CN-121769872-B - Power fluctuation smooth control method and system for photovoltaic power station cluster

CN121769872BCN 121769872 BCN121769872 BCN 121769872BCN-121769872-B

Abstract

The invention discloses a power fluctuation smooth control method and a system of a photovoltaic power station cluster, in particular relates to the technical field of power system control, and is used for solving the problems of power oscillation and instability caused by neglecting dynamic characteristics of a power grid in the cluster in the conventional smooth control method; by acquiring historical power flow and historical smooth control instructions, analyzing information transfer entropy distribution between instructions and power flows under different frequencies to identify dominant oscillation mode frequency, evaluating nonlinear interaction intensity of smooth control loops of all sub-power stations, calculating space-time distribution inertia centers of node oscillation energy, constructing a dynamic instability criterion of a system based on the nonlinear interaction intensity and inertia center deviation track, constructing a dynamic safety constraint set aiming at inhibiting the oscillation mode when the criterion is met, and solving a smooth control instruction optimization model under constraint set limitation to generate and execute new control instructions to all the sub-power stations.

Inventors

WANG JUN
DAI YUXING
ZHENG ZIJIE
Ke Zipeng

Assignees

湖南大学

Dates

Publication Date: 20260508
Application Date: 20260304

Claims (9)

1. The power fluctuation smooth control method of the photovoltaic power station cluster is characterized by comprising the following steps of: S1, acquiring historical power flow and a corresponding historical smoothing control instruction of a power grid in a photovoltaic power station cluster; s2, analyzing information transfer entropy distribution between the historical smoothing control instruction and the historical power flow under different frequencies based on the historical power flow and the historical smoothing control instruction, wherein the information transfer entropy distribution comprises the following steps: Respectively carrying out frequency domain transformation on the historical power flow and the historical smoothing control instruction to obtain a power flow spectrum and a control instruction spectrum; Dividing a power flow spectrum and a control instruction spectrum on a frequency domain according to frequency resolution, respectively calculating conditional probability transfer entropy from a historical smoothing control instruction to a historical power flow for each frequency point, and obtaining information transfer entropy values of each frequency point by comparing a logarithm expected value of conditional probability and unconditional probability; combining the information transfer entropy values corresponding to all frequencies to form information transfer entropy distribution; S3, identifying that the frequency corresponding to the information transfer entropy extremum is the dominant oscillation modal frequency of the power grid in the photovoltaic power station cluster according to the information transfer entropy distribution; s4, evaluating nonlinear interaction strength of smooth control loops of different sub-power stations in the photovoltaic power station cluster under the dominant oscillation mode frequency, and calculating space-time distribution inertia centers of oscillation energy of all nodes under the dominant oscillation mode; s5, constructing a dynamic instability criterion of the system based on the nonlinear interaction strength and the offset track of the space-time distribution inertia center; and S6, when the dynamic instability criterion of the system meets the preset condition, constructing a dynamic safety constraint set aiming at restraining the oscillation mode, solving an optimization model of the smooth control instruction under the limit of the dynamic safety constraint set, and generating and executing a new smooth control instruction to each sub-power station in the photovoltaic power station cluster.
2. The method for smoothing power fluctuations of a photovoltaic power plant cluster according to claim 1, wherein obtaining a historical power flow of a power grid inside the photovoltaic power plant cluster and a corresponding historical smoothing control instruction comprises: Collecting historical power flow from a monitoring device of a power grid in a photovoltaic power station cluster; extracting a historical smooth control instruction from the control instruction record; The historical power flow and the historical smoothing control instruction are matched and aligned through the time stamp, so that the time correspondence and the integrity of the data are guaranteed.
3. The method for smoothing power fluctuations of a photovoltaic power plant cluster according to claim 1, wherein identifying, according to the information transfer entropy distribution, a frequency corresponding to an extremum of information transfer entropy as a dominant oscillation mode frequency of a power grid inside the photovoltaic power plant cluster comprises: positioning all local extreme points in the information transfer entropy distribution; screening extreme points meeting preset saliency conditions from the local extreme points to serve as information transfer entropy extreme values; And determining the frequency of the information transfer entropy extreme value corresponding to the information transfer entropy distribution transverse axis as the dominant oscillation modal frequency of the power grid in the photovoltaic power station cluster.
4. The method for smoothing control of power fluctuations of a photovoltaic power plant cluster according to claim 1, wherein evaluating the nonlinear interaction strength of smoothing control loops of different sub-power plants in the photovoltaic power plant cluster at a dominant oscillation mode frequency and calculating a spatiotemporal distribution inertia center of oscillation energy of each node in the dominant oscillation mode comprises: Extracting dynamic response signals of each sub-power station smoothing control loop near the dominant oscillation mode frequency; Evaluating the nonlinear interaction strength of the smooth control loop under the dominant oscillation mode frequency by calculating nonlinear correlation coefficients between dynamic response signals; meanwhile, node oscillation energy is calculated according to oscillation components of voltage and current of each node; And carrying out weighted average on the node position coordinates by taking the node oscillation energy as a weight to obtain the space-time distribution inertia center of each node oscillation energy in the dominant oscillation mode.
5. The method for smoothing power fluctuations of a photovoltaic power plant cluster according to claim 4, wherein evaluating the nonlinear interaction strength of the smoothing control loop at the dominant oscillation mode frequency by calculating the nonlinear correlation coefficient between dynamic response signals comprises performing phase space reconstruction on the dynamic response signals of each sub-power plant smoothing control loop, calculating the mutual information amount between every two loops based on the reconstructed state vector, obtaining the nonlinear correlation coefficient by normalizing the mutual information amount, and synthesizing the nonlinear correlation coefficients between all loops to obtain the nonlinear interaction strength.
6. The method for smoothing power fluctuation of a photovoltaic power plant cluster according to claim 4, wherein the step of weighted averaging node position coordinates with node oscillation energy as a weight to obtain a space-time distribution inertia center of each node oscillation energy in a dominant oscillation mode comprises the steps of firstly calculating instantaneous oscillation energy of each node in the dominant oscillation mode, taking the instantaneous oscillation energy as a weight coefficient, weighted summing the position coordinates of the node in a power grid topology, and dividing the sum by the total oscillation energy to obtain coordinate values of the space-time distribution inertia center.
7. The method for smoothing power fluctuations of a photovoltaic power plant cluster according to claim 1, wherein constructing a system dynamic instability criterion based on a nonlinear interaction strength and an offset trajectory of a spatiotemporal distribution inertia center comprises: Comparing the nonlinear interaction strength with a preset interaction threshold value, and simultaneously analyzing whether the offset track of the space-time distribution inertia center presents a continuous divergence characteristic or not; When the nonlinear interaction strength exceeds a preset interaction threshold value and the offset tracks of the space-time distribution inertia centers simultaneously present continuous divergence characteristics, the dynamic instability criterion of the system is judged to be established.
8. The method for smoothing power fluctuation of a photovoltaic power plant cluster according to claim 1, wherein when a system dynamic instability criterion satisfies a preset condition, constructing a dynamic safety constraint set targeting to suppress an oscillation mode, and solving an optimization model of a smoothing control instruction under the limitation of the dynamic safety constraint set, generating and executing a new smoothing control instruction to each sub-power plant in the photovoltaic power plant cluster, comprising: when the dynamic instability criterion of the system is established, a dynamic safety constraint set for limiting the change rate of the control instruction is constructed based on the dominant oscillation mode frequency; under the limitation of a dynamic safety constraint set, solving an optimization model of the smooth control instruction by taking the minimized cluster total power fluctuation as an objective function to obtain a new smooth control instruction of each sub-power station; and issuing a new smoothing control instruction to each sub-power station in the photovoltaic power station cluster for execution.
9. A power fluctuation smoothing control system of a photovoltaic power plant cluster, configured to implement the power fluctuation smoothing control method of a photovoltaic power plant cluster according to any one of claims 1 to 8, comprising: the information acquisition module is used for acquiring the historical power flow of the power grid in the photovoltaic power station cluster and a corresponding historical smoothing control instruction; The distribution analysis module is used for analyzing information transfer entropy distribution between the historical smoothing control instruction and the historical power flow under different frequencies based on the historical power flow and the historical smoothing control instruction; The dominant recognition module is used for recognizing that the frequency corresponding to the information transfer entropy extremum is the dominant oscillation modal frequency of the power grid in the photovoltaic power station cluster according to the information transfer entropy distribution; The information analysis module is used for evaluating nonlinear interaction strength of smooth control loops of different sub-power stations in the photovoltaic power station cluster under the frequency of the dominant oscillation mode and calculating the space-time distribution inertia center of oscillation energy of each node under the dominant oscillation mode; The criterion construction module is used for constructing a dynamic instability criterion of the system based on the nonlinear interaction strength and the offset track of the space-time distribution inertia center; and the instruction execution module is used for constructing a dynamic safety constraint set aiming at restraining an oscillation mode when the dynamic instability criterion of the system meets a preset condition, solving an optimization model of the smooth control instruction under the limit of the dynamic safety constraint set, and generating and executing a new smooth control instruction to each sub-power station in the photovoltaic power station cluster.

Description

Power fluctuation smooth control method and system for photovoltaic power station cluster Technical Field The invention relates to the technical field of power system control, in particular to a power fluctuation smooth control method and system for a photovoltaic power station cluster. Background The photovoltaic power station cluster is used for realizing the smoothing of the total power fluctuation by coordinately controlling the output of a plurality of geographically dispersed power stations, is an effective way for improving the large-scale photovoltaic grid-connected performance, generally generates and distributes a smoothing control instruction to each sub-power station according to the deviation between the total power of the cluster and a target value through a centralized or distributed control architecture in the prior art, and is characterized in that the fluctuation of the total power is stabilized in the time dimension through an optimization algorithm or a control strategy so as to meet the grid-connected requirement. The design and analysis of the control strategy focuses on the response characteristics of the control strategy on time sequence and the influence of the control strategy on the external characteristics at the public connection point of the power grid. However, the existing smoothing control methods have limitations when being applied to large photovoltaic clusters with tight electrical connection, the methods generally consider an internal power grid of the clusters as an ideal power transmission channel, neglect the inherent dynamic characteristics of the internal power grid as a complex electromagnetic network, and when a rapidly-changing smoothing control command acts on each power station, the complex circulation and power oscillation are induced in the internal power grid of the clusters, and the power oscillation mode excited by the control command is mutually coupled with the inherent resonance characteristics of the internal power grid of the clusters, so that the oscillation can be amplified and even unstable, the safe operation of the internal power grid of the clusters can be endangered, and the failure of the smoothing control target can be finally caused. Disclosure of Invention Aiming at the technical problems in the prior art, the invention provides a power fluctuation smooth control method and a power fluctuation smooth control system for a photovoltaic power station cluster. The technical scheme for solving the technical problems is as follows: the power fluctuation smooth control method of the photovoltaic power station cluster comprises the following steps: S1, acquiring historical power flow and a corresponding historical smoothing control instruction of a power grid in a photovoltaic power station cluster; S2, analyzing information transfer entropy distribution between the historical smoothing control instruction and the historical power flow under different frequencies based on the historical power flow and the historical smoothing control instruction; S3, identifying that the frequency corresponding to the information transfer entropy extremum is the dominant oscillation modal frequency of the power grid in the photovoltaic power station cluster according to the information transfer entropy distribution; s4, evaluating nonlinear interaction strength of smooth control loops of different sub-power stations in the photovoltaic power station cluster under the dominant oscillation mode frequency, and calculating space-time distribution inertia centers of oscillation energy of all nodes under the dominant oscillation mode; s5, constructing a dynamic instability criterion of the system based on the nonlinear interaction strength and the offset track of the space-time distribution inertia center; and S6, when the dynamic instability criterion of the system meets the preset condition, constructing a dynamic safety constraint set aiming at restraining the oscillation mode, solving an optimization model of the smooth control instruction under the limit of the dynamic safety constraint set, and generating and executing a new smooth control instruction to each sub-power station in the photovoltaic power station cluster. Further, obtaining the historical power flow of the power grid in the photovoltaic power station cluster and the corresponding historical smoothing control instruction includes: Collecting historical power flow from a monitoring device of a power grid in a photovoltaic power station cluster; extracting a historical smooth control instruction from the control instruction record; The historical power flow and the historical smoothing control instruction are matched and aligned through the time stamp, so that the time correspondence and the integrity of the data are guaranteed. Further, based on the historical power flow and the historical smoothing control instruction, analyzing the information transfer entropy distribution between the historical smoot