Pneumatic self-cleaning method and device based on blade variable pitch

Abstract

The invention discloses a pneumatic self-cleaning method and device based on variable pitch of a blade, and relates to the technical field of wind power generation, the method comprises the following steps: after a self-cleaning mode is triggered, firstly determining a smudginess degree and a distribution atlas according to multi-sensor monitoring data on the surface of the blade, and then calculating a pneumatic disturbance demand required by stripping pollutants; secondly, the optimal frequency and amplitude combination of the blade variable pitch action is determined through gradient descent optimization or a multi-scene optimization model in combination with current wind regime parameters; then, in order to avoid resonance with the blade intrinsic mode, performing frequency-avoiding adjustment on the frequency combination, and synthesizing an instruction sequence by integrating the blade modal shape; and finally, the variable pitch system is controlled to execute the instruction sequence, so that the blades generate specific aerodynamic disturbance to remove surface pollutants. By means of the method, intelligent and low-energy-consumption online blade self-cleaning is achieved, and the power generation efficiency and operation safety of the wind turbine generator are effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wind power generation, in particular to an aerodynamic self-cleaning method and device based on blade pitch. BACKGROUND

[0002] Currently, when cleaning the blades, it often depends on periodic manual inspection and cleaning. This method has significant hysteresis and passivity. Usually, it can only be started when the pollutants have accumulated to cause significant loss to the aerodynamic performance and power generation capacity. It cannot achieve timely preventive cleaning, resulting in long-term non-optimal state of wind energy capture efficiency.

[0003] In addition, manual cleaning operation itself has great challenges and risks. It requires professional personnel to operate on huge blades in the air with a basket or rope. Not only is the operation cost high and the period long, but it is also severely limited by adverse weather conditions and requires high safety protection for personnel, thereby further exacerbating the inefficiency and high risk of cleaning operation.

[0004] The above content is only used to assist in understanding the technical solutions of the present application and does not represent the acknowledgement of the above content as prior art. SUMMARY

[0005] The main purpose of the present application is to provide an aerodynamic self-cleaning method and device based on blade pitch, aiming to solve the technical problems of low efficiency and insecurity of manual blade cleaning in the prior art.

[0006] To achieve the above-mentioned purpose, the present application provides an aerodynamic self-cleaning method based on blade pitch, the method comprising: When the self-cleaning mode is triggered, determining the aerodynamic disturbance requirement according to the monitored dirt degree of the blade surface; According to the current wind condition parameters and the aerodynamic disturbance requirement, determining the combination of the target frequency and the target amplitude of the blade pitch action; Based on the combination of the target frequency and the target amplitude, combining the blade modal shape, generating a blade pitch instruction sequence; Controlling the blade of the wind turbine generator set to execute the blade pitch action based on the blade pitch instruction sequence.

[0007] In an embodiment, the step of determining the aerodynamic disturbance requirement according to the monitored dirt degree of the blade surface when the self-cleaning mode is triggered comprises: When the self-cleaning mode is triggered, data fusion is performed on the dirt monitoring data returned by the multiple sensors arranged on the leading edge and pressure surface of the blade to obtain the dirt degree of the blade surface; According to the dirt degree of the blade surface, a dirt distribution map of the dirt degree at different spanwise positions is obtained; determining a flow stall critical point and a boundary layer thickening expectation corresponding to the dirt distribution map; determining a minimum local aerodynamic shear force of the shedding contaminant according to the flow stall critical point and the boundary layer thickening expectation, and taking the minimum local aerodynamic shear force as the aerodynamic disturbance demand.

[0008] In an embodiment, the step of determining the flow stall critical point and the boundary layer thickening expectation corresponding to the dirt distribution map comprises: identifying a key dirt area according to the dirt distribution map, and determining a position of the key dirt area in the blade spanwise and chordwise; determining an aerodynamic performance degradation caused by the key dirt area to the blade according to the airfoil of the blade and the position of the key dirt area in the blade spanwise and chordwise; determining the flow stall critical point and the boundary layer thickening expectation according to the aerodynamic performance degradation.

[0009] In an embodiment, the step of determining the combination of the target frequency and the target amplitude of the blade pitch action according to the current wind condition parameter and the aerodynamic disturbance demand comprises: determining a real-time wind speed and a wind direction turbulence intensity according to the current wind condition parameter, and classifying the wind condition into a stable condition, a gradual change condition and a turbulence condition according to the real-time wind speed and the wind direction turbulence intensity; in the stable condition, performing gradient descent optimization in the preset frequency and the preset amplitude based on the aerodynamic disturbance demand to obtain the combination of the target frequency and the target amplitude of the blade pitch action; in the gradual change condition or the turbulence condition, determining the combination of the target frequency and the target amplitude of the blade pitch action according to historical wind data, a short-term predicted wind sequence and the aerodynamic disturbance demand.

[0010] In an embodiment, the step of determining the combination of the target frequency and the target amplitude of the blade pitch action according to historical wind data, a short-term predicted wind sequence and the aerodynamic disturbance demand in the gradual change condition or the turbulence condition comprises: in the gradual change condition or the turbulence condition, constructing a multi-scenario optimization model according to the historical wind data, the short-term predicted wind sequence and the aerodynamic disturbance demand; inputting the real-time wind speed and the wind direction turbulence intensity into the multi-scenario optimization model to output a target scenario; determining a feasible search space of the target scenario, taking the target frequency and the target amplitude of the blade pitch action as optimization variables, and randomly initializing a group of particles in the feasible search space, each particle representing a combination of a frequency and an amplitude; in each iteration, calculating a cleaning index that each particle can achieve in the target scenario, and taking the minimum value in the cleaning index as a fitness value; updating the velocity and position of each particle according to the fitness value, the historical optimal position and the group historical optimal position, and determining a change value of the group optimal fitness; terminating the iteration when the number of iterations reaches a maximum value or the change value of the group optimal fitness is in a preset interval, and determining a combination of a frequency and an amplitude corresponding to the group historical optimal particle as a target frequency and a target amplitude of the blade pitch action.

[0011] In an embodiment, the step of generating the blade pitch instruction sequence based on the combination of the target frequency and the target amplitude and in combination with the blade modal shape includes: obtaining a finite element model of the blade, and extracting a flap modal frequency, a pitch modal frequency and a blade modal shape according to the finite element model; when the target frequency is the same as the flap modal frequency or the pitch modal frequency, adjusting the target frequency based on the flap modal frequency and the pitch modal frequency to obtain an adjusted target frequency; synthesizing the adjusted target frequency, the target amplitude and the blade modal shape into a blade pitch instruction sequence.

[0012] In an embodiment, the step of adjusting the target frequency based on the flap modal frequency and the pitch modal frequency to obtain an adjusted target frequency when the target frequency is the same as the flap modal frequency or the pitch modal frequency includes: when the target frequency is the same as the flap modal frequency or the pitch modal frequency, performing frequency multiplication on the target frequency within a frequency threshold interval to obtain a multiplied frequency; comparing the multiplied frequency with the flap modal frequency or the pitch modal frequency to determine whether there is a frequency conflict; when there is a frequency conflict, adjusting the target frequency to a frequency point with a maximum safety margin in the frequency threshold interval to obtain an adjusted target frequency; when there is no frequency conflict, adjusting the target frequency to the multiplied frequency to obtain a target frequency.

[0013] In an embodiment, the step of synthesizing the adjusted target frequency, the target amplitude and the blade modal shape into a blade pitch instruction sequence includes: for any single blade, synthesizing the adjusted target frequency, the target amplitude and the blade modal shape into a single blade pitch instruction sequence; associating all the single blade pitch instruction sequences to determine a resultant moment fluctuation of aerodynamic exciting forces of the blades at the hub; Adjust each single-blade variable-pitch instruction sequence based on the fluctuation of the resultant moment, to obtain a blade variable-pitch instruction sequence.

[0014] In an embodiment, the step of synthesizing the adjusted target frequency, the target amplitude, and the mode shape into a blade variable-pitch instruction sequence comprises: Perform icing detection on the blade, and determine an ice layer detachment strength threshold when the blade is in an icing condition; Determine an auxiliary vibration instruction according to the ice layer detachment strength threshold, and synthesize the auxiliary vibration instruction, the adjusted target frequency, the target amplitude, and the mode shape into a blade variable-pitch instruction sequence.

[0015] In addition, to achieve the above-mentioned purpose, the present application also provides a blade variable-pitch based aerodynamic self-cleaning device, which comprises: A demand determination module configured to determine an aerodynamic disturbance demand according to a monitored degree of dirtiness of a blade surface when a self-cleaning mode is triggered; A parameter optimization module configured to determine a combination of a target frequency and a target amplitude of a blade variable-pitch action according to current wind condition parameters and the aerodynamic disturbance demand; An instruction generation module configured to generate a blade variable-pitch instruction sequence based on the combination of the target frequency and the target amplitude in combination with a blade mode shape; A variable-pitch control module configured to control a blade of a wind turbine generator to perform a blade variable-pitch action based on the blade variable-pitch instruction sequence.

[0016] In addition, to achieve the above-mentioned purpose, the present application also provides a blade variable-pitch based aerodynamic self-cleaning device, which comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the blade variable-pitch based aerodynamic self-cleaning method as described above.

[0017] In addition, to achieve the above-mentioned purpose, the present application also provides a storage medium, which is a computer-readable storage medium, and the storage medium stores a computer program, the computer program being executable by a processor to implement the steps of the blade variable-pitch based aerodynamic self-cleaning method as described above.

[0018] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which comprises a computer program, the computer program being executable by a processor to implement the steps of the blade variable-pitch based aerodynamic self-cleaning method as described above.

[0019] The application provides a kind of based on blade variable pitch aerodynamic self-cleaning method, by in self-cleaning mode triggers first, according to the dirt degree and distribution atlas of blade surface multi-sensor monitoring data is determined, then the aerodynamic disturbance required for calculating the detachment of pollutant;Second, in combination with current wind condition parameters, by gradient descent optimization or multi-scenario optimization model, the best frequency and amplitude combination of blade variable pitch action is determined;Then, to avoid resonance with blade inherent mode, the frequency combination is adjusted to avoid frequency, and the blade modal shape synthesis instruction sequence is integrated;Finally, control variable pitch system executes the instruction sequence, so that blade produces specific aerodynamic disturbance to remove surface pollutants.By the above-mentioned mode, the method realizes intelligent, low-energy consumption online blade self-cleaning, effectively improves the power generation efficiency and operation safety of wind turbine. BRIEF DESCRIPTION OF DRAWINGS

[0020] The accompanying drawings incorporated in and forming a part of the specification illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the application.

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, those skilled in the art can obtain other drawings from these drawings without creative labor.

[0022] Figure 1 Flowchart of the aerodynamic self-cleaning method based on blade variable pitch of the present application embodiment one; Figure 2 Blade spanwise dirt degree schematic diagram of the aerodynamic self-cleaning method based on blade variable pitch of the present application one embodiment; Figure 3 Blade radial dirt degree schematic diagram of the aerodynamic self-cleaning method based on blade variable pitch of the present application one embodiment; Figure 4 Module structure schematic diagram of the aerodynamic self-cleaning device based on blade variable pitch of the present application embodiment; Figure 5 Device structure schematic diagram of the hardware running environment involved in the aerodynamic self-cleaning method based on blade variable pitch in the present application embodiment.

[0023] The implementation of the present application, functional characteristics and advantages will be further described with reference to the embodiments and accompanying drawings. DETAILED DESCRIPTION

[0024] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application, and are not used to limit the present application.

[0025] For better understanding of the technical solutions of the present application, the following will be described in detail in combination with the drawings of the specification and specific embodiments.

[0026] The main solution of the embodiment of the present application is: when the self-cleaning mode is triggered, the aerodynamic disturbance demand is determined according to the monitored dirt degree of the blade surface; According to the current wind condition parameters and the aerodynamic disturbance demand, the combination of the target frequency and the target amplitude of the blade pitch action is determined; Based on the combination of the target frequency and the target amplitude, the blade pitch instruction sequence is generated in combination with the blade modal shape; The blade pitch action of the wind turbine generator is controlled based on the blade pitch instruction sequence.

[0027] At present, when cleaning the blade, it often depends on periodic manual inspection and cleaning. This method has significant hysteresis and passivity. Usually, it can only be started when the pollutants accumulate to cause obvious loss of aerodynamic performance and power generation capacity. It cannot achieve timely preventive cleaning, resulting in long-term non-optimal state of wind energy capture efficiency.

[0028] In addition, manual cleaning operation itself has great challenges and risks. It requires professional personnel to operate on huge blades in the air with a basket or a rope, which not only has high operation cost and long period, but also is severely limited by adverse weather conditions, and puts high requirements on personnel safety protection, thereby further exacerbating the inefficiency and high risk of cleaning operation.

[0029] The present application provides a solution. After the self-cleaning mode is triggered, the dirt degree and distribution map are first determined according to the multi-sensor monitoring data of the blade surface, and then the aerodynamic disturbance demand required for stripping the pollutants is calculated. Secondly, the best frequency and amplitude combination of the blade pitch action is determined by gradient descent optimization or multi-scenario optimization model in combination with the current wind condition parameters. Then, in order to avoid resonance with the inherent modal of the blade, the frequency combination is adjusted to avoid frequency, and the instruction sequence is synthesized in combination with the blade modal shape. Finally, the instruction sequence is executed to control the pitch system to make the blade produce a specific aerodynamic disturbance to remove the surface pollutants. Through the above method, the intelligent and low-energy online blade self-cleaning is realized, and the power generation efficiency and operation safety of the wind turbine generator are effectively improved.

[0030] It should be noted that the execution subject of the present embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, an aerodynamic self-cleaning device based on blade pitch, etc. The present embodiment does not make specific limitation thereon. The present embodiment and each of the following embodiments will be described below taking the aerodynamic self-cleaning device based on blade pitch as an example.

[0031] All actions of obtaining signals, information or data in this application are carried out in compliance with the corresponding data protection regulations and policies of the country where the corresponding device owner is authorized.

[0032] The embodiment of the present application provides a kind of based on blade variable pitch aerodynamic self-cleaning method, refer to Figure 1 , Figure 1 It is the flow chart of the first embodiment of the aerodynamic self-cleaning method based on blade variable pitch of the present application.

[0033] In the embodiment, the aerodynamic self-cleaning method based on blade variable pitch includes steps S10-S40: Step S10, when the self-cleaning mode is triggered, the aerodynamic disturbance demand is determined according to the monitored degree of dirtiness of the blade surface; It should be noted that the aerodynamic disturbance demand is the minimum aerodynamic load required to effectively remove the pollutants attached to the surface of the blade. The core is to induce specific flow field instability in the blade boundary layer, so as to generate enough wall shear stress to peel off the pollutants.

[0034] It can be understood that in determining the aerodynamic disturbance demand of the degree of dirtiness of the blade surface, the data collected by the multi-sensor array deployed on the leading edge and pressure surface of the blade is relied on, the heterogeneous sensor signals are processed by data fusion algorithm, and high-precision blade surface dirtiness distribution atlas is generated. Subsequently, computational fluid dynamics model or query mechanism based on airfoil aerodynamic database is called, according to the thickness, distribution position and characteristics of dirt, the critical local shear stress required to cause the pollutants to be detached is inversely calculated, and then the mechanical quantity is converted into the preliminary demand parameters of the action amplitude and frequency of the variable pitch system.

[0035] It should be understood that the trigger condition of the self-cleaning mode at least includes: Active trigger based on direct dirt monitoring: by installing online real-time monitoring sensors at key positions of the blade (such as leading edge, pressure surface), the system can continuously obtain the thickness or attachment of pollutants on the surface of the blade. When the sensor reading exceeds the preset cleaning threshold, the system will automatically trigger the self-cleaning mode.

[0036] Predictive trigger based on performance data analysis: by analyzing the operating performance data of the wind turbine generator to indirectly judge the degree of dirt, and then trigger cleaning. The core monitoring parameters include: power curve deviation, wind energy utilization coefficient drop and / or blade force imbalance. When the deterioration trend of these performance indicators exceeds the set threshold, the system will judge that there is significant dirt, and trigger cleaning.

[0037] Intelligent triggering based on environmental conditions and operation strategy: Comprehensive consideration of external environment and grid demand, select the optimal opportunity to trigger, to maximize the benefits and ensure safety. For example, trigger at low wind speed or power limit period, to avoid power loss caused by cleaning at high wind speed and high power generation period, the system will prefer to start cleaning automatically in the idle window period when the wind speed is lower than the cut-in wind speed or the grid dispatching requirement is power limited, to minimize the power loss.

[0038] In a feasible implementation, when the self-cleaning mode is triggered, the step of determining the aerodynamic disturbance requirement according to the monitored dirt level of the blade surface includes: When the self-cleaning mode is triggered, the dirt monitoring data returned by the multiple sensors arranged on the blade leading edge and pressure surface are fused to obtain the dirt level of the blade surface; According to the dirt level of the blade surface, a dirt distribution map of the dirt level at different spanwise positions is obtained; The flow field stall critical point and boundary layer thickening expectation corresponding to the dirt distribution map are determined; The minimum local aerodynamic shear force for stripping the dirt is determined according to the flow field stall critical point and boundary layer thickening expectation, and the minimum local aerodynamic shear force is taken as the aerodynamic disturbance requirement.

[0039] It should be noted that the flow field stall critical point refers to the critical position at which the local airfoil surface flow of the blade begins to separate significantly under a certain dirt profile, which is usually represented by the generation of a leading edge separation bubble or a laminar separation point. The boundary layer thickening expectation refers to the degree of blade thickening caused by dirt in the blade region.

[0040] In a specific implementation, when the dirt distribution map is obtained , the influence of the dirt distribution map on the aerodynamic shape needs to be quantified. In addition, the boundary layer thickening expectation refers to the geometric shape change directly caused by the physical accumulation of dirt on the blade surface, i.e., the equivalent profile thickening amount. The thickening amount can be obtained by integrating the dirt distribution data or using a pre-calibrated dirt thickness-equivalent height mapping relationship through experiments. For example, for a micro-element region, the equivalent aerodynamic roughness height can be associated with the dirt monitoring signal value through an empirical formula: , wherein and are calibration coefficients. Further, the local roughness will significantly change the nominal thickness of the boundary layer.

[0041] After the geometric thickening caused by dirt and the stall critical point obtained through flow field analysis are determined, the minimum local aerodynamic shear force required for cleaning According to Prandtl's boundary layer theory, wall shear stress It is closely related to the boundary layer velocity profile. For turbulent boundary layers, It can be approximated as follows:

[0042] in air density, For the outer edge velocity of the boundary layer, The viscosity of air motion. This represents the total thickness of the boundary layer. (Dirty layer thickening) This will lead to a total thickness Increase( , (for clean state thickness), thus in its natural state It will decrease. Therefore, the aerodynamic disturbance requirement necessitates active control (such as dynamic pitch control) to locally accelerate the airflow and improve the effective flow. To compensate for Increased The stress is attenuated until it reaches the critical shear stress threshold required to strip away contaminants. The system uses iterative simulation or table lookup methods to inversely determine the conditions that satisfy... The required flow field conditions are determined and quantified into a set of specific actuator commands.

[0043] In one feasible implementation, the step of determining the flow field stall critical point and the desired boundary layer thickening in the fouling distribution map includes: The key contamination areas were identified based on the contamination distribution map, and the positions of the key contamination areas in the spanwise and tangential directions of the blades were determined. Based on the airfoil of the blade and the position of the critical contamination area in the spanwise and chordwise directions of the blade, determine the aerodynamic performance degradation caused by the critical contamination area to the blade; The flow field stall critical point and the expected boundary layer thickening are determined based on the aerodynamic performance degradation.

[0044] It should be noted that critical contamination areas refer to those locally highly contaminated areas that have a nonlinear and significant impact on aerodynamic performance. Aerodynamic performance degradation is a quantified performance loss relative to the clean airfoil baseline.

[0045] In the specific implementation, refer to Figure 2 Figure 3 First, analyze the distribution map of the contaminants. Gradient analysis and region clustering are performed to identify key contaminated areas. The algorithm sets a dynamic threshold based on location weights. The threshold value is even lower in sensitive areas such as the leading edge, thus filtering out those that meet the high level of contamination. It is located in an aerodynamically sensitive region, and its spanwise coordinates of the center of mass are accurately recorded. Chord coordinates .

[0046] Subsequently, using the geometric information of the key contaminated areas as input, a lightweight aerodynamic reduced-order model (ROM) or a pre-trained neural network surrogate model is invoked to quickly evaluate the local aerodynamic performance degradation. This model outputs the local lift loss caused by the key contaminated areas. and pressure difference resistance increase The determination of the critical stall point in the flow field is based on a modified critical stall angle of attack. Stall angle of attack of clean airfoil It decreases as the roughness of the leading edge increases, and its empirical relationship can be expressed as follows: ,in This is an empirical coefficient. For the local chord length. The system uses... The equivalent angle-of-attack loss is derived by reverse calculation, thereby determining the new stall critical point.

[0047] For the identified key areas It can be calculated directly from high-resolution dirt distribution maps through geometric relationships. For example, assuming the pollutant density is uniform, then... .

[0048] Step S20: Determine the combination of target frequency and target amplitude of blade pitching action based on the current wind condition parameters and the aerodynamic disturbance requirements; It should be noted that the target frequency refers to the number of complete oscillation cycles that the blade completes per unit time when performing periodic pitch control, and the unit is Hz. The target amplitude refers to the angular amplitude of the pitch angle as it periodically changes around its equilibrium position, and the unit is degrees.

[0049] Understandably, this involves constructing an optimized mapping relationship that takes the current wind conditions and aerodynamic disturbance requirements as input and the target frequency and target amplitude as output. This is typically based on a pre-trained lookup table or lightweight regression model, which implicitly incorporates knowledge obtained from aerodynamic simulations or field test data. Specifically, it identifies the most effective frequency-amplitude pairs required to generate specific areas of the blade (determined by the distribution of dirt) at different wind speeds.

[0050] In one feasible implementation, the step of determining the combination of the target frequency and target amplitude of the blade pitching action based on the current wind condition parameters and the aerodynamic disturbance requirements includes: The real-time wind speed and turbulence intensity in the wind direction are determined based on the current wind condition parameters, and the wind conditions are classified into steady conditions, gradual change conditions and turbulence conditions based on the real-time wind speed and the turbulence intensity in the wind direction. In the stable working condition, gradient descent optimization is performed in the preset frequency and preset amplitude based on the aerodynamic disturbance demand to obtain a combination of a target frequency and a target amplitude of the blade pitch action. In the gradual working condition or the turbulent working condition, a combination of a target frequency and a target amplitude of the blade pitch action is determined according to historical wind data, a short-term predicted wind sequence and the aerodynamic disturbance demand.

[0051] It should be noted that the stable working condition refers to an operating condition in which the change rates of the wind speed and the wind direction are both lower than a set threshold, and the flow field state is relatively stable. The gradual working condition refers to a working condition in which the wind speed or the wind direction presents a slow, continuous and trend change. The turbulent working condition refers to a working condition in which the wind speed or the wind direction exists high-frequency and irregular and violent fluctuations.

[0052] It can be understood that first, the wind speed change rate and the wind direction change rate are compared with preset thresholds and , and the turbulent intensity is combined to classify the wind condition. The classification logic can be formalized as: if and and , it is the stable working condition; otherwise, if , it is the turbulent working condition; if the turbulent intensity is not high but the change rate is out of the threshold, it is the gradual working condition.

[0053] In the stable working condition, since the external excitation is highly predictable, the system adopts a model-based gradient descent method for online optimization. It defines a cleaning performance function:

[0054] wherein is the estimated dirt removal efficiency based on the aerodynamic model, is the pitch action power, is a weight coefficient. The algorithm starts from a preset initial frequency and amplitude , iteratively calculates the partial derivatives (gradients) of to and , and updates the parameters in the opposite direction of , until the optimal combination that maximizes is found.

[0055] For the gradual or turbulent wind conditions, the online optimization converges difficultly and may fail due to the dramatic change of wind conditions. Therefore, the system adopts a data-driven predictive control strategy. It queries the historical wind database to find similar historical scenarios to the current wind conditions and short-term predicted wind sequence (e.g. the average wind speed and turbulence intensity in the next 2 minutes). Through pattern matching (e.g. k-nearest neighbor algorithm), it extracts the frequency-amplitude combination that has been verified to be effective in meeting the current aerodynamic disturbance demand under these similar historical scenarios, and uses it as the target instruction for the current control period.

[0056] In a possible implementation, the step of determining the combination of the target frequency and target amplitude of the blade pitch action according to the historical wind data, the short-term predicted wind sequence and the aerodynamic disturbance demand in the gradual or turbulent wind conditions comprises: constructing a multi-scenario optimization model according to the historical wind data, the short-term predicted wind sequence and the aerodynamic disturbance demand in the gradual or turbulent wind conditions; inputting the real-time wind speed and wind direction turbulence intensity into the multi-scenario optimization model to output a target scenario; determining a feasible search space of the target scenario, taking the target frequency and target amplitude of the blade pitch action as the variables to be optimized, randomly initializing a group of particles in the feasible search space, and each particle representing a combination of frequency and amplitude; in each iteration, calculating the cleaning index that each particle can achieve in the target scenario, and taking the minimum value in the cleaning index as the fitness value; updating the speed and position of each particle according to the fitness value, the historical optimal position and the group historical optimal position, and determining the change value of the group optimal fitness; when the number of iterations reaches the maximum value or the change value of the group optimal fitness is in the preset interval, terminating the iteration, and determining the combination of the frequency and amplitude corresponding to the group historical optimal particle as the target frequency and target amplitude of the blade pitch action.

[0057] In a specific implementation, when the system determines that it is currently in the gradual or turbulent wind condition, it will first start the multi-scenario optimization model. The core of this model is a pre-constructed scenario library, where each scenario is defined by a set of feature vectors, in the form of . Wherein represents the statistical features related to this scenario extracted from the historical wind data; represents the features of the short-term predicted wind sequence corresponding to this historical moment; corresponding historical optimal control effect record. The current real-time wind condition feature and the predicted feature are input into the model, and the Euclidean distance of each scenario in the scenario library is calculated:

[0058] And select the scene with the minimum distance as the current target scene.

[0059] Subsequently, the system determines a feasible search space, i.e. a frequency range and an amplitude range based on the wind condition characteristics of the target scene, and performs fast optimization within this space using a particle swarm optimization algorithm. The algorithm initializes a swarm of particles, each particle position representing a frequency-amplitude combination. In each iteration, the fitness value of each particle is evaluated, which is defined as the negative value of the cleaning index that the combination can achieve under the wind condition of the target scene , because PSO usually solves the minimum value problem. The cleaning index is a comprehensive index, which is directly proportional to the degree of satisfaction of the aerodynamic disturbance demand and inversely proportional to the energy consumption of the variable pitch action.

[0060] The particle updates its speed and position according to the following formula:

[0061]

[0062] wherein is the historical optimal position of particle k, is the historical optimal position of the entire group, is the inertia weight, , is the learning factor, , is a random number. The algorithm continuously monitors the change value of the optimal fitness of the group or the absolute value of of consecutive iterations is less than a very small threshold , the iteration is terminated. Finally, the frequency and amplitude combination corresponding to is output as the optimal control instruction under the current dynamic wind condition.

[0063] Step S30, based on the combination of the target frequency and the target amplitude, generate a blade pitch instruction sequence in combination with the blade modal shape.

[0064] It should be noted that the blade modal shape describes the distribution of the relative amplitude and phase of each point on the blade. The blade pitch instruction sequence contains the target trajectory of each blade pitch angle over time.

[0065] It can be understood that, based on the combination of the target frequency and the target amplitude, in combination with the blade modal shape, the blade pitch instruction sequence is generated, which is to convert the macroscopic aerodynamic control target into a specific actuator operable control instruction and avoid exciting the blade harmful structural vibration. First, the finite element analysis model or the experimental modal parameter database of the blade is queried to obtain the modal shape of the blade under the first few main modes (such as the first-order flapping and the first-order edgewise), so as to determine the sensitive area of the blade to different frequency excitations. Then, a reference sinusoidal signal is generated based on the target frequency f . Then, the phase and amplitude of the reference signal are spatially modulated by using the modal shape information. For example, the pitch action amplitude is appropriately increased near the modal node to strengthen the disturbance effect, and the amplitude is reduced at the modal anti-node to suppress the resonance risk, and finally a blade pitch instruction sequence that may be non-uniform along the blade span is generated.

[0066] In a feasible implementation, the step of generating the blade pitch instruction sequence based on the combination of the target frequency and the target amplitude, in combination with the blade modal shape, comprises: obtaining a finite element model of the blade, and extracting a flapping modal frequency, an edgewise modal frequency and a blade modal shape according to the finite element model; when the target frequency is the same as the flapping modal frequency or the edgewise modal frequency, adjusting the target frequency based on the flapping modal frequency and the edgewise modal frequency to obtain an adjusted target frequency; combining the adjusted target frequency, the target amplitude and the blade modal shape into a blade pitch instruction sequence.

[0067] In a specific implementation, first, a pre-established finite element model of the blade is called, and the flapping modal frequency and the edgewise modal frequency are extracted by solving an eigenvalue problem (wherein is a mass matrix, is a stiffness matrix, is a circular frequency, and is a modal shape). The system compares the target frequency with all the extracted modal frequencies in real time. If , , is a small tolerance percentage, usually 5%, it is considered that there is a resonance risk. At this time, the target frequency is fine-tuned according to the rule of , wherein is a pre-set safety offset, which ensures that the adjusted target frequency avoids all main modal frequencies.

[0068] Subsequently, the time-domain command is modulated according to the spatial characteristics of the blade. The underlying time signal is a sinusoidal wave However, in order to achieve differentiated but coordinated aerodynamic disturbance effects at different spanwise locations of the blade and to avoid local stress concentration, the first order flapping mode shape is used as the weight distribution function. Finally, what is generated is a sequence of instructions for a plurality of control segments on the blade (if independent pitch is supported) or the whole blade. For the jthcontrol segment, the instruction is:

[0069] where is the first order flapping mode shape amplitude at the corresponding location of the segment, is a phase offset introduced for optimization of disturbance effect.

[0070] In an implementable embodiment, the step of adjusting the target frequency based on the flapping mode frequency and the edgewise mode frequency when the target frequency is the same as the flapping mode frequency or the edgewise mode frequency to obtain an adjusted target frequency comprises: when the target frequency is the same as the flapping mode frequency or the edgewise mode frequency, performing integer frequency multiplication on the target frequency within a frequency threshold interval to obtain a multiplied frequency; comparing the multiplied frequency with the flapping mode frequency or the edgewise mode frequency to determine whether there is a frequency conflict; when there is a frequency conflict, adjusting the target frequency to a frequency point with the largest safety margin in the frequency threshold interval to obtain an adjusted target frequency; when there is no frequency conflict, adjusting the target frequency to the multiplied frequency to obtain a target frequency.

[0071] In a specific implementation, when the system detects that the target frequency differs from any flapping mode frequency or edgewise mode frequency by less than a safety threshold (i.e. | |), it means that there is a risk of resonance and adjustment must be made. In order to avoid the possibility that a simple frequency offset may still conflict with other modes at the new frequency point or the disturbance effect is poor, a frequency threshold interval centered on the original may be calculated, i.e. where is a positive integer and satisfies:

[0072] For example, if , , then the possible multiple frequencies are 0.5Hz (itself) when n=1 and 1.0Hz when n=2.

[0073] Next, the system compares each multiple frequency with all the modal frequencies that need to be avoided and performs frequency conflict detection. The conflict condition is: . The system will go through all the candidate points and record the points without conflict. If there is a point without conflict (usually the point with smaller multiple n is preferred because it causes less change to the original control target), the system will directly adjust the target frequency to this multiple frequency, i.e. If all the candidate points have conflict, the system will enable the alternative strategy: within the frequency threshold interval, calculate the minimum distance from each frequency point f to all the modal frequencies, then select the frequency point that maximizes , i.e. the frequency point with the largest safety margin, as the final .

[0074] In an embodiment, the step of synthesizing the adjusted target frequency, the target amplitude, and the blade modal shape into a blade pitch command sequence comprises: for each individual blade, synthesizing the adjusted target frequency, the target amplitude, and the blade modal shape into an individual blade pitch command sequence; associating all the individual blade pitch command sequences to determine the resultant torque fluctuation at the hub center caused by the aerodynamic excitation forces of all the blades; adjusting each individual blade pitch command sequence based on the resultant torque fluctuation to obtain a blade pitch command sequence.

[0075] In a specific implementation, based on the adjusted target frequency and the target amplitude A, and in combination with the flap modal shape of the blade (where is the spanwise position), a basic individual blade pitch command sequence is generated for each individual blade, and the mathematical model of the basic individual blade pitch command sequence can be represented as:

[0076] where is the initial phase. Subsequently, the system calculates the resultant torque fluctuation at the hub center of the wind turbine caused by the aerodynamic excitation forces of all the blades acting according to the basic command sequence by using a multi-body dynamics model or an aerodynamic load model. In order to avoid excessive fatigue load on the entire wind turbine main body structure, the system will minimize the resultant torque fluctuation as the target, and use an optimization algorithm (such as the least square method) to adjust the phase Even small adjustments in its amplitude are made to coordinate fine-tuning, and finally output a set of blade pitch command sequences that can achieve both aerodynamic targets and ensure the stability of the entire machine.

[0077] In a possible implementation, the step of synthesizing the adjusted target frequency, the target amplitude, and the mode into a blade pitch command sequence includes: Performing icing detection on the blade, and determining an ice layer detachment strength threshold when the blade is in an icing condition; Determining an auxiliary vibration command according to the ice layer detachment strength threshold, and synthesizing the auxiliary vibration command, the adjusted target frequency, the target amplitude, and the mode into a blade pitch command sequence.

[0078] In a specific implementation, when the icing detection unit (such as based on image recognition or vibration spectrum analysis) determines that the blade is in an icing condition, the system first looks up or derives an empirical formula based on the characteristics of the ice layer (such as type, thickness, usually estimated by a meteorological model or historical data) to determine the ice layer detachment strength threshold . Subsequently, the system reversely derives the vibration level required to generate an alternating aerodynamic load on the blade surface sufficient to overcome this threshold, and generates a high-frequency, small-amplitude auxiliary vibration command , where is usually higher than the main target frequency to avoid coupling. Finally, the auxiliary deicing command and the main control command based on the adjusted target frequency and target amplitude are synthesized into a composite blade pitch command sequence for the icing condition by linear superposition .

[0079] Step S40: Controlling the blade pitch action of the wind turbine generator set based on the blade pitch command sequence.

[0080] It can be understood that the main controller of the wind turbine generator set converts the calculated blade pitch command sequence into actual physical action. This process is achieved through a control loop: the controller sends the command sequence to the pitch drive of each blade, which is usually a servo motor or a hydraulic system, and the drive drives the pitch bearing to rotate, thereby accurately adjusting the blade pitch action, i.e., changing the angle of attack of the blade relative to the wind. Ensures that the blade can perform periodic pitching motion according to the preset specific mode and frequency, achieving the purpose of blade self-cleaning.

[0081] The embodiment provides a kind of based on blade variable pitch aerodynamic self-cleaning method, by first determining the dirt degree and distribution atlas according to the multi-sensor monitoring data of blade surface after self-cleaning mode triggers, then calculate the aerodynamic disturbance demand required to strip pollutant;Second, the best frequency and amplitude combination of blade variable pitch action is determined by gradient descent optimization or multi-scenario optimization model in combination with current wind condition parameters;Then, to avoid resonance with blade inherent mode, the frequency combination is adjusted to avoid frequency, and the blade modal shape synthesis instruction sequence is integrated;Finally, control variable pitch system executes the instruction sequence, so that blade generates specific aerodynamic disturbance to remove surface pollutants.By the above mode, the method realizes intelligent, low-energy consumption online blade self-cleaning, effectively improves the power generation efficiency and operation safety of wind turbine.

[0082] It should be noted that the above examples are only for understanding the present application and do not constitute a limitation on the aerodynamic self-cleaning method based on blade variable pitch of the present application, and more forms of simple transformation based on this technical concept are within the protection scope of the present application.

[0083] The present application also provides an aerodynamic self-cleaning device based on blade variable pitch, please refer to Figure 4 The aerodynamic self-cleaning device based on blade variable pitch comprises: A demand determination module 10 is configured to determine the aerodynamic disturbance demand according to the monitored blade surface dirt degree when the self-cleaning mode is triggered; A parameter optimization module 20 is configured to determine the combination of target frequency and target amplitude of blade variable pitch action according to the current wind condition parameters and the aerodynamic disturbance demand; An instruction generation module 30 is configured to generate blade variable pitch instruction sequence based on the combination of target frequency and target amplitude in combination with blade modal shape; A variable pitch control module 40 is configured to control the blade of wind turbine to execute blade variable pitch action based on the blade variable pitch instruction sequence.

[0084] In a possible implementation, the demand determination module 10 is further configured to fuse the dirt monitoring data returned by the multi-sensor arranged at the leading edge and pressure surface of the blade to obtain the blade surface dirt degree when the self-cleaning mode is triggered; Obtain the dirt distribution atlas of dirt degree at different spanwise positions according to the blade surface dirt degree; Determine the corresponding flow field stall critical point and boundary layer thickening expectation in the dirt distribution atlas; Determine the minimum local aerodynamic shear force required to strip pollutant according to the flow field stall critical point and boundary layer thickening expectation, and take the minimum local aerodynamic shear force as the aerodynamic disturbance demand.

[0085] In an embodiment, the demand determining module 10 is further configured to identify a key fouling area according to the fouling distribution map, and determine a position of the key fouling area in a spanwise direction and a chordwise direction of the blade; According to the airfoil of the blade and the position of the key fouling area in the spanwise direction and the chordwise direction, determine an aerodynamic performance degradation caused by the key fouling area to the blade; According to the aerodynamic performance degradation, determine a flow field stall critical point and a boundary layer thickening period.

[0086] In an embodiment, the parameter optimization module 20 is further configured to determine a real-time wind speed and a wind direction turbulence intensity according to current wind condition parameters, and classify wind conditions into stable conditions, gradual change conditions and turbulence conditions according to the real-time wind speed and the wind direction turbulence intensity; In the stable conditions, perform gradient descent optimization in a preset frequency and a preset amplitude based on the aerodynamic disturbance demand, to obtain a combination of a target frequency and a target amplitude of the blade pitch action; In the gradual change conditions or the turbulence conditions, determine the combination of the target frequency and the target amplitude of the blade pitch action according to historical wind data, a short-term predicted wind sequence and the aerodynamic disturbance demand.

[0087] In an embodiment, the parameter optimization module 20 is further configured to, in the gradual change conditions or the turbulence conditions, construct a multi-scenario optimization model according to the historical wind data, the short-term predicted wind sequence and the aerodynamic disturbance demand; Input the real-time wind speed and the wind direction turbulence intensity into the multi-scenario optimization model, and output a target scenario; Determine a feasible search space of the target scenario, take the target frequency and the target amplitude of the blade pitch action as to-be-optimized variables, randomly initialize a group of particles in the feasible search space, and each particle represents a combination of a frequency and an amplitude; In each iteration, calculate a cleaning index that can be achieved by each particle in the target scenario, and take a minimum value in the cleaning index as a fitness value; According to the fitness value, a historical optimal position and a group historical optimal position, update a speed and a position of each particle, and determine a change value of a group optimal fitness; When a maximum number of iterations is reached or the change value of the group optimal fitness is in a preset interval, terminate the iteration, and determine a combination of a frequency and an amplitude corresponding to a group historical optimal particle as the target frequency and the target amplitude of the blade pitch action.

[0088] In an embodiment, the instruction generating module 30 is further configured to obtain a finite element model of the blade, and extract a flapwise modal frequency, a edgewise modal frequency and a blade modal shape according to the finite element model; adjust the target frequency based on the flap mode frequency and the edgewise mode frequency when the target frequency is the same as the flap mode frequency or the edgewise mode frequency, to obtain an adjusted target frequency; synthesize the adjusted target frequency, the target amplitude, and the blade mode shape into a blade pitch instruction sequence.

[0089] In an implementable embodiment, the instruction generation module 30 is further configured to, when the target frequency is the same as the flap mode frequency or the edgewise mode frequency, multiply the target frequency by an integer within a frequency threshold interval to obtain a frequency multiplication frequency; compare the frequency multiplication frequency with the flap mode frequency or the edgewise mode frequency to determine whether there is a frequency conflict; when there is a frequency conflict, adjust the target frequency to a frequency point with the largest safety margin in the frequency threshold interval to obtain an adjusted target frequency; when there is no frequency conflict, adjust the target frequency to the frequency multiplication frequency to obtain a target frequency.

[0090] In an implementable embodiment, the instruction generation module 30 is further configured to, for any single blade, synthesize the adjusted target frequency, the target amplitude, and the blade mode shape into a single blade pitch instruction sequence; associate all the single blade pitch instruction sequences to determine a resultant moment fluctuation of the aerodynamic exciting force of each blade at the hub; adjust each single blade pitch instruction sequence based on the resultant moment fluctuation to obtain a blade pitch instruction sequence.

[0091] In an implementable embodiment, the instruction generation module 30 is further configured to perform icing detection on the blade, and determine an ice layer detachment strength threshold when the blade is in an icing working condition; determine an auxiliary vibration instruction according to the ice layer detachment strength threshold, and synthesize the auxiliary vibration instruction, the adjusted target frequency, the target amplitude, and the mode shape into a blade pitch instruction sequence.

[0092] The aerodynamic self-cleaning device based on blade pitch provided in the application adopts the aerodynamic self-cleaning method based on blade pitch in the above embodiments, and can solve the technical problems of low efficiency and insecurity of manual blade cleaning. Compared with the prior art, the aerodynamic self-cleaning device based on blade pitch provided in the application has the same beneficial effects as the aerodynamic self-cleaning method based on blade pitch provided in the above embodiments, and other technical features of the aerodynamic self-cleaning device based on blade pitch are the same as the features disclosed in the above method, which will not be described here.

[0093] The application provides a blade-pitching-based aerodynamic self-cleaning device, which comprises at least one processor and a memory in communication connection with the at least one processor; the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the blade-pitching-based aerodynamic self-cleaning method in the above embodiment one.

[0094] Reference will be made to the following Figure 5 which shows a structural schematic diagram of the blade-pitching-based aerodynamic self-cleaning device suitable for being used to implement the embodiments of the application. The blade-pitching-based aerodynamic self-cleaning device in the embodiments of the application can include, but is not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (Personal Digital Assistant), PADs (Portable Application Description), PMPs (Portable Media Player), vehicle-mounted terminals (for example, vehicle-mounted navigation terminals), and the like, and fixed terminals such as digital TVs, desktop computers, and the like. Figure 5 The blade-pitching-based aerodynamic self-cleaning device shown is only an example, and should not bring any limitation to the functions and use range of the embodiments of the application.

[0095] As Figure 5As shown, the blade pitch based aerodynamic self-cleaning device can include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to programs stored in a ROM (Read Only Memory) 1002 or programs loaded from a storage device 1003 into a RAM (Random Access Memory) 1004. In the RAM 1004, various programs and data required for the blade pitch based aerodynamic self-cleaning device to operate are also stored. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, an LCD (Liquid Crystal Display), a speaker, a vibrator, etc.; the storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the blade pitch based aerodynamic self-cleaning device to communicate with other devices wirelessly or by wire to exchange data. Although the blade pitch based aerodynamic self-cleaning device with various systems is shown in the figure, it should be understood that all the systems shown are not required to be implemented or possessed. More or fewer systems can be alternatively implemented or possessed.

[0096] In particular, according to embodiments of the present disclosure, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program code for executing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network by a communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the methods of the embodiments of the present disclosure are performed.

[0097] The blade pitch based aerodynamic self-cleaning device provided by the present disclosure adopts the blade pitch based aerodynamic self-cleaning method in the above-mentioned embodiments, and can solve the technical problem of blade pitch based aerodynamic self-cleaning. Compared with the prior art, the blade pitch based aerodynamic self-cleaning device provided by the present disclosure has the same beneficial effects as the blade pitch based aerodynamic self-cleaning method provided by the above-mentioned embodiments, and other technical features in the blade pitch based aerodynamic self-cleaning device are the same as the features disclosed in the previous embodiment method, which will not be repeated here.

[0098] It should be understood that various aspects of the disclosure can be implemented in hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0099] The above merely provides a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0100] The present application provides a computer readable storage medium having stored thereon computer readable program instructions (i.e. computer programs) for performing the above-mentioned blade-pitch-based aerodynamic self-cleaning method.

[0101] The computer readable storage medium provided by the present application may, for example, be a U disk, but is not limited to an electric, magnetic, optical, electromagnetic, infrared, or semiconductor system, system or device, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to: an electric connection with one or more conductive wires, a portable computer disk, a hard disk, a RAM (Random Access Memory), a ROM (Read Only Memory), an EPROM (Erasable Programmable Read Only Memory or flash memory), an optical fiber, a CD-ROM (CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present embodiment, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system or device. The program code contained on the computer readable storage medium can be transmitted by any suitable medium, including but not limited to: an electric wire, an optical cable, an RF (Radio Frequency), etc., or any suitable combination of the above.

[0102] The above-mentioned computer readable storage medium can be contained in the blade-pitch-based aerodynamic self-cleaning device; or can exist separately without being assembled into the blade-pitch-based aerodynamic self-cleaning device.

[0103] The computer readable storage medium described above carries one or more programs, when the one or more programs are executed by the blade-pitching based aerodynamic self-cleaning device, cause the blade-pitching based aerodynamic self-cleaning device to: determine an aerodynamic disturbance demand according to a monitored dirt level of a blade surface when a self-cleaning mode is triggered; determine a combination of a target frequency and a target amplitude of a blade-pitching action according to current wind condition parameters and the aerodynamic disturbance demand; generate a blade-pitching instruction sequence based on the combination of the target frequency and the target amplitude in combination with a blade modal shape; control the blade-pitching action of a blade of a wind turbine generator based on the blade-pitching instruction sequence.

[0104] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0105] The flow diagrams and the block diagrams in the drawings are illustrations of architectures, functionalities, and operations of possible implementations of systems, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flow diagrams or block diagrams can represent a module, a segment, or a portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may

[0106] The modules described in the embodiments of the present application can be implemented in the form of software or in the form of hardware. In some cases, the names of the modules do not constitute a limitation on the modules themselves.

[0107] The readable storage medium provided by the present application is a computer readable storage medium, which stores computer readable program instructions (i.e., a computer program) for executing the above-mentioned blade-pitch-based aerodynamic self-cleaning method, and can solve the technical problem of blade-pitch-based aerodynamic self-cleaning. Compared with the prior art, the computer readable storage medium provided by the present application has the same beneficial effects as the blade-pitch-based aerodynamic self-cleaning method provided by the above-mentioned embodiments, and will not be described here.

[0108] The present application also provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the steps of the blade-pitch-based aerodynamic self-cleaning method as described above.

[0109] The computer program product provided by the present application can solve the technical problem of blade-pitch-based aerodynamic self-cleaning. Compared with the prior art, the computer program product provided by the present application has the same beneficial effects as the blade-pitch-based aerodynamic self-cleaning method provided by the above-mentioned embodiments, and will not be described here.

[0110] The above is only some embodiments of the present application, and does not limit the patent scope of the present application. Any equivalent structural transformation made by using the content of the present application specification and drawings, or direct / indirect application in other related technical fields is included in the patent protection scope of the present application.

Claims

1. A pneumatic self-cleaning method based on blade pitch control, characterized in that, The aerodynamic self-cleaning method based on blade pitch control includes: When the self-cleaning mode is triggered, the aerodynamic disturbance requirement is determined based on the monitored degree of dirt on the blade surface. Based on the current wind conditions and the aerodynamic disturbance requirements, determine the combination of the target frequency and target amplitude of the blade pitching action; Based on the combination of the target frequency and target amplitude, and combined with the blade mode shape, a blade pitching command sequence is generated; The blades of the wind turbine generator are controlled to perform blade pitching actions based on the blade pitching command sequence.

2. The method as described in claim 1, characterized in that, The step of determining the aerodynamic disturbance requirement based on the monitored degree of dirt on the blade surface when the self-cleaning mode is triggered includes: When the self-cleaning mode is triggered, the degree of dirt on the blade surface is obtained by fusing data returned by multiple sensors set on the leading edge and pressure surface of the blade. Based on the degree of dirt on the blade surface, dirt distribution maps of different spanwise positions were obtained; Determine the flow field stall critical point and the expected boundary layer thickening in the aforementioned fouling distribution map; The minimum local aerodynamic shear force for stripping contaminants is determined based on the flow field stall critical point and the expected boundary layer thickening, and this minimum local aerodynamic shear force is used as the aerodynamic disturbance requirement.

3. The method as described in claim 2, characterized in that, The steps for determining the flow field stall critical point and the desired boundary layer thickening in the fouling distribution map include: The key contamination areas were identified based on the contamination distribution map, and the positions of the key contamination areas in the spanwise and tangential directions of the blades were determined. Based on the airfoil of the blade and the position of the critical contamination area in the spanwise and chordwise directions of the blade, determine the aerodynamic performance degradation caused by the critical contamination area to the blade; The flow field stall critical point and the expected boundary layer thickening are determined based on the aerodynamic performance degradation.

4. The method as described in claim 1, characterized in that, The step of determining the combination of the target frequency and target amplitude of the blade pitching action based on the current wind condition parameters and the aerodynamic disturbance requirements includes: The real-time wind speed and turbulence intensity in the wind direction are determined based on the current wind condition parameters, and the wind conditions are classified into steady conditions, gradual change conditions and turbulence conditions based on the real-time wind speed and the turbulence intensity in the wind direction. Under stable operating conditions, gradient descent optimization is performed on a preset frequency and preset amplitude based on the aerodynamic disturbance requirements to obtain a combination of target frequency and target amplitude for blade pitching action. In gradual or turbulent conditions, the combination of target frequency and target amplitude of blade pitching action is determined based on historical wind data, short-term predicted wind sequences, and the aerodynamic disturbance requirements.

5. The method as described in claim 4, characterized in that, The step of determining the combination of target frequency and target amplitude of blade pitching action based on historical wind data, short-term predicted wind sequences, and the aerodynamic disturbance requirements under gradually changing or turbulent conditions includes: In the case of gradual or turbulent conditions, a multi-scenario optimization model is constructed based on historical wind data, short-term predicted wind sequences, and the aerodynamic disturbance requirements. The real-time wind speed and wind direction turbulence intensity are input into the multi-scenario optimization model, and the target scenario is output. Determine the feasible search space of the target scenario, take the target frequency and target amplitude of the blade pitching action as variables to be optimized, and randomly initialize a set of particles in the feasible search space, each particle representing a combination of frequency and amplitude. In each iteration, the cleanliness index that each particle can achieve in the target scene is calculated, and the minimum value of the cleanliness index is used as the fitness value. Based on the fitness value, the historical best position, and the population's historical best position, update the velocity and position of each particle to determine the change in the population's optimal fitness. When the number of iterations reaches the maximum value or the change value of the optimal fitness of the population is within a preset range, the iteration is terminated, and the combination of frequency and amplitude corresponding to the historical best particle of the population is determined as the target frequency and target amplitude of the blade pitching action.

6. The method as described in claim 1, characterized in that, The step of generating the blade pitch command sequence based on the combination of the target frequency and target amplitude, combined with the blade mode shape, includes: Obtain the finite element model of the blade, and extract the flapping mode frequency, oscillation mode frequency, and blade mode shape based on the finite element model; When the target frequency is the same as the waving mode frequency or the swing mode frequency, the target frequency is adjusted based on the waving mode frequency and the swing mode frequency to obtain the adjusted target frequency; The adjusted target frequency, the target amplitude, and the blade mode shape are combined into a blade pitching command sequence.

7. The method as described in claim 6, characterized in that, The step of adjusting the target frequency based on the waving mode frequency and the oscillating mode frequency to obtain the adjusted target frequency when the target frequency is the same as the waving mode frequency or the oscillating mode frequency includes: When the target frequency is the same as the waving mode frequency or the swing mode frequency, the target frequency is multiplied by an integer within the frequency threshold range to obtain the harmonic frequency; The harmonic frequency is compared with the waving mode frequency or the swing mode frequency to determine whether there is a frequency conflict; When the frequency conflict occurs, the target frequency is adjusted to the frequency point with the maximum safety margin in the frequency threshold range to obtain the adjusted target frequency; When there is no frequency conflict, the target frequency is adjusted to the harmonic frequency to obtain the target frequency.

8. The method as described in claim 6, characterized in that, The step of synthesizing the adjusted target frequency, the target amplitude, and the blade mode shape into a blade pitch command sequence includes: For any single blade, the adjusted target frequency, the target amplitude, and the blade mode shape are synthesized into a single blade pitch command sequence; By associating all the single-blade pitch command sequences, the resultant torque fluctuation formed at the hub by the aerodynamic excitation force of each blade is determined. The blade pitch command sequence is obtained by adjusting the single blade pitch command sequence based on the resultant torque fluctuation.

9. The method as described in claim 6, characterized in that, The step of synthesizing the adjusted target frequency, the target amplitude, and the mode shape into a blade pitch command sequence includes: The blades are subjected to icing detection, and the ice desorption strength threshold is determined when the blades are in icing condition. The auxiliary vibration command is determined based on the ice desorption strength threshold, and the auxiliary vibration command, the adjusted target frequency, the target amplitude, and the mode shape are combined into a blade pitching command sequence.

10. A pneumatic self-cleaning device based on blade pitch control, characterized in that, The pneumatic self-cleaning device based on blade pitch control includes: The demand determination module is used to determine the aerodynamic disturbance demand based on the monitored degree of dirt on the blade surface when the self-cleaning mode is triggered. The parameter optimization module is used to determine the combination of the target frequency and target amplitude of the blade pitching action based on the current wind condition parameters and the aerodynamic disturbance requirements. The instruction generation module is used to generate a blade pitching instruction sequence based on the combination of the target frequency and the target amplitude, combined with the blade mode shape; The pitch control module is used to control the blades of the wind turbine generator to perform blade pitching actions based on the blade pitching command sequence.

Images