verification cost; based on the rescheduling verification cost, determining the valid violation scenario, and outputting the installation node, rated power, energy capacity, day-ahead operation plan, and real-time rescheduling strategy.

[0053] In one possible implementation, the process of determining valid violation scenarios based on the rescheduling verification cost and outputting the installation node, rated power, energy capacity, day-ahead operating plan, and real-time rescheduling strategy includes: if the rescheduling verification cost is greater than a preset cost threshold, identifying the corresponding candidate worst-case scenario as the valid violation scenario; if the rescheduling verification cost is less than or equal to the preset cost threshold, adding the corresponding candidate worst-case scenario to the scenario exclusion set for a typical day, resolving the worst-case scenario selection problem, and outputting the installation node, rated power, energy capacity, day-ahead operating plan, and real-time rescheduling strategy. For example, the preset cost threshold can be the current maximum real-time rescheduling cost threshold. Scene generation tolerance sum.

[0054] In one possible implementation, the above-mentioned resolving of the worst-case scenario selection problem and outputting the installation node, rated power, energy capacity, day-ahead operation plan, and real-time rescheduling strategy includes: selecting a second preset number of target valid violation scenarios from multiple valid violation scenarios according to the rescheduling verification cost from high to low; adding the new energy output uncertainty variable, real-time rescheduling variable copy, real-time operation constraint, frequency security constraint, and real-time rescheduling cost constraint corresponding to the target valid violation scenario to the outer master problem; repeatedly solving the adjusted outer master problem and the worst-case scenario selection problem to determine a new rescheduling verification cost until the new rescheduling verification cost is less than or equal to the preset cost threshold, or the number of iterations is greater than or equal to the preset number of iterations, and then outputting the installation node, rated power, energy capacity, day-ahead operation plan, and real-time rescheduling strategy.

[0055] For example, the compact form of the outer master problem [MP] is: ; in, This serves as an auxiliary variable for maximizing real-time rescheduling cost. A collection of the worst historical scenarios; For the first The worst scenario in history; For the corresponding real-time reschedulable variable replicas; matrix The real-time rescheduling feasible domain includes real-time operation constraints, frequency change rate constraints, and frequency minimum point constraints.

[0056] For each typical day, the real-time charge and discharge state of the grid-type energy storage is fixed as the day-ahead charge and discharge state obtained from the outer master problem, transforming the second-stage inner minimization problem into a continuous linear programming problem. Based on strong duality theory, a worst-case scenario selection problem [DP] is established: ; In the formula, For the nonnegative dual variable corresponding to the continuous real-time rescheduling problem; and These are the network-type energy storage planning solution and the day-ahead dispatch solution obtained from the current outer master problem, respectively.

[0057] Candidate scenarios are obtained by filtering the worst-case scenario. Substituting the original real-time rescheduling model, we remove the fixed real-time charging and discharging state of grid-type energy storage and establish a real-time flexible rescheduling verification problem [RP-flex]: .

[0058] When the rescheduling verification cost of the real-time flexible rescheduling verification problem exceeds the current maximum real-time rescheduling cost threshold. Scene generation tolerance When the sum is greater than (i.e., greater than) When the candidate scenario is determined as a valid violation scenario, and the rescheduling verification cost is less than or equal to the time limit, the candidate scenario is determined as a valid violation scenario; when the rescheduling verification cost is less than or equal to the time limit, the candidate scenario is determined as a valid violation scenario. When the time comes, the candidate scenarios are added to the scenario exclusion set of the corresponding typical day, and the worst-case scenario selection problem is solved again.

[0059] From the valid violation scenarios of each typical day, a second preset number of scenarios are selected according to the rescheduling verification cost from high to low. The corresponding uncertainties in new energy output, real-time rescheduling variable replicas, real-time operational constraints, frequency security constraints, and real-time rescheduling cost constraints are added to the outer master problem. The expanded outer master problem is solved repeatedly, and worst-case scenario screening and real-time flexible rescheduling verification are performed until no rescheduling verification cost exceeds a certain threshold. The effective violation scenario, or reaching the preset number of iterations.

[0060] Based on the same concept, this application also provides an electronic device, including: a memory storing a computer program thereon; and a processor for executing the computer program in the memory to implement the steps of the robust planning method for low-inertia AC / DC system grid-type energy storage provided in this embodiment.

[0061] Figure 2 This is a block diagram of an electronic device provided in an embodiment of this application. (For example...) Figure 2 As shown, the electronic device 200 may include a processor 201 and a memory 202. The electronic device 200 may also include one or more of a multimedia component 203, an input / output interface 204, and a communication component 205.

[0062] The processor 201 controls the overall operation of the electronic device 200 to complete all or part of the steps in the image pair construction method and model training method described above. The memory 202 stores various types of data to support the operation of the electronic device 200. This data may include, for example, instructions for any application or method operating on the electronic device 200, and application-related data such as contact data, sent and received messages, images, audio, video, etc. The memory 202 can be implemented by 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. The multimedia component 203 may include a screen and audio components. The screen may be, for example, a touchscreen, and the audio component is used for outputting and / or inputting audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in memory 202 or transmitted via communication component 205. The audio component also includes at least one speaker for outputting audio signals. Input / output interface 204 provides an interface between processor 201 and other interface modules, such as a keyboard, mouse, buttons, etc. These buttons may be virtual or physical buttons. Communication component 205 is used for wired or wireless communication between the electronic device 200 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, 4G, NB-IoT, eMTC, or other 5G technologies, or combinations thereof, is not limited here. Therefore, the corresponding communication component 205 may include: a Wi-Fi module, a Bluetooth module, an NFC module, etc.

[0063] In an exemplary embodiment, the electronic device 200 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 execute the aforementioned robust planning method for low-inertia AC / DC system grid-type energy storage.

[0064] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided. When executed by a processor, these program instructions implement the steps of the robust planning method for grid-based energy storage in low-inertia AC / DC systems described above. For example, the computer-readable storage medium may be the memory 202 including program instructions described above, which may be executed by the processor 201 of the electronic device 200 to complete the robust planning method for grid-based energy storage in low-inertia AC / DC systems described above.

[0065] The various embodiments in this specification are described in a progressive manner. For the same or similar parts between the various embodiments, please refer to each other. Each embodiment focuses on describing the differences from other embodiments.

[0066] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of this application.

Claims

1. A robust planning method for grid-based energy storage in low-inertia AC / DC systems, characterized in that, The method includes: Obtain network topology parameters, system operating parameters, frequency safety parameters, new energy output parameters, and grid-type energy storage planning parameters for low-inertia AC / DC systems; Based on the inertia response of synchronous machines and grid-type energy storage in low-inertia AC / DC systems, the primary frequency response of synchronous machines and the fast frequency response of grid-type energy storage, frequency change rate constraints and frequency minimum point constraints are constructed. Based on the synchronous machine and the electrical coupling relationship between the synchronous machine and the access node, the computing node inertia of the access node is determined; the computing node inertia is sorted to identify nodes with weak inertia and to construct a set of candidate nodes for grid-type energy storage. Construct an uncertain set of wind power and photovoltaic power output based on the aforementioned new energy output parameters; Based on the frequency change rate constraint, the frequency minimum point constraint, the uncertain set of wind power and photovoltaic output, and the candidate node set of grid-type energy storage, a two-stage robust planning model for grid-type energy storage is established. The nested column and constraint generation algorithm is used to solve the two-stage robust programming model of the grid-type energy storage, and the solution results are obtained. The installation nodes, rated power, energy capacity, day-ahead operation plan and real-time rescheduling strategy of the grid-type energy storage are output.

2. The method according to claim 1, characterized in that, The system operating parameters are line parameters, node load parameters, synchronous machine operating parameters, and DC transmission power parameters; the frequency safety parameters are the system rated frequency, equivalent active power disturbance, maximum allowable frequency change rate, maximum allowable frequency deviation, synchronous machine primary frequency complete response time, and grid-type energy storage fast frequency complete response time. The new energy output parameters are the predicted output of wind power and photovoltaic power, the down-biased ratio of the maximum allowable output, and the output uncertainty budget; the grid-type energy storage planning parameters are the candidate installation nodes, the upper and lower limits of power capacity, the upper and lower limits of energy capacity, the charging and discharging efficiency, the range of inertia time constant values, the fast frequency response capability, and the investment and maintenance cost parameters.

3. The method according to claim 1, characterized in that, The method constructs frequency change rate constraints and frequency minimum point constraints based on the inertia response of synchronous machines and grid-type energy storage in low-inertia AC / DC systems, the primary frequency response of synchronous machines, and the fast frequency response of grid-type energy storage, including: Ignoring load damping and secondary frequency regulation during the transient frequency response process in the low-inertia AC / DC system, a dynamic model of system frequency deviation is constructed based on the primary frequency response of the synchronous machine, the fast frequency response of the grid-type energy storage, and the equivalent active power deficit. The total system inertia is determined based on the physical inertia response of the synchronous machine in operation and the virtual inertia response of the grid-type energy storage. Based on the dynamic model of the system frequency deviation, the total inertia requirement of the system is determined by combining the total inertia of the system and the disturbance power, and the frequency change rate constraint is constructed. A segmented response model is constructed based on the primary frequency response of the synchronous machine and the fast frequency response of the grid-type energy storage. Based on the piecewise response model, the minimum frequency constraint is constructed.

4. The method according to claim 1, characterized in that, The computational node inertia of the access node is determined based on the synchronization machine and the electrical coupling relationship between the synchronization machine and the access node. The inertia of the computing nodes is sorted to identify nodes with weak inertia and to construct a set of candidate nodes for grid-type energy storage, including: Before planning a grid-type energy storage system, the synchronous machine in operation is identified as an existing inertial source; the inertia of the computational node is determined based on the node disturbance power and the initial frequency change rate of the node at the access node. The inertia of the computing nodes are arranged in ascending order. A first preset number of nodes whose inertia values ​​are less than a preset inertia threshold are identified as weak inertia nodes. The set of candidate nodes for grid-type energy storage is constructed based on the weak inertia nodes.

5. The method according to claim 1, characterized in that, The two-stage robust planning model for grid-based energy storage includes an objective function, a first-stage planning and day-ahead operation constraint, and a second-stage real-time rescheduling constraint. The first-stage planning and day-ahead operation constraint is determined based on the grid-based energy storage installation status, power capacity, energy capacity, and number of installations, and is used to determine the installation location, rated power, energy capacity, and day-ahead operation plan of the grid-based energy storage. The second-stage real-time rescheduling constraint is determined based on the real-time charging and discharging power and real-time charging and discharging status of the grid-based energy storage, and is used to determine the real-time rescheduling strategy under uncertain wind and solar power output scenarios.

6. The method according to claim 5, characterized in that, The first phase of planning and day-ahead operation constraints also includes day-ahead start-stop constraints of synchronous machines, day-ahead output constraints of synchronous machines, ramping constraints, upper and lower reserve constraints, primary frequency regulation reserve constraints, node power balance constraints, DC power flow constraints, line transmission capacity constraints, day-ahead charge and discharge mutual exclusion constraints of grid-type energy storage, day-ahead energy balance constraints, state of charge constraints, and new energy utilization constraints. The second-stage real-time rescheduling constraints also include real-time rescheduling constraints for synchronous machines, real-time utilization constraints for new energy sources, load reduction constraints, real-time power balance constraints for nodes, DC power flow constraints, line transmission capacity constraints, frequency change rate constraints, and frequency minimum point constraints.

7. The method according to claim 1, characterized in that, The nested column and constraint generation algorithm is used to solve the two-stage robust programming model of the grid-type energy storage, and the solution results are output, including the installation nodes, rated power, energy capacity, day-ahead operation plan, and real-time rescheduling strategy of the grid-type energy storage. Establish an outer master problem; the outer master problem is used to determine the planning variables, day-ahead scheduling variables, and auxiliary variables for the maximum real-time rescheduling cost of grid-type energy storage. Initialize the set of worst-case scenarios in history and solve the outer master problem to obtain the grid-type energy storage planning solution, the day-ahead scheduling solution, and the maximum real-time rescheduling cost threshold. Establish a worst-case scenario screening problem to determine candidate worst-case scenarios from the uncertain set of wind power and photovoltaic output; When the objective function value of the worst-case scenario selection problem is greater than the preset function threshold, the candidate worst-case scenario is substituted into the original real-time rescheduling model to re-optimize the grid-type energy storage charging and discharging state, charging and discharging power and fast frequency response power to obtain the rescheduling verification cost. Based on the rescheduling verification cost, valid violation scenarios are determined, and the installation node, rated power, energy capacity, day-ahead operating plan, and real-time rescheduling strategy are output.

8. The method according to claim 7, characterized in that, The process of determining valid violation scenarios based on the rescheduling verification cost and outputting the installation node, rated power, energy capacity, day-ahead operating plan, and real-time rescheduling strategy includes: If the rescheduling verification cost is greater than a preset cost threshold, the corresponding worst-case candidate scenario is determined as the valid violation scenario. If the rescheduling verification cost is less than or equal to a preset cost threshold, the corresponding candidate worst-case scenario is added to the scenario exclusion set of a typical day, the worst-case scenario screening problem is solved again, and the installation node, rated power, energy capacity, day-ahead operation plan and real-time rescheduling strategy are output.

9. The method according to claim 8, characterized in that, The process of resolving the worst-case scenario selection problem and outputting the installation node, rated power, energy capacity, day-ahead operating plan, and real-time rescheduling strategy includes: From the multiple valid violation scenarios, a second preset number of target valid violation scenarios are selected in descending order of the rescheduling verification cost; The outer master problem is to incorporate the uncertain variables of new energy output, the real-time rescheduling variable replicas, the real-time operation constraints, the frequency security constraints, and the real-time rescheduling cost constraints corresponding to the scenario in which the objective is effectively violated. Repeatedly solve the adjusted outer master problem and the worst-case scenario screening problem to determine a new rescheduling verification cost until the new rescheduling verification cost is less than or equal to the preset cost threshold, or until the number of iterations is greater than or equal to the preset number of iterations. Output the installation node, rated power, energy capacity, day-ahead operation plan and real-time rescheduling strategy.

10. An electronic device, characterized in that, include: A memory on which computer programs are stored; A processor for executing the computer program in the memory to implement the steps of the method according to any one of claims 1 to 9.

Images