CN-122000933-A - Wind farm primary frequency modulation method and device

CN122000933ACN 122000933 ACN122000933 ACN 122000933ACN-122000933-A

Abstract

The invention relates to the technical field of power system control, in particular to a method and a device for primary frequency modulation of a wind farm. The method comprises the steps of obtaining historical data of primary frequency modulation of a wind power plant, wherein the historical data comprise wind speed data, wind power set operation parameters, power grid frequency data, frequency deviation data and primary frequency modulation trigger signal data, inputting the historical data into a wind power output and power grid frequency mutation prediction model to obtain a wind power output predicted value and a power grid frequency mutation amplitude predicted value of a preset short time period in the future, obtaining corresponding wind power plant frequency modulation control parameters by means of the wind power output predicted value, the power grid frequency mutation amplitude predicted value and a preset multi-objective solving function, and carrying out primary frequency modulation on the wind power plant based on electromagnetic torque adjustment quantity and rotor kinetic energy adjustment quantity. Therefore, the control precision of primary frequency modulation of the wind power plant can be met on the premise of ensuring economy.

Inventors

SU GUOWEI
ZHANG PENGFEI
WANG BO
SONG JU
ZHANG HONGMIN
FAN ZHIHUI
SHI ZHIHUA
WANG BO
ZHAO AQIN
YAN LIWEI
LIU ZHIQIANG

Assignees

华润新能源投资有限公司山西分公司

Dates

Publication Date: 20260508
Application Date: 20260206

Claims (10)

1. A method for primary frequency modulation in a wind farm, comprising: Acquiring historical data of primary frequency modulation of a wind power plant, wherein the historical data comprises wind speed data, wind power unit operation parameters, power grid frequency data, frequency deviation data and primary frequency modulation trigger signal data; The historical data is input into a wind power output and power grid frequency mutation prediction model to obtain a wind power output predicted value and a power grid frequency mutation amplitude predicted value of a preset short time period in the future, wherein the wind power output predicted value comprises maximum active support power and predicted reference output, and the wind power output and power grid frequency mutation prediction model is constructed according to a long-short-period memory network; obtaining a wind power plant frequency modulation control parameter by using the wind power output predicted value, the power grid frequency mutation amplitude predicted value and a preset multi-objective solving function, wherein the wind power plant frequency modulation control parameter comprises an electromagnetic torque adjusting quantity and a rotor kinetic energy adjusting quantity; and performing primary frequency modulation on the wind farm based on the electromagnetic torque adjustment quantity and the rotor kinetic energy adjustment quantity.
2. The method of claim 1, wherein the multi-objective solution function is constructed by the formula: in the formula, For the function values of the multi-objective solution function, As a first weight coefficient, a first set of weights, For the cost of primary frequency modulation operation, As a result of the second weight coefficient, For the cost of the kinetic energy loss of the rotor, As a result of the third weight coefficient, For the fourth weight coefficient, the first weight coefficient, For the unit safety punishment item, The kinetic energy of the rotor is regulated to be the same, For the amount of electromagnetic torque adjustment, Is the cost coefficient of the unit active support energy consumption, For the total active support power quantity, For the maximum support power loss factor, For the maximum active support power, For the final rotational speed of the motor, For the purpose of the transmission efficiency of the unit, For the predicted reference output force, As a function of the frequency-power response coefficient, Is a predicted value of the sudden change amplitude of the power grid frequency, Is the cost coefficient of loss of unit kinetic energy, For the initial rotational speed of the motor, Is the moment of inertia of the rotor.
3. The method of claim 2, wherein the unit security penalty term is determined by the formula: in the formula, For the first rotational speed out-of-limit penalty factor, For a safe rotational speed boundary of the first rotor, For the second rotational speed out-of-limit penalty factor, For a safe rotational speed boundary of the second rotor, The amount of kinetic energy for the first rotor is regulated, The boundary is adjusted for the kinetic energy of the first rotor, For the current kinetic energy of the rotor, The amount of kinetic energy of the second rotor is regulated, The boundary is adjusted for the second rotor kinetic energy.
4. The method according to claim 1, wherein the obtaining the wind power plant frequency modulation control parameter by using the wind power output predicted value, the power grid frequency mutation amplitude predicted value and a preset multi-objective solving function includes: Taking the electromagnetic torque adjustment quantity and the rotor kinetic energy adjustment quantity as particle groups, randomly generating fifty particles to form an initial population, and setting the maximum iteration number as one hundred; substituting the wind power output predicted value and the power grid frequency mutation amplitude predicted value into the multi-objective solving function for iteration, adjusting the particle speed and the position by dynamic inertia weight for each iteration, and simultaneously reserving each generation of optimal particles; and stopping iteration when the iteration times reach one hundred or more function values of the objective solving function and converging, and storing the final optimal particles as the wind power plant frequency modulation control parameters.
5. The method according to claim 1, wherein when the absolute value of the grid frequency mutation amplitude prediction value is greater than a preset frequency mutation amplitude value, the following operations are preferably performed: Determining a first virtual inertia control additional power command based on the initial virtual inertia coefficient; The first virtual inertia control additional power instruction is brought into a wind turbine generator simulation model to obtain simulation data, wherein the simulation data comprise simulation full-period frequency change rate, simulation full-period rotor rotating speed and simulation full-period frequency deviation; The simulation data are carried into a first objective function to be optimized, and an optimal virtual inertia coefficient corresponding to the first objective function value when the first objective function value is minimum is obtained; Determining an optimal virtual inertia control additional power instruction based on the optimal virtual inertia coefficient; and controlling the wind turbine based on the optimal virtual inertia control additional power instruction.
6. The method of claim 5, wherein the first virtual inertial control additional power command is determined by the following equation: in the formula, Additional power commands for the first virtual inertial control, For the initial virtual inertia coefficient in question, For the initial rate of change of the frequency, Is the derivative of the initial frequency rate of change.
7. The method of claim 6, wherein the wind turbine simulation model is constructed by the following equation: in the formula, Is the rotational inertia of a rotor of the wind turbine generator, Is the mechanical torque of the wind turbine generator, Is the electromagnetic torque of the wind turbine generator, Is the damping torque of the wind turbine generator, Is the electromagnetic power of the wind turbine generator, For the simulated full period rotor speed, The power is the basic output power of the unit, For the equivalent inertia of the power grid, For the power difference of the power grid, For the power of the grid load fluctuation, For the simulation of the full period frequency deviation, As the actual value of the time-series frequency, For the rated frequency of the power grid, And simulating the full period frequency change rate for the simulation.
8. A wind farm primary frequency modulation apparatus, comprising: The system comprises an acquisition unit, a control unit and a control unit, wherein the acquisition unit is configured to acquire historical data of primary frequency modulation of a wind farm, and the historical data comprise wind speed data, wind turbine running parameters, power grid frequency data, frequency deviation data and primary frequency modulation trigger signal data; the first data processing unit is configured to input the historical data into a wind power output and power grid frequency mutation prediction model to obtain a wind power output predicted value and a power grid frequency mutation amplitude predicted value of a preset short time period in the future, wherein the wind power output predicted value comprises maximum active support power and predicted reference output, and the wind power output and power grid frequency mutation prediction model is constructed according to a long-short-term memory network; The second data processing unit is configured to obtain corresponding wind farm frequency modulation control parameters by using the wind power output predicted value, the power grid frequency mutation amplitude predicted value and a preset multi-objective solving function, wherein the wind farm frequency modulation control parameters comprise electromagnetic torque adjustment quantity and rotor kinetic energy adjustment quantity; And a third data processing unit configured to primary tune the wind farm based on the electromagnetic torque adjustment amount and the rotor kinetic energy adjustment amount.
9. An electronic device comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the method of any of claims 1-7 when the computer program is executed.
10. A computer readable storage medium, having stored thereon a computer program which, when executed in a computer, causes the computer to perform the method of any of claims 1-7.

Description

Wind farm primary frequency modulation method and device Technical Field The invention relates to the technical field of power system control, in particular to a method and a device for primary frequency modulation of a wind farm. Background With the continuous improvement of the permeability of wind power, the participation of wind power stations in grid frequency modulation has become a key requirement for guaranteeing the frequency stability of a power system. The primary frequency modulation is used as a first defense line for stabilizing the frequency of a power grid, belongs to millisecond-second inertial response, and aims to inhibit frequency mutation by rapidly releasing/absorbing rotor kinetic energy, slow down the frequency change rate and maintain the stability of the frequency of the system. The conventional method for optimizing primary frequency modulation of the wind power plant based on traditional numerical values, such as dynamic planning, genetic algorithm and the like, has the defects that firstly, the conventional simulation method has poor adaptability to wind power output fluctuation and load time variability, and is difficult to characterize a frequency modulation dynamic process under complex working conditions, and secondly, the economic dispatching target and the frequency modulation performance have insufficient collaborative optimization capability, and the cost is the lowest or the frequency deviation is the smallest due to multiple single consideration, so that the dynamic balance of multiple targets is not realized. Based on the above, the invention provides a method and a device for primary frequency modulation of a wind farm to solve the technical problems. Disclosure of Invention The invention describes a method and a device for primary frequency modulation of a wind power plant, which can meet the control precision of the primary frequency modulation of the wind power plant on the premise of ensuring economy. According to a first aspect, the invention provides a method for primary frequency modulation of a wind farm, comprising: Acquiring historical data of primary frequency modulation of a wind power plant, wherein the historical data comprises wind speed data, wind power unit operation parameters, power grid frequency data, frequency deviation data and primary frequency modulation trigger signal data; The historical data is input into a wind power output and power grid frequency mutation prediction model to obtain a wind power output predicted value and a power grid frequency mutation amplitude predicted value of a preset short time period in the future, wherein the wind power output predicted value comprises maximum active support power and predicted reference output, and the wind power output and power grid frequency mutation prediction model is constructed according to a long-short-period memory network; obtaining a wind power plant frequency modulation control parameter by using the wind power output predicted value, the power grid frequency mutation amplitude predicted value and a preset multi-objective solving function, wherein the wind power plant frequency modulation control parameter comprises an electromagnetic torque adjusting quantity and a rotor kinetic energy adjusting quantity; and performing primary frequency modulation on the wind farm based on the electromagnetic torque adjustment quantity and the rotor kinetic energy adjustment quantity. According to a second aspect, the invention provides a wind farm primary frequency modulation device, comprising: The system comprises an acquisition unit, a control unit and a control unit, wherein the acquisition unit is configured to acquire historical data of primary frequency modulation of a wind farm, and the historical data comprise wind speed data, wind turbine running parameters, power grid frequency data, frequency deviation data and primary frequency modulation trigger signal data; the first data processing unit is configured to input the historical data into a wind power output and power grid frequency mutation prediction model to obtain a wind power output predicted value and a power grid frequency mutation amplitude predicted value of a preset short time period in the future, wherein the wind power output predicted value comprises maximum active support power and predicted reference output, and the wind power output and power grid frequency mutation prediction model is constructed according to a long-short-term memory network; The second data processing unit is configured to obtain corresponding wind farm frequency modulation control parameters by using the wind power output predicted value, the power grid frequency mutation amplitude predicted value and a preset multi-objective solving function, wherein the wind farm frequency modulation control parameters comprise electromagnetic torque adjustment quantity and rotor kinetic energy adjustment quantity; And a third data processing unit configured to pri