Search

CN-122018301-A - Self-adaptive yaw control method, device and medium based on fuzzy logic

CN122018301ACN 122018301 ACN122018301 ACN 122018301ACN-122018301-A

Abstract

The invention belongs to the technical field of intelligent control of wind generating sets, and particularly relates to a self-adaptive yaw control method, equipment and medium based on fuzzy logic. And (3) constructing a virtual yaw system, taking wind direction deviation, wind speed and turbulence intensity as inputs, taking a yaw control instruction as output, and building an initial fuzzy rule base and a membership function. And adopting a genetic algorithm to encode and iterate the fuzzy rule combination and membership function parameters, and obtaining optimal control parameters by taking the maximum wind energy utilization and meeting yaw frequency constraint as targets. And finally, transferring the optimal parameters to an actual fan controller to realize real-time self-adaptive yaw control. According to the yaw control method and the yaw control device, the yaw control instruction can be accurately generated according to the real-time wind conditions and the unit state, the wind energy utilization rate is improved under the condition that the yaw frequency constraint is met, the yaw control of the wind turbine unit is realized, and the power generation efficiency is improved.

Inventors

  • CHEN PANPAN
  • YI CHUNLIN

Assignees

  • 中车山东风电有限公司

Dates

Publication Date
20260512
Application Date
20260410

Claims (10)

  1. 1. An adaptive yaw control method based on fuzzy logic, which is characterized by comprising the following steps: S1, collecting operation data of a wind turbine generator, and storing the collected operation data into a database; S2, cleaning the stored operation data to obtain cleaned effective operation data; s3, preprocessing the cleaned effective operation data, and respectively calculating absolute wind direction, wind direction deviation and turbulence intensity to generate a preprocessed data set; S4, constructing a virtual yaw system, performing yaw simulation by using the generated preprocessing data set, determining wind direction deviation, filtered wind speed and turbulence intensity as input variables, and yaw control instructions as output variables, dividing the domain of each input and output variable, defining corresponding fuzzy subsets, designing initial membership functions of each fuzzy subset, constructing an IF-THEN initial fuzzy rule base based on the fuzzy subsets of the input variables, and building a fuzzy control model to be optimized; S5, optimizing a fuzzy control model to be optimized based on a genetic algorithm, encoding a combination mode of an initial fuzzy rule base and initial membership function parameters into chromosome individuals, performing yaw simulation operation on each chromosome individual by utilizing a pretreatment data set in a virtual yaw system, and screening out an optimal fuzzy rule combination and optimal membership function parameters with highest wind energy utilization rate under the constraint condition of meeting yaw times by iterative calculation based on an evaluation function of a simulation operation result; And S6, migrating the optimal fuzzy rule combination and the optimal membership function parameters to a fan control system, and performing fuzzy reasoning by applying the optimal fuzzy rule combination and the optimal membership function parameters according to the determined fuzzy reasoning mechanism and the determined fuzzy solution method and according to the operation data acquired in real time to generate a yaw control instruction to drive a yaw system to execute yaw action.
  2. 2. The adaptive yaw control method according to claim 1, wherein in step S3, an absolute wind direction is calculated from the relative wind direction and the nacelle yaw position, and a wind direction deviation is calculated from the absolute wind direction and the nacelle yaw position, comprising the steps of: s31 based on relative wind direction And nacelle yaw position The absolute wind direction is calculated through vector synthesis and circumference reduction operation Adding the relative wind direction and the azimuth angle of the cabin, and performing remainder operation taking 2 pi as a mode on the result; S32, defining a set of discrete yaw control instructions x (k) in the virtual yaw system, wherein the discrete yaw control instructions x (k) are used for simulating the yaw action of the actual system; S33, in the virtual yaw system, calculating the absolute wind direction for each simulation time i Yaw position with nacelle Angle difference Δθ (i) = |between them - |; S34, correlating the processing procedures of the steps S31 to S33 for each data point i in the preprocessing data set to generate a wind direction deviation sequence { theta ew (1),θ ew (2),…,θ ew (N) } containing N time points; the wind direction deviation sequence and the corresponding filtered wind speed sequence { V w (1),V w (2),…,V w (N) } together form a preprocessing data set for evaluating the performance of the fuzzy control strategy.
  3. 3. The adaptive yaw control method based on fuzzy logic according to claim 1, wherein in step S4, a virtual yaw system is constructed, yaw simulation is performed by using the generated preprocessing data set, it is determined that the wind direction deviation, the filtered wind speed and the turbulence intensity are taken as input variables, the yaw control command is taken as output variables, the domain of each input and output variable is divided, corresponding fuzzy subsets are defined, and the initial membership function of each fuzzy subset is designed, comprising the following steps: s41, defining a wind direction deviation sequence { theta ew (1),θ ew (2),…,θ ew (N) } in the preprocessing dataset as a first input variable of the fuzzy controller as wind direction deviation; Defining a filtered wind speed sequence { V w (1),V w (2),…,V w (N) } as a second input variable, defining a turbulence intensity sequence I (I) as a third input variable, defining a turbulence intensity, defining an output variable of the fuzzy controller as a yaw control command; s42, setting a basic argument for wind direction deviation variable as [ -pi, pi ] radian, and defining a fuzzy subset based on language values on the basic argument; Setting a basic discourse domain as [ Vcut-in, vrated ] for the filtered wind speed variable, wherein Vcut-in is the cut-in wind speed of the unit, vrated is the rated wind speed, and defining a language value; setting a basic domain of the turbulence intensity variable as [0, imax ], wherein Imax is a preset upper limit value, and defining a language value; Setting a basic domain of arguments as a discrete set of { -1,0,1}, and defining a language value for the yaw control command variable; S43, selecting a triangle function as an initial form of a membership function for each language value of all input and output variables; S44, establishing a set of fuzzy rules in the form of IF-THEN, and mapping the combination of the input language values to the output language values.
  4. 4. The adaptive yaw control method according to claim 3, wherein in step S4, the step of constructing an IF-THEN initial fuzzy rule base based on the fuzzy subset of the input variables, and the step of constructing a fuzzy control model to be optimized specifically includes the steps of: S411, substituting a group of input values extracted from the preprocessing data set at a certain simulation moment I, namely wind direction deviation theta ew (I), filtered wind speed V w (I) and turbulence intensity I, into membership functions of fuzzy subsets defined for each input variable respectively for calculation; S412, traversing each rule Rj in the initial fuzzy rule base established in the step S44, wherein for one rule with the form IFWEisAANDWSisBANDWIisCTHENYRisD; Extracting membership degrees of input values to the precondition subset A, B, C from the membership degree vector calculated in the step S411, wherein A, B, C is a fuzzy subset of the precondition part, and D is a conclusion subset, and the membership degrees are respectively marked as mu A (i), mu B (i) and mu C (i); Calculating the activation intensity alpha j (i) of the current rule under the current input by fuzzy AND operation, wherein alpha j (i) =min (mu A (i), mu B (i), mu C (i)); S413, operating the output fuzzy subset of the rule conclusion part according to the calculated activation intensity alpha j (i) of each rule; and S414, performing defuzzification calculation on the output fuzzy set mu out (y) obtained in the step S413 to obtain a yaw control instruction x (i).
  5. 5. The adaptive yaw control method based on fuzzy logic of claim 1, wherein S5 specifically comprises the steps of: S51, performing mixed coding on membership function parameters of a fuzzy controller and rule indexes in a fuzzy rule base to form a chromosome individual containing floating point numbers and integers; S52, randomly generating a plurality of chromosome individuals meeting parameter boundaries and sequence constraints based on the set population scale and chromosome coding space to form an initial population; s53, decoding each chromosome individual in the current population, configuring the chromosome individual into a virtual yaw system, performing simulation operation by utilizing a preprocessing data set, and calculating an adaptability value of wind energy capturing quantity and yaw frequency constraint corresponding to each individual according to an evaluation function; s54, selecting, crossing and mutating the current population according to the fitness value of the individual to generate a next generation population meeting the constraint; and S55, repeatedly executing fitness calculation and genetic operation until the maximum evolution algebra is reached, screening out chromosomes with the highest fitness from the calendar individuals, and decoding to obtain the optimal membership function parameters and the optimal fuzzy rule base.
  6. 6. The method for adaptive yaw control based on fuzzy logic of claim 5, wherein S53 comprises the steps of, For each chromosomal individual Indk in the current population P (g), k=1, 2,..m, the following operations are performed: decoding the chromosome, recovering to obtain a specific membership function parameter and a complete fuzzy rule base, and instantiating a fuzzy controller in a virtual yaw system; Using a fuzzy controller obtained by individual decoding to generate a control instruction x (i) time by time according to historical wind condition data, and recursively calculating a simulated cabin yaw position theta y (i) and accumulated yaw times D; after the simulation is finished, calculating total available wind energy E under the section of historical data based on the following formula; In the formula, To and air density The wind sweeping area A of the unit has a related function; calculating the total simulation yaw times D based on the following formula; Substituting E and D into the evaluation function: ; For errors of various moments Is a sum of the sums of (1); indicating when the actual yaw number D exceeds the design value Introducing a second penalty term at 95% of (2); The penalty weight coefficient; Fitness value Fitnessk of chromosome individual Indk is calculated, and fitness value Fitnessk is used as the value of J.
  7. 7. The adaptive yaw control method based on fuzzy logic of claim 6, wherein in step S54, Performing selection, crossover and mutation operations on the current population P (g) based on the fitness value calculated in the step S53 to generate a next generation population P (g+1); Selection, using roulette selection, the probability that individual Indk is selected is proportional to its Fitness Fitnessk, i.e. Pselect (k) = Fitnessk/Σfitness; repeatedly selecting M times, and selecting M parent individuals to enter a mating pool; the crossing operation, namely, pairing individuals in a mating pool in pairs randomly, and determining whether to execute crossing or not according to crossing probability Pc for each pair of parent individuals; If the method is executed, simulating binary crossover is adopted for a real-value coding part, single-point crossover is adopted for a rule part of integer coding, partial gene information is exchanged, and two sub-generation individuals are generated; a mutation operation, in which mutation is performed on each gene position by using mutation probability Pm on the offspring individuals generated after crossing; The method comprises the steps of generating a small random disturbance in a value range of a real-valued gene by using polynomial variation, randomly resetting the real-valued gene to a new effective index value in the value range of an integer gene, enabling all variation operations to meet parameter boundary constraint and sequence constraint of a fuzzy subset, and forming a child population P (g+1) by a new individual set generated by crossing and variation.
  8. 8. The adaptive yaw control method based on fuzzy logic of claim 1, wherein S6 specifically comprises the steps of: s61, acquiring second-level operation data of a wind turbine generator through a SCADA system, preprocessing, calculating real-time absolute wind direction, real-time wind direction deviation and real-time turbulence intensity, and generating a real-time preprocessing data set; S62, an optimal fuzzy rule combination and an optimal membership function parameter are derived from the virtual yaw system; s63, invoking real-time wind direction deviation, real-time filtering wind speed and real-time turbulence intensity in the real-time preprocessing data set, and calculating membership values of each input variable to the corresponding fuzzy subset based on the optimal membership function; s64, invoking a Mamdani fuzzy reasoning mode, and reasoning an input membership value based on an optimal fuzzy rule combination to obtain an output fuzzy set; s65, outputting a discrete yaw control instruction to a fan control system, controlling a yaw motor to execute corresponding yaw motions, and collecting yaw position feedback signals to update unit state data.
  9. 9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the fuzzy logic based adaptive yaw control method of any one of claims 1 to 8 when the program is executed.
  10. 10. A storage medium having stored thereon a computer program, which when executed by a processor, implements the steps of the fuzzy logic based adaptive yaw control method of any one of claims 1 to 8.

Description

Self-adaptive yaw control method, device and medium based on fuzzy logic Technical Field The invention belongs to the technical field of intelligent control of wind generating sets, and particularly relates to a self-adaptive yaw control method, equipment and medium based on fuzzy logic. Background The yaw system is a key component of the horizontal axis wind turbine generator system and functions to drive the nacelle in alignment with the wind direction to maximize wind energy capture efficiency. At present, a dead zone control strategy is adopted by the wind turbine generator, so that the wind turbine generator cannot adapt to complex and changeable natural wind conditions, and the wind turbine generator is often in dilemma that a yaw motor is easy to cause frequent actions under the working conditions of severe turbulence and changeable wind direction, the mechanical abrasion of bearings, gears and drivers in a yaw system is aggravated, and the service life of the wind turbine generator is shortened. When the wind direction is stable, the unit is in a non-optimal wind angle for a long time due to slow response, and energy loss is caused. In addition, the input and output variables of fuzzy control in the prior art are not fully selected, and the wind conditions with different fluctuation degrees cannot be adapted. The partitioning of the fuzzy parameter set and membership degree design are not combined with virtual simulation to carry out early-stage debugging, so that the model suitability is poor, complex and changeable wind conditions are difficult to deal with, and the control precision is insufficient. In the effective optimization process, excessive yaw easily causes excessive loss of components such as a motor, a gear box and the like, the optimization algorithm is unreasonable to select, the population parameter setting lacks basis, premature convergence easily occurs, and globally optimal control parameters cannot be found, so that the optimization effect is poor. Disclosure of Invention The invention provides a self-adaptive yaw control method based on fuzzy logic, which can improve the accuracy of wind aiming of a unit under different wind speeds and turbulence and reduce the action frequency of a yaw system by collecting meteorological and unit operation data in real time and combining fuzzy reasoning to dynamically adjust the yaw control method, thereby taking the wind energy capturing efficiency and the service life of equipment into account. The method comprises the following steps: S1, collecting operation data of a wind turbine generator, and storing the collected operation data into a database; S2, cleaning the stored operation data to obtain cleaned effective operation data; s3, preprocessing the cleaned effective operation data, and respectively calculating absolute wind direction, wind direction deviation and turbulence intensity to generate a preprocessed data set; S4, constructing a virtual yaw system, performing yaw simulation by using the generated preprocessing data set, determining wind direction deviation, filtered wind speed and turbulence intensity as input variables, and yaw control instructions as output variables, dividing the domain of each input and output variable, defining corresponding fuzzy subsets, designing initial membership functions of each fuzzy subset, constructing an IF-THEN initial fuzzy rule base based on the fuzzy subsets of the input variables, and building a fuzzy control model to be optimized; S5, optimizing a fuzzy control model to be optimized based on a genetic algorithm, encoding a combination mode of an initial fuzzy rule base and initial membership function parameters into chromosome individuals, performing yaw simulation operation on each chromosome individual by utilizing a pretreatment data set in a virtual yaw system, and screening out an optimal fuzzy rule combination and optimal membership function parameters with highest wind energy utilization rate under the constraint condition of meeting yaw times by iterative calculation based on an evaluation function of a simulation operation result; And S6, migrating the optimal fuzzy rule combination and the optimal membership function parameters to a fan control system, and performing fuzzy reasoning by applying the optimal fuzzy rule combination and the optimal membership function parameters according to the determined fuzzy reasoning mechanism and the determined fuzzy solution method and according to the operation data acquired in real time to generate a yaw control instruction to drive a yaw system to execute yaw action. According to another embodiment of the application, an electronic device is provided comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the adaptive yaw control method based on fuzzy logic when executing the program. According to yet another embodiment of the present application,