Search

CN-121976908-A - Multi-variable cooperative control method and system for wind generating set

CN121976908ACN 121976908 ACN121976908 ACN 121976908ACN-121976908-A

Abstract

The invention discloses a multivariable cooperative control method and a multivariable cooperative control system for a wind generating set, which relate to the field of wind generating sets, and comprise a fusion module, a control module and a control module, wherein the fusion module is used for acquiring wind speed, rotating speed, pitch angle, cabin vibration and grid voltage signals of the wind generating set and carrying out normalization processing on the signals so as to construct a multidimensional data matrix; the invention realizes scientific disassembly of a total power regulation target by accurately collecting and optimizing multiple types of operation signals and deeply mining the nonlinear association relation among variables, effectively improves the accuracy of multi-variable cooperative control and greatly reduces the influence of variable coupling interference on the regulation effect.

Inventors

  • Li Jiankan
  • LIN WU
  • WANG YUXI
  • CHEN WEICONG
  • ZHENG WEIJIAN

Assignees

  • 华能湛江风力发电有限公司

Dates

Publication Date
20260505
Application Date
20260311

Claims (10)

  1. 1. A multivariable coordinated control system of a wind turbine, comprising: The fusion module is used for collecting wind speed, rotating speed, pitch angle, cabin vibration and power grid voltage signals of the wind generating set, and carrying out normalization processing on the signals to construct a multidimensional data matrix; the analysis module is used for receiving the multidimensional data matrix, mining nonlinear association relations among variables in the multidimensional data matrix, and outputting a quantized coupling coefficient table and an association direction identifier; The distribution module is used for acquiring a coupling coefficient table, applying the coupling coefficient table and the current operating condition threshold of the unit, and disassembling the total power adjustment target into independent control sub-targets of pitch angle, rotating speed and exciting current; the generation module is used for introducing variable coupling compensation factors to each control sub-target, constructing a time-varying parameter control equation and generating real-time control instructions of pitch adjustment, rotation speed adjustment and excitation adjustment; the scheduling module is used for receiving real-time control instructions, sequencing the response priorities of the execution mechanisms corresponding to the instructions, and controlling the action time sequence and the amplitude of the pitch drive, the converter and the braking system through a time sequence allocation strategy; The correction module is used for collecting the actual output parameters of each actuating mechanism after action, carrying out deviation calculation with the control sub-target, and feeding back the deviation value to the generation module to update the control equation parameters.
  2. 2. The multi-variable cooperative control system of a wind turbine according to claim 1, wherein the fusion module, when normalizing each signal, obeys: ; wherein: the output value after normalization of the ith class of signals; The original acquisition value of the i-th signal; for the signal in the last sampling period A minimum value in; for the signal in the last sampling period An inner maximum value; 、 Respectively a preset upper limit and a preset lower limit of the normalized output value; The sampling period And dynamically adjusting the current operation condition of the unit, wherein the dimension of the multidimensional data matrix is m multiplied by n, m is the number of signal categories, and n is the number of sampling points in a single sampling period.
  3. 3. The multi-variable cooperative control system of a wind generating set according to claim 1, wherein the analysis module digs a nonlinear association relation stage between variables, and adopts a mutual information calculation mode based on an improved kernel function, and the improved kernel function is: ; wherein: To improve the kernel function output value; the Euclidean distance between the ith variable and the jth variable; Is a dynamic kernel width parameter; the time sequence fluctuation coefficient of the ith variable and the jth variable at the time t is used as the time sequence fluctuation coefficient; each coupling coefficient in the output coupling coefficient table ; Wherein: the coupling coefficient of the ith variable and the jth variable; the mutual information value of the ith variable and the jth variable is obtained; Time sequence association weights of the ith class variable and the jth class variable; a self-information value for a class i variable; a self-information value for a j-th class variable; The associated direction identification adopts a bidirectional array The representation is made of a combination of a first and a second color, =1 Indicates that the class i variable has a positive correlation effect on the class j variable, = -1 Indicates that the i-th variable has a reverse-associated influence on the j-th variable, =0 Indicates that the two variables have no direct correlation effect, and the correlation direction is represented by a cross kernel covariance matrix The sign of the trace of (c) is determined.
  4. 4. The multi-variable cooperative control system of a wind turbine generator system according to claim 1, wherein when the distribution module disassembles the total power adjustment target into independent control sub-targets, a variable association influence matrix M is constructed based on a coupling coefficient table, wherein the dimension of M is 3×3, and the matrix elements correspond to three control variables of pitch angle, rotation speed and exciting current Where p, q ε {1,2,3}, In order to resolve the coupling coefficient output by the module, Influencing the weight for the interaction between the control variables; based on the safety constraint condition corresponding to the current operation condition threshold of the unit, the unit is disassembled through a layered target allocation strategy and matrix solution, and the mapping relation between the total power regulation target and each control sub-target meets the formula: ; wherein: a pitch angle control sub-target, a rotation speed control sub-target and an excitation current control sub-target; an inverse matrix of the correlation influence matrix M; 、 、 weight is distributed to the power of the pitch angle, the rotating speed and the exciting current; Adjusting a target for the total power; the sum of p-th row off-diagonal elements of the association influence matrix; The vector is modified for the security constraint.
  5. 5. The multi-variable cooperative control system of a wind generating set according to claim 1, wherein the variable coupling compensation factor in the generating module is a time-varying parameter which dynamically varies with time, and the value of the variable coupling compensation factor is determined by a coupling coefficient, a variable deviation rate and a set running state evaluation value which are output by the analyzing module together: ; wherein: An output value for a k-th type control instruction; 、 、 is a dynamic proportional coefficient, a dynamic integral coefficient and a dynamic differential coefficient; Real-time deviation values for the k-th class control sub-targets; Is the current moment; Is that Real-time deviation value of k-th control sub-target at moment; a coupling compensation factor for the k-th class control instruction relative to the j-th class associated variable; real-time bias values for class j associated variables.
  6. 6. The multi-variable cooperative control system of a wind turbine according to claim 5, wherein the ; Wherein: for control variables corresponding to control instructions of the kth class coupling coefficients to the j-th class associated variable; real-time deviation rate of the j-th class associated variable; A real-time running state evaluation value of the unit; Adjusting the coefficient for the deviation sensitivity; is a dynamic correction term.
  7. 7. The multi-variable cooperative control system of a wind generating set according to claim 1, wherein the time sequence allocation strategy in the scheduling module is constructed based on response characteristic parameters of an executing mechanism and priority coefficients of control instructions; the response characteristic parameters of the actuating mechanism comprise response delay time, action speed upper limit, load tolerance threshold and interaction interference coefficient, and the priority coefficient ; Wherein: is a preset weight coefficient, which is positive number and the sum of the addition is 1; The deviation weight corresponding to the control instruction is obtained; the maximum value of the coupling coefficient is output by the analysis module; the unit safety priority weight is given; in the time sequence distribution process, the priority coefficient is firstly used Arranging all real-time control instructions in descending order, preferentially distributing an execution window for the high-priority instruction, establishing an interference matrix based on the interaction interference coefficient of the execution mechanism, determining the mutual exclusion interval of the actions of each execution mechanism according to the interference matrix, enabling the action time sequence difference of the adjacent execution mechanisms to be not smaller than the interference avoidance time calculated based on the interference matrix, and enabling the action amplitude superposition value corresponding to the control instructions executed in all the simultaneous periods to not exceed a preset safety threshold value determined based on the safety design standard of the unit; The starting time and the duration of the execution window are dynamically adjusted by the target response speed of the control instruction, the upper limit of the action speed of the execution mechanism and the current load state, and the action amplitude is subjected to self-adaptive amplitude limiting according to the load tolerance threshold of the execution mechanism.
  8. 8. The multivariable coordinated control system of a wind turbine according to claim 1, wherein the deviation calculation stage in the correction module decomposes the deviation value of the actual output parameter and the control sub-target into a static deviation component, a dynamic deviation component and a coupling deviation component, and calculates the integrated deviation value by the following formula: ; wherein: a comprehensive deviation value for a k-th control sub-target; 、 、 is a static deviation weight, a dynamic deviation weight and a coupling deviation weight; Actual output parameters of the k-th control sub-target; setting values for a k-th control sub-target; the change rate of the actual output parameter; setting a change rate for the control sub-target; The actual output value of the related variable of the j-th class; target values for the j-th class associated variables; the integrated deviation value After being fed back to the generation module, the time-varying parameter control equation is synchronously updated 、 、 And 。
  9. 9. The multi-variable cooperative control system of a wind generating set according to claim 1, wherein the fusion module is interactively connected with the analysis module through a wireless network, the analysis module is interactively connected with the distribution module through a wireless network, the distribution module is interactively connected with the generation module through a wireless network, the generation module is interactively connected with the scheduling module through a wireless network, and the scheduling module is interactively connected with the correction module through a wireless network.
  10. 10. A method of multi-variable cooperative control of a wind turbine, the method being implemented in a multi-variable cooperative control system of a wind turbine according to any of claims 1 to 9, comprising: collecting wind speed, rotating speed, pitch angle, cabin vibration and power grid voltage signals of a wind generating set, carrying out normalization processing, and constructing a multidimensional data matrix to integrate multisource operation information; mining nonlinear association among variables in the multidimensional data matrix, and outputting a quantized coupling coefficient table and a bidirectional array for identifying association directions; constructing a variable association influence matrix based on a coupling coefficient table, and disassembling a total power regulation target into an independent control sub-target through a layered target allocation strategy by combining the safety constraint of the unit operation working condition; Introducing variable coupling compensation factors which dynamically change along with time, constructing a time-varying control equation, and generating three types of real-time control instructions for adjustment; according to response characteristics of the execution mechanisms and the priority coefficient sequencing of the control instructions, controlling action time sequence and amplitude of each execution mechanism through a time sequence distribution strategy; and acquiring actual output parameters after the actuating mechanism acts, decomposing the deviation component, calculating a comprehensive deviation value, and synchronously feeding back to a control equation to update equation parameters.

Description

Multi-variable cooperative control method and system for wind generating set Technical Field The invention relates to the technical field of wind generating sets, in particular to a multivariable cooperative control method and a multivariable cooperative control system for a wind generating set. Background The wind generating set captures wind energy through the blades and converts the wind energy into mechanical energy, the driving system drives the generator to generate electricity, the control system is matched to adjust the angle and the rotating speed of the blades, the wind speed change is adapted, and stable and efficient electric energy output is realized. The invention patent application with the application number 202210412210.6 discloses a diagnosis method of a wind driven generator, and aims to solve the problem that the components of the wind driven generator are affected and even damaged to different degrees under the complex working condition of alternating load, so that the state monitoring and the real-time fault diagnosis of the wind driven generator have great significance for monitoring and preventing the damage of all parts and reducing the operation and maintenance cost. However, in the prior art, although each parameter of the wind generating set can be intelligently regulated and controlled, which variables are not considered to be coupled and the regulation precision is insufficient, so that the cooperative regulation and control of each parameter of the wind generating set is poor. Therefore, we propose a multivariable cooperative control method and system for a wind generating set. Disclosure of Invention Aiming at the defects existing in the prior art, the invention provides a multivariable cooperative control method and a multivariable cooperative control system for a wind generating set, which can effectively solve the problems in the prior art. In order to achieve the above object, the present invention is achieved by the following technical scheme; the invention discloses a multivariable cooperative control system of a wind generating set, which comprises: The system comprises a wind generating set, a fusion module, a resolving module, an allocation module, a generation module, a scheduling module, a correction module, a control module and a control module, wherein the wind generating set is used for acquiring wind speed, rotating speed, pitch angle, cabin vibration and power grid voltage signals of the wind generating set, carrying out normalization processing on all signals to construct a multidimensional data matrix, the resolving module is used for receiving the multidimensional data matrix, excavating nonlinear association relation among variables in the multidimensional data matrix, outputting a quantized coupling coefficient table and an association direction mark; The fusion module is interactively connected with an analysis module through a wireless network, the analysis module is interactively connected with a distribution module through the wireless network, the distribution module is interactively connected with a generation module through the wireless network, the generation module is interactively connected with a scheduling module through the wireless network, and the scheduling module is interactively connected with a correction module through the wireless network. Furthermore, when the fusion module performs normalization processing on each signal, the following is: ; wherein: the output value after normalization of the ith class of signals; The original acquisition value of the i-th signal; for the signal in the last sampling period A minimum value in; for the signal in the last sampling period An inner maximum value;、 Respectively a preset upper limit and a preset lower limit of the normalized output value; The sampling period And dynamically adjusting the current operation condition of the unit, wherein the dimension of the multidimensional data matrix is m multiplied by n, m is the number of signal categories, and n is the number of sampling points in a single sampling period. Furthermore, the analysis module digs nonlinear association relation stages among variables, and adopts a mutual information calculation mode based on an improved kernel function, wherein the improved kernel function is as follows: ; wherein: To improve the kernel function output value; the Euclidean distance between the ith variable and the jth variable; Is a dynamic kernel width parameter; the time sequence fluctuation coefficient of the ith variable and the jth variable at the time t is used as the time sequence fluctuation coefficient; each coupling coefficient in the output coupling coefficient table ; Wherein: the coupling coefficient of the ith variable and the jth variable; the mutual information value of the ith variable and the jth variable is obtained; Time sequence association weights of the ith class variable and the jth class variable; a self-information valu