A vehicle model self-adaptive based intelligent front wheel spoiler system
By using an intelligent front wheel deflector system that combines deep reinforcement learning and adaptive curved surface materials, the problems of poor adaptability and response delay of traditional deflectors are solved. This enables the optimization of wind resistance and lift across multiple vehicle models, reducing development costs and time.
Patent Information
- Application Number
- CN202610618676.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-05-07
- Publication Date
- 2026-07-10
AI Technical Summary
Traditional automotive wind deflectors cannot adapt to the geometric differences in wheel arches of different car models, resulting in large differences in drag coefficients, delayed adjustment response, long development cycles, and high costs.
The system employs a vehicle-adaptive intelligent front wheel deflector system, which combines a parameter perception layer, an intelligent decision-making layer, and an execution control layer. It utilizes deep reinforcement learning algorithms and adaptive curved surface materials to achieve dynamic adjustment to optimize wind resistance and lift.
Achieve optimal drag and lift control under different vehicle models and complex operating conditions, reduce response time, lower development costs and cycle time, and improve vehicle performance.
Smart Images
Figure CN122354658A_ABST
Abstract
Description
Technical Field
[0001] This application belongs to the field of automotive aerodynamic optimization technology, and in particular relates to an intelligent front wheel wind deflector system based on vehicle model adaptation. Background Technology
[0002] Traditional fixed wind deflectors for automobiles cannot adapt to the differences in wheel arch geometry between different models, resulting in a drag coefficient difference of up to 10% to 15% between sedans and SUVs; the response delay of passive adjustment mechanisms when the vehicle speed changes exceeds 500ms, which cannot meet the requirements of dynamic operating conditions; moreover, each new model requires 150h to 200h of wind tunnel testing, resulting in a long development cycle and high costs.
[0003] Actual testing revealed significant performance deficiencies in the traditional solution:
[0004] Therefore, it is necessary to provide a new intelligent front wheel deflector system based on vehicle model adaptation to solve the above-mentioned technical problems. Summary of the Invention
[0005] The purpose of this disclosure is to provide a vehicle-type adaptive intelligent front wheel deflector system to solve the above-mentioned problems.
[0006] This disclosure achieves the above objectives through the following technical solutions: A vehicle-vehicle-adaptive intelligent front wheel deflector system includes: The parameter sensing layer is used to collect sensor parameters in real time. The intelligent decision-making layer is used to calculate the minimum drag coefficient and maximize downforce based on the sensor parameters and using a deep reinforcement learning algorithm, given vehicle characteristics and operating conditions, and to complete the adjustment within the response time constraint and generate control commands. An execution control layer is used to execute the control commands; The front wheel dam is dynamically adjusted under the control of the aforementioned execution layer.
[0007] As a further optimization of this disclosure, the sensing parameters include wheel arch inner diameter, tire contact patch, and front suspension displacement.
[0008] As a further optimization of this disclosure, a deep learning database is also included, which contains vehicle feature data of multiple benchmark vehicle models.
[0009] As a further optimization of this disclosure, the vehicle features include vehicle type, wheel arch geometry, suspension height, and track width.
[0010] As a further optimization of this disclosure, the operating conditions include high-speed cruising, urban driving conditions, strong crosswinds, braking, and low-temperature environments.
[0011] As a further optimization of this disclosure, the execution control layer controls the front wheel choke plate through a three-degree-of-freedom electric adjustment mechanism to execute the control command.
[0012] As a further optimization of this disclosure, the execution control layer enables the front wheel choke to adaptively adjust in three directions: angle of attack, chord length, and curvature.
[0013] As a further optimization of this disclosure, the front wheel baffle adopts a deformable curved surface design formed by a shape memory alloy and carbon fiber composite structure.
[0014] As a further optimization of this disclosure, the deep reinforcement learning algorithm adopts the NSGA-II algorithm to solve the Pareto front, and the optimization objective includes a multi-objective trade-off between minimizing the drag coefficient and maximizing downforce.
[0015] As a further optimization of this disclosure, the execution control layer also includes a heating module for preheating the front wheel choke.
[0016] The beneficial effects of this disclosure are as follows: Through a three-layer optimization architecture of multi-dimensional parameter perception, intelligent decision-making and execution control, combined with adaptive curved surface materials (SMA and carbon fiber composite structure), the dynamic adjustment of parameters such as curvature, thickness and chord length is achieved, thereby realizing optimal wind resistance and lift control under various vehicle models and complex working conditions. Attached Figure Description
[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 This is a system structure block diagram of an embodiment of the present disclosure. Detailed Implementation
[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0020] like Figure 1 As shown, an intelligent front wheel wind deflector system based on vehicle model adaptation includes a three-layer optimization architecture: a parameter perception layer (perception and acquisition), an intelligent decision-making layer (optimization and strategy), and an execution control layer (execution drive and mechanism). The front wheel wind deflector, combined with adaptive curved surface material (shape memory alloy SMA and carbon fiber composite structure), realizes dynamic adjustment of parameters such as curvature, thickness, and chord length, thereby achieving optimal wind resistance and lift control under various vehicle models and complex working conditions.
[0021] The parameter perception layer collects key sensing parameters, including wheel arch inner diameter (measurement accuracy ±0.5 mm, update frequency 10 Hz), tire contact patch (measurement accuracy ±2 mm², update frequency 20 Hz), and front suspension displacement (measurement accuracy ±0.3 mm, update frequency 50 Hz). The sensor data such as tire contact patch and front suspension displacement need to be fused within the vehicle control system to ensure robustness.
[0022] The intelligent decision-making layer employs a deep reinforcement learning (DRL) algorithm to minimize the drag coefficient Cd and maximize downforce under given vehicle characteristics and operating conditions, while completing the adjustment within response time constraints. Specifically, the NSGA-II algorithm is used to solve the Pareto front, and the optimization objective includes a multi-objective trade-off between minimizing Cd and maximizing downforce. The DRL model should be fully validated in simulation and testing environments to ensure that uncontrollable dam failure does not occur under extreme conditions. The action execution must have a fault detection and rollback strategy, which should roll back to a safe dam geometry state when sensor anomalies or mechanical failures are detected.
[0023] The control layer can adaptively adjust in three directions: angle of attack, chord length, and curvature. It employs a three-degree-of-freedom electric adjustment mechanism. Example parameters: Angle of attack: -8° to +25°, resolution 0.1°; Chord length: 80 mm to 150 mm, resolution 0.5 mm; Curvature: 0.05 to 0.2 m. - ¹, resolution 0.01 m - ¹. Control objective: To quickly and accurately adjust the shape of the baffle plate under different operating conditions to achieve the optimal wind resistance and lift.
[0024] The front wheel dam uses adaptive curved surface technology: 1. Materials and Structure: Combining shape memory alloy (SMA) with a carbon fiber composite structure to form a deformable curved surface, exhibiting high specific stiffness and low weight. When SMA is used as the primary drive material, operating temperature range, cycle life, response speed, and fatigue characteristics must be considered. Combining it with a carbon fiber composite base reduces weight and increases strength. To avoid thermal stress and fatigue issues, the design should include thermal management and redundancy mechanisms (such as redundant drive channels and limit protection).
[0025] 2. Dynamic adjustment capability: The curvature dynamic adjustment strain can reach 2.5% to achieve a rapid response to bending and curvature.
[0026] 3. Vibration characteristics: The design goal is to keep the modal frequencies outside the vibration frequency range of 45 Hz to 50 Hz to reduce the risk of resonance.
[0027] 4. Weight comparison: The weight is reduced by about 40% compared to traditional metal structures.
[0028] The database contains 200-250 benchmark vehicle models, covering features such as different vehicle types, wheel arch geometry, suspension height, and track width. The database should support continuous expansion, capable of incorporating new vehicle parameters and wind tunnel / CFD validation results. The system interface should be compatible with existing vehicle control architectures, ensuring seamless integration with the powertrain, suspension, and other systems. The database is used for offline training; in the online phase, it performs policy inference based on perception layer data and current operating conditions, outputting optimized deflector parameters. Its dynamic response time is less than 80 ms.
[0029] Five operating modes were developed, dynamically adjusting the deflector parameters under five conditions: high-speed cruising, urban driving, strong crosswinds, braking, and low-temperature environments, to achieve optimal performance under the target conditions. This system achieves the best aerodynamic performance for different vehicle models under various driving conditions through multi-dimensional parameter collaborative optimization, and is applicable to various vehicle types including passenger cars, SUVs, and commercial vehicles, as shown in the table below:
[0030] Taking a certain SUV as an example, describe the adjustment process of the intelligent front wheel deflector system: I. Vehicle model characteristic parameters (matched from the database) Baseline parameters of a certain SUV in the system database: (1) Wheel assembly geometry: front wheel arch inner diameter 720mm, wheel arch depth 150mm; (2) Suspension system: CDC electromagnetic suspension, standard ground clearance 180mm; (3) Track width: Front track width 1620mm, rear track width 1620mm; (4) Tire specifications: 245 / 45 R20; (5) Vehicle dimensions: Length 4781mm × Width 1920mm × Height 1671mm; (6) Aerodynamic reference: basic drag coefficient Cd0.32.
[0031] II. System Adjustment Completion Process Phase 1: Parameter Sensing and Data Acquisition (0-20ms) 1. Measurement of wheel arch inner diameter: (1) The laser rangefinder measures the inner diameter of the wheel arch with an accuracy of ±0.5 mm; (2) Confirm the current value is 720.3 mm (within the tolerance range); (3) Update frequency: 10Hz.
[0032] 2. Tire contact mark monitoring: (1) Pressure distribution sensors monitor tire contact area in real time; (2) Current measured value: 245 mm2 (under standard load); (3) Update frequency: 20Hz; 3. Front suspension displacement monitoring: (1) Displacement sensors monitor the status of the CDC electromagnetic suspension; (2) Current front suspension compression: 2.1mm (normal driving condition); (3) Update frequency: 50Hz; 4. Data fusion and preprocessing: (1) The vehicle control system performs Kalman filtering fusion on multi-source sensor data; (2) Eliminate sensor noise and ensure data robustness; (3) Generate the comprehensive working condition parameter vector.
[0033] Phase Two: Intelligent Decision Making and Optimization Calculation (20-60ms) 1. Operating mode recognition: (1) GPS vehicle speed: 105km / h → trigger high-speed cruise mode; (2) Crosswind sensor: Crosswind angle 8° (<15°) → Strong crosswind mode not triggered; (3) Temperature sensor: Ambient temperature 18°C → Low temperature mode not triggered; (4) Braking pressure sensor: deceleration 0.2g (<0.4g) → braking mode not triggered; 2. DRL Model Inference: (1) Input: Vehicle model characteristic parameters + real-time operating condition data; (2) Deep reinforcement learning models call upon pre-trained NSGA-II Pareto frontiers; (3) Multi-objective optimization: Objective 1: Minimize the drag coefficient Cd; Objective 2: Maximize front axle downforce; Constraints: Response time < 80ms, limited range of motion of the mechanism.
[0034] 3. Optimal parameter calculation: (1) Based on the high-speed cruise mode strategy: Target angle of attack: 23.5° (within the range of 22°-25°); String length optimization: Maintain the baseline value of 120mm; Curvature optimization: 0.12m -1 (Balancing wind resistance and downforce); DRL output: Angle of attack 23.5°, chord length 120mm, curvature 0.12m -1 .
[0035] Phase 3: Execution Control and Organizational Adjustment (60-75ms) 1. Operation of the three-degree-of-freedom electric adjustment mechanism: (1) Angle of attack adjustment: Current value: 5° (default urban driving conditions); Target value: 23.5°; Execution: The electric servo mechanism drives the wind baffle to rotate around the shaft; Resolution: 0.1°, Adjustment time: 15ms.
[0036] (2) String length adjustment: Current value: 120mm; Target value: 120mm (no adjustment required); Maintain the status.
[0037] (3) Curvature adjustment: Current value: 0.08m -1 ; Target value: 0.12m -1 ; Execution: SMA drive unit activated, carbon fiber composite structure bending; Strain variation: 1.2% (within 2.5% of capacity); Resolution: 0.01m -1 Adjustment time: 10ms.
[0038] 2. Adaptive surface material response: (1) SMA drive: Current passes through the shape memory alloy, and the temperature rises to 65°C; (2) Carbon fiber composite structure: provides structural support and limits excessive deformation; (3) Thermal management: The miniature heat sink maintains the operating temperature within a safe range of 40-70°C; (4) Redundancy mechanism: standby drive channel, limit protection activated.
[0039] Phase 4: Verification and Feedback (75-80ms) 1. Position sensor verification: (1) Actual angle of attack: 23.4° (error 0.1°, within the allowable range); (2) Actual chord length: 120.1 mm (error 0.1 mm); (3) Actual curvature value: 0.121m -1 (Error 0.001m) -1 ).
[0040] 2. Wind tunnel effect simulation verification: (1) Real-time simulation verification of computational fluid dynamics (CFD); (2) Expected improvement in drag coefficient: Cd will decrease from 0.32 to 0.295; (3) Expected increase in downforce: 12% increase in downforce on the front axle.
[0041] 3. Fault detection and safety rollback: (1) Sensor health status: All are normal; (2) Mechanism movement range: all within safe limits; (3) Temperature monitoring: SMA temperature 62°C, within the safe range; (4) If an abnormality is detected, immediately retreat to a safe geometric state (angle of attack 5°, curvature 0.08m). -1 ).
[0042] System performance indicators achieved: 1. Total response time: 78ms (<80ms target); 2. Accuracy achieved: (1) Angle of attack error: 0.1° (<0.5° allowable error); (2) Chord length error: 0.1mm (<0.5mm allowable error); (3) Curvature error: 0.001m -1 (<0.005m -1 (Tolerance)
[0043] 3. Energy consumption: Approximately 15W.s of energy is consumed per adjustment; 4. Weight advantage: 43% weight reduction compared to traditional metal structures.
[0044] Performance in adaptability to operating conditions: The system dynamically adjusts based on five operating modes:
[0045] System collaboration: 1. In coordination with CDC electromagnetic suspension: The status of the wind deflector is fed back to the suspension control to optimize the vehicle's attitude; 2. Integration with ESP system: Works in conjunction with ESP system under strong crosswind conditions to improve vehicle stability; 3. Energy recovery optimization: Reduced wind resistance indirectly increases the range of electric vehicles by approximately 1.2%.
[0046] The embodiments described above are merely examples of several implementations of this disclosure, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent disclosure. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this disclosure, and these modifications and improvements all fall within the protection scope of this disclosure.
Claims
1. A vehicle-vehicle-adaptive intelligent front wheel wind deflector system, characterized in that, include: The parameter sensing layer is used to collect sensor parameters in real time. The intelligent decision-making layer is used to calculate the minimum drag coefficient and maximize downforce based on the sensor parameters and using a deep reinforcement learning algorithm, given vehicle characteristics and operating conditions, and to complete the adjustment within the response time constraint and generate control commands. An execution control layer is used to execute the control commands; The front wheel dam is dynamically adjusted under the control of the aforementioned execution layer.
2. The intelligent front wheel wind deflector system based on vehicle model adaptation according to claim 1, characterized in that, The sensing parameters include wheel arch inner diameter, tire contact patch, and front suspension displacement.
3. The intelligent front wheel wind deflector system based on vehicle model adaptation according to claim 1, characterized in that, It also includes a deep learning database, which contains vehicle feature data of multiple benchmark vehicle models.
4. The intelligent front wheel wind deflector system based on vehicle model adaptation according to claim 1, characterized in that, The vehicle characteristics include vehicle type, wheel arch geometry, suspension height, and track width.
5. The intelligent front wheel wind deflector system based on vehicle model adaptation according to claim 1, characterized in that, The operating conditions include high-speed cruising, urban driving, strong crosswinds, braking, and low-temperature environments.
6. The intelligent front wheel wind deflector system based on vehicle model adaptation according to claim 1, characterized in that, The execution control layer controls the front wheel choke plate through a three-degree-of-freedom electric adjustment mechanism to execute the control command.
7. The intelligent front wheel wind deflector system based on vehicle model adaptation according to claim 1, characterized in that, The execution control layer enables the front wheel dam to adaptively adjust in three directions: angle of attack, chord length, and curvature.
8. The intelligent front wheel wind deflector system based on vehicle model adaptation according to claim 1, characterized in that, The front wheel baffle adopts a deformable curved surface design formed by a composite structure of shape memory alloy and carbon fiber.
9. The intelligent front wheel wind deflector system based on vehicle model adaptation according to claim 1, characterized in that, The deep reinforcement learning algorithm uses the NSGA-II algorithm to solve the Pareto front, and the optimization objective includes a multi-objective trade-off between minimizing the drag coefficient and maximizing downforce.
10. A vehicle-adaptive intelligent front wheel wind deflector system according to claim 1, characterized in that, The execution control layer also includes a heating module for preheating the front wheel choke.