CN-122001033-A - Inverter control method and system based on edge rolling optimization

CN122001033ACN 122001033 ACN122001033 ACN 122001033ACN-122001033-A

Abstract

The invention discloses an inverter control method and system based on edge rolling optimization, which are applied to the technical field of power system control, wherein the method comprises the steps of collecting voltage and current of an inverter port and actual state data of a power grid in real time; and then, constructing a short-time domain rolling optimization model, generating an initial control instruction by solving a power optimal allocation problem in a four-quadrant operation domain, and issuing and executing the initial control instruction. After the instruction is executed, the system calculates the voltage deviation rate by comparing the actual voltage with a preset reference value, and triggers different correction mechanisms according to the deviation rate, namely, if the deviation rate exceeds a preset threshold, the system directly carries out rolling optimization again, and if the deviation rate is smaller than the preset threshold, the system carries out online correction on the weight coefficient in the rolling optimization model by utilizing an improved sensitivity analysis method. And re-solving the optimization model based on the corrected weight coefficient, outputting a final control instruction, and enabling the power system to accurately control the inverter by the method so as to ensure the stable operation of the power system.

Inventors

LUO QING
Bi Shuixian
XU JINGWEN
ZHANG JIARUI
XIANG XINYU
CHEN YI
CAI YINUO
GE WEIWEI
WANG YANG
SHEN JIAJING

Assignees

国网浙江省电力有限公司杭州供电公司

Dates

Publication Date: 20260508
Application Date: 20260410

Claims (10)

1. An inverter control method based on edge rolling optimization, comprising: acquiring voltage data and current data of an inverter port and actual state data of a power grid; constructing a short-time-domain rolling optimization model, and solving the short-time-domain rolling optimization model based on the voltage data and the current data to obtain a power optimal allocation result in a four-quadrant operation domain; Obtaining a control instruction according to the optimal power distribution result, so that the power system controls the inverter to operate according to the control instruction, and obtains actual voltage data after the control instruction is executed; Obtaining a voltage deviation rate according to the actual voltage data and a preset voltage reference value, if the voltage deviation rate is larger than a preset threshold value, optimizing the short-time-domain rolling optimization model again, and if the voltage deviation rate is smaller than the preset threshold value, correcting a weight coefficient of the short-time-domain rolling optimization model by using an improved sensitivity analysis method based on the actual state data of the power grid to obtain a corrected weight coefficient; and solving the short-time domain rolling optimization model again according to the corrected weight coefficient to obtain a final control instruction, so that the power system controls the inverter to operate according to the final control instruction.
2. The inverter control method based on edge rolling optimization according to claim 1, wherein the solving the short-time domain rolling optimization model based on the voltage data and the current data to obtain the power optimal allocation result in the four-quadrant operation domain comprises: according to the voltage data and the current data, predicting by using a Kalman filtering algorithm to obtain a voltage data predicted value and a current data predicted value; And solving by adopting the short-time domain rolling optimization model and four-quadrant constraint conditions based on the voltage data predicted value and the current data predicted value to obtain a power optimal distribution result in a four-quadrant operation domain.
3. The inverter control method based on edge rolling optimization according to claim 1, wherein the correcting the weight coefficient of the short time domain rolling optimization model based on the actual state data of the power grid by using an improved sensitivity analysis method to obtain the corrected weight coefficient comprises: Establishing a multi-objective optimization function and a multi-objective conflict factor matrix, wherein the multi-objective optimization function comprises a voltage deviation rate minimization objective function, an economic benefit maximization objective function and a device loss minimization objective function; Calculating gradient norms of all objective functions in the multi-objective optimization function; Obtaining sensitivity coefficients of the objective functions according to gradient norms of the targets; And correcting the weight coefficient of the short-time domain rolling optimization model based on the sensitivity coefficient of each objective function and the actual state data of the power grid to obtain the corrected weight coefficient.
4. The inverter control method based on edge rolling optimization according to claim 1, wherein the correcting the weight coefficient of the short-time domain rolling optimization model based on the sensitivity coefficient of each objective function and the actual state data of the power grid to obtain the corrected weight coefficient comprises: judging whether the voltage deviation rate is larger than or equal to a first preset voltage deviation rate threshold value, and if so, correcting the weight coefficient of the short-time-domain rolling optimization model according to a voltage priority mode to obtain a corrected weight coefficient; judging whether the electricity price in the peak time is greater than or equal to a preset electricity price threshold value, and if so, correcting the weight coefficient of the short time domain rolling optimization model according to an economic priority mode to obtain the corrected weight coefficient; judging whether the temperature rise value is greater than or equal to a preset temperature rise threshold value, and if so, correcting the weight coefficient of the short-time-domain rolling optimization model according to the equipment protection mode to obtain a corrected weight coefficient; And if the voltage deviation rate is smaller than the first preset voltage deviation rate threshold value and larger than or equal to the second preset deviation rate threshold value, and the peak period electricity price is smaller than the preset electricity price threshold value, correcting the weight coefficient of the short time domain rolling optimization model by using the sensitivity coefficient of each objective function to obtain the corrected weight coefficient.
5. The edge roll optimization-based inverter control method of claim 1, wherein the re-optimizing the short-time domain roll optimization model comprises: The method comprises the steps of obtaining voltage improvement quantity and renewable energy source power consumption increment of an electric power system after controlling an inverter to operate according to a control instruction; obtaining a contribution factor according to the voltage improvement quantity and the renewable energy power consumption increment, and determining a learning rate according to the contribution factor; according to the learning rate, updating parameters of the equivalent power grid model by using a recursive least square algorithm with forgetting factors to obtain an updated equivalent power grid model; and optimizing the short-time domain rolling optimization model based on the updated equivalent power grid model.
6. The edge roll optimization-based inverter control method of claim 5, wherein the deriving a contribution factor from the voltage improvement amount and the renewable energy power consumption increment comprises: Obtaining a voltage support contribution rate according to the voltage improvement amount and the rated voltage reference value; Obtaining renewable energy consumption contribution rate according to the renewable energy power consumption increment and the rated capacity of the inverter; And obtaining a contribution factor based on the voltage support contribution rate and the renewable energy consumption contribution rate.
7. An inverter control system based on edge roll optimization, comprising: the acquisition module is used for acquiring voltage data and current data of the inverter port and actual state data of the power grid; The power optimal allocation result determining module is used for constructing a short-time domain rolling optimization model, and solving the short-time domain rolling optimization model based on the voltage data and the current data to obtain a power optimal allocation result in a four-quadrant operation domain; The first execution module is used for obtaining a control instruction according to the optimal power distribution result so that the power system can control the inverter to operate according to the control instruction and obtain actual voltage data after the control instruction is executed; The adjustment module is used for obtaining a voltage deviation rate according to the actual voltage data and a preset voltage reference value, if the voltage deviation rate is larger than a preset threshold value, optimizing the short-time-domain rolling optimization model again, and if the voltage deviation rate is smaller than the preset threshold value, correcting the weight coefficient of the short-time-domain rolling optimization model by using an improved sensitivity analysis method based on the actual state data of the power grid to obtain a corrected weight coefficient; And the second execution module is used for solving the short-time domain rolling optimization model again according to the corrected weight coefficient to obtain a final control instruction so that the electric power system controls the inverter to operate according to the final control instruction.
8. The inverter control system based on edge rolling optimization of claim 7 wherein the power optimal allocation result determination module comprises a prediction unit, a prediction window determination unit, and a solution unit, wherein, The prediction unit is used for predicting by using a Kalman filtering algorithm according to the voltage data and the current data to obtain a voltage data predicted value and a current data predicted value; And the solving unit is used for solving by adopting the short-time domain rolling optimization model and the four-quadrant constraint condition based on the voltage data predicted value and the current data predicted value to obtain a power optimal distribution result in the four-quadrant operation domain.
9. A computer device, comprising: A memory for storing a computer program; A processor for implementing the edge roll optimization based inverter control method according to any one of claims 1 to 6 when executing the computer program.
10. A computer storage medium, characterized in that the storage medium has stored thereon a computer program which, when executed by a processor, implements the steps of the edge roll optimization based inverter control method according to any one of claims 1 to 6.

Description

Inverter control method and system based on edge rolling optimization Technical Field The invention relates to the technical field of power system control, in particular to an inverter control method and system based on edge rolling optimization. Background The traditional inverter control mostly adopts local rules based on fixed threshold values or relies on cloud centralized optimization scheduling, the former is difficult to adapt to multi-target dynamic scenes and has limited adjustment precision, and the latter is difficult to meet the rapid disturbance response requirement of a power grid due to communication delay. In recent years, an edge computing technology provides distributed decision support for mass terminals of a power distribution network, but traditional four-quadrant control depends on fixed rules, such as Q-V sagging curves, a dynamic weight adjustment mechanism is lacked, pareto optimality is difficult to achieve among conflict targets of voltage support, economy maximization, equipment loss minimization and the like, accurate control instructions are difficult to obtain, and resource utilization efficiency is low. Disclosure of Invention In order to solve the technical problems, the embodiment of the invention provides an inverter control method and an inverter control system based on edge rolling optimization, which are used for solving the technical problems that the existing inverter control method cannot dynamically coordinate active/reactive output and voltage quality and other multi-objective conflicts, so that accurate control instructions are difficult to obtain. A first aspect of an embodiment of the present invention provides an inverter control method based on edge rolling optimization, the method including: acquiring voltage data and current data of an inverter port and actual state data of a power grid; Constructing a short-time-domain rolling optimization model, and solving the short-time-domain rolling optimization model based on voltage data and current data to obtain a power optimal allocation result in a four-quadrant operation domain; obtaining a control instruction according to the optimal power distribution result, so that the power system controls the inverter to operate according to the control instruction, and obtains actual voltage data after the control instruction is executed; Obtaining a voltage deviation rate according to the actual voltage data and a preset voltage reference value, if the voltage deviation rate is larger than a preset threshold value, optimizing the short-time-domain rolling optimization model again, and if the voltage deviation rate is smaller than the preset threshold value, correcting the weight coefficient of the short-time-domain rolling optimization model by using an improved sensitivity analysis method based on the actual state data of the power grid to obtain a corrected weight coefficient; And solving the short-time domain rolling optimization model again according to the corrected weight coefficient to obtain a final control instruction so that the power system controls the inverter to operate according to the final control instruction. In a possible implementation manner of the first aspect, the solving the short-time domain rolling optimization model based on the voltage data and the current data to obtain a power optimal allocation result in the four-quadrant operation domain includes: according to the voltage data and the current data, predicting by using a Kalman filtering algorithm to obtain a voltage data predicted value and a current data predicted value; and solving by adopting a short-time domain rolling optimization model and four-quadrant constraint conditions based on the voltage data predicted value and the current data predicted value to obtain a power optimal distribution result in the four-quadrant operation domain. In a possible implementation manner of the first aspect, based on the actual state data of the power grid, the modifying the weight coefficient of the short-time domain rolling optimization model by using the improved sensitivity analysis method to obtain the modified weight coefficient includes: Establishing a multi-objective optimization function and a multi-objective conflict factor matrix, wherein the multi-objective optimization function comprises a voltage deviation rate minimization objective function, an economic benefit maximization objective function and a device loss minimization objective function; Calculating gradient norms of all objective functions in the multi-objective optimization function; Obtaining sensitivity coefficients of the objective functions according to gradient norms of the targets; And correcting the weight coefficient of the short-time domain rolling optimization model based on the sensitivity coefficient of each objective function and the actual state data of the power grid to obtain the corrected weight coefficient. In a possible implementation manner of the first aspe