A method and system for structural and aerodynamic optimization control of a morphing wing

Abstract

The application discloses a kind of structure aerodynamic optimization control method and system of retractable wing plate, the method includes: reinforcement learning SAC algorithm is deployed on server, forms master control module;Current wind direction angle and surface pressure data of control system structure are read as state value input, the ratio of the change of lift and drag coefficient after intelligent control of reinforcement learning agent model compared to basic lift and drag coefficient is taken as reward value, according to reward value situation, wing plate retraction strategy is formulated as action value, new wing plate retraction information is transmitted to motor;The environmental parameters of control system are obtained and returned to master control module for interactive iteration, and the optimal structure aerodynamic optimization control strategy is obtained.The application can adjust aerodynamic performance in real time, so as to optimize the aerodynamic performance of structure under different inflow conditions and different task backgrounds.The application is a kind of structure aerodynamic optimization control method and system of retractable wing plate, which can be widely applied in the field of structure aerodynamic performance control technology.

Description

Technical Field

[0001] This invention relates to the field of structural aerodynamic performance control technology, and in particular to a structural aerodynamic optimization control method and system for a retractable airfoil. Background Technology

[0002] Aerodynamic performance describes the force and moment characteristics of a structure in airflow and its response to the external environment. In long-span structures and high-rise buildings, the impact of aerodynamic performance is crucial to structural design and safety, requiring the minimization of lift drag. Conversely, in some wind energy harvesting structures, it is necessary to increase lift drag to maximize wind energy utilization. Therefore, controlling the aerodynamic performance of structures has always been a significant challenge in the field of wind engineering. Against the backdrop of global climate change and the development of new energy sources, aerodynamic optimization of structures has become an urgent issue. Some traditional methods include improving structural shape design and optimizing building complex layouts; however, these methods only consider single environments or tasks and cannot achieve optimization goals under complex environments or tasks. Summary of the Invention

[0003] To address the aforementioned technical problems, the present invention aims to provide a structural aerodynamic optimization control method and system for a retractable airfoil, which can adjust aerodynamic performance in real time, thereby optimizing the aerodynamic performance of the structure under different incoming flow conditions and different mission backgrounds.

[0004] The first technical solution adopted in this invention is: a structural aerodynamic optimization control method for a retractable airfoil, comprising the following steps: The control system uses the PyTorch framework to deploy the reinforcement learning SAC algorithm on the server, forming the main control module. The main control module communicates with the stepper motor that controls the retractable wing plate through an Arduino microcontroller, controlling the motor's speed and direction to drive the retractable wing plate to extend and retract. The main control module reads the current wind direction angle and surface pressure data of the control system structure as the state value input, takes the ratio of the change in lift and drag coefficients after intelligent control by the reinforcement learning agent model to the change in the basic lift and drag coefficients as the reward value, and formulates the wing extension and retraction strategy as the action value based on the reward value, and generates new wing extension and retraction information and transmits it to the motor. The environmental parameters of the control system are obtained and sent back to the main control module for interactive iteration to obtain the optimal structural aerodynamic optimization and control strategy.

[0005] Furthermore, the lift and drag coefficients are determined by extracting data features from sensors, and their expressions are as follows: ; In the above formula, , This indicates the air resistance and lift experienced by the structure. Indicates air density, Indicates wind speed. Indicates the area of ​​the windward side. Indicates the drag coefficient. This represents the lift coefficient.

[0006] The second technical solution adopted in this invention is: a structural aerodynamic optimization control system for a retractable wingplate, comprising a retractable wingplate, a control system, a power supply system, and sensors. The control system is electrically connected to the sensors and the retractable wingplate, and the power supply system is electrically connected to the control system. The retractable wing panels are arranged around the perimeter of the control system structure; The control system is used to control the extension and retraction of the retractable wing panel; The power system is used to provide power; The sensor is used to acquire environmental parameters of the control system.

[0007] Furthermore, the control system structure is a square column with slots cut at the four corners and retractable wing plates installed. The retractable wing plates can freely extend and retract along the slots, and each retractable wing plate is connected to two synchronous stepper motors at the top and bottom to control the extension and retraction of the retractable wing plates.

[0008] Furthermore, the side length of the square column is D, the thickness of the retractable wing is e, the maximum extension length of the retractable wing is L, and the structural airflow angle of the square column is... Depending on the angle of the incoming wind direction, the extension and retraction of different retractable airfoils are controlled, where L / D = 1 / 5.

[0009] Furthermore, the sensors include a pressure sensor and a temperature sensor. The pressure sensor is mounted on the surface of the control system structure to analyze the current incoming flow environment and the pressure distribution on the structure surface, and the temperature sensor is used to measure the ambient temperature.

[0010] The beneficial effects of the method and system of this invention are as follows: This invention reads the current wind direction angle and surface pressure data of the control system structure as state value input, and uses the ratio of the change in lift and drag coefficients after intelligent control by the reinforcement learning agent model to the change in the basic lift and drag coefficients as reward value. According to different incoming wind direction angles, the extension and retraction of different airfoils can be controlled to achieve aerodynamic optimization of the structure. Based on the reward value, an airfoil extension and retraction strategy is formulated as action value, and new airfoil extension and retraction information is generated and transmitted to the motor. By controlling the extension and retraction of the airfoils, the aerodynamic performance can be adjusted in real time, thereby optimizing the aerodynamic performance of the structure under different incoming flow conditions and different task backgrounds. Attached Figure Description

[0011] Figure 1 This is a flowchart of the steps of a structural aerodynamic optimization control method for a retractable wingplate according to the present invention; Figure 2 This is a structural block diagram of a retractable wingplate aerodynamic optimization control system according to the present invention; Figure 3 This is a schematic diagram of the arrangement of square column telescopic wing plates provided in a specific embodiment of the present invention. Detailed Implementation

[0012] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments. The step numbers in the following embodiments are only for ease of explanation and do not limit the order of the steps. The execution order of each step in the embodiments can be adapted according to the understanding of those skilled in the art.

[0013] Reference Figure 1 This invention provides a structural aerodynamic optimization control method for a retractable airfoil, the method comprising the following steps: S100 and the control system use the PyTorch framework to deploy the reinforcement learning SAC algorithm on the server to form the main control module; The S200 main control module communicates with the stepper motor that controls the retractable wing plate via an Arduino microcontroller, controlling the motor's speed and direction to drive the retractable wing plate to extend and retract. S300: The main control module reads the current wind direction angle and surface pressure data of the control system structure as the state value input, takes the ratio of the change in lift and drag coefficients after intelligent control by the reinforcement learning agent model to the change in the basic lift and drag coefficients as the reward value, and formulates the wing extension and retraction strategy as the action value based on the reward value, and generates new wing extension and retraction information and transmits it to the motor. S400 acquires environmental parameters from the control system and sends them back to the main control module for interactive iteration to obtain the optimal structural aerodynamic optimization and control strategy.

[0014] Specifically, the control system uses the PyTorch framework to deploy the reinforcement learning SAC algorithm on a server, forming a main control module. The main control module communicates with the stepper motor controlling the wingplate via an Arduino microcontroller, controlling the motor's speed / direction to drive the wingplate's extension and retraction. During the control phase, the main control module reads the current wind direction angle and surface pressure data of the structure as state value inputs, uses the ratio of the lift and drag coefficients after intelligent control by the reinforcement learning agent model to the changes in the base lift and drag coefficients as reward values, and formulates the wingplate extension and retraction strategy as action values ​​based on the reward values. That is, new wingplate extension and retraction information is generated and transmitted to the motor to achieve structural flow field environment control. The sensor information acquired after control is fed back to the main control module for interactive iteration, and finally the optimal structural aerodynamic optimization and control strategy is obtained.

[0015] Reference Figure 2 A structural aerodynamic optimization control system for a retractable wingplate includes a retractable wingplate, a control system, a power supply system, and sensors. The control system is electrically connected to the sensors and the retractable wingplate, and the power supply system is electrically connected to the control system. The retractable wing panels are arranged around the perimeter of the control system structure; The control system is used to control the extension and retraction of the retractable wing panel; The power system is used to provide power; The sensor is used to acquire environmental parameters of the control system.

[0016] Specifically, it includes retractable wing panels arranged around the structure, a control system for controlling the extension and retraction of the wing panels, a power system for the wing panel control system, and sensors distributed around the structure. The control system is connected to the sensors and the retractable wing panels, and the power system is connected to the wing panel control system to provide power support for it.

[0017] Furthermore, the winglets located around the structure are connected to the stepper motors of the control system and have controllable extension / retraction lengths. The control system monitors relevant parameters such as pressure distribution on the structural surface and adjusts the extension / retraction length of the winglets according to these parameters. A power system provides power to the control motors of the winglets. Sensors monitor the pressure distribution and aerodynamic performance of the structural surface in real time. The control system reads the pressure distribution parameters of the structure fed back by the sensors, and then calculates the optimal extension / retraction length of each winglet based on these parameters to achieve aerodynamic optimization, increase or decrease drag and lift. Finally, the control system issues commands to adjust the extension / retraction of the winglets, control the pressure distribution on the structural surface, and optimize the aerodynamic performance of the structure.

[0018] like Figure 3 As shown, a square column has slots cut at its four corners and wing plates installed. The wing plates can freely extend and retract along the slots. The side length of the square column is D, and the thickness of the wing plates is e. Each extendable wing plate is connected to two synchronous stepper motors at the top and bottom to control the extension and retraction of the wing plate. The maximum extension length of the wing plate is L, where L / D = 1 / 5. The incoming wind direction angle of the square column structure is... Depending on the incoming airflow angle, the extension and retraction of different airfoils can be controlled to achieve aerodynamic optimization of the structure. According to experimental tests, using the above aerodynamic drag reduction scheme, in... At °, with winglets 3 and 4 fully extended, the Karman vortex street of the square column wake can be greatly reduced, resulting in a 20% reduction in the overall drag coefficient and an 80% reduction in the RMS value of the lift coefficient. If winglets 1 and 2 are fully extended, the average drag coefficient of the overall structure can be increased by 30%, and the RMS value of the lift coefficient can be increased by 60%.

[0019] The sensors primarily consist of pressure and temperature sensors. These pressure sensors help the system analyze the current incoming flow environment and pressure distribution on the structural surfaces, thus influencing the airfoil's operating mode. Different wind angles may require different airfoil extension / retraction patterns to achieve optimal aerodynamic performance. Pressure sensors can be mounted on the structural surface to monitor the surface pressure distribution, guiding the system to adjust the airfoil's extension / retraction method to optimize aerodynamic performance. Temperature sensors are used to measure ambient temperature, as temperature significantly affects air density and viscosity. This data can be used to calibrate measurements from other sensors, ensuring the system's accuracy under varying temperatures. The vehicle drag coefficient is determined using data features extracted from the sensors, expressed as: ; In the above formula, , This represents the air resistance and lift experienced by the structure. Indicates air density, Indicates wind speed. Indicates the area of ​​the windward side. Indicates the drag coefficient. This represents the lift coefficient.

[0020] Therefore, the specific working method of this invention is as follows: 1) Data acquisition and monitoring are achieved by sensors. These data include relevant parameters such as the direction of the incoming flow, the pressure and temperature of the structural surface, and the status information of the airfoil. Data acquisition is real-time, and the sensors need to continuously monitor and update the data in order to make timely responses. 2) The control system processes and analyzes data to make decisions regarding winglet extension and retraction adjustments. It uses advanced algorithms to process and analyze sensor data, determining current lift and drag levels and adjusting winglet extension and retraction under different wind directions and pressure conditions. Based on the data processing and analysis results, the control system makes decisions on how to adjust winglet extension and retraction to ensure optimal aerodynamic performance under various operating conditions. Specifically, the control system uses the PyTorch framework to deploy the reinforcement learning SAC algorithm on a server, forming the main control module. This main control module utilizes Arduin... The microcontroller controls the stepper motor of the winglet to communicate and control the motor's speed / direction, thereby driving the extension and retraction of the winglet. During the control phase, the main control module reads the current wind direction angle and surface pressure data of the structure as state value inputs. It uses the ratio of the change in lift and drag coefficients after intelligent control by the reinforcement learning agent model to the change in the basic lift and drag coefficients as reward values. Based on the reward values, it formulates the winglet extension and retraction strategy as action values, that is, generates new winglet extension and retraction information and transmits it to the motor to realize the structural flow field environment control. The sensor information acquired after control is sent back to the main control module for interactive iteration, and finally obtains the optimal structural aerodynamic optimization and control strategy.

[0021] 3) Wing control: The control system uses the power system to adjust the rotation of the motor to drive the extension and retraction of the wing. Based on the decision, the control system can extend or shorten a specific wing, change the pressure distribution on the structural surface, and thus control the lift and drag coefficients.

[0022] 4) Real-time monitoring and feedback: The sensors continuously monitor the system's operating status and performance. The sensors can provide real-time feedback information so that users can understand the system's performance and status, and confirm whether the structure is in a better aerodynamic performance state than before the adjustment. If an abnormality occurs or the sensor data changes, the control system can make timely adjustments to ensure the normal operation of the system.

[0023] In summary, the embodiments of the present invention can solve the problem of structural aerodynamic optimization in a more efficient and comprehensive way, introducing an innovative technology to the field of wind engineering; by controlling the extension and retraction of different airfoils, lift and drag can be adjusted in real time, thereby optimizing the aerodynamic performance of the structure under different environments and tasks, and achieving the purpose of reducing structural wind load or increasing energy utilization.

[0024] The content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0025] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.

Claims

1. A structural aerodynamic optimization control system for a retractable wingplate, characterized in that, The system includes a retractable wing panel, a control system, a power supply system, and sensors. The control system is electrically connected to the sensors and the retractable wing panel, and the power supply system is electrically connected to the control system. The retractable wing panels are arranged around the perimeter of the control system structure; The control system is used to control the extension and retraction of the retractable wing panel; The control system is a square column with slots cut at the four corners and retractable wing plates installed. The retractable wing plates can freely extend and retract along the slots, and each retractable wing plate is connected to two synchronous stepper motors at the top and bottom to control the extension and retraction of the retractable wing plates. The square column has a side length of D, a retractable wing plate thickness of e, a maximum retractable wing plate length of L, and a structural airflow angle of the square column. The extension and retraction of different retractable airfoils are controlled according to the different angles of the incoming wind direction, where L / D=1 / 5; The control system uses the PyTorch framework to deploy the reinforcement learning SAC algorithm model on the server to form the main control module. The main control module communicates with the stepper motor that controls the retractable wing plate through the Arduino microcontroller to control the speed and direction of the motor, thereby driving the retractable wing plate to extend and retract. The power system is used to provide power; The sensor is used to acquire environmental parameters of the control system; The main control module reads the current wind direction angle and surface pressure data of the control system structure as state value inputs, and uses the ratio of the lift and drag coefficients after intelligent control by the reinforcement learning SAC algorithm model to the changes in the basic lift and drag coefficients as reward values. Based on the reward values, it formulates the extension and retraction strategy of the retractable airfoil as action values, generates new retractable airfoil extension and retraction information, and transmits it to the stepper motor. The sensor information acquired after control is sent back to the main control module for interactive iteration, and finally obtains the optimal structural aerodynamic optimization and control strategy.

2. The structural aerodynamic optimization control system for a retractable winglet according to claim 1, characterized in that, The sensors include a pressure sensor and a temperature sensor. The pressure sensor is mounted on the surface of the control system structure to analyze the pressure distribution on the surface of the current incoming flow environment and the control system structure. The temperature sensor is used to measure the ambient temperature.

3. The control method for the structural aerodynamic optimization control system of a retractable airfoil according to any one of claims 1 to 2, characterized in that, Includes the following steps: The control system uses the PyTorch framework to deploy the reinforcement learning SAC algorithm on the server, forming the main control module. The main control module communicates with the stepper motor that controls the retractable wing plate through an Arduino microcontroller, controlling the motor's speed and direction to drive the retractable wing plate to extend and retract. The main control module reads the current wind direction angle and surface pressure data of the control system structure as the state value input, takes the ratio of the change in lift and drag coefficients after intelligent control by the reinforcement learning agent model to the change in the basic lift and drag coefficients as the reward value, and formulates the wing extension and retraction strategy as the action value based on the reward value, and generates new wing extension and retraction information and transmits it to the motor. The environmental parameters of the control system are obtained and sent back to the main control module for interactive iteration to obtain the optimal structural aerodynamic optimization and control strategy.

4. The control method for the structural aerodynamic optimization control system of a retractable wingplate according to claim 3, characterized in that, The lift and drag coefficients are determined by extracting data features from sensors, and their expressions are as follows: ； In the above formula, This indicates the air resistance and lift experienced by the structure. Indicates air density, Indicates wind speed. Indicates the area of ​​the windward side. Indicates the drag coefficient. This represents the lift coefficient.

Images