Rocof safety budget based on double-fed wind farm speed recovery peak shaving scheduling method

Abstract

The application discloses a double-fed wind farm rotating speed recovery staggered peak regulation method based on RoCoF safety budget, and belongs to the technical field of new energy grid connection and power system control. The application introduces a dynamic RoCoF safety budget function at the field level, and performs staggered peak regulation on the recovery starting time and recovery duration of each unit in the field. The method calculates the field-level available RoCoF budget in real time, and each wind turbine is started in a staggered peak manner according to the single-machine occupation intensity from large to small, so that the total occupation of each active unit to the RoCoF budget at any time does not exceed the safety boundary. The application can significantly reduce the instantaneous RoCoF impact during the recovery stage of the wind farm without relying on additional energy storage devices, effectively suppress the secondary frequency drop, and improve the frequency stability of the multi-unit wind farm in the high penetration rate scenario.

Description

Technical Field

[0001] This invention belongs to the field of new energy grid connection and power system control technology, specifically involving a method for peak-shifting scheduling of multiple units in a doubly fed wind farm based on RoCoF safety budget. Background Technology

[0002] With the continuous increase in wind power penetration, doubly-fed induction generators (DFIGs) are playing an increasingly important role in power systems. To compensate for the decrease in system inertia and frequency regulation capability after the replacement of synchronous generators, DFIGs are typically equipped with virtual inertia and droop control to briefly release rotor kinetic energy to support the grid frequency during frequency disturbance events. However, the support process causes a significant drop in turbine speed, requiring a speed recovery phase to return to the maximum power point.

[0003] Existing technologies include dynamic recovery strategies for individual wind turbines, such as deriving the minimum safe recovery time based on the swing equation. Calculate the adaptive recovery time based on the rotor kinetic energy state The total recovery time was determined by combining both factors. This type of method can better balance grid security and generator capacity.

[0004] However, when the scenario is extended to the wind farm level (i.e., N wind turbines operating simultaneously), the existing technology has the following significant problems:

[0005] 1. RoCoF Instantaneous Impact Superposition Problem: When all wind turbines in the field simultaneously enter the recovery phase, the RoCoF disturbances of each turbine to the system will be superimposed, resulting in an instantaneous RoCoF impact that is much greater than that of a single turbine. Even if each turbine itself... The calculation is correct; the cumulative field-level impact may still exceed the RoCoF safety limit, triggering a secondary frequency drop.

[0006] 2. Standalone methods cannot be directly extended to the field level: Existing standalone methods... The formula only guarantees that a single unit does not violate RoCoF constraints. However, when multiple units operate concurrently, each wind turbine cannot be aware of the real-time occupancy status of other units, therefore each turbine calculates its own independently. When combined, they no longer constitute a field-level safe solution.

[0007] 3. Unit heterogeneity is overlooked: Due to factors such as wind speed distribution, historical output, and ranking, the individual wind turbines within a wind farm exhibit different characteristics. , , Significant heterogeneity exists. If all units passively and synchronously recover, the total recovery time at the field level will be stalled by the slowest unit, while the recovery of the fastest unit will cause unnecessary instantaneous impact on the power grid.

[0008] 4. Lack of farm-level coordination mechanism: Existing wind farm control mostly adopts the mode of "each unit executes independently", lacking a dynamic perception and allocation mechanism for the available RoCoF budget at the farm level. Summary of the Invention

[0009] The purpose of this invention is to solve the problem of increased risk of secondary frequency drop caused by the instantaneous impact of RoCoF due to the synchronous recovery of multiple wind turbines in the prior art, and to provide a method for staggered scheduling of doubly fed wind farm speed recovery based on RoCoF safety budget.

[0010] The core concept of this invention lies in: making a standalone machine The power utilization ≤ RoCoF budget inequality is improved to the field-level multi-unit total utilization ≤ dynamic budget constraint, and the use of budget by each unit is distributed in the time dimension through peak scheduling, thereby avoiding the superposition of instantaneous impacts.

[0011] A method for peak-shaving scheduling of doubly-fed induction generator (DFIG) wind farms based on RoCoF safety budget includes the following steps:

[0012] a) When N doubly-fed induction generator (DFIG) wind turbines in the wind farm complete the frequency support phase and enter the speed recovery phase, the farm-level coordination controller calculates the RoCoF safety budget function in real time:

[0013] ;

[0014] This function characterizes the maximum power surge introduced by the wind power side that the system can still tolerate at time t. During the recovery phase, this function can be approximated as linearly increasing. ,in Reflects the self-healing rate of the synchronous generator unit;

[0015] b) For each wind turbine in the site Calculate the power recovery requirements of a single machine. Minimum recovery time for a single machine and single-machine occupancy intensity , Penetration rate weighting The sampling period;

[0016] c) Press The wind turbines on site should be sorted from largest to smallest, with priority given to starting and restoring the units that are under the most heavy load.

[0017] d) Determine the restart time of each fan in sequence. and recovery time This ensures that the field-level RoCoF budget constraint holds at any time:

[0018] ;

[0019] in for A constantly active fleet;

[0020] Determine the start-up time of each unit :

[0021] Unit 1: =0;

[0022] Unit k≥2: Select the one that satisfies The earliest moment;

[0023] And ensure field-level budget constraints This holds true for all t;

[0024] e) Each wind turbine independently performs recovery according to its assigned start-up time and recovery duration, and the wind farm controller continuously updates the power reference value:

[0025] ;

[0026] in The pullback step size per unit sampling period; when When this happens, the current unit exits recovery and returns to MPPT operation.

[0027] The beneficial effects of this invention are as follows:

[0028] 1) Ensuring field-level power grid security: through dynamic RoCoF budgeting Explicit modeling of field-level safety margins avoids the risk of secondary frequency drops caused by the superposition of recovery impacts from multiple units.

[0029] 2) Orderly coordination of heterogeneous units: according to Prioritizing startup allows units under high load to start first, preventing the slowest units from dragging down the entire plant, while maintaining plant-level safety.

[0030] 3) High adaptability: Utilizing It reflects the speed regulation response of synchronous generator units, allowing for more compact peak shifting when the system has strong speed regulation capability, and automatically extending the recovery window when the speed regulation capability is weak, thus exhibiting strong operating condition adaptability.

[0031] 4) No additional energy storage required: This method is a pure scheduling-level improvement. It can reduce field-level RoCoF impact simply by coordinating the start-up time, without increasing hardware costs.

[0032] 5) Compatible with single-machine control strategies: As a field-level upper-layer scheduling method, the present invention is compatible with any single-machine level. , It is compatible with computational strategies and has good scalability. Attached Figure Description

[0033] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0034] Figure 1 This is a schematic diagram of the system topology and control architecture of the doubly fed wind farm involved in this invention.

[0035] Figure 2 This is a flowchart of the overall control logic of the method of the present invention. Detailed Implementation

[0036] To better understand the technical solution of this application, the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0037] This invention proposes a method for peak-shifting scheduling of doubly-fed wind farms based on RoCoF safety budget for speed recovery. Figure 1 The diagram illustrates the topology and control architecture of the doubly-fed induction generator (DFIG) wind farm system involved in this invention. The wind farm comprises N DFIG wind turbines (WT1 ~ WTN), each equipped with a local controller to perform low-level control for single-unit frequency support and speed recovery. A farm-level coordination controller is configured at the upper level of the wind farm. This controller exchanges the following information with the local controllers via a communication network:

[0038] Local delivery: , , ;

[0039] • Field-level distribution: , .

[0040] The field-level coordination controller simultaneously receives grid-side measurement information. , Used to calculate field-level security budgets.

[0041] Figure 2 The overall control logic flow of the method of the present invention is illustrated, and the method specifically includes the following steps:

[0042] 1) Frequency Disturbance Detection and Support Phase: After a frequency disturbance occurs in the power grid, each wind turbine in the field enters the frequency support phase according to its local control strategy, releasing kinetic energy to support the frequency. During the support process, each wind turbine records... , Status information, etc.

[0043] 2) Switching condition judgment; each fan is independently judged to see if the switching conditions are met. or Approaching When any wind turbine meets the switching conditions, it reports to the plant-level coordination controller, which then plans and schedules the recovery in a unified manner.

[0044] 3) Field-level parameter calculation; the field-level coordination controller calculates parameters in each sampling period. The following calculations are performed internally: (a) First, the grid information is read, and the field-level RoCoF security budget function is calculated. : During the recovery phase It approximates a linear growth form: ,in For the initial budget, RoCoF at the start of the recovery phase, The system self-recovery rate is determined by the response rate of the synchronous generator governor. (b) Receive the power supplied by each wind turbine. , ; Calculate the cost of each wind turbine and The equivalent budget occupancy of the i-th wind turbine within one sampling period is: The proportion of the rated capacity of the i-th fan to the total system capacity. Minimum recovery time for a single machine The calculations are performed by the local controller, which can be implemented using the adaptive calculation method based on rotor kinetic energy state proposed earlier by the applicant. As a field-level coordination layer, this invention is not limited to... The specific calculation form is only used as an input parameter for field-level scheduling; (c) according to Sort by size from largest to smallest.

[0045] 4) Staggered start-up time allocation; (a) Based on the sorting results of single-unit occupancy intensity from largest to smallest, first determine the recovery start-up time of the first unit after sorting. The first unit with the highest priority is initialized first, letting... Secondly, regarding the The field-level coordination controller calculates the recovery start-up time that satisfies the constraints of the allocated units based on the recovery intervals and the current budget constraints. This ensures that after the k-th unit is put into operation and restored, the total occupancy of active units at any given time still meets the farm-level budget limit: in (a) Set of active units at time t, (b) Start-up time The solution can be obtained using one of the following two methods: Method 1: Strict non-overlapping method: Option 2: Budget-feasible approach: Take the satisfaction (c) The earliest time for each unit to recover. Its value should not be less than the lower limit of the recovery time sent by the local controller. .

[0046] 5) Distribution and execution of allocation results; the field-level coordination controller will... Distributed to each local controller. Each wind turbine... Maintain the support phase at its end, in When the recovery phase begins, the local controller will proceed according to the assigned... The additional output is gradually reduced with a preset slope until the maximum power point tracking curve is restored. The field-level scheduling method of this invention is compatible with any single-machine-level speed restoration law and is not limited to specific local restoration implementations. In one optional embodiment, the local controller can... Use rolling updates The single-unit recovery is achieved in the form of evenly distributing the power deficit to each sampling period according to the recovery time and then withdrawing it; this specific implementation has been reflected in the single-unit control scheme proposed by the applicant in the past. As a field-level coordination layer, this invention further realizes the peak-shifting coordination of multiple units on its basis.

[0047] 6) Recovery completion judgment; when a certain fan... When the wind turbine exits recovery mode, it returns to MPPT operation. Once all wind turbines in the field have completed recovery, the frequency event processing ends, and the system returns to normal MPPT operation.

[0048] The method of this invention can dynamically characterize the change of the field-level RoCoF safety budget over time, and orderly stagger the start-up and recovery times of each unit according to the individual characteristics of each unit; ensure that the total field-level occupancy does not exceed the budget boundary at any time; and suppress secondary frequency drops in high-penetration wind power scenarios.

[0049] Through the above methods, the present invention can ensure that the cumulative RoCoF impact at the wind farm level is always within the safety budget during the recovery process in the later stage of multi-unit frequency regulation, avoiding the risk of secondary frequency drop caused by concurrent recovery of multiple units in the prior art, and realizing an intelligent, coordinated and safe frequency response post-processing mechanism at the wind farm level.

Claims

1. A method for peak-shaving scheduling of doubly-fed induction generator (DFIG) wind farms based on RoCoF safety budget, characterized in that, Includes the following steps: a) When N doubly-fed induction generator (DFIG) wind turbines in the wind farm complete the frequency support phase and enter the speed recovery phase, calculate the farm-level RoCoF safety budget function in real time. ; b) Calculate the single-unit power recovery requirement for each wind turbine i in the field. Minimum recovery time for a single machine and single-machine occupancy intensity ,in , As per penetration rate weight, The sampling period; c) Based on the occupancy intensity of a single machine Arrange the wind turbines in the field from largest to smallest; d) Determine the restart time of each fan in sequence. and recovery time This ensures that the field-level RoCoF budget constraint holds at any time: ； in for A constantly active fleet; e) Each fan is allocated according to its specifications. Independent speed recovery is performed, and the wind farm controller continuously updates the power reference values ​​of each unit.

2. The method according to claim 1, characterized in that, The formula for calculating the mid-range RoCoF safety budget function B(t) in step a) is as follows: ； in Let be the system's inertial constant. The maximum allowable rate of frequency change of the system. for The measured rate of change of frequency at any given time.

3. The method according to claim 2, characterized in that, During the recovery phase It approximates a linear growth form: ； in For the initial budget, RoCoF at the start of the recovery phase, The system's self-recovery rate is determined by the response rate of the synchronous generator's speed governor. in, This is due to a power deficit caused by disturbance. This is the response time constant of the synchronous generator speed governor.

4. The method according to claim 1, characterized in that, The lower limit of single-machine recovery time in step b) The local controller of each wind turbine determines the value based on its own rotor kinetic energy state and sends it to the field-level coordination controller as one of the inputs for the field-level peak-shaving scheduling of this invention.

5. The method according to claim 1, characterized in that, The method for determining the start time in step d) is as follows: according to Starting time of the first unit, in descending order of size. =0; For the k≥2th unit, its start-up time Determined by the field-level budget constraint, such that The total active units at any given time shall not exceed .

6. The method according to claim 5, characterized in that, The start time The solution can be obtained using one of the following two methods: 1) Strictly non-overlapping method: ; 2) Budgetary feasible approach: Take the satisfaction The earliest moment.

7. The method according to claim 1, characterized in that, In step e), each wind turbine... Based on the allocated recovery time The additional output during the frequency support phase is gradually reduced with a preset slope until it returns to the maximum power tracking curve; At that time, the first The power reference value of the typhoon turbine remains unchanged at the end of the frequency support phase.

8. The method according to claim 1, characterized in that, The single-machine occupancy intensity The rated capacity of the i-th wind turbine accounts for the proportion of the total system capacity, reflecting the weight of the unit's influence on the field-level RoCoF.

9. The method according to claim 1, characterized in that, This is implemented in the central coordination controller of the wind farm, which receives status information from each wind turbine within the farm. , , ) and grid-side information ( , ), centralized computing { , The information was then distributed to each unit.

Images