Lightweight embedded rotor active noise reduction integrated protection device

Abstract

This invention discloses a lightweight embedded rotor active noise reduction integrated protection device, belonging to the field of aircraft structural design and acoustic control hardware integration technology. It includes a ring-shaped rotor protective shell, a speaker array module, a motor bracket, and a distributed intelligent control module. The speaker array module is mounted inside the ring-shaped rotor protective shell to form a quiet zone. The motor bracket is arranged radially along the ring-shaped rotor protective shell. An error sensor array module, a reference sensor module, and a multi-source sensing module are installed at the bottom of the motor bracket. The distributed intelligent control module includes a central master controller and local slave controllers, used to generate a global noise reduction control strategy and drive the speaker array module, respectively. The distributed intelligent control module incorporates a geometric estimation algorithm to quickly deduce acoustic path parameters based on a spatial geometric model. Using this device, the integration of a multi-rotor low-altitude aircraft and an active noise reduction system is achieved, reducing the computational load of the algorithm and improving the noise reduction adaptability and acoustic airworthiness under all operating conditions.

Description

Technical Field

[0001] This invention relates to the field of aircraft structural design and acoustic control hardware integration technology, and in particular to a lightweight embedded rotor active noise reduction integrated protection device. Background Technology

[0002] Multirotor aircraft are widely used in logistics, power line inspection, and urban security. However, the intense aerodynamic noise generated by the rotors has become a core bottleneck restricting their large-scale commercialization and airworthiness certification. Noise reduction can be broadly categorized into two technical approaches: passive noise reduction and active noise control. Passive noise reduction, the current mainstream technology, primarily uses aerodynamic design techniques such as rotor blade airfoil optimization, tip structure modification, and reduced rotor operating speed to weaken the aerodynamic noise radiation generated by rotor rotation at its source. Active noise control (ANC), based on the core principle of destructive interference of sound waves, employs a closed-loop control logic—using microphones to collect raw noise signals, controllers to generate inverse control signals, and loudspeakers to output signals that cancel out the sound waves—to effectively address low-frequency noise that passive noise reduction methods struggle to suppress. ANC is currently a core research and development direction in the field of noise reduction for low-altitude aircraft.

[0003] Currently, the practical application of traditional active noise control technology on multirotor low-altitude aircraft faces multiple challenges. First, the open free-field environment has a complex sound field with no fixed echo path, making it difficult for secondary sound source layouts to meet noise reduction causality requirements, resulting in extremely low sound field coverage and noise reduction efficiency in open spaces. Second, the computational load of traditional centralized multi-channel active noise reduction algorithms increases quadratically with the number of channels, making real-time operation on low-power embedded platforms on aircraft difficult, highlighting a significant computational bottleneck. Third, the drastic changes in motor speed, flight attitude, and flight conditions during multirotor operation cause insufficient tracking response speed in traditional algorithms, easily leading to phase mismatch problems and even the counterproductive effect of noise amplification. Fourth, traditional active noise reduction systems often employ external mounting layouts, which significantly increase the computational load of the aircraft. The flight load and wind resistance of the aircraft disrupt its original aerodynamic layout. There are also problems such as limited installation space, complex wiring, and insufficient protection and reliability of acoustic components. At the same time, conventional active noise cancellation schemes rely on a single acoustic signal to achieve closed-loop control, which is a passive "reactive" noise cancellation that cannot adapt to the dynamic changes in the flight environment and operating conditions. The noise cancellation stability and robustness are extremely poor throughout the entire flight envelope. Fifth, the secondary acoustic path modeling of traditional active noise cancellation systems relies heavily on complex online identification or time-consuming offline measurement. In the complex open sound field and severe vibration environment of multi-rotor aircraft, model mismatch is very likely to occur, resulting in slow algorithm convergence or even divergence. At the same time, there is a lack of fast acoustic estimation methods based on the real geometric relationship of physical space, which seriously restricts the real-time response capability of the noise cancellation system. Summary of the Invention

[0004] The purpose of this invention is to provide a lightweight embedded rotor active noise reduction integrated protection device to solve the above-mentioned technical problems.

[0005] To achieve the above objectives, the present invention provides a lightweight embedded rotor active noise reduction integrated protection device, comprising an annular rotor protective shell arranged around the periphery of the rotor, and further comprising: A loudspeaker array module is embedded in the inner side of the annular rotor protective shell to emit controllable sound waves that form a quiet zone around the rotor. The motor brackets are multiple in number and arranged radially along the annular rotor protective shell. One end of each motor bracket is circumferentially fixed to the inner wall of the annular rotor protective shell, and the other end converges at the axis of the annular rotor protective shell and is fixed with a motor base for placing the motor. The bottom of the motor brackets is embedded with an error sensor array module for monitoring and evaluating the noise reduction performance of the target area, a reference sensor module for capturing the original aerodynamic noise signal generated by the rotor rotation in real time, and a multi-source sensing module for collecting multi-dimensional information required for the noise reduction strategy. The distributed intelligent control module includes a central main controller and local slave controllers located on the fuselage. The central main controller is communicatively connected to the multi-source sensing module and is used to generate a global noise reduction control strategy. The local slave controller is mounted on the annular rotor protective shell. The control input terminal of the local slave controller is communicatively connected to the central main controller, and the control output terminal is electrically connected to the speaker array module, used to drive the speaker array module to output and form an acoustic quiet zone. The distributed intelligent control module has a built-in geometric estimation unit, used to calculate the amplitude attenuation and phase delay of the acoustic transmission path in real time based on the spatial three-dimensional geometric relationship between the sound source, the speaker array module, and the error sensor array module, in order to optimize the noise reduction control strategy.

[0006] Preferably, the inner side of the annular rotor protective shell has a groove that matches the shape of the speaker array module, and the speaker array module is fixedly enclosed in the groove; the groove integrates an anti-aliasing filter and a signal shielding layer to suppress electromagnetic interference of the aircraft's power cables to acoustic signals.

[0007] Preferably, the loudspeaker array module consists of multiple ultra-thin strong magnetic monopole loudspeakers uniformly assembled circumferentially within the groove. The multiple ultra-thin strong magnetic monopole loudspeakers work together to output controllable sound waves to achieve omnidirectional sound field coverage of the target noise reduction area.

[0008] Preferably, each of the motor brackets has a mounting slot at its bottom, and multiple sets of reference sensor modules are fixedly embedded in the mounting slots to collect the original aerodynamic noise signal generated by the rotor rotation with a high signal-to-noise ratio; each error sensor array module includes multiple miniature silicon microphones, which are respectively embedded in the mounting slots to collect the residual noise signal of the target noise reduction area in real time to evaluate the noise reduction performance.

[0009] Preferably, each of the motor brackets has wiring vias on both sides to facilitate the access of connecting cables. The speaker array module, the error sensor array module, and the reference sensor module are all connected to the central main controller through the connecting cables.

[0010] Preferably, it also includes a signal conditioning and interface module, which is mounted in the mounting slot. The signal input terminal of the signal conditioning and interface module is electrically connected to the reference sensor module and the error sensor array module, respectively, and the signal output terminal is electrically connected to the speaker array module and the distributed intelligent control module, respectively. The signal conditioning and interface module includes a feedforward amplification circuit, a power amplification circuit, a filtering circuit, and a multi-core data interface, which is used to pre-amplify and filter the weak acoustic signals collected by the reference sensor module and the error sensor array module, and simultaneously amplify the power of the control signal output by the distributed intelligent control module to drive the speaker array module to work. The signal conditioning and interface module is connected to the central main controller through the connecting cable.

[0011] Preferably, the multi-source sensing module has three signal terminals. The first signal terminal is connected to the signal conditioning and interface module to obtain acoustic sensing data. The second signal terminal is connected to the data bus of the aircraft flight control system to obtain real-time flight status data of the aircraft. The third signal terminal is connected to the environmental sensing unit of the aircraft to obtain flight environment sensing data.

[0012] Preferably, the central main controller has a built-in noise reduction strategy engine based on a deep neural network. The noise reduction strategy engine is established through offline training using acoustic test data from all flight conditions and scenarios of the aircraft and corresponding optimal active noise control parameters. When running online, the noise reduction strategy engine maps and outputs a set of noise reduction control parameters adapted to the current operating conditions based on the fused data acquired in real time by the multi-source sensing module. The noise reduction control parameter set includes initial filter values, secondary path model corrections, and control bandwidth.

[0013] Preferably, the central main controller can combine the aircraft's flight path planning information to predict changes in the flight environment, so as to preload noise reduction strategies adapted to the target environment in advance, thereby achieving smooth switching of noise reduction modes; at the same time, it can fine-tune the parameters of the noise reduction strategy engine online based on the feedback signal of the error sensor array module to adapt to the performance changes during long-term use of the aircraft.

[0014] Preferably, the algorithm steps based on the geometric estimation unit in the above-described device include: S1. Based on the rotor spatial layout and acoustic module installation position, establish a three-dimensional spatial rectangular coordinate system with the rotor axis as the origin, and obtain the radiation point coordinates of the sound-generating unit of the loudspeaker array module, the receiving point coordinates of the error sensor array module, and the spatial coordinates of the preset target noise reduction area. S2. Based on the real-time physical spatial relationship between the multi-rotor aircraft and the target point, the straight-line distance is calculated by the geometric estimation unit to obtain the straight-line spatial geometric distance from the radiation point of the sound-emitting unit of the speaker array module to the receiving point of the error sensor array module and the noise reduction area of ​​the target below. S3. Based on the current air speed of sound and the geometric divergence law of free field sound waves, the pure propagation delay time difference and amplitude attenuation ratio of sound waves reaching the target noise reduction area are obtained by extrapolation and calculation through spatial distance difference. S4. Based on the pure propagation delay time difference and amplitude attenuation ratio, the geometric compensation control parameters for the adaptive noise reduction algorithm are obtained through phase and amplitude parameter compensation calculations. S5. Substitute the geometric compensation control parameters into the active noise reduction algorithm to update the phase and amplitude of the control weights, thereby projecting the acoustic quiet domain onto the target noise reduction area.

[0015] Therefore, the present invention employs the above-mentioned lightweight embedded rotor active noise reduction integrated protection device and the steps of the built-in algorithm based on the geometric estimation unit, which have the following beneficial effects: 1. This invention solves the shortcomings of traditional external noise reduction systems by deeply coupling the core acoustic components of active noise reduction with the original rotor protective shell and motor support structure of the multirotor aircraft. By creating a precision recessed embedded solid-sealed speaker array inside the rotor protective shell and integrating various sensing modules and signal processing units in the motor support, all noise reduction components are installed in a hidden and modular manner without altering the original aerodynamic design of the aircraft, thus achieving the integration of noise reduction and rotor protection functions. At the same time, the rotor protective shell is used as a physical barrier to provide reliable mechanical protection for the acoustic components, improving the impact resistance and environmental adaptability of the device. The integrated modular structural design also enables independent and rapid disassembly and assembly of the noise reduction unit, improving the maintainability and scalability of the system.

[0016] 2. This invention adopts a master-slave distributed intelligent control architecture consisting of a central master controller and local slave controllers on the rotors, achieving hierarchical decoupling and efficient operation of noise reduction computation tasks. By delegating the generation of audio waveforms with extremely high real-time requirements and the operation of local lightweight active noise reduction algorithms to the local slave controllers of the corresponding single rotor units, the central master controller is only responsible for the scheduling, optimization, and parameter distribution of the global noise reduction strategy. This reduces the computational load caused by centralized computation, improves the response speed of the noise reduction system, and avoids the phase mismatch problem under the condition of rapid changes in the rotational speed of multi-rotor aircraft. This distributed architecture can also realize independent control and modular expansion of single rotor units, and can flexibly adapt to multi-rotor low-altitude aircraft with different numbers of rotors and different load levels, possessing strong versatility.

[0017] 3. By constructing a multi-source perception system that integrates acoustic perception, aircraft state perception, and environmental perception, "proactive" noise reduction based on full-dimensional information is achieved, improving the system's adaptability to the environment and operating conditions. Through access to the flight control system data bus, core state data such as aircraft altitude, airspeed, attitude angle, and motor speed are acquired in real time. This is combined with environmental sensing units such as aircraft vision, radar, and altimeter to obtain environmental information such as spatial terrain, reflectors, and ground materials. Combined with noise source and residual noise signals collected by acoustic sensors, this provides a full-dimensional decision-making basis for generating noise reduction strategies. The system can accurately predict changes in noise frequency and sound field propagation characteristics based on multi-source fusion data, dynamically correcting the acoustic transfer function, adjusting speaker output weights and algorithm convergence step size. This solves the problems of fluctuating noise reduction effects and insufficient robustness caused by complex sound fields and drastic changes in the operating conditions of multi-rotor aircraft in open free-field environments, ensuring stable noise reduction performance across the entire flight envelope, including hovering, vertical takeoff and landing, and horizontal cruise.

[0018] 4. By implementing a high-fidelity signal processing and anti-interference scheme across the entire chain, the effectiveness and stability of noise reduction in open-field environments are significantly improved. A circumferentially distributed array of ultra-thin, high-strength magnetic monopole loudspeakers achieves omnidirectional sound field coverage of the target noise reduction area, meeting the causal requirements of active noise reduction in open fields. The reference sensor module and error sensor array are embedded below the motor bracket, effectively reducing turbulent noise generated by direct airflow through structural shielding, thus improving the signal-to-noise ratio of the original aerodynamic noise and residual noise signals. Simultaneously, the integrated anti-aliasing filter, signal shielding layer, and integrated signal conditioning and interface module suppress electromagnetic interference and vibration interference caused by aircraft power cables and airframe vibrations. This enables local pre-amplification of acoustic signals, filtering pre-processing, and power amplification of control signals. The entire chain from signal acquisition and transmission to output ensures the fidelity of acoustic signals and the accuracy of control signals, enhancing the suppression of low-frequency discrete line spectrum noise from the rotor. It can accurately construct an acoustic quiet zone within the aircraft cabin and ground personnel activity areas.

[0019] 5. An integrated noise reduction strategy engine based on deep neural networks enables adaptive intelligent noise reduction across all operating conditions and scenarios. This engine completes offline training using massive amounts of acoustic test data from all flight conditions and scenarios of the aircraft, along with corresponding optimal noise reduction parameters. It establishes a high-order nonlinear mapping from environmental and operating condition characteristics to optimal noise reduction control parameters. During online operation, it can map and output a set of noise reduction parameters, such as initial filter values, secondary path model corrections, and control bandwidth, adapted to the current operating condition within milliseconds based on real-time fusion data from multi-source sensing modules. This enhances the algorithm's ability to track rapidly changing operating conditions. Simultaneously, the engine can predict changes in the flight environment by combining aircraft trajectory planning information and preload noise reduction strategies adapted to the target environment in advance. This enables smooth and seamless switching of noise reduction modes. Furthermore, it can fine-tune model parameters online based on residual noise feedback signals from the error sensor array, adapting to long-term performance changes caused by battery consumption and component aging. This improves the all-scenario adaptability and long-term stability of the noise reduction system.

[0020] 6. By introducing a geometric estimation method, the problem of spatial deviation between physical measurement points and actual noise reduction target areas is solved. By performing precise three-dimensional geometric modeling of the speaker array, error sensor, and target noise reduction area below, the system can directly calculate the pure phase delay and amplitude attenuation of sound waves propagating to the target point using the known physical spatial distance. This geometric parameter is injected into the algorithm as a compensation term, which accurately focuses the acoustic quiet domain at the physical sensor onto the distant target noise reduction area, improving the actual noise reduction effectiveness for ground personnel or payload compartments.

[0021] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0022] Figure 1 The present invention provides an overall structural diagram of a lightweight embedded rotor active noise reduction integrated protection device; Figure 2 for Figure 1 A bottom view; Figure 3 The overall architecture block diagram of the intelligent active noise reduction system of the lightweight embedded rotor active noise reduction integrated protection device provided by the present invention; Figure 4 A schematic diagram of a multi-source sensing module for a lightweight embedded rotor active noise reduction integrated protection device provided by the present invention; Figure 5 A flowchart illustrating the generation and updating of a noise reduction strategy based on a deep neural network for a lightweight embedded rotor active noise reduction integrated protection device provided by the present invention; Figure 6A schematic diagram of the geometric estimation algorithm for a lightweight embedded rotor active noise reduction integrated protection device provided by the present invention; Figure 7 The flowchart shows the geometric estimation algorithm update process for a lightweight embedded rotor active noise reduction integrated protection device provided by the present invention.

[0023] Figure Labels 1. Speaker array module; 11. Groove; 2. Annular rotor protective shell; 3. Motor bracket; 31. Assembly slot; 4. Motor base; 5. Wiring via; 6. Error sensor array module; 7. Signal conditioning and interface module; 8. Multi-source sensing module; 9. Reference sensor module; 10. Distributed intelligent control module. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages disclosed in the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the embodiments of the present invention and are not intended to limit the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of this application. Examples of embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout.

[0025] It should be noted that the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion, such as a process, method, system, product, or server that includes a series of steps or units, not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product, or device.

[0026] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0027] In the current field of noise reduction for multi-rotor low-altitude aircraft, conventional passive noise reduction schemes have inherent limitations such as poor suppression of low-frequency discrete line spectrum noise and easy sacrifice of aerodynamic and load performance. On the other hand, traditional active noise reduction technologies face problems such as low coverage efficiency of open free field sound field, prominent computing power bottleneck of centralized multi-channel algorithm, poor adaptability to dynamic flight conditions, and external structure disruption of aerodynamic layout and additional load. It is difficult to balance noise reduction stability, aircraft aerodynamic performance and engineering feasibility across the entire flight envelope, which has become the core bottleneck restricting the acoustic airworthiness certification and large-scale commercial application of low-altitude aircraft.

[0028] Based on the above analysis, this invention is designed, see appendix. Figures 1-7A lightweight embedded rotor active noise reduction integrated protection device includes: an annular rotor protective shell 2 arranged around the periphery of the rotor, and further includes: Speaker array module 1 is embedded in the inner side of the annular rotor protective shell 2 to emit sound waves that form a quiet zone around the rotor. The motor brackets 3 are multiple and arranged radially along the annular rotor protective shell 2. One end of each motor bracket 3 is circumferentially fixed to the inner wall of the annular rotor protective shell 2, and the other end converges at the axis of the annular rotor protective shell 2 and is fixed with a motor base 4 for placing the motor. The bottom of the motor bracket 3 is embedded with an error sensor array module 6 for monitoring and evaluating the noise reduction performance of the target area, a reference sensor module 9 for capturing the original aerodynamic noise signal generated by the rotor rotation in real time, and a multi-source sensing module 8 for collecting multi-dimensional information required for the noise reduction strategy. The distributed intelligent control module 10 includes a central main controller and local slave controllers located inside the fuselage. The central main controller is communicatively connected to the multi-source sensing module 8 and is used to generate a global noise reduction control strategy. The local slave controller is mounted on the annular rotor protective shell 2. The control input terminal of the local slave controller is communicatively connected to the central main controller, and the control output terminal is electrically connected to the speaker array module 1, which is used to drive the speaker array module 1 to output and form an acoustic quiet domain. The distributed intelligent control module 10 has a built-in geometric estimation unit, which is used to calculate the amplitude attenuation and phase delay of the acoustic transmission path in real time based on the spatial three-dimensional geometric relationship between the sound source, the speaker array module 1 and the error sensor array module 6, so as to optimize the noise reduction control strategy.

[0029] In a specific embodiment of the annular rotor protective shell 2 in this invention, the annular rotor protective shell 2 is made of lightweight and high-strength materials, such as T700 / T800 grade carbon fiber reinforced epoxy resin matrix composite material, 7075-T6 aluminum alloy or LA103Z ultralight magnesium-lithium alloy. The annular rotor protective shell 2 has a groove 11 on its inner side that matches the shape of the speaker array module 1. The speaker array module 1 is fixed in the groove 11 by laser welding or epoxy resin sealing process. The groove 11 integrates an anti-aliasing filter and a signal shielding layer to suppress electromagnetic interference of the aircraft's power cable to the acoustic signal. The annular rotor protective shell 2 is made of the aforementioned lightweight and high-strength material. By creating a groove 11 on the inner side that precisely matches the shape of the speaker array module 1, and using laser welding or epoxy resin sealing technology to fix the speaker array module 1 in the groove 11, an embedded and hidden installation of the speaker array is achieved. This does not disrupt the original aerodynamic layout of the aircraft, nor does it significantly increase flight load and wind resistance. At the same time, the annular rotor protective shell 2, which originally only had physical collision protection function, is upgraded to a noise reduction core structure that also functions as an acoustic baffle, achieving deep coupling between the protective structure and the acoustic components. The anti-aliasing filter and signal shielding layer integrated inside the groove 11 can effectively suppress electromagnetic interference from the aircraft's power cables to the acoustic signals, ensuring the fidelity of the acoustic signals. Meanwhile, the sealed structure can prevent airflow, dust, and moisture from corroding the speaker array module 1, improving the operational stability and environmental adaptability of the acoustic system. The integrated modular design also enables independent and rapid disassembly and assembly of the noise reduction unit and subsequent maintenance and upgrades.

[0030] In a specific embodiment of the speaker array module 1 in this invention, the speaker array module 1 is composed of multiple ultra-thin strong magnetic monopole speakers uniformly assembled circumferentially within a groove 11. These multiple ultra-thin strong magnetic monopole speakers collaboratively output controllable sound waves to achieve omnidirectional sound field coverage of the target noise reduction area. The speaker array module 1, composed of multiple ultra-thin strong magnetic monopole speakers uniformly assembled circumferentially within a groove 11, can adapt to the omnidirectional radiation aerodynamic noise characteristics generated by rotor rotation through its uniform array layout. When multiple units collaboratively output controllable anti-phase sound waves, it can achieve omnidirectional sound field coverage of the target noise reduction area, fully meeting the causal requirements of active noise reduction in open free-field environments. This solves the problems of limited sound field coverage and low noise reduction efficiency in open spaces associated with traditional secondary sound sources. The selection of ultra-thin strong magnetic monopole speakers can fully adapt to the embedded and concealed installation requirements of the groove 11, without occupying additional installation space. By disrupting the original aerodynamic layout and streamlined design of the aircraft, the strong magnetic drive structure endows the loudspeaker with excellent low-frequency response performance, which can accurately cancel the low-frequency discrete line spectrum noise that is difficult to suppress effectively by passive noise reduction schemes in rotor aerodynamic noise. The monopole radiation characteristics improve the sound wave radiation efficiency, which can efficiently build a stable acoustic quiet domain in the aircraft cabin and ground personnel activity areas. The multi-unit distributed array layout can also be adapted to the multi-channel independent control of the distributed intelligent control module, flexibly adjust the sound wave output weight in different directions, and further improve the noise reduction stability and environmental adaptability of the aircraft under all flight conditions and all flight attitudes.

[0031] In a specific embodiment of the motor bracket 3 in this invention, each motor bracket 3 has a mounting slot 31 at its bottom. Multiple reference sensor modules 9 are fixedly embedded in the mounting slots 31 for high signal-to-noise ratio acquisition of the original aerodynamic noise signal generated by rotor rotation. The error sensor array module 6 includes multiple miniature silicon microphones, each embedded in a corresponding mounting slot 31, for real-time acquisition of residual noise signals in the target noise reduction area to evaluate noise reduction performance. By opening dedicated mounting slots 31 at the bottom of each motor bracket 3, multiple reference sensor modules 9 and the miniature silicon microphones of the error sensor array module 6 are fixedly embedded in the mounting slots 31, achieving integrated embedded installation of acoustic sensing elements without the need for additional mounting brackets. This does not disrupt the original aerodynamic layout of the aircraft, nor does it increase flight drag or load. Simultaneously, the main structure of the motor bracket 3 effectively physically shields the embedded sensors, reducing turbulence noise generated by airflow directly hitting the sensors during flight, improving the signal-to-noise ratio of the original aerodynamic noise signal acquired by the reference sensor module 9 from rotor rotation, and ensuring the accuracy of the error sensor array module. The accuracy of residual noise signal acquisition in the target noise reduction area provides a high-fidelity input signal and a reliable basis for noise reduction performance evaluation for the active noise reduction algorithm. The embedded layout of multiple sensors distributed along the three circumferences of the motor bracket can realize the acquisition of rotor noise signals without dead angles, accurately capture the aerodynamic noise changes and sound field distribution characteristics under different flight attitudes, improve the closed-loop control accuracy and all-condition adaptability of the multi-channel active noise reduction system, and at the same time, the embedded mounting design of the mounting slot 31 can also provide stable mechanical protection for the sensors, reduce the impact of flight vibration, outdoor dust and water vapor on the performance of sensing elements, and improve the long-term operational stability and environmental adaptability of the acoustic sensing system.

[0032] In the above embodiment, each motor bracket 3 has wiring vias 5 on both sides to facilitate the access of connecting cables. The speaker array module 1, error sensor array module 6, and reference sensor module 9 are all connected to the central main controller via connecting cables. The wiring vias 5 on both sides of each motor bracket 3 provide a standardized centralized routing channel for the connecting cables of the speaker array module 1, error sensor array module 6, and reference sensor module 9, realizing a concealed and neat routing of the entire noise reduction system cable. This avoids the problems of exposed cables damaging the original aerodynamic layout of the aircraft and increasing flight drag and turbulence, without affecting the original aerodynamic performance of the aircraft. At the same time, it solves the problems of messy wiring and difficult routing caused by multiple sets of distributed acoustic components, simplifying the system assembly process and the difficulty of subsequent maintenance and repair. The main structure of the motor bracket 3 can be through-hole The connecting cables located within the vias 5 form a stable physical protection, avoiding cable damage and signal interruption caused by airflow scouring, foreign object scraping, and mechanical collisions during flight. This improves the long-term reliability of the electrical connection. At the same time, the centralized wiring design, combined with the built-in shielding structure, further suppresses electromagnetic interference from the aircraft's power cables to the acoustic signal transmission, ensuring high fidelity in the transmission of acoustic signals collected by the reference sensor module 9 and the error sensor array module 6, and control signals from the speaker array module 1. This provides a stable and reliable electrical transmission foundation for the precise closed-loop control of the multi-channel active noise reduction system.

[0033] In another embodiment of the present invention, a signal conditioning and interface module 7 is also included. The signal conditioning and interface module 7 is assembled in the mounting slot 31. The signal input terminal of the signal conditioning and interface module 7 is electrically connected to the reference sensor module 9 and the error sensor array module 6, respectively. The signal output terminal is electrically connected to the speaker array module 1 and the distributed intelligent control module 10, respectively. The signal conditioning and interface module 7 includes a feedforward amplification circuit, a power amplification circuit, a filtering circuit, and a multi-core data interface. It is used to pre-amplify and filter the weak acoustic signals collected by the reference sensor module 9 and the error sensor array module 6, and at the same time, to amplify the control signal output by the distributed intelligent control module 10 to drive the speaker array module 1 to work. The signal conditioning and interface module 7 is connected to the central main controller through a connecting cable. The signal conditioning and interface module 7, installed in the assembly slot 31, achieves close integration of the signal processing unit and acoustic sensing element, shortening the transmission path of the acoustic signal and reducing the risk of signal attenuation and electromagnetic interference caused by long-distance transmission. Its signal input terminals are electrically connected to the reference sensor module 9 and the error sensor array module 6, respectively. Through the built-in feedforward amplification circuit and filtering circuit, it can perform real-time pre-amplification and filtering pre-processing of the weak acoustic signals collected by the two types of sensors, effectively eliminating noise signals caused by body vibration and electromagnetic interference, improving the signal-to-noise ratio and fidelity of the acoustic input signal, and providing accurate and reliable input data for the noise reduction algorithm of the distributed intelligent control module 10. Its signal output terminals are connected to the speaker, respectively. The array module 1 and the distributed intelligent control module 10 are electrically connected. The built-in power amplifier circuit can efficiently amplify the noise reduction control signal output by the distributed intelligent control module 10, providing stable and compatible driving capability for the speaker array module 1, ensuring the phase and amplitude accuracy of the anti-phase sound wave output and the consistency of multiple units. At the same time, it is connected to the central main controller through the built-in multi-core data interface and connecting cable, realizing high-speed interaction and unified scheduling of the entire link signal of the noise reduction system, realizing local preprocessing of acoustic signals, reducing the computing load of the central main controller, simplifying the wiring and assembly process of the system, and improving the integration, real-time closed-loop control accuracy, operational stability and maintenance of the multi-channel active noise reduction system.

[0034] In a specific embodiment of the multi-source sensing module 8 in this invention, the multi-source sensing module 8 has three signal terminals. The first signal terminal is communicatively connected to the signal conditioning and interface module 7 to obtain acoustic sensing data. The second signal terminal is connected to the data bus of the aircraft's flight control system to obtain real-time flight status data of the aircraft. The third signal terminal is communicatively connected to the environmental sensing unit of the aircraft to obtain flight environment sensing data. The multi-source sensing module 8 constructs a three-dimensional fusion sensing system integrating acoustic sensing, flight status sensing, and environmental sensing by setting up three signal terminals. Its first signal terminal, communicatively connected to the signal conditioning and interface module 7, obtains high-fidelity acoustic sensing data, which can synchronize the original aerodynamic noise characteristics of the rotor with the residual noise signal of the target noise reduction area in real time. This provides accurate original noise input for the noise reduction algorithm and reliable acoustic feedback for real-time evaluation and closed-loop optimization of noise reduction performance. The second signal terminal, connected to the data bus of the aircraft's flight control system, obtains real-time flight status data of the aircraft, which can accurately capture the dynamic changes of core operating conditions such as motor speed, flight altitude, airspeed, and attitude angle. It can predict instantaneous changes in noise characteristics in advance, providing a preliminary operating condition basis for the dynamic adjustment of noise reduction strategies, and solving the problems of traditional active noise reduction systems lagging in tracking drastic changes in flight conditions and being prone to phase mismatch. The third signal terminal communicates with the environmental sensing unit of the aircraft to obtain flight environment perception data, which can identify key environmental factors affecting sound field propagation in real time, such as reflectors in the surrounding space, ground materials, and terrain undulations. This provides environmental data support for the dynamic correction of the secondary path acoustic model, so as to adapt to the complex and ever-changing sound field environment of the open free field. It improves the noise reduction stability, environmental adaptability and system robustness of the noise reduction system throughout the entire flight envelope, including hovering, cruise, takeoff and landing.

[0035] In the above embodiments, the real-time flight status data acquired by the multi-source sensing module 8 includes the aircraft's altitude, airspeed, attitude angle, real-time speed of each motor, and current load. The flight environment perception data acquired by the multi-source sensing module 8 includes the ground material below the aircraft, terrain undulations, spatial reflector information, and altitude above the ground. The environmental sensing units include the aircraft's visual sensors, ultrasonic altimeters, lidar, or millimeter-wave radar. This module can directly reuse the aircraft's existing environmental sensing units such as visual sensors, ultrasonic altimeters, lidar, and millimeter-wave radar to acquire environmental data, without the need for additional redundant sensing hardware, thus improving the versatility of the noise reduction system with different aircraft models.

[0036] Specifically, the multi-source sensing module 8 uses a miniature silicon microphone located below the motor bracket 3 to capture the raw aerodynamic noise generated by the rotor, extracting the blade passing frequency and its higher-order harmonic signals; through the interface with the flight control system, it acquires real-time motor speed, current load, and flight attitude data to predict instantaneous changes in noise frequency; it uses millimeter-wave radar and visual sensors to scan the physical boundaries around the aircraft to identify whether it is currently in a free field or has a reflecting surface, such as when flying near a building or in a tunnel; it uses an altimeter and GPS to determine the aircraft's altitude above the ground, directly affecting the sound field propagation characteristics and noise reduction weight allocation of the target noise reduction area below; and it fuses speed and altitude data to determine the aircraft's current position. The system is in a "hovering," "vertical takeoff and landing," or "horizontal cruise" state. Based on spatial reflection information fed back by radar, it dynamically corrects the acoustic transfer function of the secondary path to compensate for sound field changes caused by obstacle reflections. Combining flight attitude, it calculates the spatial directivity of noise radiation and automatically adjusts the output weights of speakers at different positions on the rotor protective shell. The fused environmental feature vector is output to the distributed control unit, and the system matches the closest filtering parameters from the noise reduction benchmark library generated by the neural network based on the current features. If it senses severe speed fluctuations, the module instructs the controller to increase the convergence step size of the adaptive algorithm to improve tracking performance. Furthermore, it automatically selects the acoustic path library based on environmental characteristics.

[0037] In the above embodiments, the signal conditioning and interface module 7 establishes a database through the distributed intelligent controller module using the data collected by the multi-source sensing module 8. Specifically, the database is an environment-policy knowledge base: systematically collecting data from the aircraft under different combined operating conditions in an anechoic chamber, a semi-anechoic chamber, and a real outdoor field. Operating condition variables include: various altitudes / speeds, various ground materials, various typical obstacle scenarios, and various motor loads. Under each operating condition, the reference signal and error signal are measured, the corresponding multi-source sensing layer data is recorded, and the optimal set of controller parameters and the driving mode of the secondary sound source array that achieve the best noise reduction effect under this environment are experimentally determined.

[0038] In one specific embodiment of the central main controller in this invention, the central main controller has a built-in noise reduction strategy engine based on a deep neural network. The noise reduction strategy engine is established through offline training using acoustic test data from all flight conditions and all scenarios of the aircraft and the corresponding optimal active noise control parameters. When running online, the noise reduction strategy engine maps and outputs a set of noise reduction control parameters adapted to the current operating conditions based on the fused data acquired in real time by the multi-source sensing module 8. The noise reduction control parameter set includes the initial value of the filter, the correction amount of the secondary path model, and the control bandwidth. The pre-trained filter initial values ​​can shorten the convergence time of the adaptive algorithm, effectively avoiding noise reduction lag or even noise amplification when the motor speed changes drastically. The secondary path model correction can adapt to changes in sound field propagation characteristics under different environments in real time, accurately compensating for acoustic transfer function distortion caused by spatial reflection and terrain undulation. The dynamically adjusted control bandwidth can effectively cope with Doppler frequency shift under high-speed flight, ensuring stable noise reduction effect across the entire frequency band. This engine realizes the upgrade of the noise reduction strategy from the traditional fixed parameter, manual debugging mode to intelligent adaptive, full-condition autonomous matching mode. It eliminates the need for manual repeated debugging of algorithm parameters for different scenarios, improving the engineering implementation efficiency of the noise reduction system. At the same time, it can complete the inference and distribution of optimal parameters in milliseconds, taking into account both the real-time performance and accuracy of noise reduction control, ensuring that the aircraft maintains stable and excellent noise reduction performance throughout the entire flight envelope, including hovering, take-off and landing, and cruise. It can also continuously complete model iteration and optimization by supplementing test data, possessing strong scalability and long-term adaptability.

[0039] The specific working principle of the central main controller includes: the central main controller loads a pre-trained deep neural network model and a noise reduction strategy database, and then inputs the fusion feature vector generated by the multi-source sensing module 8 based on the collected data into the pre-trained deep neural network model. The model quickly infers and generates the optimal noise reduction parameter set adapted to the current flight conditions and environment based on the mapping relationship established by learning from massive data of all flight conditions and all scenarios in the offline stage. This parameter set specifically includes the initial weights of the FxLMS filters of each channel, the secondary path model correction coefficients suitable for flat hard ground scenarios, and the wide control bandwidth parameters that can cope with the Doppler frequency shift under high-speed flight. Then, the central main controller distributes the inferred optimal noise reduction parameter set to the local slave controllers corresponding to each rotor unit through a high-speed communication bus to complete the distributed deployment of the global noise reduction strategy. The local controller initializes and starts running the local active noise reduction algorithm using the optimal noise reduction parameter set issued by the central main controller. The local active noise reduction algorithm takes the original noise reference signal collected by the reference sensor module 9 as the core input, and combines it with the real-time feedback signal of the residual noise in the target noise reduction area collected by the error sensor array module 6 to generate an anti-phase sound wave control signal that cancels the interference of the rotor aerodynamic noise in real time. This signal drives the speaker array module 1 to output the corresponding controllable sound wave, thus completing the real-time active noise reduction closed-loop control of the single rotor unit.

[0040] Specifically, to address the spatial misalignment between the physical error sensor installation location and the target noise reduction area, a geometric estimation method is deeply integrated into the local active noise reduction algorithm of the controller. The specific steps include: S1. Based on the rotor spatial layout and acoustic module installation position, establish a three-dimensional spatial rectangular coordinate system with the rotor axis as the origin, and obtain the radiation point coordinates of the sound-generating unit of the loudspeaker array module, the receiving point coordinates of the error sensor array module, and the spatial coordinates of the preset target noise reduction area. S2. Based on the real-time physical spatial relationship between the multi-rotor aircraft and the target point, the straight-line distance is calculated by the geometric estimation unit to obtain the straight-line spatial geometric distance from the radiation point of the sound-emitting unit of the speaker array module to the receiving point of the error sensor array module and the noise reduction area of ​​the target below. S3. Based on the current air speed of sound and the geometric divergence law of free field sound waves, the pure propagation delay time difference and amplitude attenuation ratio of sound waves reaching the target noise reduction area are obtained by extrapolation and calculation through spatial distance difference. S4. Based on the pure propagation delay time difference and amplitude attenuation ratio, the geometric compensation control parameters for the adaptive noise reduction algorithm are obtained through phase and amplitude parameter compensation calculations. S5. Substitute the geometric compensation control parameters into the active noise reduction algorithm to update the phase and amplitude of the control weights, thereby projecting the acoustic quiet domain onto the target noise reduction area.

[0041] Therefore, the system forces the speaker to output geometrically compensated sound waves, achieving precise noise reduction of the target area without adding virtual error synthesis calculation load.

[0042] The noise reduction effect feedback signal collected by the error sensor array module 6 is synchronously transmitted to the multi-source sensing module 8. Based on this feedback data, the central main controller performs long-term online evaluation and adaptive fine-tuning of the core parameters of the deep neural network noise reduction strategy engine, enabling the system to continuously adapt to the slow performance changes of the aircraft caused by factors such as battery consumption and component aging, and ensuring the stability of noise reduction performance throughout the entire life cycle of the aircraft.

[0043] In the above embodiments, the central main controller can combine the aircraft trajectory planning information to predict changes in the flight environment, so as to preload noise reduction strategies adapted to the target environment in advance, thereby achieving smooth switching of noise reduction modes; at the same time, it can fine-tune the parameters of the noise reduction strategy engine online based on the feedback signal of the error sensor array module 6 to adapt to the performance changes during long-term use of the aircraft. The central main controller predicts changes in the flight environment by combining the aircraft's trajectory planning information and preloads noise reduction strategies adapted to the target environment in advance. This enables smooth and seamless switching of noise reduction modes, overcoming the technical limitations of traditional active noise reduction systems that can only passively respond and adjust after environmental changes. It avoids the risks of noise reduction effect fluctuations, algorithm phase mismatch, and even noise amplification during cross-scenario flight, improving the system's environmental adaptability and noise reduction stability in urban low-altitude cross-scenario operations. At the same time, the central main controller can perform online adaptive fine-tuning of the core parameters of the noise reduction strategy engine based on the residual noise feedback signal of the target area collected in real time by the error sensor array module 6. This can accurately adapt to the noise characteristics and system performance changes caused by long-term use of the aircraft, such as battery consumption, component aging, and blade wear. It solves the problem of continuous decay of noise reduction effect after long-term use of traditional fixed-parameter noise reduction schemes, ensuring the performance stability of the noise reduction system throughout the entire life cycle of the aircraft. In addition, this online closed-loop fine-tuning mechanism can also compensate for subtle operating conditions and environmental changes not covered by the offline training model, further improving the control accuracy and all-condition robustness of the multi-channel active noise reduction system.

[0044] The working principle of the lightweight embedded rotor active noise reduction integrated protection device of the present invention is as follows: 1) Through acoustic testing and flight data acquisition under full operating conditions and in all scenarios, the training of the deep neural network noise reduction strategy engine is completed, the mapping relationship between scene features and optimal noise reduction parameters is established, and a lightweight noise reduction strategy database is solidified; after the aircraft is powered on, the central main controller completes the loading of the pre-trained deep neural network model and the noise reduction strategy database, and completes the system initialization configuration. 2) The multi-source sensing module synchronously collects full-dimensional data through three signal terminals: the first channel obtains the original aerodynamic noise signal from the reference sensor module and the residual noise signal from the error sensor array module through the signal conditioning and interface module; the second channel connects to the flight control bus to obtain real-time flight status data such as aircraft altitude, airspeed, and motor speed; the third channel links with the aircraft environment sensing unit to obtain environmental perception data such as ground material and space reflectors, and completes feature extraction and fusion of multi-source data to generate standardized scene feature vectors; 3) The central main controller inputs the fused feature vector output by the multi-source sensing module into the pre-trained deep neural network noise reduction strategy engine. Through rapid model inference, it generates the optimal noise reduction parameter set that is adapted to the current flight conditions and environment. The core parameters include the initial weights of the FxLMS filters of each channel, the correction coefficients of the secondary path model, and the control bandwidth matching the flight conditions. 4) The central main controller distributes the optimal noise reduction parameter set to the local slave controllers corresponding to each rotor unit through a high-speed communication bus; each local slave controller completes the initial configuration of the local lightweight active noise control algorithm based on the distributed parameter set and locks the algorithm running benchmark adapted to the current working conditions. 5) The built-in geometric estimation unit is started locally from the controller. Based on the three-dimensional physical coordinates of the speaker unit, physical error sensing unit and target noise reduction area, which are pre-calibrated with high precision, and combined with the air speed of the current environment, the geometric distance difference of the sound wave transmitted to the target point in the physical space is calculated in real time. Then, the pure phase delay compensation amount and the amplitude attenuation compensation amount based on the spatial divergence law are directly derived. 6) The local controller directly feeds back the derived phase delay and amplitude attenuation parameters to the active noise reduction algorithm as spatial compensation terms; then, using the original noise signal collected by the reference sensor module as the core input, the noise reduction algorithm generates an anti-phase sound wave control signal, which, after signal conditioning and power amplification by the interface module, drives the speaker array module to output controllable sound waves; at the same time, combined with the real-time feedback of residual noise from the physical error sensor array module, the algorithm's filtering weights are finely adjusted online, and under the guidance of geometric parameters, the acoustic quiet domain is accurately projected onto the target noise reduction area below; 7) Based on long-term feedback data on the noise reduction effect of the error sensor array module, the central main controller performs online adaptive fine-tuning of the core parameters of the noise reduction strategy engine to adapt to the slow performance changes caused by aircraft battery consumption and component aging; at the same time, it combines the aircraft trajectory planning information to predict changes in the flight environment and preload the adapted noise reduction strategy in advance to achieve smooth switching of noise reduction modes across scenarios and ensure the noise reduction stability of the system throughout its entire life cycle and flight envelope.

[0045] In summary, this invention integrates a speaker array into a ring rotor protective shell, embeds multiple types of sensors and sensing modules in the motor bracket, and combines a distributed intelligent control module and a built-in geometric estimation unit to achieve accurate acoustic signal acquisition, interference suppression, real-time adaptation and predictive switching of noise reduction strategies. While reducing computational load, it accurately projects the acoustic quiet domain to the target area, improves noise reduction performance, takes into account rotor protection and the reasonable layout of each module, adapts to all flight conditions of the aircraft, and ensures stable and reliable noise reduction effect in the target area.

[0046] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A lightweight embedded rotor active noise reduction integrated protection device, comprising an annular rotor protective shell (2) arranged around the periphery of the rotor, characterized in that, Also includes: A loudspeaker array module (1) is embedded in the inner side of the annular rotor protective shell (2) to emit sound waves that form a quiet zone around the rotor. The motor brackets (3) are multiple and arranged radially along the annular rotor protective shell (2). One end of each motor bracket (3) is circumferentially fixed to the inner wall of the annular rotor protective shell (2), and the other end converges at the axis of the annular rotor protective shell (2) and is fixed with a motor base (4) for placing the motor. The bottom of the motor brackets (3) is embedded with an error sensor array module (6) for monitoring and evaluating the noise reduction performance of the target area, a reference sensor module (9) for capturing the original aerodynamic noise signal generated by the rotor rotation in real time, and a multi-source sensing module (8) for collecting multi-dimensional information required for the noise reduction strategy. The distributed intelligent control module (10) includes a central main controller and a local slave controller located on the fuselage. The central main controller is communicatively connected to the multi-source sensing module (8) and is used to generate a global noise reduction control strategy. The local slave controller is mounted on the annular rotor protective shell (2). The control input terminal of the local slave controller is communicatively connected to the central main controller, and the control output terminal is electrically connected to the loudspeaker array module (1) and is used to drive the loudspeaker array module (1) to output and form an acoustic quiet domain. The distributed intelligent control module (10) has a built-in geometric estimation unit, which is used to calculate the amplitude attenuation and phase delay of the acoustic transmission path in real time based on the spatial three-dimensional geometric relationship between the sound source, the loudspeaker array module (1) and the error sensor array module (6) in order to optimize the noise reduction control strategy.

2. The lightweight embedded rotor active noise reduction integrated protection device according to claim 1, characterized in that: The inner side of the annular rotor protective shell (2) is provided with a groove (11) that matches the shape of the speaker array module (1), and the speaker array module (1) is sealed in the groove (11); the groove (11) integrates an anti-aliasing filter and a signal shielding layer to suppress electromagnetic interference of the aircraft's power cable to the acoustic signal.

3. The lightweight embedded rotor active noise reduction integrated protection device according to claim 2, characterized in that: The loudspeaker array module (1) consists of multiple ultra-thin strong magnetic monopole loudspeakers uniformly assembled in the groove (11) in the circumferential direction. The multiple ultra-thin strong magnetic monopole loudspeakers work together to output controllable sound waves to achieve omnidirectional sound field coverage of the target noise reduction area.

4. The lightweight embedded rotor active noise reduction integrated protection device according to claim 1, characterized in that: Each of the motor brackets (3) has an assembly slot (31) at its bottom. The reference sensor module (9) consists of multiple sets and is fixedly embedded in the assembly slot (31) for high signal-to-noise ratio acquisition of the original aerodynamic noise signal generated by the rotor rotation. The error sensor array module (6) includes multiple miniature silicon microphones, which are respectively embedded in the assembly slot (31) for real-time acquisition of residual noise signal in the target noise reduction area to evaluate the noise reduction performance.

5. The lightweight embedded rotor active noise reduction integrated protection device according to claim 4, characterized in that: Each of the motor brackets (3) has wiring vias (5) on both sides to facilitate the access of connecting cables. The speaker array module (1), the error sensor array module (6) and the reference sensor module (9) are all connected to the central main controller through the connecting cables.

6. The lightweight embedded rotor active noise reduction integrated protection device according to claim 5, characterized in that: It also includes a signal conditioning and interface module (7), which is installed in the mounting slot (31). The signal input terminal of the signal conditioning and interface module (7) is electrically connected to the reference sensor module (9) and the error sensor array module (6), respectively, and the signal output terminal is electrically connected to the loudspeaker array module (1) and the distributed intelligent control module (10), respectively. The signal conditioning and interface module (7) includes a feedforward amplifier circuit, a power amplifier circuit, a filter circuit and a multi-core data interface, which are used to pre-amplify and filter the weak acoustic signals collected by the reference sensor module (9) and the error sensor array module (6), and at the same time, to amplify the control signal output by the distributed intelligent control module (10) to drive the loudspeaker array module (1) to work. The signal conditioning and interface module (7) is connected to the central main controller through the connecting cable.

7. The lightweight embedded rotor active noise reduction integrated protection device according to claim 6, characterized in that: The multi-source sensing module (8) is provided with three signal terminals. The first signal terminal is connected to the signal conditioning and interface module (7) to obtain acoustic sensing data. The second signal terminal is connected to the data bus of the aircraft flight control system to obtain real-time flight status data of the aircraft. The third signal terminal is connected to the environmental sensing unit of the aircraft to obtain flight environment sensing data.

8. The lightweight embedded rotor active noise reduction integrated protection device according to claim 7, characterized in that: The central main controller has a built-in noise reduction strategy engine based on deep neural networks. The noise reduction strategy engine is established through offline training using acoustic test data from all flight conditions and scenarios of the aircraft and the corresponding optimal active noise control parameters. When running online, the noise reduction strategy engine maps and outputs a noise reduction control parameter set adapted to the current working condition based on the fusion data obtained in real time by the multi-source perception module (8). The noise reduction control parameter set includes the initial value of the filter, the correction amount of the secondary path model, and the control bandwidth.

9. A lightweight embedded rotor active noise reduction integrated protection device according to claim 8, characterized in that: The central main controller can combine the flight path planning information of the aircraft to predict changes in the flight environment, so as to preload noise reduction strategies adapted to the target environment in advance, thereby achieving smooth switching of noise reduction mode; at the same time, it can fine-tune the parameters of the noise reduction strategy engine online based on the feedback signal of the error sensor array module (6) to adapt to the performance changes during long-term use of the aircraft.

10. A lightweight embedded rotor active noise reduction integrated protection device according to claim 9, characterized in that, The algorithm steps based on the geometric estimation unit include: S1. Based on the rotor spatial layout and acoustic module installation position, establish a three-dimensional spatial rectangular coordinate system with the rotor axis as the origin, and obtain the radiation point coordinates of the sound-generating unit of the loudspeaker array module, the receiving point coordinates of the error sensor array module, and the spatial coordinates of the preset target noise reduction area. S2. Based on the real-time physical spatial relationship between the multi-rotor aircraft and the target point, the straight-line distance is calculated by the geometric estimation unit to obtain the straight-line spatial geometric distance from the radiation point of the sound-emitting unit of the speaker array module to the receiving point of the error sensor array module and the noise reduction area of ​​the target below. S3. Based on the current air speed of sound and the geometric divergence law of free field sound waves, the pure propagation delay time difference and amplitude attenuation ratio of sound waves reaching the target noise reduction area are obtained by extrapolation and calculation through spatial distance difference. S4. Based on the pure propagation delay time difference and amplitude attenuation ratio, the geometric compensation control parameters for the adaptive noise reduction algorithm are obtained through phase and amplitude parameter compensation calculations. S5. Substitute the geometric compensation control parameters into the active noise reduction algorithm to update the phase and amplitude of the control weights, thereby projecting the acoustic quiet domain onto the target noise reduction area.

Images