Adaptive directional active noise reduction device and method for phase modifier based on multi-modal perception

Abstract

The embodiment of the invention provides a phase modifier adaptive directional active noise reduction device and method based on multi-modal sensing, and the device comprises a multi-modal sensing unit which is used for collecting a noise signal and a vibration signal of a phase modifier; the signal processing and control unit is connected with the multi-mode sensing unit and is used for receiving and processing the noise signal and the vibration signal and generating a control instruction; the directional execution unit is connected with the signal processing and control unit and used for emitting anti-phase noise reduction sound waves based on the control instruction; and the feedback optimization unit is connected with the signal processing and control unit and is used for collecting the residual noise signal and feeding back the residual noise signal to the signal processing and control unit for dynamic optimization. According to the technical scheme provided by the invention, efficient, accurate and self-adaptive control on the noise of the phase modifier can be realized.

Description

Technical Field

[0001] This invention belongs to the field of noise control technology, and particularly relates to an adaptive directional active noise reduction device and method for camera adjustment based on multimodal sensing. Background Technology

[0002] Synchronous condensers, as key equipment in power systems, generate significant noise during operation, primarily from electromagnetic excitation and the aerodynamic effects of rotor rotation. This broadband noise not only pollutes the plant environment and affects personnel health but also poses a potential threat to the safe operation of the equipment. Traditional noise reduction methods struggle to achieve efficient, precise, and adaptive noise control while ensuring the normal operation of the equipment.

[0003] In existing technologies, noise control of synchronous condensers mainly relies on passive sound insulation and basic active noise reduction. Passive sound insulation methods block noise propagation by adding soundproof enclosures or sound-absorbing materials, but they have inherent drawbacks such as large size, impaired heat dissipation, and poor suppression of low-frequency noise. Basic active noise reduction technology typically uses only a single type of acoustic sensor, which cannot accurately distinguish complex mixed noise sources, resulting in a lack of targeted noise reduction strategies. In addition, existing active noise reduction systems lack real-time linkage with the operating conditions of the synchronous condenser, and cannot dynamically adjust according to changes in equipment load and speed, resulting in unstable noise reduction effects. At the same time, the energy directivity of the noise reduction waves is poor, often causing energy waste or even interference with the sound field in non-target areas, resulting in low energy efficiency and a lack of closed-loop control capabilities for self-optimization based on noise reduction effects. Summary of the Invention

[0004] The adaptive directional active noise reduction device and method for camera condensers based on multimodal perception provided in this application embodiment can achieve efficient, accurate and adaptive control of camera condenser noise.

[0005] In a first aspect, embodiments of this application provide a camera-adaptive directional active noise reduction device based on multimodal perception, comprising: A multimodal sensing unit is used to acquire noise and vibration signals from a synchronous condenser. The signal processing and control unit, connected to the multimodal sensing unit, is used to receive and process noise and vibration signals and generate control commands. The directional execution unit, connected to the signal processing and control unit, is used to transmit inverted noise-reducing waves based on control commands; The feedback optimization unit, connected to the signal processing and control unit, is used to collect residual noise signals and feed them back to the signal processing and control unit for dynamic optimization.

[0006] In one alternative implementation, the multimodal sensing unit includes a broadband acoustic sensing array and a vibration sensing subunit; the broadband acoustic sensing array includes multiple broadband acoustic sensors arranged circumferentially on the housing of the condenser; the vibration sensing subunit includes a vibration sensor mounted on the stator or bearing housing of the condenser.

[0007] In one alternative implementation, the signal processing and control unit includes a noise feature analysis module; the noise feature analysis module is configured to perform frequency domain analysis on noise signals acquired by a broadband acoustic sensor and to perform correlation analysis in conjunction with vibration signals acquired by a vibration sensor to identify the dominant noise source and its type.

[0008] In one optional implementation, the signal processing and control unit further includes an operating condition linkage module; the operating condition linkage module communicates with the synchronous condenser control system via an industrial bus to obtain the real-time operating parameters of the synchronous condenser and establish a mapping model between the operating parameters and noise characteristics.

[0009] In one alternative implementation, the signal processing and control unit further includes an inverse signal generation module for generating an inverse noise reduction signal with the same amplitude but opposite phase as the original noise.

[0010] In one alternative implementation, the directional execution unit includes an array-type sound-generating assembly and a direction adjustment mechanism; the array-type sound-generating assembly includes a plurality of broadband loudspeakers arranged corresponding to broadband acoustic sensors; the direction adjustment mechanism is connected to at least one broadband loudspeaker for adjusting the direction of the sound waves emitted by the broadband loudspeaker.

[0011] In one alternative implementation, the direction adjustment mechanism includes a stepper motor and an acoustic reflector driven by the stepper motor; the acoustic reflector is disposed in the sound wave emission path of the broadband loudspeaker.

[0012] In one optional implementation, the feedback optimization unit includes at least one error sensor disposed in the noise reduction target area for collecting residual noise signals after the directional execution unit has worked.

[0013] In one alternative implementation, the signal processing and control unit is configured to receive a residual noise signal and dynamically adjust the phase and amplitude of the inverted noise-reducing signal sent to the directional execution unit based on the residual noise signal, while adjusting the output power of the directional execution unit according to the intensity of the residual noise signal.

[0014] Secondly, embodiments of this application provide a camera-adaptive directional active noise reduction method based on multimodal perception, including: Simultaneously collect noise signals and vibration signals from different positions of the synchronous condenser; By jointly analyzing noise and vibration signals, the dominant noise source is identified, and corresponding anti-phase noise reduction signals and control commands are generated. Based on control commands, an anti-phase noise reduction signal is emitted in the direction of the dominant noise source; The residual noise signal after denoising is collected, and the inverse denoising signal and transmission strategy are dynamically optimized based on the residual noise signal.

[0015] Thirdly, embodiments of this application provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements the method provided in embodiments of this application.

[0016] Fourthly, embodiments of this application provide a computer-readable storage medium, characterized in that it stores a computer program thereon, which, when executed in a computer, causes the computer to perform the method provided in embodiments of this application.

[0017] The technical solution provided in this application has the following beneficial effects: This invention utilizes a multimodal sensing unit to comprehensively leverage broadband acoustic sensors arranged around the circumference of the camera condenser and vibration sensors installed at key parts of the equipment. Based on the principles of multi-source information fusion and correlation analysis, it achieves accurate identification and location of different noise sources, such as electromagnetic noise and mechanical noise, overcoming the limitations of single acoustic sensing and laying a reliable data foundation for subsequent precise noise reduction. Relying on a signal processing and control unit, it innovatively combines noise feature analysis, real-time equipment operating parameters, and advanced adaptive filtering algorithms. Based on the operating condition-noise feature mapping model and the principle of an improved adaptive algorithm, the system can dynamically... The dynamic adjustment noise reduction strategy significantly improves the system's noise reduction adaptability and convergence speed under different operating conditions, achieving a leap from "static noise reduction" to "dynamic intelligent noise reduction." Through the collaborative work of the directional execution unit and the feedback optimization unit, the system utilizes arrayed loudspeakers and directionally adjustable acoustic reflectors to focus sound wave energy. Based on the feedback signal from the error sensor, the noise reduction signal is finely adjusted in real time. Based on the principles of acoustic wave interference cancellation and closed-loop control, this not only greatly improves the noise reduction efficiency in the target area and avoids energy dispersion, but also achieves a dynamic balance between noise reduction effect and energy consumption, significantly improving the overall energy efficiency and long-term stability of the system. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the structure of the adaptive directional active noise reduction device for camera adjustment based on multimodal perception provided in the embodiments of this application; Figure 2 This is a flowchart illustrating the adaptive directional active noise reduction method for camera adjustment based on multimodal perception provided in this application embodiment. Detailed Implementation

[0019] The present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] This application provides a camera-adaptive directional active noise reduction device based on multimodal sensing. Figure 1 This is a schematic diagram of the adaptive directional active noise reduction device for camera adjustment based on multimodal perception provided in the embodiments of this application, as shown below. Figure 1 As shown, the technical solution provided in this application includes the following modules: A multimodal sensing unit is used to acquire noise and vibration signals from a synchronous condenser. The signal processing and control unit, connected to the multimodal sensing unit, is used to receive and process noise and vibration signals and generate control commands. The directional execution unit, connected to the signal processing and control unit, is used to transmit inverted noise-reducing waves based on control commands; The feedback optimization unit, connected to the signal processing and control unit, is used to collect residual noise signals and feed them back to the signal processing and control unit for dynamic optimization.

[0021] In some embodiments, the multimodal sensing unit includes a broadband acoustic sensing array and a vibration sensing subunit.

[0022] The broadband acoustic sensing array consists of multiple broadband acoustic sensors, exemplarily at least six, and these sensors have broadband frequency response characteristics, capable of covering the main noise frequency range generated during the operation of the camera condenser (e.g., frequency response 20Hz-10kHz). These sensors are mounted on the housing of the camera condenser in a circumferential arrangement. This surround layout aims to achieve omnidirectional spatial acquisition of camera condenser noise, ensuring effective sensing regardless of the direction from which the noise primarily originates.

[0023] All sensors acquire data synchronously, allowing the acquisition of noise signals from different spatial locations at the same time. This is crucial for subsequent analysis of the spatial distribution characteristics of the noise and for pinpointing the specific location of the dominant noise source. For example, by comparing the strength and phase difference of the same noise signal received by different sensors, the direction of the noise source can be determined.

[0024] Wideband acoustic sensing arrays provide information on the sound pressure level, frequency, and spatial orientation of noise.

[0025] The vibration sensing subunit contains at least one vibration sensor, whose installation location is based on the consideration that the noise source and the vibration source are of the same origin, and is located on the key mechanical structure of the synchronous condenser, such as the stator (core electromagnetic component) or the bearing housing (core rotating support component).

[0026] Vibration sensors directly collect mechanical vibration signals generated by electromagnetic forces and mechanical imbalance forces within the synchronous condenser. These vibrations are the physical source of noise. Since different types of faults or operating conditions produce vibrations with different characteristics (e.g., electromagnetic noise is related to the power grid frequency and its harmonics, and rotor aerodynamic noise is related to the rotational speed's multiples), by analyzing the frequency characteristics of the vibration signals and performing correlation analysis with acoustic signals, the physical type of the dominant noise source can be effectively distinguished and identified. When the vibration signal is highly correlated with a noise signal of a specific frequency, it can be determined that the noise originates from that vibration location.

[0027] The vibration sensing subunit links sound signals with physical vibrations, solving the problem that it is difficult to accurately determine whether noise is electromagnetic, mechanical, or aerodynamic noise by relying solely on sound, thus enabling precise source tracing of noise.

[0028] The acoustic array is responsible for depicting the external manifestations of noise; the vibrating sub-unit is responsible for revealing the internal causes of noise. This multimodal perception fusion is one of the prerequisites and core foundations for the entire device to achieve adaptive, directional, and precise noise reduction.

[0029] In some embodiments, the signal processing and control unit includes a noise feature analysis module; the noise feature analysis module is configured to perform frequency domain analysis on noise signals acquired by broadband acoustic sensors and correlation analysis in conjunction with vibration signals acquired by vibration sensors to identify the dominant noise source and its type.

[0030] The noise feature analysis module first performs a frequency domain transformation on the time-domain noise signal acquired by the broadband acoustic sensor. This can be achieved using digital signal processing algorithms such as Fourier Transform (FFT) based on a multimodal sensing-based adaptive directional active noise reduction device or wavelet analysis. Through this transformation, the module can extract the frequency components of the noise signal, identifying the main frequencies contained in the noise (such as the 50Hz power frequency, 100Hz, 150Hz, etc. harmonics), as well as the sound pressure level (intensity) and spatial distribution characteristics of each frequency component.

[0031] While obtaining the noise spectrum, the noise feature analysis module processes the vibration signal collected by the vibration sensor in parallel (usually also performs frequency domain analysis), and then performs correlation analysis to determine the strength of the correlation between noise and vibration.

[0032] Correlation analysis is used to measure the degree of linear correlation between two signals. For example, the correlation coefficient between the amplitude sequence of a noise component at a specific frequency (such as 150Hz noise) and the amplitude sequence of a vibration component at the same frequency can be calculated. When the correlation between the vibration signal and the noise signal at a specific frequency exceeds a preset threshold, it can be determined that the noise at that frequency mainly originates from this vibration location. For example, the preset threshold can be set to 0.8, but it can be set according to the actual situation.

[0033] For example, if a 150Hz noise is highly correlated with a 150Hz vibration at the bearing housing, the noise is determined to be mechanical noise caused by rotor rotation (such as rotor aerodynamic noise); if a 100Hz noise is highly correlated with electromagnetic vibration, it may be determined to be electromagnetic noise.

[0034] The noise feature analysis module integrates the analysis results of the above two steps to make a comprehensive decision and finally outputs the judgment of the dominant noise source at the current moment, which may include: the type of dominant noise source (e.g., rotor aerodynamic noise is dominant), the location of dominant noise source (located by the signal difference of each sensor in the acoustic array), and the precise frequency of dominant noise.

[0035] In some embodiments, the signal processing and control unit further includes a working condition linkage module; the working condition linkage module communicates with the synchronous condenser control system via an industrial bus to obtain the real-time operating parameters of the synchronous condenser and establish a mapping model between the operating parameters and noise characteristics.

[0036] The operational linkage module establishes a communication connection with the core control system of the synchronous condenser via an industrial bus, such as RS485 or Ethernet. Specifically, the Modbus RTU protocol can be used. This is a standard communication protocol widely used in industrial control, ensuring the reliability and real-time performance of data exchange.

[0037] Through this industrial bus link, the operational status linkage module can acquire key operating parameters of the synchronous condenser in real time. These parameters mainly include: Rotational speed: The rotational speed of the phase converter rotor directly affects the frequency of aerodynamic and mechanical noise; Load: The amount of reactive power output by the synchronous condenser reflects its electromagnetic load level and is closely related to the intensity of electromagnetic noise.

[0038] The core of the operating condition linkage module is to build and apply an operating condition-noise feature mapping model. This model is essentially a knowledge base or algorithm model based on a large amount of experimental or historical operating data. It describes the correspondence between different operating conditions (combinations of speed and load) and typical noise characteristics (dominant noise frequency, sound pressure level).

[0039] During system debugging or learning, this mapping model is established by analyzing noise data collected under different speeds and loads. For example, the model records that when the load reaches a high range, electromagnetic noise in a specific frequency band (e.g., 50-200Hz) is usually significantly enhanced. In real-time operation, once the operating condition linkage module obtains the current speed and load values ​​from the synchronous condenser control system, it immediately queries the mapping model to predict the most likely prominent noise characteristics under the current operating condition. Based on the prediction results of the mapping model, the operating condition linkage module generates control commands to dynamically adjust the noise reduction strategy of the signal processing and control unit. For example, when the module determines that the current synchronous condenser load is in a high-load state (e.g., ≥80%), it instructs the noise feature analysis module and the inverted signal generation module to increase the noise reduction weight of the frequency band strongly correlated with this operating condition (e.g., 50-200Hz).

[0040] Noise reduction weights can be understood as the computational resources and output priorities allocated by the system when processing multiple noise sources. Increasing the weights means that the system will invest more resources in generating and optimizing the inverted signal for that frequency band, ensuring that it is suppressed more effectively.

[0041] In some embodiments, the signal processing and control unit further includes an inverse signal generation module for generating an inverse noise reduction signal with the same amplitude but opposite phase as the original noise.

[0042] The core function of the reverse signal generation module is to generate an anti-phase noise reduction signal with the same amplitude but opposite phase as the original noise signal. This goal is based on the physical principle of sound wave interference: when two sound waves with the same frequency and amplitude but a phase difference of 180 degrees meet in space, they will cancel each other out due to complete destructive interference, thereby stopping the vibration of air particles at that location and achieving the purpose of noise reduction.

[0043] To achieve this goal, the reverse signal generation module can employ adaptive filtering algorithms such as the Least Mean Square Error (LMS) algorithm and its improved version (the improved LMS algorithm has a convergence speed that is twice that of the traditional algorithm), as well as the Recursive Least Squares (RLS) algorithm.

[0044] The inverse signal generation module receives instructions from the noise feature analysis module to determine which dominant noise frequency(s) need to be generated with an inverse signal; it runs an adaptive algorithm to calculate and generate a noise reduction signal with accurate inverse characteristics in real time using the acquired noise reference signal; the generated inverse noise reduction signal (electrical signal) is sent to the directional execution unit to drive the corresponding speaker to emit sound waves.

[0045] In some embodiments, the directional execution unit includes an array-type sound-generating assembly and a direction adjustment mechanism; the array-type sound-generating assembly includes a plurality of broadband loudspeakers arranged corresponding to broadband acoustic sensors; the direction adjustment mechanism is connected to at least one broadband loudspeaker for adjusting the direction of the sound waves emitted by the broadband loudspeaker.

[0046] The array-type sound-generating assembly includes multiple wideband loudspeakers, exemplarily six in number, each with a wideband frequency response, such as 20Hz-15kHz, capable of covering the noise frequency range to be canceled, ensuring accurate reproduction of the inverse noise-canceling signal generated by the control unit. These loudspeakers are arranged in a spatial one-to-one correspondence with the wideband acoustic sensor array in the multimodal sensing unit. For example, each loudspeaker faces one or more acoustic sensors in a specific direction. This one-to-one correspondence allows the signal processing and control unit to selectively drive one or more loudspeakers in a specific direction based on the noise source localization results. When a certain direction (such as the driver end of a phase shifter) is identified as the dominant noise source, the system primarily activates the loudspeakers in that direction, rather than having all loudspeakers operate equally. This arrangement achieves initial spatial selectivity in noise reduction, laying the foundation for subsequent, more refined sound energy focusing and avoiding ineffective dissipation of noise reduction energy in non-target areas.

[0047] The directional adjustment mechanism enhances the functionality of the array-type sound-emitting components, further optimizing the propagation direction of the noise-reducing waves for more precise "direction." This mechanism connects to at least one wideband loudspeaker and physically adjusts the main axis direction of the sound wave beam emitted by that loudspeaker. By fine-tuning the radiation direction of the sound waves, the noise-reducing waves can be more concentratedly projected and cover identified dominant noise source areas or specific spaces requiring focused noise reduction (such as maintenance personnel passageways). Acoustically, this significantly enhances the noise reduction pressure level in the target area, improving noise reduction efficiency.

[0048] The array-type sound-generating components and the direction adjustment mechanism together constitute an acoustic execution system that enables precise spatial intervention, ensuring that the noise reduction energy can be efficiently applied to the target point. This is one of the core technical means to achieve efficient and energy-saving noise reduction.

[0049] Furthermore, the direction adjustment mechanism includes a stepper motor and an acoustic reflector driven by the stepper motor; the acoustic reflector is positioned in the sound wave emission path of the broadband loudspeaker.

[0050] The direction adjustment mechanism consists of a stepper motor and an acoustic reflector driven by the stepper motor, which work together to achieve precise control of the sound wave direction. A stepper motor is an actuator that precisely converts electrical pulse signals into angular displacement, enabling precise open-loop position control. Each input pulse rotates the motor shaft by a fixed angle (i.e., the step angle, exemplarily 0.9°), thus allowing for precise control of the rotation angle. The stepper motor acts as a drive source, receiving a sequence of pulses representing the target angle from a signal processing and control unit. For example, if the control unit calculates that a sound wave needs to be redirected by 30 degrees, it will send the corresponding number of pulses to the stepper motor.

[0051] An acoustic reflector is a physical component with a specific acoustic reflective surface. It is positioned in the sound wave emission path of a broadband loudspeaker, and its position and orientation are directly controlled by a stepper motor. The adjustable angle range of the acoustic reflector is 0-90°, with a response time ≤500ms. Its working principle is based on the law of sound wave reflection. When the noise-reducing wave emitted by the broadband loudspeaker propagates to the surface of the acoustic reflector, it is reflected. By precisely changing the angle of the acoustic reflector through the stepper motor, the reflection path of the sound wave can be altered, thus effectively changing the direction in which the final noise-reducing wave propagates into space. This is similar to using a mirror to change the direction of light.

[0052] Collaborative workflow: The signal processing and control unit calculates the target angle that the acoustic reflector needs to rotate based on the noise source location determined by the multimodal sensing unit.

[0053] The control unit sends a corresponding pulse signal to the stepper motor, driving the stepper motor to rotate, which in turn drives the acoustic reflector mechanically connected to it to rotate to the target angle.

[0054] The wideband loudspeaker continuously emits anti-phase noise reduction waves. After being reflected by an acoustic reflector at a fixed angle, these waves are directed to the location of the dominant noise source, focusing the noise reduction energy and preventing energy dispersion.

[0055] In some embodiments, the feedback optimization unit includes at least one error sensor disposed in the noise reduction target area for collecting residual noise signals after the directional execution unit has worked.

[0056] The core function of the error sensor is to collect the residual noise signal after the directional execution unit has worked. That is, it does not measure the original environmental noise, but the noise level that remains in a specific area after intervention by the active noise reduction system.

[0057] Error sensors are deployed in key noise reduction areas (such as near the maintenance corridor 3 meters from the synchronous condenser) to collect residual noise signals in real time. The choice of location is based on the ultimate purpose of the noise reduction device, which is usually to protect personnel or equipment in a specific area. For example, it could provide maintenance personnel with an operating environment below harmful noise levels or protect noise-sensitive critical auxiliary equipment. Therefore, placing error sensors directly in these critical areas that need protection will most accurately and directly reflect the actual effect of the noise reduction system at that location.

[0058] Technical benefits: This deployment strategy ensures that the system optimization goals align with the final application objectives. The direct basis for system optimization is the residual noise in the target area, thus guaranteeing the effectiveness and relevance of the optimization actions.

[0059] The residual noise signal collected by the error sensor is used as the feedback signal for the entire adaptive active noise reduction closed-loop control system.

[0060] Workflow: The system generates an inverse noise reduction signal based on the input of the multimodal sensing unit and sends it out through the directional execution unit (which is the forward control channel of the system).

[0061] The error sensor then measures the sound pressure in the target area where it is located to obtain the residual noise signal.

[0062] The residual noise signal is fed back to the signal processing and control unit in real time.

[0063] By introducing an error sensor, the system can provide objective and quantitative basis for subsequent dynamic corrections, enabling the system to have the ability to self-monitor and self-adjust, ensuring that the noise reduction device can operate stably for a long time and adapt to changes in working conditions.

[0064] In some embodiments, the signal processing and control unit is configured to receive a residual noise signal and dynamically adjust the phase and amplitude of the inverted noise-reducing signal sent to the directional execution unit based on the residual noise signal, while adjusting the output power of the directional execution unit according to the intensity of the residual noise signal.

[0065] The signal processing and control unit continuously receives residual noise signals from the error sensor. Internally, the signal processing and control unit compares this signal with the desired noise reduction target in real time, calculating the error value of the residual noise, especially for the identified dominant noise frequency components.

[0066] Based on this error value, the signal processing and control unit uses an adaptive algorithm to perform calculations, iteratively minimizing the error. The algorithm dynamically adjusts the phase of the inverted noise reduction signal sent to the directional execution unit (accuracy ±1°). This is because the effectiveness of acoustic interference cancellation is highly dependent on whether the phase difference between the inverted wave and the original noise wave is precisely maintained at 180 degrees. Any tiny deviation will lead to a decrease in the cancellation effect. Through feedback, the system can compensate in real time for phase drift caused by changes in ambient temperature and humidity or slight equipment displacement, ensuring the accuracy of the inversion. At the same time, the algorithm also dynamically adjusts the amplitude (gain) of the inverted signal. The goal is to precisely match the sound pressure level of the inverted noise reduction wave with the sound pressure level of the original noise. If the amplitude is too low, the cancellation will be insufficient; if it is too high, it will overcompensate, generating new noise.

[0067] This continuous fine-tuning of phase and amplitude ensures that the noise reduction system can always lock onto the noise target and maintain the best noise cancellation effect in changing environments, achieving adaptive tracking and steady-state high-precision noise reduction.

[0068] The signal processing and control unit simultaneously monitors the overall intensity of the residual noise signal (typically assessed as A-weighted sound pressure level). When the system determines that the noise reduction effect has reached a very ideal level, i.e., the residual noise intensity is lower than or equal to a preset decibel threshold, the signal processing and control unit will activate the energy efficiency management strategy. At this time, the signal processing and control unit will automatically adjust the output power of the directional actuator (i.e., the wideband loudspeaker), appropriately reducing it from full-load operation. For example, when the residual noise is ≤65dB(A), the loudspeaker output power will be automatically reduced to 30%-50% of the rated power, achieving a balance between energy saving and noise reduction.

[0069] This mechanism achieves an intelligent balance between noise reduction effect and energy consumption. When the noise is high, the system works at full capacity to quickly suppress the noise; when the noise has been effectively suppressed, the system automatically enters a low-power maintenance state, thereby significantly reducing the energy consumption of the system during long-term operation, demonstrating the intelligence and economy of the device.

[0070] Figure 2 This is a flowchart illustrating the adaptive directional active noise reduction method for camera adjustment based on multimodal sensing provided in this application. The method can be executed by a multimodal sensing-based adaptive directional active noise reduction device for camera adjustment. This device can be implemented by software and / or hardware and can be configured in electronic devices such as computers. See [link to relevant documentation]. Figure 2 The method includes: S110, synchronously acquires noise signals and vibration signals of the synchronous condenser from different positions.

[0071] Specifically, a broadband acoustic sensing array arranged circumferentially along the housing of the camera condenser is used to synchronously collect noise signals from different locations of the camera condenser at a preset frequency (e.g., once every 10ms); vibration sensors synchronously collect equipment vibration data.

[0072] The acquired analog signals are digitized after being converted to digital form by a 16-bit A / D converter and then transmitted to the signal processing and control unit for further analysis. The signal processing and control unit uses an STM32H743 microcontroller with a main frequency of 400MHz and an integrated FFT hardware accelerator, achieving a processing speed of ≥1024 points / 1ms, ensuring the real-time performance of subsequent complex signal processing.

[0073] This step enables multimodal and multi-directional synchronous perception of noise and its physical sources, providing a comprehensive, synchronous, and spatially correlated raw data foundation for subsequent joint analysis, ensuring the accuracy and timeliness of the analysis.

[0074] S120: Based on the joint analysis of noise and vibration signals, identify the dominant noise source and generate the corresponding anti-phase noise reduction signal and control command.

[0075] Specifically, firstly, the noise signal undergoes frequency domain transformation (such as FFT) to extract its frequency components and spatial distribution characteristics, identifying the dominant noise frequency, such as the 50Hz power frequency or the 150Hz rotor harmonic frequency. Correlation analysis is then performed on the vibration signal to determine the noise source type. By considering the sound pressure level of the noise and its correlation with vibration, the most significant and urgent noise source to be identified, and its dominant frequency and spatial location are determined. For example, when the correlation between the vibration signal and the 150Hz noise is greater than 0.8, it is determined to be dominated by rotor aerodynamic noise, achieving precise source tracing.

[0076] The system invokes a pre-stored operating condition-noise feature mapping model. For example, if the current synchronous condenser load is 80%, the system automatically increases the noise reduction weight of the 150Hz frequency band to 1.5 times the default value, making the noise reduction strategy adaptively linked with the equipment operating status.

[0077] Based on the recognition results, an adaptive filtering algorithm (such as the LMS algorithm) is used to generate an inverse noise reduction signal with the same amplitude but opposite phase as the original dominant noise. At the same time, control commands are generated according to the location of the noise source to control the directional execution unit (such as selecting a specific speaker or adjusting the reflector angle).

[0078] S130: Based on control commands, transmit an anti-phase noise reduction signal in the direction of the dominant noise source.

[0079] Specifically, the directional execution unit, based on the received control commands, drives one or more wideband speakers corresponding to the noise source location to convert the inverted noise reduction signal in electrical form into actual sound waves for transmission. Simultaneously, the control direction adjustment mechanism (such as an acoustic reflector driven by a stepper motor) rotates to the optimal angle (e.g., 30° for rotor-side noise), focusing the energy of the noise reduction wave onto the identified dominant noise source area, thereby focusing the noise reduction energy, enhancing the cancellation effect at the noise source location, and preventing energy dispersion.

[0080] S140. Collect the residual noise signal after noise reduction, and dynamically optimize the inverse noise reduction signal and transmission strategy based on the residual noise signal.

[0081] Specifically, the feedback optimization unit uses error sensors deployed in the noise reduction target area to collect residual noise signals after active noise reduction processing. If the noise in the dominant frequency band (e.g., 150Hz) is still higher than the threshold (e.g., >70dB(A)), the signal is fed back to the signal processing and control unit. The system dynamically adjusts the phase of the inverted signal (with an accuracy of ±1°) to achieve the best cancellation effect. Based on the difference between the residual noise and the desired target, the signal processing and control unit dynamically adjusts the phase and amplitude of the next generated inverted noise reduction signal to optimize the cancellation effect. Simultaneously, if the residual noise is already at a low level, the transmission strategy is optimized, such as reducing the speaker output power to achieve energy saving. For example, when the residual noise drops to a satisfactory level (e.g., ≤65dB(A)), the system automatically reduces the speaker power from the rated value (e.g., 50W) to a lower level (e.g., 20W), achieving a balance between noise reduction effect and energy consumption. This completes one closed-loop optimization cycle. This method continuously executes all the above steps in a loop to adapt to changes in the working conditions of the camera and the noise environment in real time, maintaining a long-term, stable, and efficient optimal noise reduction state, thereby improving the method's adaptive and self-optimizing capabilities.

[0082] This application also provides an electronic device, including: at least one processor; a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to execute any of the new energy-based power grid multi-resource coordinated control methods.

[0083] This application also proposes a computer storage medium storing a computer program, which, when executed by a processor, implements any one of the new energy-based power grid multi-resource coordinated control methods.

[0084] Computer storage media may be simply referred to as media. Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Dual Data SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchlink DRAM (SLDRAM), Rambus Direct RAM (RDRAM), Direct Memory Bus Dynamic RAM (DRDRAM), and Memory Bus Dynamic RAM (RDRAM). The various embodiments described in this specification are presented in a progressive manner, and similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, for embodiments of apparatus, devices, and non-volatile computer storage media, since they are substantially similar to the method embodiments, the description is relatively simple, and relevant parts can be referred to the description of the method embodiments.

[0085] The above embodiments are merely illustrative examples and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations. However, obvious variations or modifications derived therefrom are still within the scope of protection of this application.

Claims

1. A camera adaptive directional active noise reduction device based on multimodal sensing, characterized in that, include: A multimodal sensing unit is used to acquire noise and vibration signals from a synchronous condenser. A signal processing and control unit, connected to the multimodal sensing unit, is used to receive and process the noise signal and the vibration signal, and generate control commands; A directional execution unit, connected to the signal processing and control unit, is used to transmit an anti-phase noise-reducing wave based on the control command; The feedback optimization unit, connected to the signal processing and control unit, is used to collect residual noise signals and feed them back to the signal processing and control unit for dynamic optimization.

2. The apparatus according to claim 1, characterized in that, The multimodal sensing unit includes a broadband acoustic sensing array and a vibration sensing subunit; the broadband acoustic sensing array contains multiple broadband acoustic sensors arranged circumferentially on the housing of the phase converter; the vibration sensing subunit contains a vibration sensor mounted on the phase converter stator or bearing housing.

3. The apparatus according to claim 2, characterized in that, The signal processing and control unit includes a noise feature analysis module; the noise feature analysis module is configured to perform frequency domain analysis on the noise signal collected by the broadband acoustic sensor and perform correlation analysis in combination with the vibration signal collected by the vibration sensor to identify the dominant noise source and its type.

4. The apparatus according to claim 3, characterized in that, The signal processing and control unit also includes an operating condition linkage module; the operating condition linkage module communicates with the synchronous condenser control system via an industrial bus to obtain the real-time operating parameters of the synchronous condenser and establish a mapping model between the operating parameters and noise characteristics.

5. The apparatus according to claim 4, characterized in that, The signal processing and control unit also includes an inverse signal generation module, used to generate an inverse noise reduction signal with the same amplitude but opposite phase as the original noise.

6. The apparatus according to claim 2, characterized in that, The directional execution unit includes an array-type sound-emitting component and a direction adjustment mechanism; the array-type sound-emitting component includes multiple broadband loudspeakers arranged corresponding to the broadband acoustic sensor; the direction adjustment mechanism is connected to at least one of the broadband loudspeakers and is used to adjust the direction of the sound waves emitted by the broadband loudspeakers.

7. The apparatus according to claim 6, characterized in that, The direction adjustment mechanism includes a stepper motor and an acoustic reflector driven by the stepper motor; the acoustic reflector is disposed on the sound wave emission path of the broadband loudspeaker.

8. The apparatus according to claim 1, characterized in that, The feedback optimization unit includes at least one error sensor located in the noise reduction target area, used to collect residual noise signals after the directional execution unit has worked.

9. The apparatus according to claim 7, characterized in that, The signal processing and control unit is configured to receive the residual noise signal, dynamically adjust the phase and amplitude of the inverted noise reduction signal sent to the directional execution unit based on the residual noise signal, and adjust the output power of the directional execution unit according to the intensity of the residual noise signal.

10. A camera-adaptive directional active noise reduction method based on multimodal sensing, characterized in that, include: Simultaneously collect noise signals and vibration signals from different positions of the synchronous condenser; Based on the joint analysis of the noise signal and the vibration signal, the dominant noise source is identified, and the corresponding anti-phase noise reduction signal and control command are generated. Based on the control command, the anti-phase noise reduction signal is emitted in the direction of the dominant noise source; The residual noise signal after noise reduction is collected, and the inverse noise reduction signal and transmission strategy are dynamically optimized based on the residual noise signal.

Images