Target region aerodynamic noise active control method, device, medium and product for tilt-rotor transition flight

Abstract

This application discloses a method, device, medium, and product for active control of aerodynamic noise in the target area during tiltrotor transition flight, relating to the field of noise control. The method includes: during tilt transition conditions, constructing a target noise control area within a set region based on flight mission and environmental requirements, discretizing it into multiple spatial observation points and assigning weights; parameterizing the controllable aerodynamic loads on the blades to obtain control load variables, establishing the relationship between the control load variables and the acoustic response of the target noise control area, determining unknowns, and establishing a global acoustic optimization objective function; using the DDQN algorithm as a solver, optimizing the global acoustic optimization objective function under the premise of satisfying actuator physical constraints to obtain the optimal control law, performing feedforward execution and feedback correction to achieve active control of aerodynamic noise in the target area. This application can solve the main shortcomings and derived problems caused by focusing on single-point noise control.

Description

Technical Field

[0001] This application relates to the field of noise control, and in particular to a method, device, medium and product for active control of aerodynamic noise in the target area for tiltrotor transition flight. Background Technology

[0002] Tiltrotor aircraft combine the vertical takeoff and landing capabilities of helicopters with the cruise efficiency of fixed-wing aircraft, typically undergoing multiple flight states including helicopter mode, tilt transition mode, and fixed-wing mode. During the tilt transition flight phase, due to continuous changes in tilt angle, rapid switching of inflow direction, and significantly unsteady blade loads, the radiation direction, peak characteristics, and spatial distribution of rotor noise change drastically within a short period. This is particularly problematic in urban low-altitude operations, where noise constraints often target sensitive ground areas (such as residential areas, schools, hospitals, and the vicinity of takeoff and landing points). Traditional noise reduction approaches focused on single-direction or single-point assessments are insufficient to meet regional compliance requirements. Therefore, there is an urgent need for an active rotor aerodynamic noise control technology that adapts to tilt transition conditions and targets specific areas, effectively suppressing noise in critical areas while ensuring flight and actuator feasibility.

[0003] Existing technologies mainly focus on two areas: active control of rotor aerodynamic noise and inversion control load solution based on acoustic models. Their typical implementation ideas are as follows: 1) Noise active control scheme based on blade active device.

[0004] This approach typically involves arranging active control units (such as controllable flaps or trailing edge controllable devices) on the rotor blades. By modulating the local aerodynamic loads on the blades, a certain degree of noise reduction is achieved at the target observation position. This type of approach generally requires outputting periodic or segmented control commands during rotor rotation based on pre-set control laws or parameters to change the amplitude or phase characteristics of rotor noise.

[0005] 2) Single-point inversion solution of control load scheme based on acoustic model.

[0006] In this approach, an aeroacoustic model is often used to establish the relationship between blade load and acoustic response at the observation point. The acoustic parameters at a target observation point are then used as the control objective, and the time history of the required control load (or equivalent control force) is obtained through inversion. This type of method generally aims to minimize the sound pressure at a single point, obtains a set of control load solutions, and then maps these control loads into actuator control commands and applies them to the blades.

[0007] 3) Noise control or operation optimization solutions for tilt rotors.

[0008] To address the noise problem of tiltrotors, existing solutions tend to employ passive structural noise reduction (such as shape optimization and blade / blade tip shape treatment) or operational strategy adjustments (such as speed / tilt law optimization and trajectory planning) to reduce noise radiation in specific scenarios. However, such solutions typically struggle to suppress rapidly changing unsteady noise in real time during the tilt transition phase and often cannot provide precise control over ground target areas.

[0009] In summary, existing technologies have the foundation for "achieving noise reduction through active blade load modulation" and "solving control loads based on acoustic model inversion." However, significant shortcomings remain under tilt transition conditions and target area constraints. For example, while existing technologies can achieve a certain degree of noise reduction through active blade load modulation, their core problem lies in the fact that the active noise control targets of existing technologies are usually based on a single observation point or fixed spatial location, lacking noise control capabilities oriented towards the target area.

[0010] Because rotor noise exhibits significant directionality and regional variations in space, especially during the transitional flight phase of tilt-rotor aircraft, the direction and intensity of noise radiation change rapidly with the tilt angle and flight state. Minimizing noise at a single point cannot effectively represent the overall noise level within sensitive ground or space areas. Therefore, even if existing technologies achieve noise reduction at a specific observation point, it may lead to increased noise at other locations within the target area, making it difficult to meet the requirements for regional noise constraints in actual operational scenarios.

[0011] The aforementioned problems constitute the most significant drawback of existing technologies when applied to the transitional flight phase of tiltrotor aircraft. In addition to these major drawbacks, existing technologies also raise the following secondary issues: (1) The control strategy is not adaptable to changes in flight conditions.

[0012] Because the control objective is focused on a single observation point, existing control strategies are often designed or tuned for specific operating conditions. When the tilt angle, flight speed, and inflow conditions change, the single-point control objective cannot reflect the overall changes in the spatial distribution of noise, resulting in unstable control performance during the tilt transition phase.

[0013] (2) The engineering feasibility of the control load solution is insufficient.

[0014] In the inversion process aimed at minimizing single-point noise, the obtained control load may be overly biased towards the acoustic requirements of the observation point without fully considering the physical constraints of the actuator and the regional noise uniformity, resulting in excessively large or drastic control load amplitudes, which increases the difficulty of engineering implementation.

[0015] (3) It is difficult to meet the noise control requirements under multiple regions or multiple constraints.

[0016] In actual operation, noise constraints often arise from multiple sensitive areas or various regulatory requirements. Existing control methods based on single-point targets are difficult to extend through a unified framework and lack the ability to coordinate regional noise control.

[0017] It should be noted that the aforementioned minor drawbacks all stem from the fundamental problem that existing technologies have failed to extend noise control targets from a single point to a target area. Summary of the Invention

[0018] To address the main drawbacks and derived problems caused by single-point noise control, this application provides a method, device, medium, and product for active aerodynamic noise control in the target area during tiltrotor transition flight.

[0019] To achieve the above objectives, this application provides the following solution: In a first aspect, this application provides an active aerodynamic noise control method for the target area during tiltrotor transition flight, including: Acquire key state parameters of the rotorcraft during flight, and determine the flight condition type based on the key state parameters; When the determined flight condition type is tilt transition condition, a target noise control area is constructed in a set area according to the flight mission and environmental requirements, and the target noise control area is discretized into multiple spatial observation points to form an observation point set, while assigning weights to different spatial observation points. The controllable aerodynamic loads on the blades are parameterized, and the aerodynamic load changes generated by the active control unit in the rotorcraft are uniformly represented as control load variables in order to establish the relationship between control load variables and acoustic response of the target noise control area, and to determine the unknowns. A global acoustic optimization objective function is established based on the set of observation points, the assigned weights, and the unknowns. Using the DDQN algorithm in deep reinforcement learning as a solver, the global acoustic optimization objective function is optimized and solved under the premise of satisfying the physical constraints of the actuator, so as to obtain the optimal control law; Feedforward execution and feedback correction are performed based on the optimal control law to achieve active control of aerodynamic noise in the target area.

[0020] Secondly, this application provides an active aerodynamic noise control device for target areas during tiltrotor transition flight, comprising: The flight status and tilt condition acquisition module is used to acquire key status parameters of the rotorcraft during flight and determine the flight condition type based on the key status parameters. The target region construction and weight allocation module is used to construct a target noise control region within a set area according to the flight mission and environmental requirements when the determined flight condition type is tilt transition condition, and to discretize the target noise control region into multiple spatial observation points to form an observation point set, while assigning weights to different spatial observation points. The load modulation parameterization module is used to parameterize the controllable aerodynamic loads on the blades, and to uniformly represent the aerodynamic load changes generated by the active control unit in the rotorcraft as control load variables, so as to establish the relationship between the control load variables and the acoustic response of the target noise control area, and to determine the unknowns. The acoustic mapping and inversion calculation module is used to establish a global acoustic optimization objective function based on the set of observation points, the assigned weights, and the unknowns. The optimization and control law generation module uses the DDQN algorithm in deep reinforcement learning as a solver to optimize the global acoustic optimization objective function and obtain the optimal control law under the premise of satisfying the physical constraints of the actuator. The actuator drive and feedback update module is used to perform feedforward execution and feedback correction based on the optimal control law to achieve active control of aerodynamic noise in the target area.

[0021] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described active aerodynamic noise control method for target area during tiltrotor transition flight.

[0022] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described active aerodynamic noise control method for target area during tiltrotor transition flight.

[0023] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described active aerodynamic noise control method for target areas during tiltrotor transition flight.

[0024] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a method, device, medium, and product for active aerodynamic noise control in target areas during tiltrotor transition flight. It abandons generalized global noise reduction or single reference point noise reduction methods, employing a combination of spatial discretization and sensitivity weighting to discretize the target noise control area into spatial observation points with different weights. This expands the control target from a single point to a regional constraint, enabling on-demand noise reduction and addressing the fundamental problem of existing technologies failing to extend the noise control target from a single point to the target area. By establishing a global acoustic optimization objective function based on the observation point set, assigned weights, and unknowns, and utilizing the DDQN algorithm from deep reinforcement learning as the solver, the global acoustic optimization objective function is optimized and solved under the premise of satisfying the actuator's physical constraints. This process forms a closed-loop computational link of parameterized load definition, acoustic model driving, and inversion solution, improving control stability. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 A flowchart illustrating an active aerodynamic noise control method for target area during tiltrotor transition flight, provided in an embodiment of this application. Figure 2 A schematic diagram of control load variables obtained under transitional operating conditions provided in an embodiment of this application; Figure 3 A schematic diagram of a control law provided for an embodiment of this application; Figure 4 This is a schematic diagram illustrating the noise reduction effect provided in an embodiment of this application; Figure 5 A schematic diagram of the functional modules of an active aerodynamic noise control device for target area during tiltrotor transition flight, provided in an embodiment of this application; Figure 6 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0027] 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, and 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.

[0028] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0029] The terms used in this application are explained as follows: (1) Tilting rotor / tilt transition stage: refers to the stage in which the rotor axis tilts continuously relative to the airframe / ground coordinate system and the flight mode changes from helicopter to fixed wing (or vice versa).

[0030] (2) Tilting angle: refers to the angle between the axis of the tilt rotor and the aircraft body or the ground reference coordinate system.

[0031] (3) Target area / sensitive area: refers to the spatial area where noise needs to be controlled (such as ground residential areas, the area around the take-off and landing point, and specific corridors), which can be represented by a set of discrete observation points.

[0032] (4) Sound pressure time history: The sound pressure signal at the observation point in the target area changes over time, which can be measured by the microphone or predicted based on the simulation model.

[0033] (5) Acoustic inversion / inverse problem: Based on the acoustic model, establish a mapping relationship between the sound pressure in the target area and the blade load disturbance, and solve the load modulation amount that satisfies the noise reduction target.

[0034] (6) Load modulation / equivalent control load: The process of changing the local aerodynamic load (lift / normal force, etc.) of the blade through an active device; in the model layer, it can be equivalent to applying control force / load distribution at a specific position of the blade.

[0035] (7) Active control unit (actuator): a device used to achieve load modulation, which may include, but is not limited to, active flaps, controllable Gurney flaps, piezoelectric trailing edge deformation, microjets, etc.

[0036] (8) Control Law / Solver: refers to the algorithm module that calculates the actuator control command based on the noise index of the target area (which can be linear least squares, constrained optimization, adaptive update, etc.).

[0037] In one exemplary embodiment, this application provides an active aerodynamic noise control method for target areas during tiltrotor transition flight. This method is executed by a computer device, specifically a terminal or server, or both. In this embodiment, the method is described using a server as an example. Figure 1 As shown, the method includes: Step 100: Obtain key state parameters of the rotorcraft during flight and determine the flight condition type based on the key state parameters; Step 101: When the determined flight condition type is tilt transition condition, construct the target noise control area within the set area according to the flight mission and environmental requirements, and discretize the target noise control area into multiple spatial observation points to form an observation point set, while assigning weights to different spatial observation points. Step 102: Parametrically describe the controllable aerodynamic loads on the blades, and uniformly represent the aerodynamic load changes generated by the active control unit in the rotorcraft as control load variables, so as to establish the relationship between the control load variables and the acoustic response of the target noise control area, and determine the unknowns. Step 103: Establish a global acoustic optimization objective function based on the observation point set, assigned weights, and unknowns; Step 104: Using the DDQN algorithm in deep reinforcement learning as a solver, the global acoustic optimization objective function is optimized and solved under the premise of satisfying the physical constraints of the actuator to obtain the optimal control law; Step 105: Perform feedforward execution and feedback correction based on the optimal control law to achieve active control of aerodynamic noise in the target area.

[0038] In an exemplary embodiment of this application, to address the problem that existing control strategies are often designed or tuned for specific operating conditions because the control target is concentrated at a single observation point, and when tilt angle, flight speed, and inflow conditions change, the single-point control target cannot reflect the overall change in the spatial distribution of noise, resulting in unstable control performance during the tilt transition phase, this embodiment combines flight state and tilt condition identification to enable the active noise reduction process to be dynamically started and updated during the tilt transition phase, thereby improving the adaptability of the control strategy to unsteady operating conditions. Based on this, in this embodiment, the key state parameters include at least one or more of the following: 1) Tilting angle and its rate of change: can be obtained by the angle sensor, rotary encoder or position feedback unit of the tilting mechanism; 2) Rotor speed: can be obtained from the feedback signal of the rotor shaft speed sensor or the motor controller; 3) Propeller performance parameters (collective pitch or equivalent pitch parameters): can be obtained from the feedback signal of the pitch actuator in the rotor control system; 4) Flight speed and attitude parameters: can be obtained by inertial measurement unit (IMU), airspeed indicator, global navigation satellite system (GNSS) or a combination thereof.

[0039] Based on the key state parameters obtained above, the implementation process of step 100 provided in this application can be described as follows: When the tilt angle is in the continuous variation range between the unstable helicopter mode and the unstable fixed wing mode, and the rate of change of the tilt angle shows a monotonic trend in multiple consecutive control cycles, the flight condition type is determined to be the tilt transition condition; under this condition, the target area aerodynamic noise active control process implemented in steps 101-105 above is initiated.

[0040] Furthermore, in other embodiments, the flight condition type can be determined by comprehensively judging the flight mode switching command of the flight control system in the rotorcraft, or by the rotor speed, rotor rotation speed, flight speed and attitude parameters and propeller performance parameters.

[0041] In an exemplary embodiment of this application, to address the problem that in the inversion process aimed at minimizing single-point noise in existing technologies, the obtained control load may be overly biased towards the acoustic requirements of that observation point, without fully considering the physical constraints of the actuator and the regional noise balance, resulting in excessively large or drastic control load amplitudes and increased engineering implementation difficulty, a target noise control region can be constructed. By introducing multiple observation points and a weight allocation mechanism, the active control target for rotor noise is expanded from a single observation point to the overall constraint of the target region, thereby achieving effective suppression of noise in sensitive areas during the tilt transition flight phase. Based on this, the implementation process of step 101 can be described as follows: a target noise control region is constructed around the rotorcraft or on the ground, and this region is discretized into multiple spatial observation points to form an observation point set; simultaneously, weights are assigned to different observation points to reflect the sensitivity of different regions to noise control.

[0042] The target noise control area can be pre-defined as a spatial region located around the rotorcraft or at a designated location on the ground, based on flight mission requirements. This area is used to characterize sensitive areas where noise control is crucial. In some implementations, the target noise control area can also be dynamically adjusted according to the current flight attitude and acoustic propagation characteristics to cover the main radiation direction of rotor noise or areas where sound energy easily converges, such as selecting areas like school zones or areas directly in front of the tiltrotor aircraft's flight direction for noise control.

[0043] The process of dynamically adjusting based on the current flight attitude and acoustic propagation characteristics includes: Step (1): Based on the flight mission and environmental requirements, acquire the acoustic propagation characteristics in real time, and combine key state parameters to determine the radiation direction of the current rotor noise and the area where sound energy easily converges. Step (2): Based on the radiation direction of the current rotor noise and the area where sound energy easily converges, redetermine the spatial range of the target noise control area and update the corresponding observation point set and weight allocation.

[0044] Since the target noise control area is usually a continuous region in space, while active control calculations need to be completed within a finite dimension, the target noise control area is discretized into multiple spatial observation points. The noise characteristics of the region are represented by a set of observation points, thereby reducing computational complexity while ensuring control accuracy. The spatial observation points can be selected using a uniform distribution, a non-uniform distribution, or a denser distribution based on acoustic propagation characteristics.

[0045] To reflect the varying sensitivity of different target noise control areas to noise control, corresponding weights are assigned to each spatial observation point in the observation point set. These weights can be determined based on the flight site policy, regional sensitivity level, spatial location, or acoustic response characteristics, and can be updated as flight conditions change. Whenever a key state parameter (tilt angle, flight speed, attitude) changes beyond a preset threshold, a weight update is triggered: ① The contribution of each spatial observation point to noise in the target noise control area is reassessed based on the latest acoustic propagation characteristics; ② Observation points in highly sensitive areas are assigned greater weights, considering relevant flight site policies and regulations or regional sensitivity levels; ③ The updated weight vector is passed to subsequent processes, ensuring that the control strategy matches the current flight conditions in real time.

[0046] In an exemplary embodiment of this application, in order to address the problem that even if noise reduction is achieved at a certain observation point, noise may still increase at other locations within the target area, making it difficult to meet the requirements for regional noise constraints in actual operating scenarios, the implementation process of step 102 provided above in this application can be described as follows: Step 1: Within one rotation cycle, express the controllable aerodynamic load at a radial position on the blade as a function of time, and uniformly express the aerodynamic load changes generated by the active control unit in the rotorcraft as control load variables; whereby the control load variables are expressed as... , T Indicates the rotation period.

[0047] Controlling the load variables to satisfy periodic constraints: The periodic control load variable, as a direct unknown in the F-1A inverse problem, is used to describe the equivalent load modulation (i.e., aerodynamic load variation) applied by the active control unit within one rotational cycle.

[0048] Step 2: Under fixed geometric configuration and spatial observation points, the relationship between the control load variables and the acoustic response of the target noise control region is expressed as a first-order linear inverse problem based on the acoustic model and a function of time; wherein, the first-order linear inverse problem is expressed as: .

[0049] In the formula, p ( t) represents a function term related to the controllable aerodynamic load and its time variation. r ( t ) represents the driving terms related to the function terms and rotor motion parameters.

[0050] Step 3: Based on the first-order linear inverse problem, the F-1A equation in the acoustic analogy method is used to express the acoustic control requirements at the space observation point as a function term related to the controllable aerodynamic load and its time variation through the equivalent transformation in the aeroacoustic model; where this function term is introduced as a known coefficient term in the load inverse solution process.

[0051] Step 4: After structuring the acoustic relations into an inverse problem solution form, we can further construct driving terms related to the function terms and rotor motion parameters; the construction form of these driving terms is as follows: .in, It is the sound pressure time history at the target observation point, which can be obtained in two ways. In the simulation scheme, it can be analyzed by the aeroacoustic solution module of CLORNS software; in the measurement scheme, it can be measured by the microphone placed at the observation point. The physical meaning and solution method of the terms can be obtained from the F-1A equation.

[0052] Step 5: Establish the inverse solution relationship between the control load variable and the driving term, and treat the control load variable as an unknown.

[0053] Under the above conditions, by establishing control load variables y ( t ) and driving items r ( t By solving the inverse relationship between the two, the control load varying with time over one rotation cycle is obtained, ensuring that the sound pressure level at the space observation point meets the expected noise suppression requirements after the control load is applied. Under the above conditions, the control load variable at a specific radial position of the blade is... y ( t As an unknown quantity, it is solved by the driving term. r ( t This involves solving a related first-order linear inverse problem to obtain a time-varying control load over one rotation cycle, ensuring that the control load variables acoustically satisfy the noise suppression requirements of the target observation point. For example, under transient conditions, a defined control load variable is applied at different blade positions, such as... Figure 2 As shown.

[0054] In an exemplary embodiment of this application, to address the issue that noise constraints in actual operation often originate from multiple sensitive areas or various regulatory requirements, and the existing single-point target-based control methods are difficult to extend through a unified framework and lack the ability to coordinate regional noise control, the engineering feasibility of control load modulation commands can be improved by introducing actuator physical constraints during the control load inversion and optimization process. Based on this, the implementation process of steps 103-104 above can be described as follows: based on the constructed observation point set and weights, a global acoustic optimization objective function is established, and the DDQN algorithm in deep reinforcement learning is used as the solver to intelligently optimize and generate the optimal control law (i.e., load modulation command) while satisfying the actuator physical constraints.

[0055] To achieve effective noise reduction across the entire target noise control area, rather than being limited to a single point, a global acoustic optimization objective function incorporating spatial weighting information is first constructed. Weighted sound pressure level calculation: The set of observation points received from the output... and its corresponding weight vector The predicted sound pressure levels at each spatial observation point are calculated using an aeroacoustic model, and the weighted total sound pressure level is calculated by combining the weights.

[0056] The constructed global acoustic optimization objective function is expressed as: .

[0057] Among them, the second item The stationarity penalty term is used to limit the range of change in the control load and prevent sudden changes in the control quantity. This represents the update increment of the control load parameters (i.e., control load variables) relative to the previous time step. In the DDQN framework, this increment is directly determined by the action commands output by the neural network. This is achieved through constraints... This avoids the drastic load fluctuations that are physically impossible to achieve in the pursuit of noise minimization. It is a pre-set or dynamically adjusted weighting factor. The value of this factor determines the tendency of the optimization strategy: The larger the value, the more it tends to output a smooth and conservative control law; The smaller the value, the more aggressive the noise reduction strategy tends to be. In practical applications, The specific values ​​can be obtained through ground testing.

[0058] Given the highly nonlinear aerodynamic environment during the tilt transition, the DDQN algorithm is used instead of the traditional gradient descent method to achieve a more efficient and robust inversion solution. The solution process is modeled as a Markov decision process (MDP): State space Define the current environmental state vector, with input features including the current tilt angle. Tilting angle change rate The residual noise level at the current observation point and the control action at the previous time.

[0059] Action space ): Defines the action output of the agent, corresponding to the control load variable. The adjustment amount (such as the increment of Fourier coefficients or the correction value of the discrete sequence). In order to adapt to the discrete action characteristics of DDQN, the continuous load parameter space is discretized into a finite set of action commands.

[0060] Reward function Design and Global Objective Function Negative correlation reward mechanism (i.e.) When the weighted total noise decreases, the agent is given a positive reward; when physical constraints are violated or noise increases, a penalty is imposed.

[0061] DDQN Network Architecture and Updates: A dual-network structure is constructed, consisting of a current evaluation network and a target network. The current network selects actions (decoupling action selection), while the target network evaluates the value of the actions (decoupling value evaluation), thereby eliminating suboptimal solutions caused by Q-value overestimation. An experience replay mechanism is introduced to store historical flight data or offline simulation data. The neural network is trained through random sampling to learn the optimal load-noise reduction mapping strategy under different operating conditions.

[0062] After the DDQN outputs the optimal action suggestion, constraint verification and signal transformation are required: hard constraint logic is added after the neural network output layer. If the control load variable output by the network causes the actuator command to exceed the amplitude limit (Saturation) or rate limit (Rate Limit), it is forcibly pruned to within the allowable boundary to ensure the physical feasibility of the output command. Finally, the constrained control load variable is reconstructed into a control waveform in the time domain and packaged into an optimal control law. The solved optimal control law is as follows: Figure 3 As shown.

[0063] In an exemplary embodiment of this application, since the control command is in the form of aerodynamic load, while the actuators, such as blade trailing edge winglets and multi-degree-of-freedom drag plates, receive displacement / voltage signals, the implementation process in step 106 above can be described as follows: the optimal control law is input as the feedforward basic command into the feedforward execution link, and the actuator is driven after multi-level mapping; at the same time, the feedback correction link generates a compensation amount according to the actual acoustic residual, and the compensation amount is superimposed on the feedforward basic command to achieve active control of aerodynamic noise in the target area.

[0064] The process of signal conversion through multi-level mapping includes: (1) First-level mapping (aerodynamic-geometry): Based on the airfoil aerodynamic database or CFD calibration table, the optimal control law (i.e. the optimized aerodynamic load command) is converted into the geometric motion quantity of the actuator; (2) Second-level mapping (geometric-electrical signal): Based on the transmission ratio and servo characteristics of the actuator, the geometric motion quantity is converted into the underlying electrical drive signal. Before output, the drive signal is subjected to D / A conversion and low-pass filtering to filter out high-frequency noise interference and ensure smooth operation.

[0065] To address the nonlinear errors in the aerodynamic model and the abrupt changes in the flight environment, a prediction-correction mechanism can be introduced into the feedback correction link, as follows: Real-time acquisition of actual sound pressure signals from microphones in the target noise control area and the expected target sound pressure Compare and calculate acoustic residuals Adaptive filtering algorithms (such as the LMS algorithm) or PID correction logic are employed, based on the acoustic residuals. Size and sign, generating compensation amount .

[0066] compensation amount Feedforward control commands superimposed on the next cycle middle( ,in To update the step size, new control commands are obtained. This eliminates model errors and maintains the long-term stability of the noise reduction effect.

[0067] Furthermore, when the actuator temperature is detected to be too high, the current is abnormal, or the vibration exceeds the limit, the control bypass mode can be forcibly triggered to reset the control command to zero or lock it in a safe position, thus prioritizing flight safety.

[0068] Based on the above description, this application achieves stable and feasible active noise control of tiltrotors under complex flight conditions by constructing a closed-loop control link of target area noise—blade load modulation—actuator control. The noise reduction effect of this application is as follows: Figure 4 As shown.

[0069] Based on the same inventive concept, this application also provides an active aerodynamic noise control device for the target area of ​​tiltrotor transition flight, used to implement the above-described active aerodynamic noise control method for the target area of ​​tiltrotor transition flight. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the active aerodynamic noise control device for the target area of ​​tiltrotor transition flight provided below can be found in the limitations of the active aerodynamic noise control method for the target area of ​​tiltrotor transition flight described above, and will not be repeated here.

[0070] In one exemplary embodiment, such as Figure 5 As shown, an active aerodynamic noise control device for target area during tilt rotor transition flight is provided, comprising: a flight state and tilt condition acquisition module, a target area construction and weight allocation module, a load modulation parameterization module, an acoustic mapping and inversion calculation module, an optimization solution and control law generation module, and an actuator driving and feedback update module.

[0071] The flight status and tilt condition acquisition module is used to acquire key status parameters of the rotorcraft during flight and determine the flight condition type based on the key status parameters. The target area construction and weight allocation module is used to construct a target noise control area within a set area according to the flight mission and environmental requirements when the determined flight condition type is tilt transition condition. The target noise control area is then discretized into multiple spatial observation points to form an observation point set, and weights are assigned to different spatial observation points. The load modulation parameterization module is used to parameterize the controllable aerodynamic loads on the blades, and uniformly represent the aerodynamic load changes generated by the active control unit in the rotorcraft as control load variables, so as to establish the relationship between the control load variables and the acoustic response of the target noise control area, and determine the unknowns. The acoustic mapping and inversion calculation module is used to establish a global acoustic optimization objective function based on the observation point set, assigned weights, and unknowns; The optimization and control law generation module uses the DDQN algorithm in deep reinforcement learning as a solver to optimize the global acoustic optimization objective function under the premise of satisfying the physical constraints of the actuator, and obtain the optimal control law. The actuator drive and feedback update module is used to perform feedforward execution and feedback correction based on the optimal control law to achieve active control of aerodynamic noise in the target area.

[0072] Based on the above description, compared to existing technologies, this application actually provides an on-demand active noise reduction framework for tiltrotor aircraft. Its core lies in: using the tilt angle to determine the timing, using a weighted point set to quantify the noise reduction demand, and using an acoustic model inversion to calculate the optimal blade control force, thereby achieving precise noise management during complex transitional flight processes. Compared to existing technologies, this application mainly has the following differences and advantages: 1. Dynamic triggering and scheduling logic for tilting transition conditions.

[0073] Unlike the all-time or manual control of traditional helicopters or fixed-wing aircraft, this application establishes an automatic triggering mechanism with tilt angle and its rate of change as the core, specifically for the transitional state unique to tilt rotor aircraft.

[0074] This application uses transition condition identification as the trigger / update condition for the control process and allows the target area and weights to be adjusted according to attitude / propagation characteristics. Therefore, when the tilt angle and noise radiation direction change rapidly, the control target can still remain consistent with the actual constraints, resulting in a more stable control effect and better adaptability to the unsteady changes in the tilt transition phase.

[0075] 2. A method for constructing a target field using a discrete set of observation points and a dynamic weight matrix.

[0076] This application abandons general global denoising or single reference point denoising, and proposes a modeling method that combines spatial discretization and sensitivity weighting. It transforms physical spatial locations into mathematically optimized weights. The key protection weighting mechanism allows different weights to be assigned to different points based on regulatory requirements or physical characteristics. This is a crucial means of achieving on-demand denoising.

[0077] Because this application defines the target noise control object as the target region-observation point set-weight, and the control target is expanded from a single point to a region-wide constraint, it can avoid the problem of "the noise decreases at a single point but increases elsewhere in the region" and achieve collaborative noise reduction at the target region level.

[0078] 3. A closed-loop computational chain of parameterized load definition, acoustic model-driven inversion solution.

[0079] This is the core of the algorithm in this application. The innovation lies in doing the opposite (inverse problem solving): instead of calculating the noise for the load first, a noise target is set first, and the required control load variables are deduced by using an acoustic analogy model (such as F-1A) as the transfer function. y ( tThe key protection of the solution path is: defining the unknown (load) - introducing the transfer function (acoustic model) - matching the target quantity (weighted sound pressure) - inversely solving the unknown, which is a complete mathematical and physical process.

[0080] Furthermore, because this application explicitly introduces physical constraints such as actuator amplitude, rate of change, and working range during the control law generation stage, and outputs drive signals through load-actuator mapping, the solution results are more easily executed by actual actuators, reducing the risk of "calculating but not being able to do it", and making the project more feasible.

[0081] Furthermore, based on the above description of this application, the actuator mentioned in this application can be replaced by other actuators. In addition to multi-degree-of-freedom drag plates, there are other actuation methods such as active flaps, controllable Gurney flaps, and piezoelectric trailing edges that can achieve force application.

[0082] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 6 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores active aerodynamic noise control data for target areas during tiltrotor transition flight. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements an active aerodynamic noise control method for target areas during tiltrotor transition flight.

[0083] Those skilled in the art will understand that Figure 6 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer equipment to which the present application is applied. Specific computer equipment may include, for example, [the following is a list of possible additional structures]. Figure 6 The diagram shows more or fewer components, or combinations of certain components, or different component arrangements.

[0084] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0085] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0086] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0087] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0088] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (RRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0089] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0090] The technical features of the above embodiments can be combined in any way. For the sake of brevity, 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.

[0091] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for active control of aerodynamic noise in the target area during tiltrotor transition flight, characterized in that, include: Acquire key state parameters of the rotorcraft during flight, and determine the flight condition type based on the key state parameters; When the determined flight condition type is tilt transition condition, a target noise control area is constructed in a set area according to the flight mission and environmental requirements, and the target noise control area is discretized into multiple spatial observation points to form an observation point set, while assigning weights to different spatial observation points. The controllable aerodynamic loads on the blades are parameterized, and the aerodynamic load changes generated by the active control unit in the rotorcraft are uniformly represented as control load variables in order to establish the relationship between control load variables and acoustic response of the target noise control area, and to determine the unknowns. A global acoustic optimization objective function is established based on the set of observation points, the assigned weights, and the unknowns. Using the DDQN algorithm in deep reinforcement learning as a solver, the global acoustic optimization objective function is optimized and solved under the premise of satisfying the physical constraints of the actuator, so as to obtain the optimal control law; Based on the optimal control law, feedforward execution and feedback correction are performed to achieve active control of aerodynamic noise in the target area.

2. The active aerodynamic noise control method for the target area during tiltrotor transition flight according to claim 1, characterized in that, Key state parameters of a rotorcraft during flight include one or more of the following: tilt angle and rate of change of tilt angle, rotor speed, flight speed and attitude parameters, and propeller performance parameters. Propeller performance parameters are collective pitch or equivalent pitch parameters; The flight condition type is determined based on the key state parameters, including: When the tilt angle is in the continuous variation range between the unstable helicopter mode and the unstable fixed wing mode, and the rate of change of the tilt angle shows a monotonic variation trend in multiple consecutive control cycles, the flight condition type is determined to be a tilt transition condition. Alternatively, the flight mode type can be determined by comprehensively judging the flight mode switching command of the flight control system in the rotorcraft, combined with rotor speed, rotor rotation speed, flight speed and attitude parameters and propeller performance parameters.

3. The active aerodynamic noise control method for the target area of ​​tiltrotor transition flight according to claim 1, characterized in that, Based on flight mission and environmental requirements, a target noise control region is constructed within a designated area. This target noise control region is then discretized into multiple spatial observation points, forming an observation point set. Weights are assigned to different spatial observation points, including: Based on flight mission and environmental requirements, the acoustic propagation characteristics are acquired in real time, and the radiation direction of the current rotor noise and the area where sound energy easily converges are determined in combination with the key state parameters. Based on the radiation direction of the current rotor noise and the area where sound energy easily converges, the spatial range of the target noise control area is redefined, and the corresponding set of observation points and weight allocation are updated.

4. The active aerodynamic noise control method for the target area of ​​tiltrotor transition flight according to claim 1, characterized in that, The controllable aerodynamic loads on the rotor blades are parameterized, and the changes in aerodynamic loads generated by the active control unit in the rotorcraft are uniformly represented as control load variables. This establishes the relationship between the control load variables and the acoustic response of the target noise control region, and determines the unknowns, including: Within one rotation cycle, the controllable aerodynamic load at a radial position on the blade is expressed as a function of time, and the aerodynamic load changes generated by the active control unit in the rotorcraft are uniformly expressed as control load variables; Under fixed geometric configuration and spatial observation point conditions, the relationship between the control load variables and the acoustic response of the target noise control area is expressed as a first-order linear inverse problem based on the acoustic model and a function of time. Based on the aforementioned first-order linear inverse problem, the F-1A equation in the acoustic analogy method is used to express the acoustic control requirements at the space observation point as a function term related to the controllable aerodynamic load and its time variation through the equivalent transformation in the aeroacoustic model. Construct driving terms related to the function terms and rotor motion parameters; Establish the inverse solution relationship between the control load variable and the driving term, and treat the control load variable as the unknown quantity.

5. The active aerodynamic noise control method for the target area of ​​tiltrotor transition flight according to claim 1, characterized in that, Using the DDQN algorithm from deep reinforcement learning as the solver, the global acoustic optimization objective function is optimized under the premise of satisfying the physical constraints of the actuator, to obtain the optimal control law, including: In the solver, a dual-network structure of a current evaluation network and a target network is constructed. The current evaluation network is used to select actions, and the target network is used to evaluate the value of the actions. An experience playback mechanism is introduced to store historical flight data or offline simulation data. The dual-network structure is trained by random sampling so that it learns the optimal load-noise reduction mapping strategy under different working conditions. After obtaining the optimal control load variable based on the optimal load-noise reduction mapping strategy, hard constraint logic is added after the output layer of the dual network structure. If the output control load variable causes the actuator command to exceed the amplitude limit or the rate of change limit, the control load variable is forcibly clipped to the allowable boundary. The trimmed control load variables are reconstructed into control waveforms in the time domain to obtain the optimal control law.

6. The active aerodynamic noise control method for the target area of ​​tiltrotor transition flight according to claim 1, characterized in that, Based on the optimal control law, feedforward execution and feedback correction are performed to achieve active control of aerodynamic noise in the target area, including: The optimal control law is used as the feedforward basic instruction and input into the feedforward execution link. After multi-level mapping, it drives the actuator. At the same time, the feedback correction link generates a compensation amount based on the actual acoustic residual and adds the compensation amount to the feedforward basic instruction to achieve active control of aerodynamic noise in the target area.

7. An active aerodynamic noise control device for target areas during tiltrotor transition flight, characterized in that, include: The flight status and tilt condition acquisition module is used to acquire key status parameters of the rotorcraft during flight and determine the flight condition type based on the key status parameters. The target region construction and weight allocation module is used to construct a target noise control region within a set area according to the flight mission and environmental requirements when the determined flight condition type is tilt transition condition, and to discretize the target noise control region into multiple spatial observation points to form an observation point set, while assigning weights to different spatial observation points. The load modulation parameterization module is used to parameterize the controllable aerodynamic loads on the blades, and to uniformly represent the aerodynamic load changes generated by the active control unit in the rotorcraft as control load variables, so as to establish the relationship between the control load variables and the acoustic response of the target noise control area, and to determine the unknowns. The acoustic mapping and inversion calculation module is used to establish a global acoustic optimization objective function based on the set of observation points, the assigned weights, and the unknowns. The optimization and control law generation module uses the DDQN algorithm in deep reinforcement learning as a solver to optimize the global acoustic optimization objective function and obtain the optimal control law under the premise of satisfying the physical constraints of the actuator. The actuator drive and feedback update module is used to perform feedforward execution and feedback correction based on the optimal control law to achieve active control of aerodynamic noise in the target area.

8. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that the processor executes the computer program to implement the active aerodynamic noise control method for target area transition flight of tiltrotor as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the active aerodynamic noise control method for target area of ​​tiltrotor transition flight as described in any one of claims 1-6.

10. A computer program product, comprising a computer program, characterized in that, When executed by a processor, the computer program implements the active aerodynamic noise control method for target area of ​​tiltrotor transition flight as described in any one of claims 1-6.

Images