A variable speed and variable pitch cooperative optimization method for noise constraint of wind turbine under yaw condition

Abstract

The present application relates to the technical field of wind power generation, specifically to a variable-speed variable-pitch collaborative optimization method for wind turbine noise constraint under yaw condition, comprising the following steps: 1. Establishing a wind turbine aerodynamic noise calculation model under yaw condition, segmenting the blade along the span, calculating the aerodynamic noise at different positions respectively, and synthesizing the whole machine noise; 2. Selecting the wind turbine speed ω and the pitch angle β as the collaborative control variables; 3. Constructing a multi-objective function, adjusting the priority of each target through the weight coefficient; 4. Setting the constraint conditions; 5. For various working conditions of different yaw angles γ, intelligent optimization algorithm is used for iterative optimization to obtain the optimal operating parameter combination; 6. The optimal operating parameter combination is transmitted to the wind turbine control system to adjust the unit operating state in real time. The present scheme can continuously and stably realize wind turbine noise reduction under various yaw conditions, and balance the noise reduction effect, power generation income and unit operation safety. The scheme has low implementation cost, small difficulty and strong applicability.

Description

Technical Field

[0001] This invention relates to the technical field of wind power generation, specifically to a variable speed and variable pitch collaborative optimization method for wind turbine noise constraints under yaw conditions. Background Technology

[0002] Wind turbines are the core equipment for wind energy development and utilization. They rely on blades to capture natural wind energy and convert it into mechanical energy, which is then transmitted through a transmission system and a generator to output electrical energy. They are an important component of large-scale clean energy power generation systems.

[0003] During operation, the interaction between the rotating blades and the air generates aerodynamic noise, primarily consisting of high-speed airflow at the blade tips and airflow disturbances between the blades and the tower. This noise is characterized by low-frequency vibration, long propagation distance, and slow attenuation. It not only affects the daily rest and living environment of residents near the wind turbines, but long-term exposure to high noise levels can also damage hearing. Furthermore, the noise from the wind farm disturbs birds and wildlife, negatively impacting their normal habitat, foraging, and migration activities, thus affecting the ecological environment. Therefore, wind turbine noise reduction is a critical technical issue that must be addressed in the design and operation and maintenance phases of wind power projects.

[0004] The aerodynamic noise of wind turbines mainly originates from the interaction between the rotating blades and the airflow. To reduce this noise, current research focuses primarily on modifying and optimizing the blade structure itself. For example, the existing technology "A Noise Reduction Improvement Method for Wind Turbine Blades" (Publication No.: CN116816587A) optimizes the blade tip structure based on the fact that the main area affecting wake flow and sound radiation is the blade tip. Another example is the existing technology "A Low-Noise Blade Design Method for Wind Turbines" (Publication No.: CN119416703), which installs a serrated structure resembling an owl feather on the leeward side of the blade, causing the air vibrations on both sides of the blade to differ by half a cycle, thus canceling out the amplitudes and achieving noise reduction. Yet another example is the existing technology "Wind Turbine and Its High-Performance Noise-Reducing Blades" (Publication No.: CN117823328A), which optimizes parameters such as the blade's spanwise length, the position of the flaps on the blade, and the chordal length of the flaps. The aforementioned existing technologies all start with blade structure modification and have shown positive noise reduction effects under theoretical analysis and specific test conditions. However, when these solutions are applied in practical engineering, the following technical problems still exist: 1. Weak noise reduction capability under yaw conditions. In real-world applications, wind turbines are often in yaw condition due to random wind direction changes. Yaw refers to the deviation of the rotor's orientation from the wind direction. Only when the blades face the oncoming wind can wind energy be maximized. Therefore, under yaw conditions, the wind turbine's yaw control system adjusts the rotor's direction to keep it always facing the wind. However, existing improvements to the blades and related structures are based on simulation optimization under steady-state conditions with the nacelle facing the wind. Under yaw conditions, the nacelle rotates horizontally to adjust the rotor's direction, causing the angle at which the wind blows onto the blades to constantly change. This renders the existing blade optimization structures unsuitable, resulting in weak noise reduction capabilities.

[0005] 2. High improvement costs and significant modification difficulties. Existing technologies are passive modifications to the blade structure hardware, requiring the physical processing of the blades, the addition of components, or the re-molding of new components. Therefore, they are only suitable for simultaneous improvements during the new turbine manufacturing stage. For existing wind turbines already connected to the grid, modifications require shutting down the entire turbine, hoisting and disassembling the blades, and performing hoisting, disassembly, and secondary processing at field sites. This process is not only cumbersome but also involves high costs for labor, equipment, and materials, making large-scale, batch modifications extremely difficult and hindering practical application.

[0006] Therefore, there is an urgent need for a method to reduce noise in wind turbines even under yaw conditions to solve the above problems. Summary of the Invention

[0007] This invention provides a variable speed and pitch collaborative optimization method for noise constraints of wind turbines under yaw conditions, which can solve the problems that existing noise reduction schemes have weak noise reduction capabilities when the wind turbine is under yaw conditions, and are costly and difficult to improve.

[0008] This application provides the following technical solution: a variable speed and variable pitch collaborative optimization method for noise constraint of wind turbine under yaw conditions; comprising the following steps: Step 1: Establish a wind turbine aerodynamic noise calculation model under yaw conditions. Divide the blades into segments along the spanwise direction, calculate the aerodynamic noise at different locations, and synthesize the overall noise. Step 2: Select the wind turbine speed With pitch angle As a collaborative control variable, it serves as the object of optimization and adjustment; Step 3: Construct a multi-objective function and adjust the priority of each objective by weighting coefficients; the objectives include the overall noise level, power generation, and the degree of fluctuation of control parameters (i.e., the stability of control parameters). Step 4: Set constraints to ensure that the optimization results meet environmental protection requirements, power generation requirements, and equipment safety requirements; Step 5: For various operating conditions with different yaw angles γ, an intelligent optimization algorithm is used to iteratively find the optimal combination of operating parameters for each yaw condition. Step six: Transmit the optimal combination of operating parameters to the wind turbine control system to adjust the unit's operating status in real time.

[0009] Beneficial effects: 1. Automatically adjust and match the optimal operating parameters under yaw conditions to maximize noise reduction capability.

[0010] Under yaw conditions, the rotor is frequently oriented by the control system to capture more wind energy. Therefore, existing technologies that simply modify the blade structure cannot adapt to the variable yaw environment, resulting in limited noise reduction capabilities. This method first establishes a yaw-specific noise calculation model, superimposing the airfoil self-noise model (BPM model) and the turbulent inflow noise model to synthesize the overall noise. This allows for a more accurate characterization of the overall aerodynamic noise under yaw conditions, providing reliable noise quantification support for subsequent multi-objective collaborative optimization. Then, using the rotor speed ω and pitch angle β selected in step two as core control variables, and constructing the multi-objective function in step three, the algorithm in step five iteratively optimizes the objective function under different yaw angle conditions, continuously screening the optimal combination of speed ω and pitch angle β that balances noise reduction, power generation, and operational stability under each yaw condition. Finally, real-time closed-loop regulation is achieved, ensuring that the wind turbine can automatically match and adjust the optimal operating parameters under yaw conditions, thereby maximizing the noise reduction capability of the wind turbine under yaw conditions.

[0011] 2. The software control scheme can adapt and iterate according to the working conditions, and there is no performance degradation problem in noise reduction capability.

[0012] In existing technologies, passive hardware modifications such as blade serration and tip trimming will experience wear and deformation over long-term operation, resulting in a gradual decline in noise reduction effectiveness. This solution relies on control algorithms to schedule operating parameters in real time, unaffected by blade aging or component wear. Facing complex field conditions such as frequent wind direction changes and continuous reciprocating yaw of the unit, the noise reduction strategy can continuously adapt to dynamic operating conditions and maintain stable noise control capabilities over the long term.

[0013] 3. To achieve a balance between noise reduction capability, power generation capacity, and unit control stability, and to realize the unity of environmental benefits, economic benefits and equipment operation safety.

[0014] If noise reduction is prioritized at the expense of lowering the rotational speed ω, power generation will decrease, impacting the economic benefits of the wind farm. Conversely, prioritizing power generation alone will easily lead to excessive noise levels during yaw conditions, failing to meet environmental regulations. Furthermore, balancing noise and power generation while neglecting control stability will result in frequent and significant adjustments to rotational speed and pitch angle, accelerating fatigue wear on blades, gears, and bearings, increasing the risk of turbine failure, drastically shortening equipment lifespan, and affecting equipment safety. Therefore, this solution flexibly allocates the priority of multiple objectives through weighting coefficients in step three, achieving a reasonable balance among these objectives. In step four, multiple constraints are set to ensure that the optimized operating parameters simultaneously meet environmental noise reduction standards, minimum power generation revenue requirements, and safe operating limits for the turbine. This ensures that the wind turbine meets noise standards during yaw conditions, minimizes power generation losses, reduces mechanical impact, and guarantees operational safety, achieving a balance between environmental benefits, economic benefits, and equipment safety.

[0015] 4. This solution is low in cost, easy to implement, highly applicable, and easy to implement quickly.

[0016] This solution is a software system-level optimization, a software control optimization strategy that eliminates the need for hardware modifications to the blades and nacelles, saving significant costs associated with parts processing, installation, and equipment procurement. The solution relies on existing sensors on the wind turbine to collect operating data, and the algorithm can be directly integrated into the existing main control system. Deployment can be completed through online debugging and remote upgrades, resulting in a low on-site implementation threshold and minimal debugging workload. Furthermore, this method is not limited by wind turbine type and is compatible with both newly built units and existing operational wind farms. It enables rapid, batch retrofitting of existing wind turbines, unaffected by turbine location distribution or site installation conditions. The optimization weights can be flexibly adjusted according to wind farm noise control requirements, adapting to various terrains and different yaw conditions, making it highly versatile for engineering projects.

[0017] Furthermore, in step three, the objective function is:

[0018] In the formula, Where N is the pitch angle, and N1, N2, and N3 are weighting coefficients; This represents the reduction in the equivalent total sound pressure level. This refers to the deviation in the aerodynamic power output of the wind turbine; This represents the variance of the rotational speed and pitch angle fluctuations.

[0019] Beneficial effects: Since the rotational speed ω and pitch angle β are the link between wind turbine noise radiation and performance, changes in these two parameters directly affect three key indicators: noise level, output power, and turbine operating parameters. This solution uses weighting coefficients to determine the optimization priority of these three indicators. By adjusting the weighting values, the optimization strategy can be adaptively tilted, allowing for flexible allocation based on different wind farm application scenarios. For example, for noise-sensitive wind farms close to residential areas, a larger N1 value is configured, prioritizing noise reduction and ensuring compliance with noise standards. For remote suburbs or sites with lax noise control, N1 is appropriately reduced and N2 is increased to prioritize controlling power generation loss and ensuring operational profitability. For turbine positions with frequent yaw, N3 can be appropriately increased to constrain drastic fluctuations in control parameters, protect turbine components, and improve wind turbine operational safety. Compared to single-objective optimization strategies, this solution can adjust the weighting ratio according to application scenarios to match differentiated control needs, achieving a scenario-based balance between noise reduction capability, power generation profitability, and turbine operational stability, thus enhancing the versatility and engineering implementation value of the entire control strategy.

[0020] Furthermore, in step three, the reduction in the equivalent total sound pressure level... In the formula, To optimize the equivalent noise value under the conditions of rotational speed ω and pitch angle β, This represents the noise equivalent value in the unoptimized state.

[0021] Beneficial effects: Reduction in equivalent total sound pressure level Characterizing the noise reduction level, it can accurately quantify the change in the equivalent noise level of the entire machine before and after the adjustment of this solution; when the reduction in equivalent total sound pressure level... The larger the value, the greater the noise reduction achieved by the optimization strategy, and the better the noise reduction performance. Introducing this index into the objective function can provide a quantitative and objective basis for noise reduction evaluation in the optimization process, allowing the algorithm to intuitively identify the noise reduction potential corresponding to different combinations of speed and pitch angle, ensuring that the optimization results have a clear noise reduction gain, and providing reliable data support for achieving controllable and quantifiable active noise reduction under yaw conditions.

[0022] Furthermore, in step three, the deviation in the aerodynamic power output of the wind turbine... In the formula, The power value for optimized aerodynamic performance; This represents the power value of the aerodynamic performance in the unoptimized state.

[0023] Beneficial effects: Deviation in aerodynamic power output of wind turbine Characterizing power generation, it can quantify the change in wind turbine aerodynamic output power before and after noise reduction control; when the wind turbine aerodynamic power output deviation... The larger the value, the smaller the power generation loss in order to achieve the noise reduction target. Introducing this indicator into the objective function can quantitatively evaluate the economic cost of power generation incurred by the noise reduction strategy, constrain the optimization algorithm, and prevent the algorithm from blindly pursuing a larger noise reduction range and indiscriminately reducing the unit's output power. This will help the optimization scheme achieve a balance between noise suppression effect and wind farm power generation revenue, ensuring that the scheme has good engineering economics.

[0024] Furthermore, in step three, the variance of the rotational speed and pitch angle fluctuations... In the formula, Let ω be the variance of the rotational speed. Let β be the variance of the pitch angle fluctuation.

[0025] Beneficial effects: Variance of rotational speed and pitch angle Characterizing the stability of control parameters, this indicator quantifies the degree of oscillation in the two core control parameters: speed and pitch angle. A smaller value indicates smoother adjustments in speed and pitch angle, resulting in higher stability and safety of the wind turbine operation. Introducing this indicator into the objective function guides the optimization algorithm to prioritize parameter combinations with smaller adjustment amplitudes and more continuous changes. This avoids frequent jumps in speed and pitch angle to pursue noise reduction or power generation targets, reducing alternating impact loads on the pitch mechanism and transmission system, suppressing control oscillations, ensuring long-term stable operation of the wind turbine, extending the service life of turbine components, and improving the engineering feasibility of the entire control strategy.

[0026] Furthermore, in step four, the constraint conditions include power values ​​( Constraints, noise values Constraints include upper and lower limits for rotational speed (ω) and pitch angle (β);

[0027] In the formula, η is the lower limit coefficient of power. The maximum noise value, This is the maximum speed.

[0028] Beneficial effects: Power constraints rely on a lower power limit coefficient to ensure the minimum aerodynamic output power and prevent excessive power generation loss; noise constraints set a maximum noise limit to ensure that the control scheme always meets environmental noise reduction standards; speed and pitch angle range constraints match the operating limits of the unit hardware and eliminate infeasible parameters. Multiple constraints define a reasonable feasible solution space for algorithm optimization, ensuring that the obtained control parameters take into account environmental compliance, power generation benefits, and equipment operation safety, effectively improving the safety and engineering feasibility of the optimization scheme.

[0029] Furthermore, in step five, the yaw angle γ is divided into multiple groups of gradient yaw conditions.

[0030] Beneficial effects: The selection of yaw conditions is based on the natural wind direction deviation characteristics generated during actual wind turbine operation. The conditions include a baseline operating state without yaw and multiple levels of different yaw angles, closely reflecting the real-world operating environment of active wind-facing operations and yaw due to varying degrees of wind direction misalignment. By independently optimizing the speed and pitch angle matching scheme for each operating condition, the differential changes in flow field, noise increase, and power generation attenuation caused by different yaw degrees can be accurately captured. Based on the above calculation results, specific control parameters for each operating condition are configured. When different yaw disturbances occur during wind turbine operation, the optimized control scheme matching the operating condition can be directly invoked. Under various slight to large yaw conditions, noise suppression levels, power generation economic benefits, and unit operating stability are simultaneously considered, effectively enhancing the adaptability and practical value of this control strategy in complex and variable yaw operating environments.

[0031] A variable speed and pitch coordinated control system for wind turbine noise constraints under yaw conditions includes a perception layer, a decision layer, and an execution layer. The perception layer includes sensor units for acquiring external environmental data such as wind direction and speed, as well as the wind turbine's rotational speed, pitch angle, and noise operation signals, and transmitting these signals to the decision layer. The decision layer includes a noise model module, a multi-objective optimization module, an algorithm module, and a constraint verification module. The noise model module calculates the equivalent noise index of the wind turbine based on the received acquired signals. The multi-objective optimization module constructs a multi-objective function including noise reduction, power generation, and control stability objectives, and loads preset parameters. The weighting coefficients; the algorithm module is used to optimize and solve multi-objective functions based on particle swarm optimization, and output candidate speed and pitch angle parameters; the constraint verification module is used to verify the candidate speed and pitch angle parameters by power constraints, noise constraints, upper and lower limits of speed constraints, and upper and lower limits of pitch angle constraints, and to filter out the target speed and target pitch angle that meet the constraints; the execution layer includes a speed adjustment driver and a pitch angle adjustment driver; the speed adjustment driver receives the target speed and adjusts the operating speed of the wind turbine body; the pitch angle adjustment driver receives the target pitch angle and adjusts the pitch angle of the wind turbine body to realize noise reduction control of the wind turbine under yaw conditions.

[0032] Beneficial effects: This system can adapt to various yaw conditions to carry out noise constraint control. It can reduce power generation loss while meeting environmental noise limits, suppress large fluctuations in speed and pitch angle, reduce alternating load on the unit, and take into account environmental protection requirements, power generation economic benefits and equipment operation safety. It has strong engineering practicality. Attached Figure Description

[0033] Figure 1 The overall optimized control flowchart of the present invention is shown below.

[0034] Figure 2 This is a schematic diagram of the closed-loop control structure of the wind turbine under yaw conditions according to the present invention.

[0035] Figure 3 This is a block diagram of the multi-objective optimization algorithm of the present invention. Detailed Implementation

[0036] The following detailed description illustrates the specific implementation method: Example 1 like Figure 1 As shown, a variable speed and variable pitch collaborative optimization method for wind turbine noise constraints under yaw conditions includes the following steps: Step 1: Establish a wind turbine aerodynamic noise calculation model under yaw conditions. Divide the blades into segments along the spanwise direction, calculate the aerodynamic noise at different locations, and synthesize the overall noise of the turbine.

[0037] The noise generated by wind turbines can be mainly divided into two categories: mechanical noise and aerodynamic noise. For megawatt-class large wind turbines, aerodynamic noise is dominant, originating from three main noise sources: turbulent inflow noise, trailing edge noise, and stall noise. Each noise source has a different formation mechanism, is located at different positions along the blade span, and does not interfere with each other. Therefore, the total noise of the wind turbine during operation can be calculated by superimposing different types of aerodynamic noise. This embodiment uses the airfoil self-noise model proposed by Brooks, Pope, and Marcolini (BPM) and the turbulent inflow noise model proposed by Amiet and modified by Lowson for noise prediction. The sound pressure level is calculated at 1 / 3 octave band frequency.

[0038] Turbulent inflow noise: At low frequencies, turbulent inflow noise is caused by the interaction between turbulent inflow generated by the atmospheric boundary layer and the leading edge of the wind turbine blade, resulting in pressure fluctuations on the blade surface.

[0039] Low-frequency sound pressure level expression: ; High-frequency sound pressure level expression: ; In the formula: l The length scale is for turbulent flow, in meters (m). I Turbulence intensity; Air density, unit: kg / m³; Speed ​​of sound, unit: m / s; L Spread length of leaf element, unit: m; K For normalized wavenumber; LFC This is a low-frequency correction factor; D L It is a directional function.

[0040] Airfoil noise: Turbulent boundary layer trailing edge noise is formed by the interaction between the turbulent boundary layer and the blade trailing edge. Laminar flow over the airfoil surface undergoes a transition, transforming into turbulence, and the resulting pressure field fluctuations affect the suction and pressure surfaces of the airfoil trailing edge.

[0041] Airfoil noise expression:

[0042]

[0043]

[0044] In the formula: SPL P , SPL S These are the noise from the pressure surface and suction surface of the trailing edge of the turbulent boundary layer, respectively. and These are the boundary layer thicknesses of the pressure and suction surfaces, respectively, and the angle of attack. It is related to the Reynolds number Re; M It is the Mach number; r Distance from the monitoring point, unit: m; St For empirical functions based on the Strouhal number ( St = f / U ); A It is the spectrum shape function; W 1 represents the amplitude function; Δ W 1 represents the sound pressure level correction function; At larger angles of attack, airfoil boundary layer separation occurs, increasing the turbulent vortices on the suction surface. When these turbulent vortices become wakes, separation flow noise is generated. It is a broadband noise.

[0045] In the formula: B For the spectrum shape function, K 2 represents the amplitude function.

[0046] Aerodynamic noise is superimposed and combined into the overall machine noise: The wind turbine blade is non-uniformly divided into multiple blade elements from the blade root region, and the above noise model is applied to each blade element according to the inflow angle of attack and relative velocity. Each noise source is independent of the others. By logarithmic superposition, the sound pressure level of each leaf element at the noise monitoring point is calculated:

[0047] By logarithmically superimposing different types of noise sources on all blade elements, the total sound pressure level of the wind turbine can be calculated. The total sound pressure level of the wind turbine can then be used to represent the magnitude of the sound pressure, thus facilitating the characterization and evaluation of the wind turbine noise level under different operating conditions. The formula for the total sound pressure level of the wind turbine is as follows:

[0048] In the formula: i The number of leaf elements; The total noise source generated by each leaf element; j These are the three noise sources mentioned above; L p,j The size of each noise source on each leaf element.

[0049] Step 2: Select the wind turbine speed ω and pitch angle β as the coordinated control variables and as the optimization adjustment objects; since the speed ω and pitch angle β are the link between the wind turbine noise radiation and performance, the changes in speed ω and pitch angle β directly affect the changes in three indicators: noise level, output power and wind turbine operating parameters.

[0050] Step 3: Construct a multi-objective function The priority of each objective is adjusted by weighting coefficients; the objectives include overall noise level, power generation, and control parameter fluctuation (i.e., control parameter stability); the logic block diagram of the multi-objective function optimization algorithm is as follows: Figure 3 As shown.

[0051] Objective function:

[0052] In the formula, Where N is the pitch angle, and N1, N2, and N3 are weighting coefficients; This represents the reduction in the equivalent total sound pressure level. This refers to the deviation in the aerodynamic power output of the wind turbine; Let V be the variance of the rotational speed and pitch angle. This objective function can be substituted into the overall noise model in step one to obtain the noise level of the wind turbine.

[0053] Reduction in equivalent total sound pressure level:

[0054] In the formula, To optimize the equivalent noise value under the conditions of rotational speed ω and pitch angle β, This represents the noise equivalent value in the unoptimized state.

[0055] Deviation in aerodynamic power output of wind turbine:

[0056] In the formula, The power value for optimized aerodynamic performance; This represents the power value of the aerodynamic performance in the unoptimized state.

[0057] Variance of rotational speed and pitch angle fluctuations:

[0058] In the formula, Let ω be the variance of the rotational speed. Let β be the variance of the pitch angle fluctuation.

[0059] The three weighting coefficients, N1, N2, and N3, determine the optimization priority of the three indicators. By adjusting the weight values, the optimization strategy can be adaptively tilted. The weighting coefficients can be selected according to the following steps: 1. Initial assignment Two scenario benchmark tables are used to assign initial weights to all units in the field, so as to quickly build an optimization framework, as shown in the table below:

[0060] It should be noted that: When adjusting the weight ratio, it is necessary to consider the control capability of the unit itself. Older units have poor control precision, so the fluctuation weight should be increased to ensure that the strategy can be implemented. Therefore, N3 should not be lower than 0.1. This is to prevent the optimization results from having frequent and large jumps in speed and pitch angle due to an excessively low weight of N3, which would increase the fatigue load on the blades and transmission chain, shorten the unit's life, and even trigger the unit's protection shutdown risk.

[0061] 2. Individual calibration, graded by unit. When configuring weights, for wind farms with multiple wind turbines, noise simulation can be performed using an aerodynamic noise model. This involves inputting parameters such as impeller speed, pitch angle, and wind speed into the model and combining them with on-site measurements to calculate the noise contribution of each wind turbine to sensitive points. This allows for the classification of core, secondary, and unaffected units, enabling more precise adjustment of the N1 and N2 ratios. Simultaneously, the fluctuation weight of N3 can be adjusted for each unit based on its operation and maintenance records, health status, and equipment age.

[0062] 3. Dynamic strategy writing Multiple fixed weighting coefficient combinations are pre-tuned for different wind speed ranges, day and night times, and wind direction conditions. The operating condition judgment logic and each set of weighting parameters are written into the wind turbine SCADA control system. During wind turbine operation, SCADA collects on-site wind speed, time, and wind direction data in real time, and automatically matches the pre-tuned weighting coefficients corresponding to the current operating condition to perform optimization calculations.

[0063] 4. Iterative optimization After the weight ratio is determined, the trial operation is carried out for 1 to 2 weeks. The weights are then fine-tuned based on the measured noise data, power generation, and unit load data. If the noise exceeds the standard, N1 is increased appropriately; if the power loss exceeds the expectation, N2 is increased appropriately; if the pitch change is frequent, N3 is increased appropriately. Finally, the weight coefficient ratio is converged to the optimal value.

[0064] Step four: Set constraints to ensure that the optimization results meet environmental protection requirements, power generation requirements, and equipment safety requirements; constraints include power values ​​( Constraints, noise values Constraints include upper and lower limits for rotational speed (ω) and pitch angle (β);

[0065] In the formula, η is the power lower limit coefficient, which ensures that the power generation performance of the wind turbine is not lower than the minimum requirement and guarantees economic benefits. In this embodiment, η is 0.8, which means that the optimized power output should not be lower than 80% of the original output. To determine the maximum noise level, noise constraints should meet local environmental protection requirements. In this embodiment, the noise threshold is set to 50 dBA. The maximum speed is determined based on the rated operating speed of the wind turbine, ensuring consistency with the unit's mechanical structural limit parameters and avoiding equipment safety risks associated with overspeed operation.

[0066] Step 5: For various operating conditions with different yaw angles γ, an intelligent optimization algorithm is used to iteratively optimize the model to obtain the optimal combination of operating parameters for each yaw condition. The yaw angle γ is divided into multiple groups of gradient yaw conditions. The selection of yaw conditions refers to the wind direction deviation characteristics naturally generated during the actual operation of the wind turbine. The conditions include the baseline operating state without yaw and multiple levels of different yaw angles, which closely match the real operating environment of the unit actively facing the wind and yawing with different degrees of wind direction misalignment during operation. By independently optimizing the optimal speed and pitch angle matching scheme for each operating condition, the differentiated changes in flow field, noise increase, and power generation attenuation caused by different degrees of yaw can be accurately captured. Based on the above calculation results, specific control parameters are configured for different operating conditions. When the wind turbine generates different yaw disturbances during on-site operation, the optimized control scheme matching the operating condition can be directly invoked. Under various slight to large yaw conditions, noise suppression level, power generation economic benefits and unit operation stability are taken into account simultaneously. This effectively enhances the adaptability and practical value of the control strategy in complex and ever-changing yaw working environments.

[0067] In this embodiment, the yaw angle γ is set to the following four operating conditions: The four yaw conditions are: no yaw γ=0°, counterclockwise yaw γ=5°, counterclockwise yaw γ=10°, and counterclockwise yaw γ=20°. These four yaw conditions are selected based on wind direction deviations that wind turbines may encounter in typical operating scenarios. This setting covers the baseline condition without yaw and yaw conditions of small, medium, and large degrees, closely matching the real-world operating scenario of wind turbines continuously facing the wind and producing different degrees of yaw. It can independently solve for the optimal combination of rotational speed and pitch angle for each yaw angle, fully capturing the differentiated characteristics of flow field distortion, noise increase, and power attenuation under different yaw degrees. This yields control parameters matched to different operating conditions. When a corresponding yaw occurs during wind turbine operation, the optimized strategy adapted to that condition can be directly invoked, ensuring a balanced achievement of noise reduction, power generation benefits, and unit operational stability regardless of whether the yaw is slight or large. This significantly improves the applicability of the entire control strategy in variable yaw environments.

[0068] The advantage of using particle swarm optimization in intelligent algorithms is that it does not require differentiation of the objective function, is suitable for the nonlinear and multi-constraint optimization model of this scheme, and has good global search capabilities, enabling it to stably solve for the optimal combination of speed and pitch angle under different yaw conditions.

[0069] Step six: Transmit the optimal combination of operating parameters to the wind turbine control system to adjust the unit's operating status in real time.

[0070] Example 2 like Figure 2 As shown in the figure, this embodiment also discloses a variable speed and pitch coordinated control system for wind turbine noise constraint under yaw conditions, including a perception layer, a decision layer and an execution layer.

[0071] The perception layer includes sensor units, which are used to collect external environmental signals such as wind direction, wind speed, wind turbine speed, pitch angle, and noise operation signals, and transmit the collected signals to the decision layer.

[0072] The decision-making layer includes a noise model module, a multi-objective optimization module, an algorithm module, and a constraint verification module. The noise model module is used to calculate the equivalent noise index of the wind turbine based on the received acquisition signals. The multi-objective optimization module is used to construct a multi-objective function that includes noise reduction, power generation, and control stability objectives, and loads preset weight coefficients. The algorithm module is used to optimize the multi-objective function based on the particle swarm optimization algorithm and output candidate speed and pitch angle parameters. The constraint verification module is used to verify the candidate speed and pitch angle parameters by power constraints, noise constraints, upper and lower limits of speed constraints, and upper and lower limits of pitch angle constraints, and select the target speed and target pitch angle that meet the constraints.

[0073] The execution layer includes a speed regulation driver and a pitch angle regulation driver; the speed regulation driver receives the target speed and regulates the operating speed of the wind turbine body; the pitch angle regulation driver receives the target pitch angle and regulates the pitch angle of the wind turbine body, thereby realizing noise reduction control of the wind turbine under yaw conditions.

[0074] The above are merely embodiments of the present invention, and the invention is not limited to the fields covered by this embodiment. Commonly known structures and characteristics in the solutions are not described in detail here. It should be noted that those skilled in the art can make various modifications and improvements without departing from the structure of the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A variable speed and variable pitch collaborative optimization method for wind turbine noise constraints under yaw conditions, characterized in that, Includes the following steps: Step 1: Establish a wind turbine aerodynamic noise calculation model under yaw conditions. Divide the blades into segments along the spanwise direction, calculate the aerodynamic noise at different locations, and synthesize the overall noise. Step 2: Select the wind turbine speed ω and the blade pitch angle β as the coordinated control variables and use them as the optimization adjustment objects; Step 3: Construct a multi-objective function and adjust the priority of each objective by weighting coefficients; the objectives include the overall noise level, power generation, and the degree of fluctuation of control parameters; Step 4: Set constraints to ensure that the optimization results meet environmental protection requirements, power generation requirements, and equipment safety requirements; Step 5: For various operating conditions with different yaw angles γ, an intelligent optimization algorithm is used to iteratively find the optimal combination of operating parameters for each yaw condition. Step six: Transmit the optimal combination of operating parameters to the wind turbine control system to adjust the unit's operating status in real time.

2. The variable speed and pitch collaborative optimization method for wind turbine noise constraint under yaw condition according to claim 1, characterized in that: In step three, the objective function is: In the formula, Where N is the pitch angle, and N1, N2, and N3 are weighting coefficients; This represents the reduction in the equivalent total sound pressure level. This refers to the deviation in the aerodynamic power output of the wind turbine; This represents the variance of the rotational speed and pitch angle fluctuations.

3. The variable speed and pitch collaborative optimization method for wind turbine noise constraint under yaw condition according to claim 2, characterized in that: In step three, the reduction in the equivalent total sound pressure level In the formula, To optimize the equivalent noise value under the conditions of rotational speed ω and pitch angle β, This represents the noise equivalent value in the unoptimized state.

4. The variable speed and pitch collaborative optimization method for wind turbine noise constraints under yaw conditions according to claim 3, characterized in that: In step three, the deviation of the wind turbine's aerodynamic power output In the formula, The power value for optimized aerodynamic performance; This represents the power value of the aerodynamic performance in the unoptimized state.

5. The variable speed and pitch collaborative optimization method for wind turbine noise constraint under yaw condition according to claim 4, characterized in that: In step three, the variance of the rotational speed and pitch angle fluctuations In the formula, Let ω be the variance of the rotational speed. Let β be the variance of the pitch angle fluctuation.

6. The variable speed and pitch collaborative optimization method for wind turbine noise constraints under yaw conditions according to claim 5, characterized in that: In step four, the constraints include power values ​​( Constraints, noise values Constraints include upper and lower limits for rotational speed (ω) and pitch angle (β); In the formula, η is the lower limit coefficient of power. The maximum noise value, This is the maximum speed.

7. The variable speed and pitch collaborative optimization method for wind turbine noise constraints under yaw conditions according to claim 6, characterized in that: In step five, the yaw angle γ is divided into multiple groups of gradient yaw conditions.

8. A variable speed and pitch coordinated control system for wind turbine noise constraint under yaw conditions, used to execute the variable speed and pitch coordinated optimization method for wind turbine noise constraint under yaw conditions as described in any one of claims 1-7, characterized in that: It includes the perception layer, decision-making layer, and execution layer; The perception layer includes sensor units; these units are used to collect external environmental signals such as wind direction and speed, as well as wind turbine speed, pitch angle, and noise operation signals, and transmit the collected signals to the decision layer. The decision-making layer includes a noise model module, a multi-objective optimization module, an algorithm module, and a constraint verification module. The noise model module is used to calculate the equivalent noise index of the wind turbine based on the received acquisition signals. The multi-objective optimization module is used to construct a multi-objective function that includes noise reduction, power generation, and control stability objectives, and loads preset weight coefficients. The algorithm module is used to optimize the multi-objective function based on the particle swarm optimization algorithm and output candidate rotational speed and pitch angle parameters. The constraint verification module is used to perform power constraints, noise constraints, upper and lower limits of speed constraints, and upper and lower limits of pitch angle constraints on candidate speed and pitch angle parameters, and to filter out target speed and target pitch angle that meet the constraint conditions. The execution layer includes a speed regulation driver and a pitch angle regulation driver; The speed regulation driver receives the target speed and adjusts the operating speed of the wind turbine body; the pitch angle regulation driver receives the target pitch angle and adjusts the pitch angle of the wind turbine body to achieve noise reduction control of the wind turbine under yaw conditions.

Images