GaN small-size charger and low-noise design method thereof

Abstract

The invention provides a GaN small-size charger and a low-noise design method thereof. Belongs to the cross technical field of high-frequency power electronics, electromagnetic compatibility (EMC) and acoustic engineering. The method comprises the following steps: performing switch transient waveform analysis on the circuit topology of the GaN charger to generate transient noise source distribution data; designing a GaN gate drive waveform shaping strategy according to the transient noise source distribution data, and generating optimized drive waveform parameters; dynamically modulating the GaN switching tube based on the optimized driving waveform parameter to generate low-noise switching behavior data; through collaborative design of GaN switch transient shaping, three-dimensional magnetic integration layout, piezoelectric active offset and acoustic metamaterial packaging, the method can reduce audible frequency band noise to below 35dBA while keeping high power density of the charger, and obviously improves mute experience of a user in use.

Description

Technical Field

[0001] This invention proposes a GaN small-volume charger and its low-noise design method, belonging to the interdisciplinary fields of high-frequency power electronics, electromagnetic compatibility (EMC) and acoustic engineering. Background Technology

[0002] With the increasing demand for fast charging in electronic devices, GaN (gallium nitride) small-size chargers have rapidly gained popularity due to their high power density. However, existing products have significant shortcomings in noise control, making it difficult to meet users' expectations for a quiet operating environment.

[0003] Traditional designs primarily focus on suppressing EMI conducted and radiated noise, neglecting mechanical vibrations and piezoelectric noise in the audible frequency range (<20kHz). While these noises have low energy, they are highly perceptible to the human ear, especially in quiet environments. Layout optimization often follows the principle of "shortening the loop," failing to consider the coupling resonance between high-frequency magnetic fields and structural modes in three-dimensional space, leading to amplification of noise in specific frequency bands. In component selection, although low-noise MLCCs or magnetic materials are used, the lack of system-level collaborative design fails to effectively control the interaction between GaN switching behavior and passive device responses. Furthermore, existing noise reduction methods rely on passive shielding or potting processes, which not only increase size but also contradict the goal of miniaturization.

[0004] Despite market claims of "silent design," these are essentially just optimizations of the shielding, failing to implement end-to-end noise suppression from the source to the perception path. Therefore, a low-noise design method that breaks with traditional thinking is urgently needed. This method should utilize multi-physics field synergy to reduce audible noise to below 35 dBA while maintaining high power density, providing technical support for the next generation of silent fast-charging products. Summary of the Invention

[0005] This invention provides a GaN small-volume charger and its low-noise design method to solve the problems mentioned in the background art above:

[0006] This invention proposes a low-noise design method for a GaN small-volume charger, the method comprising:

[0007] S1. Perform switching transient waveform analysis on the circuit topology of the GaN charger to generate transient noise source distribution data; design a GaN gate drive waveform shaping strategy based on the transient noise source distribution data to generate optimized drive waveform parameters; dynamically modulate the GaN switch based on the optimized drive waveform parameters to generate low-noise switching behavior data.

[0008] S2. Based on low-noise switching behavior data, perform three-dimensional magnetic integration layout design to generate a three-dimensional spatial distribution model of magnetic components; perform high-frequency magnetic field and structural mode coupling analysis on the three-dimensional spatial distribution model of magnetic components to generate mode decoupling parameters; adjust the position and orientation of magnetic components according to the mode decoupling parameters to generate anti-resonance magnetic integration layout data.

[0009] S3. Deploy piezoelectric sensors based on the anti-resonance magnetic integrated layout data to generate a piezoelectric noise sensing network; collect mechanical vibration and piezoelectric effect noise in the audible frequency band through the piezoelectric noise sensing network to generate raw piezoelectric noise signal data; perform phase reversal processing on the raw piezoelectric noise signal data to generate active cancellation waveform parameters; drive the piezoelectric actuator based on the active cancellation waveform parameters to generate reverse noise wave data;

[0010] S4. Design an acoustic metamaterial encapsulation structure based on the reverse noise wave data to generate metamaterial unit geometric parameters; optimize the acoustic impedance matching of the metamaterial unit geometric parameters to generate metamaterial encapsulation layer data; integrate the metamaterial encapsulation layer data with the charger shell structure to generate an integrated low-noise encapsulation model.

[0011] S5. Perform multi-physics co-simulation verification based on the integrated low-noise packaging model to generate noise suppression performance evaluation data; when the noise suppression performance evaluation data does not reach the preset threshold, iteratively adjust the parameters of S1 to S4 until the requirements are met.

[0012] This invention proposes a charger for implementing a low-noise design method for a GaN small-volume charger as described above, the charger comprising:

[0013] Data generation module: Performs switching transient waveform analysis on the circuit topology of GaN charger to generate transient noise source distribution data; designs GaN gate drive waveform shaping strategy based on transient noise source distribution data to generate optimized drive waveform parameters; dynamically modulates GaN switching transistors based on optimized drive waveform parameters to generate low-noise switching behavior data;

[0014] Coupling Analysis Module: Based on low-noise switching behavior data, a three-dimensional magnetic integration layout design is performed to generate a three-dimensional spatial distribution model of magnetic components; high-frequency magnetic field and structural mode coupling analysis is performed on the three-dimensional spatial distribution model of magnetic components to generate mode decoupling parameters; the position and orientation of magnetic components are adjusted according to the mode decoupling parameters to generate anti-resonance magnetic integration layout data;

[0015] Parameter-driven module: Deploys piezoelectric sensors based on anti-resonance magnetic integrated layout data to generate a piezoelectric noise sensing network; collects mechanical vibration and piezoelectric effect noise in the audible frequency band through the piezoelectric noise sensing network to generate raw piezoelectric noise signal data; performs phase reversal processing on the raw piezoelectric noise signal data to generate active cancellation waveform parameters; drives the piezoelectric actuator based on the active cancellation waveform parameters to generate reverse noise wave data;

[0016] Structural Fusion Module: Based on the reverse noise wave data, the acoustic metamaterial encapsulation structure is designed, generating the geometric parameters of the metamaterial unit; the acoustic impedance matching of the metamaterial unit geometric parameters is optimized, generating the metamaterial encapsulation layer data; the metamaterial encapsulation layer data is fused with the charger shell structure to generate an integrated low-noise encapsulation model;

[0017] Iterative adjustment module: Perform multi-physics co-simulation verification based on the integrated low-noise encapsulation model to generate noise suppression performance evaluation data; when the noise suppression performance evaluation data does not reach the preset threshold, iteratively adjust the parameters of S1 to S4 until the requirements are met.

[0018] The beneficial effects of this invention are as follows: Through the synergistic design of GaN switch transient shaping, three-dimensional magnetic integration layout, piezoelectric active cancellation, and acoustic metamaterial encapsulation, this method reduces audible noise to below 35 dBA while maintaining high power density in the charger, significantly improving the quiet user experience. By precisely controlling the source noise of switching behavior, the coupling interference between mechanical vibration and piezoelectric effect is reduced, effectively avoiding the noise peak problem caused by high-frequency magnetic field resonance in traditional designs. The three-dimensional layout optimization enhances the anti-resonance capability of magnetic components, reduces abnormal noise dispersion caused by structural modal coupling, and makes the noise distribution more uniform and controllable. The piezoelectric active cancellation mechanism directly neutralizes audible noise, reducing the use of passive shielding materials, avoiding both increased volume and potential heat dissipation hazards caused by potting processes. The acoustic metamaterial encapsulation further blocks the noise propagation path, achieving efficient acoustic impedance matching in a confined space. This method can meet the stringent power density requirements of fast charging devices and create a near-silent operating environment, making it particularly suitable for noise-sensitive scenarios such as libraries and conference rooms, providing core technological support for the differentiated competition of high-end consumer electronics products. Attached Figure Description

[0019] Figure 1 This is a diagram illustrating the steps of the method described in this invention;

[0020] Figure 2 This is a system module diagram of the present invention;

[0021] Figure 3 This is a detailed step diagram of S2 described in the present invention. Detailed Implementation

[0022] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0023] One embodiment of the present invention, such as Figure 1 As shown, a low-noise design method for a GaN small-volume charger is disclosed, the method comprising:

[0024] S1. Perform switching transient waveform analysis on the circuit topology of the GaN charger to generate transient noise source distribution data; design a GaN gate drive waveform shaping strategy based on the transient noise source distribution data to generate optimized drive waveform parameters; dynamically modulate the GaN switch based on the optimized drive waveform parameters to generate low-noise switching behavior data.

[0025] S2. Based on low-noise switching behavior data, perform three-dimensional magnetic integration layout design to generate a three-dimensional spatial distribution model of magnetic components; perform high-frequency magnetic field and structural mode coupling analysis on the three-dimensional spatial distribution model of magnetic components to generate mode decoupling parameters; adjust the position and orientation of magnetic components according to the mode decoupling parameters to generate anti-resonance magnetic integration layout data.

[0026] S3. Deploy piezoelectric sensors based on the anti-resonance magnetic integrated layout data to generate a piezoelectric noise sensing network; collect mechanical vibration and piezoelectric effect noise in the audible frequency band through the piezoelectric noise sensing network to generate raw piezoelectric noise signal data; perform phase reversal processing on the raw piezoelectric noise signal data to generate active cancellation waveform parameters; drive the piezoelectric actuator based on the active cancellation waveform parameters to generate reverse noise wave data;

[0027] S4. Design an acoustic metamaterial encapsulation structure based on the reverse noise wave data to generate metamaterial unit geometric parameters; optimize the acoustic impedance matching of the metamaterial unit geometric parameters to generate metamaterial encapsulation layer data; integrate the metamaterial encapsulation layer data with the charger shell structure to generate an integrated low-noise encapsulation model.

[0028] S5. Perform multi-physics co-simulation verification based on the integrated low-noise packaging model to generate noise suppression performance evaluation data; when the noise suppression performance evaluation data does not reach the preset threshold, iteratively adjust the parameters of S1 to S4 until the requirements are met; finally, generate a GaN small-volume charger design scheme that meets the audible noise standard of less than 35dBA.

[0029] The working principle and effects of the above technical solution are as follows: Through multi-stage collaborative noise reduction design, the audible noise of the GaN small-volume charger is effectively reduced to below 35 dBA, improving the accuracy and stability of noise suppression. It enhances the charger's structural anti-resonance capability and reduces coupling interference between high-frequency magnetic fields and structural modes. It avoids noise exceeding standards due to insufficient effectiveness of single noise reduction methods, preventing excessive noise from affecting the user experience. It improves overall noise reduction efficiency through a combination of active cancellation and passive packaging, while also meeting the requirements of small-volume packaging, reducing space occupation, and further enhancing the product's market competitiveness.

[0030] In one embodiment of the present invention, S1 includes:

[0031] S11. Extract the core electrical parameters of the GaN charger circuit topology and generate a basic parameter set for the circuit topology.

[0032] S12. Based on the circuit topology basic parameter set, build a switching transient simulation scenario, collect the voltage and current waveforms of the GaN switch during the entire process of turn-on and turn-off, and generate the original switching transient waveform data.

[0033] S13. Extract noise features and divide frequency bands from the original switching transient waveform data to generate transient noise source distribution data; based on the transient noise source distribution data, prioritize noise suppression and formulate a GaN gate drive waveform shaping strategy.

[0034] S14. Optimize the timing and amplitude parameters of the shaping strategy through multiple rounds of simulation iterations to generate optimized driving waveform parameters;

[0035] S15. The optimized driving waveform parameters are imported into the gate driving circuit to dynamically modulate the gate voltage change rate and on-state voltage drop of the GaN switch. The modulated switch operating status data is collected in real time to generate low-noise switching behavior data.

[0036] The working principle and effects of the above technical solution are as follows: By extracting circuit parameters step by step and capturing switching waveforms through simulation, the accuracy of transient noise source identification is improved, and the noise amplitude during the turn-on and turn-off process of the GaN switch is reduced. Multiple rounds of iterative optimization of driving parameters reduce energy waste caused by blind modulation and enhance the adaptability of the driving waveform to the characteristics of the switch. Dynamically controlling the gate voltage change rate and on-state voltage drop suppresses interference at the noise source, avoiding the instability of noise reduction effect caused by single parameter settings. It can accurately pinpoint noise sources in different frequency bands and adapt to the operating state of the switch in real time, further improving the stability of low-noise switching behavior and laying a reliable foundation for subsequent noise reduction stages.

[0037] In one embodiment of the present invention, S15 includes:

[0038] The optimized drive waveform parameters are imported into the gate drive circuit to complete the drive circuit parameter adaptation and generate drive circuit configuration data.

[0039] Based on the driving circuit configuration data, the gate voltage change rate and on-state voltage drop of the GaN switch are dynamically modulated to generate modulated gate electrical signal data.

[0040] Collect the switching state information of the switching transistor corresponding to the modulated gate electrical signal data, and integrate them to form a modulated switching behavior dataset;

[0041] The noise characteristics of the modulated switching behavior dataset are verified, and switching behavior information that meets the low noise requirements is selected to generate low-noise switching behavior data.

[0042] The working principle and effects of the above technical solution are as follows: By first adapting the driving circuit to optimized parameters, the targeting of the driving signal is improved, avoiding gate modulation failure caused by parameter mismatch. Based on the adapted data, the gate voltage change rate and on-state voltage drop are dynamically adjusted to accurately reduce the switching noise of the GaN switch and reduce energy waste caused by blind modulation. The modulated switching behavior data is collected and integrated, and noise characteristic verification is carried out to enhance the reliability of low-noise switching behavior data and prevent substandard data from entering subsequent stages, which would reduce the overall noise reduction effect. It can both track the working state of the switch in real time to achieve dynamic adaptation modulation and strictly control the quality of output data through verification and screening, laying a solid foundation for the effective implementation of subsequent noise reduction stages.

[0043] One embodiment of the present invention, such as Figure 3 As shown, S2 includes:

[0044] S21. Based on low-noise switching behavior data, clarify the power level and parasitic parameter constraints of magnetic components and generate magnetic component design indicators; select suitable magnetic materials and winding methods according to the design indicators, and construct the preliminary three-dimensional spatial layout of magnetic components.

[0045] S22. Considering the small size packaging limitations of the charger, spatial compression and interference checks are performed on the preliminary layout to generate a three-dimensional spatial distribution model of the magnetic components.

[0046] S23. Build a high-frequency magnetic field simulation module and a structural modal analysis module to conduct multi-field coupling simulation of the three-dimensional spatial distribution model of magnetic components.

[0047] S24. Extract the resonant frequency, magnetic field leakage intensity and structural vibration amplitude data from the coupled simulation, and generate modal decoupling parameters;

[0048] S25. Adjust the installation position and angle orientation of the magnetic components according to the modal decoupling parameters in different regions; perform secondary multi-field coupling verification on the adjusted layout to generate anti-resonance magnetic integrated layout data.

[0049] The working principle and effects of the above technical solution are as follows: By combining low-noise switching behavior data to determine the constraints and selection of magnetic components, the adaptability of magnetic materials and winding methods is improved, reducing magnetic field noise and energy loss caused by improper selection. The layout is optimized and interference is investigated in conjunction with small-volume packaging requirements, reducing the overall space occupied by the charger and avoiding increased vibration noise caused by component interference. Resonance and magnetic field leakage data are extracted through multi-field coupling simulation, improving the accuracy of modal decoupling and enhancing the anti-resonance capability of the magnetic component layout. The position and orientation are adjusted in different regions and secondary verification is conducted to avoid the problem of incomplete single-layout optimization. This approach can both meet the requirements of small-volume design and effectively suppress the coupling interference between high-frequency magnetic fields and structural modes, providing a stable structural foundation for subsequent noise reduction.

[0050] In one embodiment of the present invention, step S21 includes:

[0051] Import low-noise switching behavior data, extract the boundary range of power level correlation parameters and parasitic parameters of magnetic components, and generate a magnetic component constraint dataset;

[0052] Based on the magnetic component constraint dataset, power transmission requirements and noise suppression requirements are integrated to generate magnetic component design indicators;

[0053] By comparing the design specifications of magnetic components, magnetic materials and winding methods that match the high-frequency loss characteristics are selected to generate a material and winding adaptation scheme.

[0054] Based on the material and winding adaptation scheme, and combined with the preliminary planning of the internal space of the charger, a preliminary three-dimensional spatial layout of the magnetic components is constructed.

[0055] The working principle and effects of the above technical solution are as follows: By importing low-noise switching behavior data to extract magnetic component constraint parameters, the accuracy of the constraint dataset is improved, preventing subsequent design specifications from deviating from actual operating conditions. Power transmission requirements and noise suppression requirements are integrated to generate design specifications, reducing performance imbalances caused by a single requirement and ensuring the specifications better align with the overall operating needs of the charger. Materials and winding methods matching high-frequency losses are selected based on the specifications, improving their adaptability to high-frequency operating conditions and reducing energy loss and magnetic field noise during magnetic component operation. An initial layout is constructed based on internal space planning, avoiding the hassle of significant layout adjustments due to space conflicts later. This lays a smooth foundation for subsequent space compression and interference checks, while also addressing the dual requirements of noise reduction performance and small-volume packaging in advance.

[0056] In one embodiment of the present invention, step S22 includes:

[0057] Import the small-volume packaging constraint parameters of the charger, extract the spatial boundary and component spacing requirements, and generate packaging space constraint data;

[0058] Based on the packaging space constraint data and the preliminary three-dimensional spatial layout of magnetic components, a spatial topology optimization algorithm is used to compress the layout and generate preliminary compressed layout data.

[0059] Interference detection between components is performed on the preliminary compressed layout data, the interference area and interference amount are extracted, and the interference inspection results are generated. The positions of the components in the preliminary compressed layout are adjusted according to the interference inspection results, and a three-dimensional spatial distribution model of the magnetic elements is generated.

[0060] The working principle and effects of the above technical solution are as follows: By importing the small-volume packaging constraint parameters of the charger to extract spatial constraints, the accuracy of the packaging space constraint data is improved, preventing subsequent layout compression from deviating from actual packaging requirements. Based on the constraint data and preliminary layout, spatial topology optimization compression is performed to improve space utilization and effectively reduce the overall space occupation of the magnetic component layout, meeting the small-volume design requirements of the charger. Interference detection and position adjustment are performed on the preliminary compressed layout to avoid vibration noise and assembly difficulties caused by interference between components, enhancing the rationality and feasibility of the layout. It can strictly adapt to the small-volume packaging constraints and generate a regular, interference-free three-dimensional spatial distribution model of magnetic components, providing a reliable foundation for the accurate conduct of subsequent multi-field coupled simulations.

[0061] In one embodiment of the present invention, S3 includes:

[0062] S31. Based on the anti-resonance magnetic integrated layout data, identify the concentrated vibration area and noise-sensitive point inside the charger, and determine the deployment nodes of the piezoelectric sensor;

[0063] S32. Fix the piezoelectric sensors according to the deployment nodes and construct a piezoelectric noise sensing network; set the sensor acquisition frequency and range parameters, and simultaneously capture audible mechanical vibration signals and piezoelectric effect noise signals through the piezoelectric noise sensing network.

[0064] S33. Perform time axis alignment and data fusion on the two types of captured signals to generate the original piezoelectric noise signal data; perform filtering, noise reduction and amplitude calibration on the original piezoelectric noise signal data to generate a clean piezoelectric noise signal.

[0065] S34. Calculate the timing offset and amplitude matching parameters required for phase reversal based on the pure piezoelectric noise signal, and generate active cancellation waveform parameters;

[0066] S35. Input the active cancellation waveform parameters into the piezoelectric actuator drive circuit to precisely control the vibration frequency and amplitude of the actuator; drive the piezoelectric actuator to generate a wave with the opposite phase to the noise signal, generating reverse noise wave data.

[0067] The working principle and effects of the above technical solution are as follows: By determining the deployment nodes of the piezoelectric sensor based on the anti-resonance magnetic integrated layout data, the accuracy of sensor deployment is improved, avoiding the problem of incomplete noise capture caused by deployment position deviation. A sensing network is constructed to simultaneously capture two types of signals, enhancing the completeness of noise signal acquisition and reducing the possibility of missing key noise information in single signal capture. Time axis alignment, filtering, and amplitude calibration are performed on the signal to improve the purity of the noise signal and avoid interference from impurity signals in subsequent cancellation parameter calculations. Precise control of the piezoelectric actuator to generate a reverse noise wave effectively reduces mechanical vibration and piezoelectric effect noise in the audible frequency band, avoiding the limitations of passive noise reduction in dealing with dynamic noise. This solution achieves both real-time accurate capture and active cancellation of dynamic noise and synergistic noise reduction with the previous anti-resonance layout, further improving the overall noise suppression efficiency.

[0068] In one embodiment of the present invention, S33 includes:

[0069] Extract the timestamp information of the two types of captured signals to generate a signal timestamp matching dataset; based on the signal timestamp matching dataset, align the two types of captured signals on the time axis to generate aligned dual-channel signal data;

[0070] The aligned dual-channel signal data is weighted and fused to generate the original piezoelectric noise signal data.

[0071] Import the raw piezoelectric noise signal data, analyze the noise frequency band characteristics in the signal, and generate noise frequency band distribution data;

[0072] Based on the noise frequency band distribution data, an adaptive filtering algorithm is used to denoise the original piezoelectric noise signal data to generate denoised signal data; the amplitude reference is calibrated on the denoised signal data to generate a clean piezoelectric noise signal.

[0073] The working principle and effects of the above technical solution are as follows: By extracting the signal timestamp, dual-channel signal alignment is achieved, improving the synchronization of the two types of noise signals and avoiding signal fusion distortion caused by time deviation. Weighted fusion of the aligned signals enhances the integrity of the original piezoelectric noise signal data and reduces information loss due to single signal acquisition. Analysis of noise frequency band characteristics generates distribution data, making filtering more targeted, improving the denoising accuracy of adaptive filtering, and avoiding damage to effective noise signals caused by blind filtering. Amplitude calibration of the denoised signal further improves signal purity and avoids interference from impurity signals in subsequent phase inversion parameter calculations. This solution comprehensively integrates information from both types of noise signals and accurately removes invalid interference, providing reliable data support for generating accurate active cancellation waveform parameters.

[0074] In one embodiment of the present invention, step S4 includes:

[0075] S41. Determine the noise attenuation frequency band and impedance matching target of acoustic metamaterials based on reverse noise wave data, and generate metamaterial design parameters.

[0076] S42. Based on design parameters, select a combination structure of porous media and resonant units, and design the initial geometric dimensions and arrangement of metamaterial units;

[0077] S43. Perform acoustic simulation on the metamaterial unit, analyze the impedance characteristics and attenuation performance under different geometric parameters, and generate the geometric parameters of the metamaterial unit.

[0078] S44. Based on impedance characteristic data, iteratively adjust the unit size and arrangement density to optimize the overall acoustic performance of the metamaterial and generate metamaterial encapsulation layer data.

[0079] S45. Extract the structural parameters and installation interface information of the charger shell, and coordinate the metamaterial encapsulation layer data with the shell structural parameters; optimize the bonding method and fixing structure between the encapsulation layer and the shell, and generate an integrated low-noise encapsulation model.

[0080] The working principle and effects of the above technical solution are as follows: By combining reverse noise wave data to determine the attenuation frequency band and impedance matching target of the metamaterial, the metamaterial design becomes more targeted, avoiding the problem of insufficient noise attenuation caused by blind selection. Selecting suitable porous media and resonant unit combinations and optimizing geometric parameters enhances the metamaterial's attenuation efficiency for target noise, effectively reducing the penetration ability of audible noise. Iterative optimization of unit size and arrangement density through acoustic simulation improves the stability of the overall acoustic performance of the metamaterial, reducing fluctuations in noise reduction effect caused by parameter deviations. Coordinating and optimizing the bonding method between the metamaterial encapsulation layer and the shell structure avoids noise reduction failure caused by assembly gaps, while also meeting the small-volume packaging requirements of the charger. This forms a reliable passive noise reduction barrier and synergizes with the previous active noise cancellation, further improving the overall noise suppression effect and ensuring that the charger's final noise level meets standards.

[0081] In one embodiment of the present invention, S42 includes:

[0082] Import metamaterial design parameters, extract material performance parameters and structural adaptation requirements corresponding to noise attenuation frequency bands, and generate material and structure screening criteria;

[0083] Based on the material and structural selection criteria, suitable porous media types and resonant unit structures are selected to generate a combination scheme of porous media and resonant units.

[0084] Considering the space constraints of the charger packaging, the size constraint range of the metamaterial unit is extracted to generate unit size boundary data; based on the combination scheme and the unit size boundary data, the initial geometric size parameters of the metamaterial unit are set to generate the initial size dataset.

[0085] Based on the initial size dataset, the spatial arrangement logic of the metamaterial units is planned, and the initial arrangement data of the units is generated.

[0086] The working principle and effects of the above technical solution are as follows: By importing metamaterial design indicators to extract corresponding material performance and structural adaptation requirements, the targeting of material and structure selection is improved, avoiding the mismatch between porous media, resonant units, and target noise attenuation frequency bands caused by blind selection. Based on the selection criteria, a combination scheme is determined to enhance the metamaterial's adaptability to target noise attenuation and reduce insufficient noise reduction performance due to improper combination. The unit size boundary is extracted in conjunction with packaging space constraints to avoid the initial size setting exceeding the charger's space range, reducing the trouble of rework and adjustment. The initial size and arrangement logic are set according to the combination scheme and size boundaries to improve the rationality of initial parameters and lay a smooth foundation for subsequent acoustic simulation optimization. This approach can accurately match noise attenuation requirements and adapt to small-volume packaging constraints in advance, making the metamaterial design more aligned with the overall charger design requirements.

[0087] In one embodiment of the present invention, step S5 includes:

[0088] S51. Build a multi-physics field collaborative simulation platform to integrate electromagnetic simulation, structural vibration simulation and acoustic simulation modules;

[0089] S52. Import the parameters of the integrated low-noise package model into the simulation platform and set the input voltage, load conditions and environmental parameters consistent with the actual working scenario.

[0090] S53. Start the simulation platform to conduct noise suppression effectiveness simulation calculations, continuously collect noise amplitude and distribution data during the simulation process, and generate noise suppression effectiveness evaluation data;

[0091] S54. Compare and analyze the noise suppression performance evaluation data with the preset 35dBA standard threshold; when the evaluation data is higher than the threshold, locate the key link affecting noise suppression and adjust the parameters of the corresponding step in reverse.

[0092] S55. Repeat steps S1 to S4 for parameter optimization and simulation verification until the noise suppression performance evaluation data meets the threshold requirements. Integrate all optimized design parameters and verification data to generate a GaN small-volume charger design that meets the audible noise standard of less than 35 dBA.

[0093] The working principle and effects of the above technical solution are as follows: By building a multi-physics field collaborative simulation platform that integrates electromagnetic, structural vibration, and acoustic modules, the comprehensiveness of noise suppression effectiveness evaluation is improved, avoiding evaluation bias caused by the omission of multi-field coupling effects in single simulations. Simulations are conducted using parameters consistent with actual working scenarios, enhancing the authenticity and reliability of evaluation data and avoiding verification failures caused by the disconnect between simulation and actual working conditions. Continuous data collection and comparison with the 35dBA threshold quickly pinpoints key factors affecting noise suppression, reducing the time and cost waste caused by blindly adjusting parameters and preventing ineffective optimization due to a lack of understanding of the root cause. The optimization and verification process is repeated until the target is met, ensuring that the final design consistently meets low-noise standards and preventing rework due to excessive noise after product delivery. This approach comprehensively verifies the synergy of noise reduction effects across multiple stages and ensures the reliability of the design solution through precise reverse optimization, balancing the dual requirements of small size and low noise.

[0094] One embodiment of the present invention, such as Figure 2 As shown, a charger for implementing the low-noise design method of the GaN small-volume charger described above, the charger comprising:

[0095] Data generation module: Performs switching transient waveform analysis on the circuit topology of GaN charger to generate transient noise source distribution data; designs GaN gate drive waveform shaping strategy based on transient noise source distribution data to generate optimized drive waveform parameters; dynamically modulates GaN switching transistors based on optimized drive waveform parameters to generate low-noise switching behavior data;

[0096] Coupling Analysis Module: Based on low-noise switching behavior data, a three-dimensional magnetic integration layout design is performed to generate a three-dimensional spatial distribution model of magnetic components; high-frequency magnetic field and structural mode coupling analysis is performed on the three-dimensional spatial distribution model of magnetic components to generate mode decoupling parameters; the position and orientation of magnetic components are adjusted according to the mode decoupling parameters to generate anti-resonance magnetic integration layout data;

[0097] Parameter-driven module: Deploys piezoelectric sensors based on anti-resonance magnetic integrated layout data to generate a piezoelectric noise sensing network; collects mechanical vibration and piezoelectric effect noise in the audible frequency band through the piezoelectric noise sensing network to generate raw piezoelectric noise signal data; performs phase reversal processing on the raw piezoelectric noise signal data to generate active cancellation waveform parameters; drives the piezoelectric actuator based on the active cancellation waveform parameters to generate reverse noise wave data;

[0098] Structural Fusion Module: Based on the reverse noise wave data, the acoustic metamaterial encapsulation structure is designed, generating the geometric parameters of the metamaterial unit; the acoustic impedance matching of the metamaterial unit geometric parameters is optimized, generating the metamaterial encapsulation layer data; the metamaterial encapsulation layer data is fused with the charger shell structure to generate an integrated low-noise encapsulation model;

[0099] Iterative adjustment module: Based on the integrated low-noise packaging model, multi-physics co-simulation verification is performed to generate noise suppression performance evaluation data; when the noise suppression performance evaluation data does not reach the preset threshold, the parameters of S1 to S4 are iteratively adjusted until the requirements are met; finally, a GaN small-volume charger design scheme that meets the audible noise standard below 35dBA is generated.

[0100] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A low-noise design method for a GaN small-volume charger, characterized in that, The method includes: S1. Perform switching transient waveform analysis on the circuit topology of the GaN charger to generate transient noise source distribution data; design a GaN gate drive waveform shaping strategy based on the transient noise source distribution data to generate optimized drive waveform parameters; dynamically modulate the GaN switch based on the optimized drive waveform parameters to generate low-noise switching behavior data. S2. Based on low-noise switching behavior data, perform three-dimensional magnetic integration layout design to generate a three-dimensional spatial distribution model of magnetic components; perform high-frequency magnetic field and structural mode coupling analysis on the three-dimensional spatial distribution model of magnetic components to generate mode decoupling parameters; adjust the position and orientation of magnetic components according to the mode decoupling parameters to generate anti-resonance magnetic integration layout data. S3. Deploy piezoelectric sensors based on the anti-resonance magnetic integrated layout data to generate a piezoelectric noise sensing network; collect mechanical vibration and piezoelectric effect noise in the audible frequency band through the piezoelectric noise sensing network to generate raw piezoelectric noise signal data; perform phase reversal processing on the raw piezoelectric noise signal data to generate active cancellation waveform parameters; drive the piezoelectric actuator based on the active cancellation waveform parameters to generate reverse noise wave data; S4. Design an acoustic metamaterial encapsulation structure based on the reverse noise wave data to generate metamaterial unit geometric parameters; optimize the acoustic impedance matching of the metamaterial unit geometric parameters to generate metamaterial encapsulation layer data; integrate the metamaterial encapsulation layer data with the charger shell structure to generate an integrated low-noise encapsulation model. S5. Perform multi-physics co-simulation verification based on the integrated low-noise packaging model to generate noise suppression performance evaluation data; when the noise suppression performance evaluation data does not reach the preset threshold, iteratively adjust the parameters of S1 to S4 until the requirements are met.

2. The low-noise design method for a GaN small-volume charger according to claim 1, characterized in that, S1 includes: S11. Extract the core electrical parameters of the GaN charger circuit topology and generate a basic parameter set for the circuit topology. S12. Based on the circuit topology basic parameter set, build a switching transient simulation scenario, collect the voltage and current waveforms of the GaN switch during the entire process of turn-on and turn-off, and generate the original switching transient waveform data. S13. Extract noise features and divide frequency bands from the original switching transient waveform data to generate transient noise source distribution data; based on the transient noise source distribution data, prioritize noise suppression and formulate a GaN gate drive waveform shaping strategy. S14. Optimize the timing and amplitude parameters of the shaping strategy through multiple rounds of simulation iterations to generate optimized driving waveform parameters; S15. The optimized driving waveform parameters are imported into the gate driving circuit to dynamically modulate the gate voltage change rate and on-state voltage drop of the GaN switch. The modulated switch operating status data is collected in real time to generate low-noise switching behavior data.

3. The low-noise design method for a GaN small-volume charger according to claim 2, characterized in that, S15 includes: The optimized drive waveform parameters are imported into the gate drive circuit to complete the drive circuit parameter adaptation and generate drive circuit configuration data. Based on the driving circuit configuration data, the gate voltage change rate and on-state voltage drop of the GaN switch are dynamically modulated to generate modulated gate electrical signal data. Collect the switching state information of the switching transistor corresponding to the modulated gate electrical signal data, and integrate them to form a modulated switching behavior dataset; The noise characteristics of the modulated switching behavior dataset are verified, and switching behavior information that meets the low noise requirements is selected to generate low-noise switching behavior data.

4. The low-noise design method for a GaN small-volume charger according to claim 1, characterized in that, S2 includes: S21. Based on low-noise switching behavior data, clarify the power level and parasitic parameter constraints of magnetic components and generate magnetic component design indicators; select suitable magnetic materials and winding methods according to the design indicators, and construct the preliminary three-dimensional spatial layout of magnetic components. S22. Considering the small size packaging limitations of the charger, spatial compression and interference checks are performed on the preliminary layout to generate a three-dimensional spatial distribution model of the magnetic components. S23. Build a high-frequency magnetic field simulation module and a structural modal analysis module to conduct multi-field coupling simulation of the three-dimensional spatial distribution model of magnetic components. S24. Extract the resonant frequency, magnetic field leakage intensity and structural vibration amplitude data from the coupled simulation, and generate modal decoupling parameters; S25. Adjust the installation position and angle orientation of the magnetic components according to the modal decoupling parameters in different regions; perform secondary multi-field coupling verification on the adjusted layout to generate anti-resonance magnetic integrated layout data.

5. The low-noise design method for a GaN small-volume charger according to claim 4, characterized in that, S21 includes: Import low-noise switching behavior data, extract the boundary range of power level correlation parameters and parasitic parameters of magnetic components, and generate a magnetic component constraint dataset; Based on the magnetic component constraint dataset, power transmission requirements and noise suppression requirements are integrated to generate magnetic component design indicators; By comparing the design specifications of magnetic components, magnetic materials and winding methods that match the high-frequency loss characteristics are selected to generate a material and winding adaptation scheme. Based on the material and winding adaptation scheme, and combined with the preliminary planning of the internal space of the charger, a preliminary three-dimensional spatial layout of the magnetic components is constructed.

6. The low-noise design method for a GaN small-volume charger according to claim 1, characterized in that, The S3 includes: S31. Based on the anti-resonance magnetic integrated layout data, identify the concentrated vibration area and noise-sensitive point inside the charger, and determine the deployment nodes of the piezoelectric sensor; S32. Fix the piezoelectric sensors according to the deployment nodes and construct a piezoelectric noise sensing network; set the sensor acquisition frequency and range parameters, and simultaneously capture audible mechanical vibration signals and piezoelectric effect noise signals through the piezoelectric noise sensing network. S33. Perform time axis alignment and data fusion on the two types of captured signals to generate the original piezoelectric noise signal data; perform filtering, noise reduction and amplitude calibration on the original piezoelectric noise signal data to generate a clean piezoelectric noise signal. S34. Calculate the timing offset and amplitude matching parameters required for phase reversal based on the pure piezoelectric noise signal, and generate active cancellation waveform parameters; S35. Input the active cancellation waveform parameters into the piezoelectric actuator drive circuit to precisely control the vibration frequency and amplitude of the actuator; drive the piezoelectric actuator to generate a wave with the opposite phase to the noise signal, generating reverse noise wave data.

7. The low-noise design method for a GaN small-volume charger according to claim 6, characterized in that, S33 includes: Extract the timestamp information of the two types of captured signals to generate a signal timestamp matching dataset; based on the signal timestamp matching dataset, align the two types of captured signals on the time axis to generate aligned dual-channel signal data; The aligned dual-channel signal data is weighted and fused to generate the original piezoelectric noise signal data. Import the raw piezoelectric noise signal data, analyze the noise frequency band characteristics in the signal, and generate noise frequency band distribution data; Based on the noise frequency band distribution data, an adaptive filtering algorithm is used to denoise the original piezoelectric noise signal data to generate denoised signal data; the amplitude reference is calibrated on the denoised signal data to generate a clean piezoelectric noise signal.

8. The low-noise design method for a GaN small-volume charger according to claim 1, characterized in that, The S4 includes: S41. Determine the noise attenuation frequency band and impedance matching target of acoustic metamaterials based on reverse noise wave data, and generate metamaterial design parameters. S42. Based on design parameters, select a combination structure of porous media and resonant units, and design the initial geometric dimensions and arrangement of metamaterial units; S43. Perform acoustic simulation on the metamaterial unit, analyze the impedance characteristics and attenuation performance under different geometric parameters, and generate the geometric parameters of the metamaterial unit. S44. Based on impedance characteristic data, iteratively adjust the unit size and arrangement density to optimize the overall acoustic performance of the metamaterial and generate metamaterial encapsulation layer data. S45. Extract the structural parameters and installation interface information of the charger shell, and coordinate the metamaterial encapsulation layer data with the shell structural parameters; optimize the bonding method and fixing structure between the encapsulation layer and the shell, and generate an integrated low-noise encapsulation model.

9. The low-noise design method for a GaN small-volume charger according to claim 1, characterized in that, The S5 includes: S51. Build a multi-physics field collaborative simulation platform to integrate electromagnetic simulation, structural vibration simulation and acoustic simulation modules; S52. Import the parameters of the integrated low-noise package model into the simulation platform and set the input voltage, load conditions and environmental parameters consistent with the actual working scenario. S53. Start the simulation platform to conduct noise suppression effectiveness simulation calculations, continuously collect noise amplitude and distribution data during the simulation process, and generate noise suppression effectiveness evaluation data; S54. Compare and analyze the noise suppression performance evaluation data with the preset 35dBA standard threshold; when the evaluation data is higher than the threshold, locate the key link affecting noise suppression and adjust the parameters of the corresponding step in reverse. S55. Repeat steps S1 to S4 for parameter optimization and simulation verification until the noise suppression performance evaluation data meets the threshold requirements. Integrate all optimized design parameters and verification data to generate a GaN small-volume charger design scheme.

10. A charger for implementing the low-noise design method of the GaN small-volume charger as described in claim 1, characterized in that, The charger includes: Data generation module: Performs switching transient waveform analysis on the circuit topology of GaN charger to generate transient noise source distribution data; designs GaN gate drive waveform shaping strategy based on transient noise source distribution data to generate optimized drive waveform parameters; dynamically modulates GaN switching transistors based on optimized drive waveform parameters to generate low-noise switching behavior data; Coupling Analysis Module: Based on low-noise switching behavior data, a three-dimensional magnetic integration layout design is performed to generate a three-dimensional spatial distribution model of magnetic components; high-frequency magnetic field and structural mode coupling analysis is performed on the three-dimensional spatial distribution model of magnetic components to generate mode decoupling parameters; the position and orientation of magnetic components are adjusted according to the mode decoupling parameters to generate anti-resonance magnetic integration layout data; Parameter-driven module: Deploys piezoelectric sensors based on anti-resonance magnetic integrated layout data to generate a piezoelectric noise sensing network; collects mechanical vibration and piezoelectric effect noise in the audible frequency band through the piezoelectric noise sensing network to generate raw piezoelectric noise signal data; performs phase reversal processing on the raw piezoelectric noise signal data to generate active cancellation waveform parameters; drives the piezoelectric actuator based on the active cancellation waveform parameters to generate reverse noise wave data; Structural Fusion Module: Based on the reverse noise wave data, the acoustic metamaterial encapsulation structure is designed, generating the geometric parameters of the metamaterial unit; the acoustic impedance matching of the metamaterial unit geometric parameters is optimized, generating the metamaterial encapsulation layer data; the metamaterial encapsulation layer data is fused with the charger shell structure to generate an integrated low-noise encapsulation model; Iterative adjustment module: Perform multi-physics co-simulation verification based on the integrated low-noise encapsulation model to generate noise suppression performance evaluation data; when the noise suppression performance evaluation data does not reach the preset threshold, iteratively adjust the parameters of S1 to S4 until the requirements are met.

Images