Digital twin variable pitch control optimization method and system for wind turbine generator

By using digital twin technology and pitch control optimization methods, the blade angle is adjusted in real time to match the power demand, which solves the problems of low power generation efficiency and poor stability of wind turbine units, and achieves more efficient and stable wind energy conversion.

CN119021823APending Publication Date: 2024-11-26STATE GRID JIANGSU ELECTRIC POWER CO LTD NANTONG POWER SUPPLY BRANCH +3
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410970767.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-19
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

Wind turbines have low power generation efficiency and poor operational stability, which can easily lead to energy waste and equipment damage.

Method used

By establishing a digital twin model of the wind turbine using digital twin technology, and combining sensor monitoring data and meteorological platform forecast data, pitch control optimization is performed, adjusting the blade angle in real time to match power demand and optimizing the pitch strategy.

Benefits of technology

It improves the power generation efficiency and operational stability of wind turbine units and reduces maintenance costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119021823A_ABST
    Figure CN119021823A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of control optimization, and provides a digital twin variable pitch control optimization method and system for a wind turbine generator. The method comprises the following steps: carrying out digital twinborn modeling to obtain a twinborn model of the wind turbine generator; monitoring the target wind turbine generator in real time to obtain blade angle data; acquiring wind power prediction data in preset time through the meteorological platform; inputting the wind power speed prediction sequence, the wind power direction prediction sequence and the variable pitch angle data into the wind turbine generator twinborn model for operation simulation to obtain a simulated generating capacity sequence; when the difference value between the simulated generating capacity sequence and the electricity demand quantity sequence meets a preset supply and demand difference value, no control is carried out; and when the preset supply and demand difference value is not met, variable pitch optimization is conducted on the blade angle data. The technical problems that in the prior art, a wind turbine generator set is low in power generation efficiency and poor in operation stability are solved, and the technical effect of improving the operation efficiency of the wind turbine generator set is achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of control optimization technology, and in particular to a digital twin pitch control optimization method and system for a wind turbine. Background Art

[0002] Wind turbines, which convert wind energy into electrical energy, are an important area of ​​new energy application. After nearly a hundred years of development, modern wind turbine systems are constantly setting new records in tower height, impeller diameter, output power, and control complexity. Pitch control refers to the use of mechanical hydraulic or electric servo methods to control the rotation of the fan blades to adjust the pitch angle of the blades, change the angle of attack of the airflow on the blades, and then control the aerodynamic torque and aerodynamic power captured by the fan blades. Taking the rotating surface of the blade as the reference surface, when the angle between the blade and the surface and the pitch angle are 0°, the wind energy captured by the blade is the largest, and the power generated by the wind turbine at this time is also the largest. When the angle between the blade and the surface and the pitch angle are 90°, the resistance of the blade is the largest, causing the wind turbine to shut down, which is crucial to the safe operation of the entire wind turbine in some extreme weather. In the prior art, the variable pitch operation stability of wind turbines is poor, which is easy to cause energy waste and damage to wind turbines.

[0003] In summary, the prior art has the technical problems of low power generation efficiency and poor operation stability of wind turbines. Summary of the invention

[0004] Based on this, it is necessary to provide a digital twin pitch control optimization method and system for wind turbines that can improve the operating efficiency of wind turbines in response to the above technical problems.

[0005] In the first aspect, the present application provides a digital twin pitch control optimization method for a wind turbine, comprising: performing digital twin modeling through the wind power generation process of a target wind turbine to obtain a twin model of the wind turbine; performing real-time monitoring of the target wind turbine through a wind turbine sensor to obtain blade angle data; obtaining wind forecast data for a preset time through a meteorological platform, including a wind speed prediction sequence and a wind direction prediction sequence; inputting the wind speed prediction sequence, the wind direction prediction sequence, and the pitch angle data into the twin model of the wind turbine for operation simulation to obtain a simulated power generation sequence; obtaining the power demand sequence for the preset time, and when the difference between the simulated power generation sequence and the power demand sequence meets the preset supply and demand difference, no control is performed; when the preset supply and demand difference is not met, obtaining a set of pitch optimization time nodes, and performing pitch optimization on the blade angle data.

[0006] In the second aspect, the present application provides a digital twin variable pitch control optimization system for a wind turbine, including: a wind turbine twin model acquisition module, the wind turbine twin model acquisition module is used to perform digital twin modeling through the wind power generation process of the target wind turbine to obtain the wind turbine twin model; a blade angle data acquisition module, the blade angle data acquisition module is used to monitor the target wind turbine in real time through the wind turbine sensor to obtain blade angle data; a wind prediction data acquisition module, the wind prediction data acquisition module is used to obtain wind prediction data for a preset time through a meteorological platform, including a wind speed prediction sequence, a wind direction prediction sequence, and a wind speed prediction sequence. sequence; a simulated power generation sequence acquisition module, the simulated power generation sequence acquisition module is used to input the wind speed prediction sequence, the wind direction prediction sequence, and the pitch angle data into the twin model of the wind turbine for operation simulation to obtain a simulated power generation sequence; a sequence difference judgment module, the sequence difference judgment module is used to obtain the power demand sequence of the preset time, when the difference between the simulated power generation sequence and the power demand sequence meets the preset supply and demand difference, no control is performed; a pitch optimization module, the pitch optimization module is used to obtain a pitch optimization time node set when the preset supply and demand difference is not met, and to perform pitch optimization on the blade angle data.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] First, digital twin modeling is performed through the wind power generation process of the target wind turbine to obtain the twin model of the wind turbine; secondly, the target wind turbine is monitored in real time through the wind turbine sensor to obtain the blade angle data; then the wind forecast data for the preset time is obtained through the meteorological platform, including the wind speed forecast sequence and the wind direction forecast sequence; then the wind speed forecast sequence, the wind direction forecast sequence, and the pitch angle data are input into the twin model of the wind turbine for operation simulation to obtain the simulated power generation sequence; next, the power demand sequence for the preset time is obtained, and when the difference between the simulated power generation sequence and the power demand sequence meets the preset supply and demand difference, no control is performed; finally, when the preset supply and demand difference is not met, the pitch optimization time node set is obtained, and the blade angle data is optimized. This application solves the technical problems of low power generation efficiency and poor operation stability of wind turbines in the prior art, and achieves the technical effect of improving the operation efficiency of wind turbines.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 1 is a flow chart of a digital twin pitch control optimization method for a wind turbine generator system in one embodiment;

[0011] Figure 2 A schematic diagram of a flow chart of traversing a time node sequence of a digital twin pitch control optimization method for a wind turbine in one embodiment;

[0012] Figure 3 It is a structural block diagram of a digital twin pitch control optimization system for a wind turbine in one embodiment.

[0013] Explanation of the accompanying drawings: wind turbine twin model acquisition module 11, blade angle data acquisition module 12, wind force prediction data acquisition module 13, simulated power generation sequence acquisition module 14, sequence difference judgment module 15, variable pitch optimization module 16. DETAILED DESCRIPTION

[0014] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0015] like Figure 1 As shown, the present application provides a digital twin pitch control optimization method for a wind turbine, which is characterized by comprising:

[0016] Digital twin modeling is performed through the wind power generation process of the target wind turbine to obtain a twin model of the wind turbine;

[0017] A wind turbine is a system that converts the kinetic energy of wind into electrical energy. Digital twins make full use of data such as physical models, sensor updates, and operation history, integrate multi-disciplinary, multi-physical, multi-scale, and multi-probability simulation processes, and complete mapping in virtual space, thereby reflecting the entire life cycle of the corresponding physical equipment. Through digital twin technology, a digital twin model of a wind turbine is established, the pitch control process is simulated, and the pitch control strategy is optimized; this application provides a digital twin pitch control optimization method for a wind turbine, which can significantly improve the operating efficiency, reliability, and stability of a wind turbine and reduce maintenance costs.

[0018] The target wind turbine refers to the wind turbine that is confirmed to be studied, which is recorded as the target wind turbine. First, it is necessary to collect relevant data of the target wind turbine, including wind speed, wind direction, generator speed, generator current, generator voltage, etc., which can be obtained through real-time monitoring of the wind turbine sensor or through historical data records; based on the relevant data, a physical model of the wind turbine can be established, and the physical model can describe the interaction between the various components and systems of the wind turbine, including generators, gear devices, bearings, pitch systems, etc.; the physical model is mapped to a digital model, that is, a digital twin model. The digital twin model can simulate the operating state and behavior of the wind turbine, including the pitch control process, wind energy conversion process, etc., and the twin model of the active wind turbine. Through the digital twin modeling of the wind power generation process of the target wind turbine, the twin model of the wind turbine can be obtained. This model can be used for the optimization and control of wind turbines, improve wind energy conversion efficiency, and reduce operation and maintenance costs.

[0019] Collect equipment parameter information of the target wind turbine and build a 3D twin model;

[0020] Collect historical pitch angle and wind speed data to establish a pitch response model;

[0021] The three-dimensional twin model and the pitch response model are integrated to obtain the twin model of the wind turbine generator set.

[0022] The equipment parameter information of the target wind turbine is collected, including the parameter information of each component and system of the target wind turbine, such as the generator, gear device, bearing, pitch system, etc. The equipment parameter information can be obtained by consulting the equipment manual, equipment data sheet or on-site measurement; based on the collected equipment parameter information, a three-dimensional twin model of the target wind turbine can be constructed. The model can describe the geometric shape, size, position and other information of each component and system of the wind turbine. CAD software or other modeling tools can be used for modeling; historical pitch angle and historical wind data can be collected through the sensors or other data acquisition devices of the wind turbine; these data can be used to establish a pitch response model, and a pitch response model can be established based on the collected historical pitch angle and historical wind data. This model should be able to describe the pitch response behavior of the wind turbine at different wind speeds, including pitch speed, pitch angle, etc. Modeling can be performed using regression analysis, neural network and other methods; the three-dimensional twin model and the pitch response model are integrated to obtain the twin model of the target wind turbine. The model can describe the behavior of the wind turbine in real-time operation, including wind energy conversion efficiency, power generation, etc. By collecting the equipment parameter information of the target wind turbine, building a three-dimensional twin model, collecting historical pitch angle and historical wind data, building a pitch response model, and integrating the three-dimensional twin model and the pitch response model, a twin model of the wind turbine can be obtained. This model can be used for the optimization and control of the wind turbine.

[0023] The target wind turbine is monitored in real time by a wind turbine sensor to obtain blade angle data;

[0024] Wind turbine sensors can be installed at different locations throughout the wind turbine, including blades, generators, gear devices, etc. The wind turbine sensors can monitor the operating status and parameters of the wind turbine in real time, including wind speed, wind direction, generator speed, blade angle, etc. When the wind turbine is running, the wind turbine sensors will continuously collect data and transmit it to the control system or data acquisition system, which can process and analyze the data to monitor the operating status and performance of the wind turbine; among them, blade angle data is one of the important parameters of the wind turbine operating status. By real-time monitoring of blade angle data, the variable pitch control effect and wind energy conversion efficiency of the wind turbine can be understood. At the same time, the wind turbine can also be optimized and controlled according to the blade angle data to achieve higher power generation and more stable operating status, providing an important basis for the optimization and control of the wind turbine.

[0025] Obtain wind forecast data for a preset time through the meteorological platform, including wind speed forecast sequence and wind direction forecast sequence;

[0026] The meteorological platform is usually composed of a meteorological data collection system, a meteorological data analysis system and a meteorological forecast system. Among them, the meteorological data collection system can collect meteorological data in the area where the wind turbine is located, such as wind speed, wind direction, temperature, air pressure, etc. The meteorological data analysis system can process and analyze the collected data and extract information related to wind forecasting. The meteorological forecast system can perform wind forecasting based on the analyzed data and generate the wind speed forecast sequence and the wind direction forecast sequence; obtaining wind forecast data at a preset time through the meteorological platform can guide the optimization and control of wind turbines, improve wind energy conversion efficiency, and reduce operation and maintenance costs. At the same time, it also helps to better understand and master the operating laws and performance characteristics of wind turbines.

[0027] Inputting the wind speed prediction sequence, wind direction prediction sequence, and pitch angle data into the wind turbine twin model for operation simulation to obtain a simulated power generation sequence;

[0028] The wind speed prediction sequence, wind direction prediction sequence, and pitch angle data are input into the twin model of the wind turbine for operation simulation, so as to obtain the simulated power generation sequence; first, the wind speed prediction sequence and the wind direction prediction sequence are input into the twin model of the wind turbine to simulate the operation state of the wind turbine under different wind speeds and wind directions, and then, the pitch angle data is input into the twin model of the wind turbine to simulate the operation state and performance of the pitch system under different pitch angles; the simulated power generation sequence can describe the change of power generation under different pitch angles under given wind speed and wind direction conditions; the wind speed prediction sequence, the wind direction prediction sequence, and the pitch angle data are input into the twin model of the wind turbine for operation simulation, so as to obtain the simulated power generation sequence, which provides an important basis for the optimization and control of the wind turbine.

[0029] Obtaining the electricity demand sequence at the preset time, and when the difference between the simulated power generation sequence and the electricity demand sequence meets the preset supply-demand difference, no control is performed;

[0030] First, it is necessary to obtain the power demand sequence of the preset time, and the power demand sequence can describe the change of power demand of the power system within a given time period. Then, the simulated power generation sequence is compared with the power demand sequence, and the difference between the two is calculated. If the difference meets the preset supply and demand difference, it means that the power generation of the wind turbine can meet the power demand of the power system, and no additional control is required. At this time, the normal operation of the wind turbine can be maintained without any adjustment; if the difference does not meet the preset supply and demand difference, it means that the power generation of the wind turbine cannot meet the power demand of the power system, and additional control is required. At this time, corresponding measures can be taken according to the actual situation, such as adjusting the operating parameters of the wind turbine, increasing the power generation of other power generation equipment, etc., to ensure the stable operation of the power system. By comparing the difference between the simulated power generation sequence and the power demand sequence, it can be determined whether control is needed to ensure the stable operation of the power system and meet the power demand.

[0031] When the preset supply-demand difference is not met, a pitch optimization time node set is obtained, and pitch optimization is performed on the blade angle data.

[0032] First, it is necessary to determine the set of pitch optimization time nodes. This set can include multiple time nodes, each of which corresponds to a specific pitch adjustment strategy; according to the set of pitch optimization time nodes, the blade angle data is pitch optimized. The pitch angle can be adjusted according to the actual operating conditions and simulation results to achieve higher power generation and more stable operating conditions; in the process of pitch optimization, a variety of algorithms and techniques, such as genetic algorithms, particle swarm algorithms, etc., can be used to optimize the pitch angle calculation. At the same time, data mining and analysis can also be combined with the actual operating data and historical data of the wind turbine to guide the pitch optimization process; when the preset supply and demand difference is not met, the pitch optimization of the blade angle data can be performed by obtaining the pitch optimization time node set to improve the power generation and stability of the wind turbine.

[0033] like Figure 2 As shown, a time node sequence is extracted according to the electricity demand sequence to obtain a first time node;

[0034] Calculate a first difference between a first simulated power generation at a first time node and a first power demand, and determine whether the first difference satisfies a preset supply-demand difference;

[0035] If not, the first time node is added to the pitch optimization time node set, and the next time node is judged in turn until all time node sequences are traversed.

[0036] A time node sequence is extracted according to the power demand sequence. This sequence may include multiple time nodes, each of which corresponds to a specific power demand. Then, the first time node, that is, the first time node in the sequence, is obtained; the first difference between the first simulated power generation and the first power demand at the first time node is calculated, and this difference can describe the gap between the simulated power generation and the power demand at the first time node; and it is determined whether the first difference meets the preset supply and demand difference. If the difference does not meet the preset supply and demand difference, the first time node is added to the pitch optimization time node set. This set can be used for subsequent pitch optimization processes; the next time node is judged in turn until all time node sequences are traversed. For each time node, the above calculation and judgment process is repeated until the entire sequence is traversed; through the above steps, the pitch optimization time node set can be gradually determined, and the blade angle data can be pitch optimized. This process can be adjusted and optimized according to actual conditions to achieve higher power generation and a more stable operating state.

[0037] Constructing an optimization objective function and establishing constraints, wherein the constraints include a blade angle range;

[0038] Randomly generate a plurality of initial blade angles within the blade angle range as a plurality of initial particles;

[0039] Using the optimization objective function to perform fitness evaluation on the multiple initial particles to obtain a fitness evaluation result;

[0040] Screening is performed based on the fitness evaluation results to obtain a high-quality particle set and a low-quality particle set;

[0041] Performing crossover mutation on the high-quality particle set to obtain a new particle set, and replacing the low-quality particle set with the new particle set;

[0042] Iterate the optimization until a particle whose fitness meets the preset fitness is obtained, and use this particle as the optimal blade angle.

[0043] An optimization objective function is constructed, and the optimization objective function can measure the operating performance of the wind turbine, such as power generation, operating stability, etc. These performance parameters can be used as optimization targets, and the targets can be optimized by adjusting parameters such as blade angles; the constraint condition refers to the blade angle range. In actual operation, the blade angle needs to meet a certain range, and this range can be determined according to the actual operating conditions and the characteristics of the wind turbine; multiple initial blade angles are randomly generated within the blade angle range as multiple initial particles. These particles represent different combinations of blade angles and can be used in subsequent optimization processes; the optimization objective function is used to evaluate the fitness of multiple initial particles to obtain fitness evaluation results. Fitness evaluation can measure the quality of each particle for subsequent screening and optimization; screening is performed according to the fitness evaluation results to obtain a high-quality particle set and a low-quality particle set. The high-quality particle set represents a blade angle combination with a higher fitness, while the low-quality particle set represents a combination with a lower fitness; cross-mutation is performed on the high-quality particle set to obtain a new particle set. This process can combine and mutate some features of two particles by random selection to generate a new and better solution; the low-quality particle set is replaced by the new particle set. This can maintain the diversity in the population and introduce better solutions into the population; by repeating the above steps, the optimization is continuously iterated until a particle with a fitness that meets the preset fitness is obtained. This particle can be used as the optimal blade angle to guide the operation of the wind turbine. Through the above method, the optimal blade angle combination can be found while satisfying the constraints, thereby improving the operating performance and efficiency of the wind turbine.

[0044] Normalizing the fitness evaluation results, and using the results as the high-quality probabilities of the multiple initial particles;

[0045] The roulette method is used to allocate each particle to a sector of corresponding proportion according to the high-quality probability, and the high-quality particles are selected randomly by rotating the roulette wheel;

[0046] The selection process is repeated until the preset number of high-quality particles is met, and the high-quality particle set is obtained, and the remaining ones are used as the low-quality particle set.

[0047] The fitness evaluation results are normalized and the results are used as the quality probabilities of multiple initial particles. Normalization can convert the fitness evaluation results into a probability value, so that the quality of each particle can be more intuitively represented; each particle is assigned to a sector of corresponding proportion according to the quality probability. The roulette method is a commonly used particle selection strategy, which can assign different weights to particles according to their quality, so that high-quality particles are more likely to be selected during the selection process. High-quality particles are selected randomly by rotating the roulette wheel. The rotation of the roulette wheel can produce a random selection result to ensure the randomness and diversity of the selection; repeat the above selection process until the preset number of high-quality particles is met. This ensures that a certain number of high-quality particles are found within a limited number of iterations, providing a basis for subsequent crossover mutation and iterative optimization; through the above selection process, a set of high-quality particles is obtained. The remaining particles are used as a set of low-quality particles. This ensures diversity in the population, while providing more possible solutions for subsequent crossover mutation and iterative optimization, more effectively selecting high-quality particles, and providing a basis for subsequent crossover mutation and iterative optimization.

[0048] The optimal blade angle is adopted, combined with the wind speed prediction sequence and the wind direction prediction sequence, and input into the wind turbine twin model for verification;

[0049] When the verification is passed, the optimal blade angle is used for pitch control.

[0050] The optimal blade angle is adopted, and the optimal blade angle, the wind speed prediction sequence and the wind direction prediction sequence are input into the twin model of the wind turbine. The operating state of the wind turbine is simulated by the twin model, and its operating performance under different wind speed and wind direction conditions is observed. The simulation results are compared with the actual operating data to verify whether the optimal blade angle can improve the power generation and stability of the wind turbine in actual operation; when the verification is passed, that is, the performance of the optimal blade angle in the twin model is consistent with the actual operating data, indicating that it can improve the performance of the wind turbine in actual operation. At this time, the optimal blade angle can be used for variable pitch control. The effectiveness and stability of the optimal blade angle in actual operation can be ensured, and better power generation efficiency and operation stability can be provided for the wind turbine. The present application solves the technical problems of low power generation efficiency and poor operation stability of wind turbines in the prior art, and achieves the technical effect of improving the operation efficiency of wind turbines.

[0051] like Figure 3 As shown, the embodiment of the present application also provides a digital twin pitch control optimization system for a wind turbine generator set, including:

[0052] A wind turbine twin model acquisition module 11, wherein the wind turbine twin model acquisition module 11 is used to perform digital twin modeling through the wind power generation process of the target wind turbine to acquire the wind turbine twin model;

[0053] A blade angle data acquisition module 12, wherein the blade angle data acquisition module 12 is used to monitor the target wind turbine in real time through a wind turbine sensor to acquire blade angle data;

[0054] A wind prediction data acquisition module 13, which is used to acquire wind prediction data of a preset time through a meteorological platform, including a wind speed prediction sequence and a wind direction prediction sequence;

[0055] A simulated power generation sequence acquisition module 14, wherein the simulated power generation sequence acquisition module 14 is used to input the wind speed prediction sequence, the wind direction prediction sequence, and the pitch angle data into the twin model of the wind turbine to perform operation simulation and obtain a simulated power generation sequence;

[0056] A sequence difference judgment module 15, the sequence difference judgment module 15 is used to obtain the power demand sequence of the preset time, and when the difference between the simulated power generation sequence and the power demand sequence meets the preset supply and demand difference, no control is performed;

[0057] The pitch optimization module 16 is used to obtain a pitch optimization time node set and perform pitch optimization on the blade angle data when the preset supply-demand difference is not met.

[0058] Furthermore, the embodiment of the present application also includes:

[0059] A three-dimensional twin model construction module, wherein the three-dimensional twin model construction module is used to collect equipment parameter information of the target wind turbine group and construct a three-dimensional twin model;

[0060] A pitch response model establishment module, wherein the pitch response model establishment module is used to collect historical pitch angle and historical wind force data to establish a pitch response model;

[0061] A wind turbine twin model acquisition module, wherein the wind turbine twin model acquisition module is used to integrate the three-dimensional twin model and the pitch response model to acquire the wind turbine twin model.

[0062] Furthermore, the embodiment of the present application also includes:

[0063] A time node acquisition module, the time node acquisition module is used to extract a time node sequence according to the power demand sequence to obtain a first time node;

[0064] A difference judgment module, the difference judgment module is used to calculate a first difference between a first simulated power generation and a first power demand at a first time node, and judge whether the first difference meets a preset supply-demand difference;

[0065] A module for traversing a time node sequence, wherein the module for traversing a time node sequence is used to add the first time node to the pitch optimization time node set if the condition is not satisfied, and to judge the next time node in turn until all time node sequences are traversed.

[0066] Furthermore, the embodiment of the present application also includes:

[0067] A constraint condition establishment module, the constraint condition establishment module is used to construct an optimization objective function and establish constraint conditions, wherein the constraint conditions include a blade angle range;

[0068] An initial particle acquisition module, the initial particle acquisition module is used to randomly generate a plurality of initial blade angles within the blade angle range as a plurality of initial particles;

[0069] A fitness evaluation result acquisition module, wherein the fitness evaluation result acquisition module is used to perform fitness evaluation on the multiple initial particles using the optimization objective function to obtain fitness evaluation results;

[0070] A fitness evaluation result screening module, wherein the fitness evaluation result screening module is used to screen according to the fitness evaluation result to obtain a high-quality particle set and a low-quality particle set;

[0071] A new particle set acquisition module, the new particle set acquisition module is used to perform crossover mutation on the high-quality particle set to obtain a new particle set, and replace the low-quality particle set with the new particle set;

[0072] The iterative optimization module is used for iterative optimization until a particle whose fitness satisfies a preset fitness is obtained, and the particle is used as the optimal blade angle.

[0073] Furthermore, the embodiment of the present application also includes:

[0074] A normalization processing module, the normalization processing module is used to perform normalization processing on the fitness evaluation result, and use the processing result as the high-quality probability of the multiple initial particles;

[0075] A roulette method adopting module, wherein the roulette method adopting module is used to adopt a roulette method, allocate each particle to a sector of corresponding proportion according to the high-quality probability, and select high-quality particles in a random manner by rotating the roulette wheel;

[0076] The low-quality particle set module is used to repeat the selection process until a preset number of high-quality particles is met, obtain the high-quality particle set, and use the remaining ones as the low-quality particle set.

[0077] Furthermore, the embodiment of the present application also includes:

[0078] A verification module, wherein the verification module is used to adopt the optimal blade angle, in combination with the wind speed prediction sequence and the wind direction prediction sequence, and input the wind turbine twin model for verification;

[0079] A pitch control module, wherein the pitch control module is used to perform pitch control using the optimal blade angle when the verification is passed.

[0080] For the specific implementation of the digital twin pitch control optimization system for wind turbines, please refer to the implementation of the digital twin pitch control optimization method for wind turbines mentioned above, which will not be repeated here. The above modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0081] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0082] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A digital twin pitch control optimization method for a wind turbine, characterized in that: include: Digital twin modeling is performed through the wind power generation process of the target wind turbine to obtain a twin model of the wind turbine; The target wind turbine is monitored in real time by a wind turbine sensor to obtain blade angle data; Obtain wind forecast data for a preset time through the meteorological platform, including wind speed forecast sequence and wind direction forecast sequence; Inputting the wind speed prediction sequence, wind direction prediction sequence, and pitch angle data into the wind turbine twin model for operation simulation to obtain a simulated power generation sequence; Obtaining the electricity demand sequence at the preset time, and when the difference between the simulated power generation sequence and the electricity demand sequence meets the preset supply-demand difference, no control is performed; When the preset supply-demand difference is not met, a pitch optimization time node set is obtained, and pitch optimization is performed on the blade angle data.

2. The method according to claim 1, characterized in that Digital twin modeling is performed through the wind power generation process of the target wind turbine to obtain the twin model of the wind turbine, including: Collect equipment parameter information of the target wind turbine and build a 3D twin model; Collect historical pitch angle and wind speed data to establish a pitch response model; The three-dimensional twin model and the pitch response model are integrated to obtain the twin model of the wind turbine generator set.

3. The method according to claim 1, characterized in that When the preset supply-demand difference is not satisfied, the blade angle data is subjected to pitch optimization, including: Extract a time node sequence according to the electricity demand sequence to obtain a first time node; Calculate a first difference between a first simulated power generation at a first time node and a first power demand, and determine whether the first difference satisfies a preset supply-demand difference; If not, the first time node is added to the pitch optimization time node set, and the next time node is judged in turn until all time node sequences are traversed.

4. The method according to claim 3, characterized in that include: Constructing an optimization objective function and establishing constraints, wherein the constraints include a blade angle range; Randomly generate a plurality of initial blade angles within the blade angle range as a plurality of initial particles; Using the optimization objective function to perform fitness evaluation on the multiple initial particles to obtain a fitness evaluation result; Screening is performed based on the fitness evaluation results to obtain a high-quality particle set and a low-quality particle set; Performing crossover mutation on the high-quality particle set to obtain a new particle set, and replacing the low-quality particle set with the new particle set; Iterate the optimization until a particle whose fitness meets the preset fitness is obtained, and use this particle as the optimal blade angle.

5. The method according to claim 4, characterized in that According to the fitness evaluation results, we filter and obtain high-quality particle sets and low-quality particle sets, including: Normalizing the fitness evaluation results, and using the results as the high-quality probabilities of the multiple initial particles; The roulette method is used to allocate each particle to a sector of corresponding proportion according to the high-quality probability, and the high-quality particles are selected randomly by rotating the roulette wheel; The selection process is repeated until the preset number of high-quality particles is met, and the high-quality particle set is obtained, and the remaining ones are used as the low-quality particle set.

6. The method according to claim 4, characterized in that Also includes: The optimal blade angle is adopted, combined with the wind speed prediction sequence and the wind direction prediction sequence, and input into the wind turbine twin model for verification; When the verification is passed, the optimal blade angle is used for pitch control.

7. A digital twin pitch control optimization system for wind turbines, characterized in that: include: A wind turbine twin model acquisition module, wherein the wind turbine twin model acquisition module is used to perform digital twin modeling through the wind power generation process of the target wind turbine to acquire the wind turbine twin model; A blade angle data acquisition module, wherein the blade angle data acquisition module is used to monitor the target wind turbine in real time through a wind turbine sensor to acquire blade angle data; A wind prediction data acquisition module, which is used to acquire wind prediction data of a preset time through a meteorological platform, including a wind speed prediction sequence and a wind direction prediction sequence; A simulated power generation sequence acquisition module, wherein the simulated power generation sequence acquisition module is used to input the wind speed prediction sequence, the wind direction prediction sequence, and the pitch angle data into the twin model of the wind turbine to perform operation simulation and obtain a simulated power generation sequence; A sequence difference judgment module, the sequence difference judgment module is used to obtain the power demand sequence of the preset time, and when the difference between the simulated power generation sequence and the power demand sequence meets the preset supply and demand difference, no control is performed; A pitch optimization module, wherein the pitch optimization module is used to obtain a pitch optimization time node set and perform pitch optimization on the blade angle data when the preset supply and demand difference is not met.

Citation Information

Cited By

  • Photovoltaic tracking support angle optimization and power generation efficiency improvement method for meteorological prediction

    CN120257857A