Generator device and control method thereof

Abstract

The invention relates to the technical field of wind power generation control, in particular to a generator device and a control method thereof. The method comprises the following steps of performing real-time monitoring on noise frequency, amplitude and phase when a generator device operates, and generating a noise characteristic data set; extracting the main frequency and phase characteristics of a noise source from the noise characteristic data set, and performing reverse analysis on the data to generate reverse sound wave signal data opposite to the phase of the noise source; and the reverse sound wave generating device transmits the reverse sound wave signal data to the generator device, outputs reverse sound waves through a broadband loudspeaker, superposes the reverse sound waves and original noise in the space, and monitors the noise after noise reduction in real time to obtain noise reduction effect feedback data. According to the invention, noise characteristics are monitored in real time, matched reverse sound wave signals are extracted and reversely generated, reverse sound wave characteristics are dynamically adjusted, closed-loop adaptive control is formed, and efficient and stable noise reduction of a generator device is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation control, and in particular to a generator device and a control method thereof. Background Art

[0002] The generator device drives the wind wheel through wind power to convert wind energy into electrical energy. However, the generator device will generate noise problems during operation. These noise sources are mainly caused by the rotation of the wind wheel, the operation of the generator, and the disturbance of the airflow. Traditional generator devices mainly rely on the optimization of mechanical structure to reduce noise. Specific methods include optimizing the design of the wind wheel and improving the material of wind turbine blades. However, due to the high-speed rotation of the wind wheel during operation, strong friction and disturbance with the air occur, resulting in the generation of aerodynamic noise. This aerodynamic noise is difficult to completely eliminate through structural optimization. Even if the improved wind turbine blade material can alleviate the aerodynamic noise to a certain extent, it is still difficult to avoid the generation of large vibration noise under strong wind conditions. The limitations of noise control technology are also another major defect of the generator devices on the market. The traditional noise control method is mainly passive noise reduction, that is, installing a soundproof cover around the wind turbine or using noise reduction materials to absorb and block noise. This method has a certain effect on high-frequency noise, but it has limited effect on the low-frequency noise generated by the generator device. Low-frequency noise has a wide propagation range and strong penetration, and has a more significant impact on the surrounding environment. Therefore, the existing passive noise reduction technology has obvious deficiencies in reducing low-frequency noise. Summary of the invention

[0003] Based on this, it is necessary for the present invention to provide a generator device and a control method thereof to solve at least one of the above technical problems.

[0004] To achieve the above object, a control method of a generator device comprises the following steps:

[0005] Step S1: Real-time monitoring of the noise frequency, amplitude and phase of the generator device during operation to generate a noise characteristic data set;

[0006] Step S2: extracting the main frequency and phase characteristics of the noise source from the noise characteristic data set, and performing reverse analysis on the data to generate reverse sound wave signal data with a phase opposite to that of the noise source;

[0007] Step S3: transmitting the reverse sound wave signal data to the reverse sound wave generating device of the generator device, outputting the reverse sound wave through the broadband speaker to superimpose the reverse sound wave with the original noise in space, and monitoring the noise after noise reduction in real time to obtain noise reduction effect feedback data;

[0008] Step S4: dynamically adjusting the frequency, amplitude and phase of the reverse sound wave through an adaptive filtering algorithm based on the noise reduction effect feedback data to obtain adjusted reverse sound wave characteristic data;

[0009] Step S5: Apply the adjusted reverse acoustic wave characteristic data to step S3 to dynamically optimize the reverse acoustic wave and achieve closed-loop adaptive control.

[0010] The present invention monitors the noise frequency, amplitude and phase of the generator device in real time during operation, and can capture the dynamic changes of the noise source of the device. By generating a noise characteristic data set, detailed data support is provided for the subsequent precise generation and regulation of reverse sound waves, ensuring that the basic data of noise reduction is accurate and reliable. The main frequency and phase characteristics of the noise source are extracted to ensure that the reverse sound wave and the noise source match each other in frequency and phase, thereby achieving a more effective noise cancellation effect. The reverse sound wave signal data generated by reverse analysis can ensure that the phase of the reverse sound wave is opposite to that of the noise source, which helps to enhance the noise suppression effect. The reverse sound wave is output through a wide-band speaker and superimposed with the original noise in space to achieve active noise reduction. Real-time monitoring of the superimposed noise effect and obtaining noise reduction feedback data can timely evaluate the noise reduction effect, provide data support for subsequent adjustments, and make the noise reduction process more accurate and effective. The frequency, amplitude and phase of the reverse sound wave are dynamically adjusted according to the feedback data using an adaptive filtering algorithm, thereby ensuring that the reverse sound wave signal always matches the noise source and improves the noise cancellation effect. This real-time adaptive adjustment can cope with changes in noise sources and enhance the noise reduction stability of the system. By feeding back the adjusted reverse acoustic wave characteristic data to the reverse acoustic wave generating device, a closed-loop adaptive control is formed to continuously optimize the output of the reverse acoustic wave. This step helps to further improve the noise reduction efficiency, achieve a long-term and stable noise reduction effect for the generator device, and enhance the adaptability and reliability of the system.

[0011] The present invention further provides a generator device, comprising a reverse sound wave generating device, a wide-band loudspeaker and a control system, wherein the reverse sound wave generating device is disposed in the generator device, the wide-band loudspeaker is disposed in the reverse sound wave generating device, the control system is installed in the generator device, the control system is used in a control method of the generator device, and the control system comprises:

[0012] The noise characteristic monitoring module is used to monitor the noise frequency, amplitude and phase of the generator device in real time during operation and generate a noise characteristic data set;

[0013] The noise characteristic analysis and reverse signal generation module is used to extract the main frequency and phase characteristics of the noise source from the noise characteristic data set, and reversely analyze the data to generate reverse sound wave signal data with the opposite phase to the noise source;

[0014] The reverse sound wave transmission and superposition module is used to transmit the reverse sound wave signal data to the reverse sound wave generating device of the generator device, output the reverse sound wave through the wide-band speaker and superimpose it with the original noise in space, and monitor the noise after noise reduction in real time to obtain noise reduction effect feedback data;

[0015] The reverse signal adaptive adjustment module is used to dynamically adjust the frequency, amplitude and phase of the reverse sound wave through an adaptive filtering algorithm based on the noise reduction effect feedback data to obtain the adjusted reverse sound wave characteristic data;

[0016] The closed-loop control and adaptive optimization module is used to apply the adjusted reverse acoustic wave characteristic data to step S3 to dynamically optimize the reverse acoustic wave and realize closed-loop adaptive control.

[0017] The present invention can monitor the noise frequency, amplitude and phase of the generator device in real time during operation, and generate a noise characteristic data set. By real-time monitoring of the main characteristics of the noise, timely and accurate acquisition of the noise source information is ensured, and reliable data support is provided for subsequent reverse signal analysis and generation, thereby improving the accuracy of noise reduction control. By analyzing the noise characteristic data set, the main frequency and phase characteristics of the noise source are extracted, and reverse sound wave signal data with a phase opposite to the noise source is generated. Through reverse analysis, a sound wave signal with a reverse phase can be accurately generated, thereby effectively offsetting the main components of the noise source, providing a basis for achieving significant noise reduction. The reverse sound wave signal data is transmitted to the reverse sound wave generating device of the generator device, and is output through a wide-band speaker and superimposed in space with the original noise. The real-time monitoring function of this module can capture the noise reduction effect after superposition, and generate feedback data, which provides a real-time reference for the adjustment of the reverse signal, and ensures the continuous optimization and verification of the noise reduction effect. Using an adaptive filtering algorithm, this module dynamically adjusts the frequency, amplitude and phase of the reverse sound wave according to the feedback data to generate optimized reverse sound wave characteristic data. Through adaptive adjustment, the reverse acoustic wave signal can be flexibly adjusted and precisely controlled, which can better cope with the noise changes in the operation of the generator device, and further enhance the noise reduction effect and stability of the system. Closed-loop control is realized, and the adjusted reverse acoustic wave characteristic data is applied to the noise superposition process to achieve dynamic optimization of the reverse acoustic wave. Closed-loop adaptive control can ensure that the entire system automatically adjusts to the optimal noise reduction state during operation, achieving long-term and continuous noise reduction effects. The adaptive optimization characteristics of this module help the system maintain efficient noise reduction performance in a changing noise environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0019] Figure 1A schematic flow chart of the steps of the control method of the generator device of the present invention;

[0020] Figure 2 for Figure 1 Detailed step flow diagram of step S1;

[0021] Figure 3 for Figure 1 Detailed step flow chart of step S2 in FIG. DETAILED DESCRIPTION

[0022] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0023] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0024] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0025] To achieve this, please refer to Figures 1 to 3 The present invention provides a control method for a generator device, the method comprising the following steps:

[0026] Step S1: Real-time monitoring of the noise frequency, amplitude and phase of the generator device during operation to generate a noise characteristic data set;

[0027] Step S2: extracting the main frequency and phase characteristics of the noise source from the noise characteristic data set, and performing reverse analysis on the data to generate reverse sound wave signal data with a phase opposite to that of the noise source;

[0028] Step S3: transmitting the reverse sound wave signal data to the reverse sound wave generating device of the generator device, outputting the reverse sound wave through the broadband speaker to superimpose the reverse sound wave with the original noise in space, and monitoring the noise after noise reduction in real time to obtain noise reduction effect feedback data;

[0029] Step S4: dynamically adjusting the frequency, amplitude and phase of the reverse sound wave through an adaptive filtering algorithm based on the noise reduction effect feedback data to obtain adjusted reverse sound wave characteristic data;

[0030] Step S5: Apply the adjusted reverse acoustic wave characteristic data to step S3 to dynamically optimize the reverse acoustic wave and achieve closed-loop adaptive control.

[0031] In the embodiment of the present invention, reference Figure 1 The above is a schematic flow chart of the steps of a control method of a generator device of the present invention. In this example, the control method of the generator device includes the following steps:

[0032] Step S1: Real-time monitoring of the noise frequency, amplitude and phase of the generator device during operation to generate a noise characteristic data set;

[0033] The embodiment of the present invention uses a high-precision acoustic sensor array for noise monitoring. The sensor needs to have a good frequency response range and can cover various noise frequencies generated when the generator device is running. The sensor array should be arranged around the generator device, especially around the noise source and important monitoring points. These locations can be reasonably planned according to the operation mode of the wind turbine, the equipment layout and the noise propagation path to ensure that the noise characteristics of the generator device under different working conditions are captured. During the operation of the generator device, the noise signal emitted by the device is collected in real time, including the frequency, amplitude and phase characteristics of the noise. At this time, the noise signal collected by the acoustic sensor will be transmitted to the signal acquisition device for real-time data recording. The collected analog sound wave signal will be converted into a digital signal by an analog-to-digital conversion device for subsequent analysis and processing. In order to ensure the accuracy of the data, the sampling frequency should meet the requirements of the Nyquist sampling theorem, and it is usually necessary to select a sampling rate that is at least twice the highest frequency of the noise signal. The real-time collected noise signal is converted into frequency domain data through spectrum analysis techniques such as fast Fourier transform (FFT). In this process, the amplitude and phase of each frequency component during the operation of the generator device will be obtained. Based on spectrum analysis, the amplitude and phase information at different frequencies are extracted. By performing phase unwrapping and amplitude analysis on the time domain signal, detailed characteristic data of each frequency component can be obtained. The frequency, amplitude, phase and other characteristic data obtained through analysis are integrated to form a complete noise characteristic data set. These data include but are not limited to the main frequency components of the noise source, the amplitude values ​​of each frequency, phase information, etc.

[0034] Step S2: extracting the main frequency and phase characteristics of the noise source from the noise characteristic data set, and performing reverse analysis on the data to generate reverse sound wave signal data with a phase opposite to that of the noise source;

[0035] The embodiment of the present invention preprocesses the noise characteristic data set. The process includes removing the interference components in the noise data, eliminating the outliers and removing the background noise to ensure the accuracy of the data. A high-pass or low-pass filter can be used to remove the noise in the irrelevant frequency band and retain the frequency components related to the operation of the generator device. The data is normalized to ensure that the results will not be affected by excessive numerical differences in the subsequent analysis process. The noise characteristic data set is spectrally analyzed using a fast Fourier transform. By converting the time domain signal into frequency domain data, the main frequency components of the noise source of the generator device are extracted. Usually, these frequency components correspond to the rotation frequency of the fan blades, the gearbox frequency, and the vibration frequency of the mechanical parts. In addition to the frequency components, the phase information of each frequency component needs to be extracted. In order to ensure the accuracy of the phase, the phase is usually unwrapped to eliminate the error caused by the phase jump, so as to obtain the stable phase value of each frequency. The extracted phase information is analyzed to check the phase stability of each frequency component. If the phase change of some frequency components is unstable, further signal smoothing is required. According to the extracted frequency and phase characteristics, the reverse sound wave signal data with the opposite phase to the noise source is calculated. The reverse sound wave is generated by reversely calculating the phase of each frequency component to obtain a signal that is completely opposite to the phase of the noise source. For each frequency component, the reverse phase is the phase of the noise source plus π (180 degrees), so that the two cancel each other out. In order for the reverse sound wave to effectively cancel the noise source, in addition to the phase, the amplitude of the reverse sound wave needs to be calculated based on the energy contribution of each frequency component. The amplitude can be determined by adjusting the gain coefficient of each frequency component to ensure that the reverse sound wave can effectively cover the energy of the noise source. The reverse sound wave signals of all reverse frequency components are synthesized to obtain complete reverse sound wave signal data.

[0036] Step S3: transmitting the reverse sound wave signal data to the reverse sound wave generating device of the generator device, outputting the reverse sound wave through the broadband speaker to superimpose the reverse sound wave with the original noise in space, and monitoring the noise after noise reduction in real time to obtain noise reduction effect feedback data;

[0037] In the embodiment of the present invention, the reverse sound wave signal data is transmitted to the reverse sound wave generating device in the generator device. In order to enable the reverse sound wave signal to be recognized and output by the speaker, the signal needs to be converted from a digital format to an analog signal. This step is achieved by a digital-to-analog converter (DAC) to ensure that the frequency, amplitude and phase characteristics of the reverse signal can be accurately restored. After the signal is converted into an analog signal, a driving signal suitable for a wide-band speaker is further generated. The driving signal also needs to be gain-adjusted to ensure that the intensity of the output signal is sufficient to offset the noise generated by the generator device. The signal can be processed in segments so as to effectively cover all noise frequencies within the working frequency band of the speaker. The processed reverse sound wave driving signal is input into the wide-band speaker, and the speaker releases the reverse sound wave signal in the form of sound waves in space. The wide-band speaker needs to have a broadband frequency response to ensure that the main noise frequency band of the generator device can be covered to form an effective reverse sound wave. The reverse sound wave and the original noise of the generator device are superimposed in space. When the reverse sound wave has the same frequency and amplitude as the original noise but has an opposite phase, the two will interfere with each other in space, thereby achieving the effect of noise cancellation. This process mainly relies on the precise phase opposition of the reverse sound wave and the original noise, so that the two interfere with each other and partially or completely cancel each other out. In order to monitor the noise reduction effect in real time, noise monitoring sensors are arranged around the generator device and in specific noise reduction areas. The sensors should be arranged in the target noise reduction area so that the noise level after the reverse sound wave is cancelled can be accurately monitored. The noise monitoring sensor collects the superimposed spatial sound field data in real time and transmits the noise data after noise reduction to the control system. This data reflects the remaining amount of noise in the space and is a key indicator for evaluating the noise reduction effect. The noise reduction effect monitoring data is processed and analyzed to generate noise reduction effect feedback data.

[0038] Step S4: dynamically adjusting the frequency, amplitude and phase of the reverse sound wave through an adaptive filtering algorithm based on the noise reduction effect feedback data to obtain adjusted reverse sound wave characteristic data;

[0039] The embodiment of the present invention performs a detailed analysis of the residual noise energy of each frequency band according to the noise reduction effect feedback data. By quantifying the noise reduction effect of each frequency band, the noise residual value at different frequencies is determined. These data provide a target and basis for adaptive adjustment. Based on the energy distribution in the feedback data, the key frequency bands where the noise is mainly concentrated and still need to be further optimized are identified to ensure that the adjustment is more targeted. The parameters of the reverse sound wave signal are optimized using an adaptive filtering algorithm. The adaptive filtering algorithm can continuously adjust the frequency, amplitude and phase of the output signal according to the input feedback data to improve the noise reduction effect. The adaptive filter dynamically calculates and generates new frequency, amplitude and phase parameters based on real-time feedback data. Ensure that the frequency of the reverse sound wave is consistent or similar to the frequency of the noise source to achieve good frequency matching and cancellation effects. Adjust the amplitude of the reverse sound wave according to the residual energy of the noise so that its amplitude can match or slightly exceed the amplitude of the noise source. Keep the reverse sound wave in opposite phase with the noise source, and achieve effective interference of the phase in space through phase calibration. Integrate the frequency, amplitude and phase data optimized and calculated by the adaptive filter to generate adjusted reverse sound wave characteristic data.

[0040] Step S5: Apply the adjusted reverse acoustic wave characteristic data to step S3 to dynamically optimize the reverse acoustic wave and achieve closed-loop adaptive control.

[0041] The embodiment of the present invention transmits the adjusted reverse sound wave characteristic data to the reverse sound wave generating device in real time, and updates it synchronously with the current sound wave characteristic data. According to the latest data, the output parameters of the reverse sound wave are adjusted to ensure that the reverse sound wave currently output matches the latest noise source characteristics. A new reverse sound wave driving signal is generated using the updated reverse sound wave characteristic data and output through a wideband speaker. This driving signal is subjected to gain control and phase matching so that the frequency, amplitude and phase of the reverse sound wave can accurately offset the noise generated by the generator device. According to the real-time noise feedback data, the reverse sound wave generating device adjusts the output power of the speaker to maintain the best noise reduction effect. The power adjustment is mainly aimed at the amplitude change, so as to enhance the intensity of the reverse sound wave when the noise is large, and reduce the intensity when the noise is reduced to save energy. The noise reduction effect data is continuously collected through the noise monitoring sensor, the remaining noise in the superimposed sound field is analyzed in real time, and the noise reduction effect feedback data is input into the adaptive control system. The adaptive control system continuously optimizes the parameters of the reverse sound wave based on the feedback data, thereby forming a complete closed-loop control process. Whenever the noise reduction effect is insufficient or the frequency of the noise source changes, the system automatically updates the frequency, amplitude and phase of the reverse sound wave to achieve adaptive adjustment to the dynamic environment. As the system continues to adjust, the phase, frequency and amplitude of the reverse sound wave and the noise source gradually synchronize and maintain an inverse relationship, achieving the best noise reduction effect.

[0042] The present invention monitors the noise frequency, amplitude and phase of the generator device in real time during operation, and can capture the dynamic changes of the noise source of the device. By generating a noise characteristic data set, detailed data support is provided for the subsequent precise generation and regulation of reverse sound waves, ensuring that the basic data of noise reduction is accurate and reliable. The main frequency and phase characteristics of the noise source are extracted to ensure that the reverse sound wave and the noise source match each other in frequency and phase, thereby achieving a more effective noise cancellation effect. The reverse sound wave signal data generated by reverse analysis can ensure that the phase of the reverse sound wave is opposite to that of the noise source, which helps to enhance the noise suppression effect. The reverse sound wave is output through a wide-band speaker and superimposed with the original noise in space to achieve active noise reduction. Real-time monitoring of the superimposed noise effect and obtaining noise reduction feedback data can timely evaluate the noise reduction effect, provide data support for subsequent adjustments, and make the noise reduction process more accurate and effective. The frequency, amplitude and phase of the reverse sound wave are dynamically adjusted according to the feedback data using an adaptive filtering algorithm, thereby ensuring that the reverse sound wave signal always matches the noise source and improves the noise cancellation effect. This real-time adaptive adjustment can cope with changes in noise sources and enhance the noise reduction stability of the system. By feeding back the adjusted reverse acoustic wave characteristic data to the reverse acoustic wave generating device, a closed-loop adaptive control is formed to continuously optimize the output of the reverse acoustic wave. This step helps to further improve the noise reduction efficiency, achieve a long-term and stable noise reduction effect for the generator device, and enhance the adaptability and reliability of the system.

[0043] Preferably, step S1 comprises the following steps:

[0044] Step S11: Arranging an acoustic sensor array in the noise generating area of ​​the generator device and establishing sensor coordinates to record the spatial position information of each monitoring point and obtain sensor position coordinate data;

[0045] Step S12: Based on the sensor position coordinate data, the ambient background noise at each wind speed is collected when the generator device is stopped to obtain background noise spectrum data;

[0046] Step S13: Based on the sensor position coordinate data, the sound pressure signal under the operation state of the generator device is collected, and analog-to-digital conversion and digital processing are performed to obtain original time domain sound pressure data;

[0047] Step S14: performing a fast Fourier transform on the original time-domain sound pressure data and calculating the power spectrum density to obtain the runtime noise spectrum data;

[0048] Step S15: Calculate the phase information of each frequency band signal on the original time-domain sound pressure data, and perform phase unwrapping processing to obtain the runtime noise phase spectrum data;

[0049] Step S16: preprocessing the runtime noise spectrum data and the runtime noise phase spectrum data, and integrating the preprocessed data to obtain runtime noise frequency domain data;

[0050] Step S17: Integrate the sensor position coordinate data, background noise spectrum data, original time domain sound pressure data, and runtime noise frequency domain data, and timestamp the data to generate a noise characteristic data set.

[0051] As an embodiment of the present invention, refer to Figure 2 As shown, Figure 1 Detailed step flow diagram of step S1 in the embodiment of the present invention, step S1 includes the following steps:

[0052] Step S11: Arranging an acoustic sensor array in the noise generating area of ​​the generator device and establishing sensor coordinates to record the spatial position information of each monitoring point and obtain sensor position coordinate data;

[0053] The embodiment of the present invention determines the number and arrangement positions of acoustic sensors to ensure that all key points of the noise generation area of ​​the generator device are covered. Common areas include blades, towers and mechanical transmission parts, and sensors are preferably arranged in areas with high noise source intensity. A suitable sensor array layout, such as a rectangular, circular or grid array, is used to enhance the coverage of spatial noise collection and improve the accuracy of signal collection. The acoustic sensor is installed at a determined position and fixed according to the installation angle and height of each sensor to ensure stable and accurate collection of noise data. The spatial coordinates of the sensor position are accurately calibrated using a laser rangefinder, the three-dimensional spatial coordinates of each sensor are obtained, and the height, azimuth and distance of the monitoring point are recorded. The sensor array is environmentally calibrated to eliminate initial installation errors and ensure that the sensor can stably capture noise signals under actual operating conditions. The spatial coordinate information of each sensor is digitized and recorded in a database to form sensor position coordinate data for subsequent data processing. The coordinate data of all sensors are integrated into a unified spatial coordinate data set to provide accurate position information reference for subsequent noise spectrum, phase and amplitude analysis. Ensure that each device of the sensor array is connected to the data acquisition system and check the stability of signal acquisition and data recording. The recorded sensor spatial position information is timestamped so that the noise data collected subsequently can be associated with the sensor position, providing complete data support for subsequent noise characteristic analysis.

[0054] Step S12: Based on the sensor position coordinate data, the ambient background noise at each wind speed is collected when the generator device is stopped to obtain background noise spectrum data;

[0055] The embodiment of the present invention ensures that the generator device is in a shutdown state, eliminates the interference of operating noise, and makes the collected noise data only contain environmental background noise. Confirm that the acoustic sensor array is in normal working condition to ensure that all sensors can collect accurate sound pressure signals. Perform segmented tests on each target wind speed under natural conditions or wind tunnel simulation conditions. Before each collection, record the wind speed value and mark the environmental conditions for each collection to ensure the traceability of the data. Under each wind speed condition, use the sensor array to collect the sound pressure signal in the environment to obtain the original background noise time domain data. Associate the collected data of each sensor with its position coordinates. Perform FFT processing on the collected background noise time domain data under each wind speed condition, extract the spectrum information of the background noise, and obtain the corresponding background noise spectrum data.

[0056] Step S13: Based on the sensor position coordinate data, the sound pressure signal under the operation state of the generator device is collected, and analog-to-digital conversion and digital processing are performed to obtain original time domain sound pressure data;

[0057] The embodiment of the present invention ensures that the generator device has been started and is in normal operation to ensure that the collected data is operating noise data. Check whether the sensor array is operating normally, and perform necessary calibration on the sensor to ensure that the collection sensitivity of each sensor is consistent. During the normal operation of the fan, the sensor array arranged in the noise generation area is used to collect sound pressure signals in real time. Each sensor of the sensor array will collect sound pressure data at its respective spatial position. The sound pressure signal collected by each sensor is bound to the spatial position information of the sensor to facilitate the subsequent positioning of the noise source and analysis of the spatial distribution of the noise. The analog sound pressure signal collected by the sensor is converted into a digital form to convert the continuous analog signal into a discrete digital signal. Select a suitable sampling rate so that the collected digital signal can fully restore the spectral characteristics of the sound pressure signal, which is usually required to be higher than twice the frequency of the main frequency band of the noise to avoid aliasing. The discrete signal data obtained by the analog-to-digital conversion is formatted and normalized to facilitate subsequent data analysis. The digitized sound pressure signal of each sensor is saved as the original time domain sound pressure data to form a continuous time domain data stream.

[0058] Step S14: performing a fast Fourier transform on the original time-domain sound pressure data and calculating the power spectrum density to obtain the runtime noise spectrum data;

[0059] The embodiment of the present invention performs a fast Fourier transform on the original time-domain sound pressure data to convert the time-domain signal into a frequency-domain signal. This step decomposes the various frequency components of the sound pressure signal to understand the spectral characteristics of the noise. The frequency resolution is set according to the sampling rate and the data length so that the transformation result can accurately describe the frequency components of the noise. Data in the required frequency range is selected to filter out the interference components beyond the target frequency band and focus on the main frequency range of the noise. The power spectrum density is calculated based on the FFT result to determine the energy of each frequency component. The calculation of the power spectrum density helps to quantify the spectral energy distribution of the noise and find the frequency band of the main noise source. The power spectrum density curve is smoothed to remove small fluctuations and spikes, so that the noise spectrum distribution is clearer and easier to analyze. The PSD data is converted into a commonly used dB unit as needed for comparison with background noise and other spectrum data. The calculated spectrum data is stored and labeled with timestamps, sensor location coordinates, etc. to ensure the accuracy and traceability of the data in subsequent analysis, and the runtime noise spectrum data is output.

[0060] Step S15: Calculate the phase information of each frequency band signal on the original time-domain sound pressure data, and perform phase unwrapping processing to obtain the runtime noise phase spectrum data;

[0061] The embodiment of the present invention extracts the noise signal under the operating state from the original time-domain sound pressure data. The frequency band data of interest is filtered as needed to ensure that the frequency range of the phase calculation matches the main noise frequency band. The time-domain sound pressure data is subjected to a fast Fourier transform to convert the data into the frequency domain. The phase information of each frequency component is extracted from the complex output of the FFT. The phase information characterizes the initial phase difference of each frequency component and is crucial for the subsequent generation of reverse sound waves. The extracted phase information is converted into an angle for further processing. Since the phase data calculated by FFT is usually in the range of -π to π, phase unwrapping is required to make the phase value continuous to avoid mutations. An algorithm is used to smoothly transition the position where the phase change exceeds π to restore the true phase continuity. The unwrapped phase information can more accurately reflect the phase distribution of the noise. The calculated phase spectrum data is stored by frequency band classification, and the frequency and phase relationship are marked.

[0062] Step S16: preprocessing the runtime noise spectrum data and the runtime noise phase spectrum data, and integrating the preprocessed data to obtain runtime noise frequency domain data;

[0063] The embodiment of the present invention performs filtering processing on the runtime noise spectrum data and the runtime noise phase spectrum data to remove the existing environmental interference noise to ensure the purity of the data. The data of each monitoring point is normalized to reduce the amplitude deviation caused by the difference in sensitivity of the measuring equipment or the change in position. A smoothing algorithm is used to reduce sharp mutations in the noise data so as to more clearly reflect the spectral characteristics of the noise. The processed spectrum data and phase spectrum data are aligned in the frequency dimension to ensure that the two data sets have consistent information at the same frequency point. Adjust the phase information in each frequency band to ensure the consistency of the phase relationship between different frequency bands, laying the foundation for subsequent noise reduction calculations. The preprocessed noise spectrum data and phase spectrum data are integrated to form a runtime noise frequency domain data set containing frequency, amplitude, phase and other characteristics.

[0064] Step S17: Integrate the sensor position coordinate data, background noise spectrum data, original time domain sound pressure data, and runtime noise frequency domain data, and timestamp the data to generate a noise characteristic data set.

[0065] The embodiment of the present invention classifies the various types of noise data collected and processed in step S11 to step S16 according to their properties and functions. The data mainly includes sensor position coordinate data, background noise spectrum data, original time domain sound pressure data and runtime noise frequency domain data. The collected data is time synchronized to ensure the consistency of the time dimension of all data sources and avoid time deviation between data. An accurate timestamp is stamped for each set of data to track the changes of data at different time points. The timestamps are segmented and marked according to the time period of data collection for subsequent query and analysis. All marked data are stored in a unified format to ensure that the data set structure is clear and easy to access. The data with marked timestamps are integrated to form a comprehensive noise characteristic data set containing sensor coordinates, background noise, original time domain data and frequency domain data.

[0066] The present invention arranges an acoustic sensor array and establishes a coordinate system to clarify the spatial position information of each monitoring point, so that the subsequent noise monitoring data has spatial distribution characteristics, provides necessary position information support for identifying and locating noise sources, and improves the accuracy and coverage of noise monitoring. The environmental background noise under different wind speeds is collected in the shutdown state, and the spectrum characteristics of the environmental background noise can be obtained to provide benchmark data for subsequent noise processing. By eliminating the background noise interference, it is ensured that the extracted noise characteristics are more accurate, which helps to improve the accuracy of noise source identification. The sound pressure signal is collected and digitized in the running state to obtain the original time domain sound pressure data, which provides basic data for further spectrum analysis and noise characteristic extraction. The digitized sound pressure data is easier to analyze and process in the subsequent state, making the information acquisition of the noise reduction control system more reliable. The noise spectrum data during operation is obtained by fast Fourier transform, which can clarify the frequency components of the noise source and provide accurate frequency information for the generation of reverse sound waves. Spectral analysis also helps to identify noise sources with different frequency components, which is convenient for subsequent effective noise reduction processing. Calculate the phase information and perform phase unwrapping processing to obtain the phase spectrum data of the noise during operation, ensure that the reverse sound wave can achieve accurate phase confrontation and enhance the noise cancellation effect. Through the unwrapping processing of the phase information, the continuity of the phase data is ensured, laying the foundation for more accurate reverse sound wave generation. Preprocess and integrate the noise spectrum data and phase spectrum data to obtain complete frequency domain data, making the data analysis more comprehensive and accurate, providing detailed frequency domain information for the generation and dynamic adjustment of the reverse sound wave, and ensuring the effect of subsequent noise reduction control. Associate and integrate various noise data with timestamps to generate a complete noise characteristic data set, providing an accurate and traceable data basis for the entire noise reduction system. Through timestamp marking, the system can realize real-time noise monitoring and noise reduction feedback, and achieve effective closed-loop control.

[0067] Preferably, step S2 comprises the following steps:

[0068] Step S21: performing background noise elimination on the noise characteristic data set, and using a digital filter to eliminate high-frequency interference to obtain a pure noise characteristic data set;

[0069] Step S22: performing time-frequency analysis on the pure noise characteristic data set, and calculating the spectrum peak and energy distribution to identify the main frequency components, and obtain the main frequency component list data, wherein the main frequency component list data includes the energy contribution of each frequency and the frequency time-varying characteristic data;

[0070] Step S23: calculating the phase characteristics of the main frequency components of the pure noise characteristic data set based on the main frequency component list data, and performing phase continuity correction to analyze the phase change law and obtain the phase characteristic data of the main frequency;

[0071] Step S24: calculating the reverse phase value of each frequency component based on the main frequency component list and the phase characteristic data of the main frequency, and determining the amplitude ratio according to the energy contribution of each frequency to obtain the reverse sound wave frequency characteristic data and the reverse sound wave phase characteristic data;

[0072] Step S25: Using inverse Fourier transform to synthesize the reverse acoustic wave frequency characteristic data and the reverse acoustic wave phase characteristic data into a time domain signal, and performing signal smoothing processing to obtain reverse acoustic wave signal data.

[0073] As an embodiment of the present invention, refer to Figure 3 As shown, Figure 1 The detailed step flow diagram of step S2 in the embodiment of the present invention includes the following steps:

[0074] Step S21: performing background noise elimination on the noise characteristic data set, and using a digital filter to eliminate high-frequency interference to obtain a pure noise characteristic data set;

[0075] According to the background noise spectrum data collected in step S12, the embodiment of the present invention constructs a set of background noise baselines representing different wind speed conditions as a reference for subsequent background noise elimination. The current noise characteristic data set is compared with the background noise baseline data, and the signal components close to the background noise in frequency and amplitude are eliminated. The parts that overlap with the background noise baseline frequency are identified, and the amplitude of these frequency bands is attenuated. The noise data collected at each monitoring point is adjusted in amplitude, and the energy part corresponding to the background noise is subtracted to retain the characteristics of the noise source. According to the high-frequency noise interference characteristics in the noise characteristic data set, a suitable digital filter is selected to filter out unnecessary high-frequency components and retain the noise information of the main frequency band. According to the sampling frequency and the range of the noise spectrum, the cutoff frequency or bandpass frequency of the filter is set to effectively remove high-frequency interference. Usually, the main frequency of wind power generation noise is in the low frequency band, so the high-frequency part can be reduced by a low-pass filter. A filter is applied to the high-frequency components in the noise characteristic data set to gradually reduce the high-frequency interference in the noise. The purity of the data is further improved by recursive filtering or segmented filtering. The noise data after background noise elimination and high-frequency interference removal are integrated, and the filtering effect is verified to generate a pure noise characteristic data set.

[0076] Step S22: performing time-frequency analysis on the pure noise characteristic data set, and calculating the spectrum peak and energy distribution to identify the main frequency components, and obtain the main frequency component list data, wherein the main frequency component list data includes the energy contribution of each frequency and the frequency time-varying characteristic data;

[0077] The embodiment of the present invention uses methods such as wavelet transform to perform time-frequency analysis on a pure noise characteristic data set to obtain the distribution of noise signals at different times and frequencies. In the short-time Fourier transform, a suitable window function and window length are selected, and the duration and step length of the window are set according to the noise frequency characteristics to ensure that the spectrum resolution and time resolution are balanced. The frequency spectrum of pure noise is obtained by time-frequency analysis, and the frequency distribution and change law of the main noise components are determined. The peaks of the noise spectrum in different time periods are extracted from the time-frequency analysis results to identify the main frequency components with higher energy contribution. The energy distribution ratio of different frequency components is calculated to determine the energy contribution of each frequency component in the noise. Power spectral density can be used for energy calculation to quantify the influence of each frequency component. The frequency components with larger energy contribution are screened out, and the frequency components with low energy contribution are ignored to simplify the subsequent calculation of reverse sound wave generation. The time-varying characteristics of each main frequency component are recorded, and the stability and change trend of the frequency component in different time periods are obtained to facilitate the compensation of dynamic characteristics in reverse sound wave generation. The frequency, energy contribution and frequency time-varying characteristic data of each frequency component are integrated to generate a list of main frequency components, which is convenient for subsequent phase characteristic analysis and reverse acoustic wave characteristic design.

[0078] Step S23: calculating the phase characteristics of the main frequency components of the pure noise characteristic data set based on the main frequency component list data, and performing phase continuity correction to analyze the phase change law and obtain the phase characteristic data of the main frequency;

[0079] In the embodiment of the present invention, the main frequency components are selected from the main frequency component list for phase characteristic calculation. Each main frequency component corresponds to a frequency and its energy contribution in the noise. For the selected main frequency component, the corresponding phase data is extracted from the pure noise characteristic data set to ensure that the phase distribution information of the noise signal in different time and space is covered. The instantaneous phase of the main frequency component is calculated by the Hilbert transform method to obtain the phase distribution of the noise signal at each moment, so as to analyze its phase characteristics. In the calculated phase characteristic data, it is checked whether there is a phase jump. If there is, it is corrected by the phase unwrapping technology to make the phase value continuous in time. The phase fluctuation caused by noise is eliminated by smoothing processing to obtain a relatively smooth phase change curve, so as to facilitate the phase design of the subsequent reverse sound wave generation. According to the corrected phase characteristic data, the change law of the phase of the main frequency component over time is analyzed. The phase time-varying characteristics of each main frequency component are recorded to determine whether the phase presents periodic fluctuations, trend changes or random fluctuations. The analysis results are summarized as the phase characteristic data of the main frequency, including the time-varying characteristics of the phase, the phase jump correction information and the phase change mode, which provides a basis for the calculation of the reverse phase value in the subsequent steps.

[0080] Step S24: calculating the reverse phase value of each frequency component based on the main frequency component list and the phase characteristic data of the main frequency, and determining the amplitude ratio according to the energy contribution of each frequency to obtain the reverse sound wave frequency characteristic data and the reverse sound wave phase characteristic data;

[0081] The embodiment of the present invention extracts the phase value of each main frequency component from the main frequency phase characteristic data to ensure that its time-varying characteristics are included. The phase value of each frequency component is inverted, and the inverted phase value of each frequency component is calculated to generate a phase opposite to the original noise signal in the reverse sound wave to achieve a cancellation effect. If the phase value jumps or changes suddenly in time, the smooth transition of the inverted phase data is ensured by smoothing, thereby achieving more effective phase cancellation. According to the energy contribution in the main frequency component list data generated in step S22, the relative contribution of each frequency component to the overall noise is determined. According to the energy contribution of each frequency, the amplitude ratio of each frequency component in the reverse sound wave is correspondingly determined, so that the amplitude distribution of the reverse sound wave signal matches the original noise energy distribution to improve the efficiency of noise cancellation. The inverted phase value and amplitude ratio of each main frequency component are integrated to obtain the frequency characteristic data and phase characteristic data of the reverse sound wave, ensuring that it contains the reverse sound wave characteristic information of all main frequency components.

[0082] Step S25: Using inverse Fourier transform to synthesize the reverse acoustic wave frequency characteristic data and the reverse acoustic wave phase characteristic data into a time domain signal, and performing signal smoothing processing to obtain reverse acoustic wave signal data.

[0083] The embodiment of the present invention uses the frequency characteristic data of the reverse sound wave and the phase characteristic data of the reverse sound wave as input data. These data contain the amplitude and phase information of each frequency component. The frequency characteristic data of the reverse sound wave is converted from the frequency domain to the time domain through the inverse Fourier transform. The core of this process is to reconstruct the time signal through the frequency domain data to reflect the time-varying characteristics of the sound wave. Each frequency component is usually represented by a complex number in the frequency domain. The inverse Fourier transform needs to combine the amplitude and phase of the frequency component to generate a time signal. The inverse Fourier transform synthesizes the time domain signals of all frequency components to obtain a complete reverse sound wave time domain signal. Any fluctuations or instabilities that appear in the time domain are removed by signal smoothing. This process aims to reduce the noise or discontinuity generated during the conversion process from the frequency domain to the time domain, and ensure that the signal is smooth and stable. Commonly used smoothing methods include sliding average filtering, weighted average filtering, etc., and the selection of specific methods depends on actual needs. For example, a low-pass filter can be used to smooth the time domain signal to remove high-frequency noise. According to the characteristics of the signal, a suitable time window can be set to locally smooth the time domain signal to reduce volatility and ensure that the generated reverse acoustic wave signal has a stable amplitude. The smoothed time domain signal obtained after the inverse Fourier transform is the generated reverse acoustic wave signal data.

[0084] The present invention obtains a pure noise characteristic data set by eliminating background noise and filtering high-frequency interference, which helps to focus on the main noise source during subsequent analysis and avoid the influence of irrelevant noise and interference frequency. This step improves the purity and accuracy of the data and provides a reliable data basis for the subsequent identification and processing of noise frequency components. Through time-frequency analysis and energy distribution calculation, the main frequency of each noise component and its energy contribution are obtained, providing an accurate frequency target for subsequent noise reduction. Identifying the main frequency component can accurately focus on the dominant noise source, avoid consuming extra energy on irrelevant frequencies, and effectively improve the noise reduction effect and energy utilization efficiency. By calculating the phase characteristics of the main frequency components and performing phase continuity correction, accurate phase change data is obtained, so that the reverse sound waves generated subsequently can accurately counter the noise in phase. Phase correction ensures the phase accuracy of the reverse sound wave and provides key data support for enhancing the noise cancellation effect. Through the calculation of the reverse phase value and the analysis of the energy contribution, the frequency and phase characteristic data of the reverse sound wave are obtained, laying the foundation for generating reverse sound waves that are accurately matched in energy and phase. The determination of the amplitude ratio ensures the optimal energy configuration of the reverse sound wave at different frequencies, improving the efficiency and effect of the overall noise reduction. The frequency and phase characteristics of the reverse sound wave are synthesized into a time domain signal through inverse Fourier transform to generate a reverse sound wave signal that counteracts the original noise, and the continuity and stability of the signal are ensured through smoothing to reduce interference with the generator device. The generated reverse sound wave signal has the ability to counter the noise source, providing an operational signal basis for subsequent noise reduction implementation.

[0085] Preferably, step S22 includes the following steps:

[0086] Step S221: performing feature analysis on the noise signal in the pure noise characteristic data set, and selecting a wavelet basis function based on the analysis result to generate a wavelet decomposition layer number;

[0087] The embodiment of the present invention performs feature analysis on the noise signal in the pure noise characteristic data set. The goal of this process is to identify the key features of the noise signal, including the frequency distribution of the signal, the change law of the time domain and the frequency domain, the instantaneous energy, etc. Through time-frequency analysis, the distribution of the noise signal in time and frequency is studied, and the main components and frequency bandwidth of the noise signal are identified. The statistical characteristics of the noise signal such as the mean, variance, peak value, skewness, kurtosis, etc. are calculated to describe its basic waveform and change mode. The noise signal usually has non-stationary characteristics, especially the noise of wind power generation equipment. By analyzing the autocorrelation function and power spectrum density of the signal, the non-stationarity of the noise is judged, and a basis is provided for the subsequent wavelet transform. According to the results of the signal feature analysis, a suitable wavelet basis function is selected for wavelet transform. If the noise signal contains mutations or non-stationary characteristics, a wavelet basis function with good time-frequency localization capability can be selected, and different wavelet basis functions have different decomposition effects on the frequency band. According to the time-frequency characteristics of the noise signal, the wavelet basis function that best suits the signal is selected. This step can be verified by testing different wavelet basis functions, and the wavelet basis that can most effectively reflect the characteristics of the noise signal is selected. Based on the frequency band characteristics of the noise signal and the analysis results, the appropriate number of wavelet decomposition layers is determined.

[0088] Step S222: performing multi-level wavelet transform on the noise signal in the pure noise characteristic data set based on the wavelet decomposition layer number, extracting the approximate coefficients and detail coefficients of each frequency band to generate a reconstructed signal of each frequency band;

[0089] The embodiment of the present invention uses the selected wavelet basis function to perform multi-level wavelet transform on the noise signal in the pure noise characteristic data set. The wavelet transform decomposes the noise signal into multiple frequency bands so as to analyze the signal components in different frequency ranges. According to the determined wavelet decomposition layer number, the multi-level wavelet transform is performed. Each layer of decomposition is convolved and down-sampled by the wavelet basis function to decompose the signal into low-frequency and high-frequency parts. The low-frequency part of each layer represents the overall trend of the signal, while the high-frequency part represents the detail information in the signal, such as mutation or noise change. For each layer of decomposition, the wavelet transform generates a low-frequency part, and the coefficient represents the main trend of the signal. The approximate coefficient contains the main energy component of the signal in the frequency band, and is usually used to describe the general outline of the signal. The high-frequency part of each layer describes the details and changes of the signal, reflecting the rapid changes or mutation parts in the signal. The detail coefficient is often used to analyze the noise component or high-frequency change in the signal. Through the wavelet transform, the signal is divided into multiple frequency bands, and each frequency band corresponds to an approximate coefficient and a plurality of detail coefficients. According to the approximate coefficient and detail coefficient of each frequency band, the signal of each frequency band is reconstructed. The reconstructed signal contains all the information within the frequency band, which helps to analyze the contribution of different frequency bands to the overall noise signal. Through the inverse wavelet transform or synthesis process, the extracted approximate coefficients and detail coefficients of each frequency band are synthesized back. The reconstructed signal of each frequency band can independently reflect the characteristics of the signal in the frequency band. Combining the reconstructed signals of all frequency bands can obtain a complete signal reconstruction, ensuring that no frequency band information is lost, and at the same time, the contribution of each frequency band to the signal can be observed separately.

[0090] Step S223: analyzing the energy variation of the reconstructed signals of each frequency band to generate main frequency component list data.

[0091] The embodiment of the present invention performs energy calculation on each reconstructed frequency band signal. The energy of the signal is usually obtained by integrating or summing the square of the signal. By analyzing the signal energy of each frequency band, the frequency bands where the energy is concentrated or mainly distributed in the signal are identified. This can help identify which frequency ranges contribute the most to the overall noise signal. According to the energy change of each frequency band signal, the main frequency components in the signal are identified. Usually, frequency bands with higher energy correspond to the main frequency components in the noise signal, and these components have a greater impact on the overall characteristics of the noise signal. The signal of each frequency band is analyzed in the time domain and frequency domain, the time-varying characteristics of the frequency components are calculated, the change law of the frequency components over time is observed, and further analysis is performed on which frequency components have a greater contribution to the noise signal in different time periods. The main frequency components identified in all frequency bands, as well as the energy contribution and time-varying characteristics data of these components are integrated into a main frequency component list data.

[0092] The present invention selects suitable optimal wavelet basis functions and wavelet decomposition layers through characteristic analysis of noise signals, laying the foundation for subsequent multi-level wavelet decomposition. This adaptive basis function selection ensures the accuracy of the decomposition process, so that the extracted frequency characteristics can better reflect the actual noise characteristics, and improves the accuracy of noise signal decomposition. Through multi-level wavelet transform, approximate coefficients and detail coefficients on different frequency bands are extracted, so that the spectral characteristics of the noise signal are more refined. The generated reconstructed signals of each frequency band can intuitively show the energy distribution of noise in different frequency bands, which is helpful for further analysis of the main frequency components of the noise. By analyzing the energy changes of the reconstructed signal, the main frequency components can be clearly identified, and list data can be generated, so that high-energy noise components can be accurately located. This main frequency component list provides core reference data for subsequent noise reduction control, making the generation of reverse sound wave signals more effective, thereby significantly improving the overall noise reduction efficiency and effect.

[0093] Preferably, step S23 includes the following steps:

[0094] Step S231: extracting the reference phase value of the main frequency component of the pure noise characteristic data set based on the main frequency component list data, and calculating the phase stability index of each frequency to obtain the reference phase data;

[0095] The embodiment of the present invention extracts each main frequency component in the pure noise signal according to the main frequency component list data. Each frequency component corresponds to a specific signal and exhibits a certain phase characteristic in the time domain or frequency domain. For each main frequency component, a fast Fourier transform is used to extract the phase information of the frequency component. The pure noise signal is converted into the frequency domain, the amplitude and phase information of each main frequency are extracted, and the reference phase value of each frequency component is determined. Generally, the reference phase value can be determined by the initial phase of the signal or the time average value as the reference phase of the signal. The reference phase value of each frequency component is recorded and a reference phase data set is generated. These data will serve as the basis for subsequent phase analysis and adjustment. The phase stability index is used to measure whether the phase change of a frequency component is regular and stable. In order to calculate the phase stability, the standard deviation of the phase change of the frequency component over a period of time is calculated. A smaller standard deviation means that the phase change is stable, and a larger standard deviation means that the phase change is unstable. Detect whether there is a rapid mutation phenomenon. Phase jumps usually indicate that the signal noise is large or the frequency is unstable. By comparing the phase changes in different time periods, the phase stability index of each frequency component is calculated. Commonly used methods include analysis based on periodic changes, phase envelope, instantaneous frequency, etc. By comprehensively analyzing the phase stability indicators of each frequency component, the reference phase data of each frequency component is obtained.

[0096] Step S232: performing phase change analysis on the reference phase data based on the frequency time-varying characteristic data to identify the phase modulation law, and calculating the phase change rate to obtain dynamic phase characteristic data;

[0097] The embodiment of the present invention obtains a list of main frequency components through time-frequency analysis, and describes the change of the noise signal according to the time-varying characteristic data of these frequency components. The frequency time-varying characteristic data contains the evolution law of frequency in time, which can reflect the fluctuation of each frequency component at different time points. In step S231, the reference phase value of the main frequency component has been extracted, and the phase stability of each frequency component has been calculated. Now, these reference phase data are used as input to combine with the frequency time-varying characteristic data for phase change analysis. By combining the frequency time-varying characteristic data with the reference phase data, the phase value of each frequency component at different time points is obtained. This combination can help us understand the change trend of the phase of each frequency component over time. For each frequency component, the phase change is calculated on the time axis. By analyzing the phase signal of each frequency, their phase modulation law is identified. Phase modulation refers to the adjustment or fluctuation of the phase of the frequency component over time, which is closely related to factors such as the operating state of the wind turbine and wind speed changes. The phase signal is Fourier transformed to identify the low-frequency modulation component therein, which helps to identify the periodic or non-periodic changes of the phase modulation. By calculating the instantaneous phase or instantaneous frequency, the dynamic changes of the phase of the frequency component can be captured. The phase envelope is calculated to analyze the change law of the phase over time in order to identify the modulation characteristics of the phase. The phase change rate refers to the rate of change of the phase signal per unit time, which usually reflects the speed of the phase change of the frequency component. The calculation of the phase change rate helps to identify the rapidity or gentleness of the phase change, and then judge the dynamic characteristics of the noise signal. The instantaneous change rate of the phase is obtained by calculating the derivative of the phase signal with respect to time. The instantaneous phase can be obtained by analyzing the signal, and the derivative provides dynamic information about the phase change rate. In order to eliminate the influence of noise on the calculation, the phase change rate can be smoothed to remove the mutation points, so that the results are more stable and reliable. According to the size of the phase change rate, the rate of change of the working state of the generator device can be evaluated. For example, a high-speed phase change rate indicates a sudden change in wind speed or a sharp change in the speed of the wind turbine. The analyzed phase modulation law and the calculated phase change rate are sorted to generate dynamic phase characteristic data.

[0098] Step S233: integrating the reference phase data with the dynamic phase characteristic data to obtain the phase characteristic data of the main frequency.

[0099] The purpose of integrating the reference phase data with the dynamic phase characteristic data in the embodiment of the present invention is to create a more complete and accurate phase characteristic data set. This integration can not only reflect the static phase characteristics of each frequency component, but also reflect its dynamic characteristics that change over time. The reference phase and dynamic phase characteristic data are arranged in chronological order, and a time series data containing the reference phase and dynamic phase characteristics is created for each frequency component. This helps to accurately track the change process of the phase of each frequency component. For each frequency component, the reference phase and the dynamic phase characteristics are aligned in the time and frequency dimensions by corresponding data points. Specifically, the reference phase data can be used as the starting point of the time series, while the dynamic phase data describes the phase modulation process that changes over time. Ensure that the reference phase data and the dynamic phase characteristic data are synchronized in time. If the reference phase data and the dynamic phase characteristic data are not completely consistent in time sampling, interpolation technology can be used to adjust the data so that they match at the same time point. If the data has a large time span or has periodic changes, the data can be divided into smaller time windows so that data integration can be performed in each window, thereby capturing the phase characteristics more carefully. In the frequency dimension, each frequency component is matched. For each frequency component, first determine its reference phase value, and then arrange the corresponding dynamic phase characteristic data in time. In this way, the generated data set will contain the time series data of each frequency, while recording its reference phase value and dynamic change information. In order to improve the comparability of the integrated data, the reference phase data and dynamic phase data can be standardized. Especially when these data come from different data sources or have different dimensions, standardization can eliminate dimensional differences and make the final integrated data more consistent. After data integration, the phase characteristic data of the main frequency is finally obtained.

[0100] The present invention can accurately reflect the reference phase characteristics of the noise signal at the main frequency by extracting the reference phase value of the main frequency component and calculating the phase stability index. The reference phase data lays the foundation for the subsequent phase modulation and anti-phase signal generation, which helps to improve the phase matching degree of the reverse sound wave in the noise reduction system and enhance the noise reduction effect. By analyzing the dynamic change characteristics of the reference phase data, the changing law of the phase modulation can be accurately identified and the phase change rate can be calculated. This dynamic phase characteristic data helps the system to track and adapt to the frequency changes of the noise signal in real time, so that the generated reverse sound wave maintains the same dynamic modulation effect as the noise signal, and improves the adaptability and response speed of the noise reduction system. By integrating the reference phase data and the dynamic phase characteristic data, the complete phase characteristic data of the main frequency is obtained, providing accurate phase control information for the generation of the reverse sound wave. This integrated phase characteristic data ensures that the reverse sound wave can accurately match the phase characteristics of the original noise in a time-varying noise environment, minimize residual noise, and achieve a more significant noise reduction effect.

[0101] Preferably, step S25 comprises the following steps:

[0102] Step S251: based on the reverse acoustic wave frequency characteristic data and the reverse acoustic wave phase characteristic data, the frequency domain signal is converted into the time domain by using an inverse Fourier transform to generate an initial time domain signal of the reverse acoustic wave;

[0103] The inverse Fourier transform in the embodiment of the present invention is a mathematical method for converting a frequency domain signal into a time domain signal. Through this conversion, a complete representation of the reverse sound wave in the time domain can be obtained. The reverse sound wave frequency characteristic data is combined with the reverse sound wave phase characteristic data to ensure that the amplitude and phase of each frequency correspond. The amplitude and phase of each frequency component together constitute the information of the frequency in the frequency domain. Ensure that the sampling rate of all frequency data matches the sampling rate of the time domain data to ensure the correctness of the conversion. Use the numerical integration method to convert the frequency characteristic and phase characteristic data into a time domain signal. Select an appropriate step size to ensure the accuracy of the signal in the time domain. The signal generated by this process is called the "reverse sound wave initial time domain signal", which is a signal that represents the reverse sound wave characteristics in the time domain.

[0104] Step S252: smoothing the initial time domain signal of the reverse acoustic wave, and eliminating the sharp discontinuities and abnormal peaks generated in the inverse Fourier transform process, to obtain a smoothed time domain signal of the reverse acoustic wave;

[0105] The embodiment of the present invention identifies discontinuities or abnormal peaks by analyzing the waveform of the initial time domain signal of the reverse acoustic wave. Usually, these points appear to be much larger than the values ​​of the surrounding data points. A threshold can be set. When a data point exceeds the threshold, it is considered to be an abnormal point. Adjacent data points are used to interpolate and fill in the abnormal points, and the abnormal peaks are eliminated by replacing each data point with the median of its neighboring data points. The initial time domain signal of the reverse acoustic wave is input into the smoothing program. According to the characteristics of the signal, a suitable smoothing method is selected. For low-frequency signals, the moving average method is selected, and for high-frequency noise, wavelet denoising is used. The signal is processed using the selected smoothing method to remove sharp discontinuities and abnormal peaks. After smoothing, the smoothed reverse acoustic wave time domain signal obtained will have less noise and its waveform will be smoother.

[0106] Step S253: performing amplitude and phase calibration on the smoothed reverse acoustic wave time domain signal to obtain reverse acoustic wave signal data.

[0107] In the embodiment of the present invention, when the amplitude of the signal is too large or too small, the signal can be normalized. The normalization method adjusts the amplitude of the signal to a set range. After normalization, the amplitude of the signal will be compressed or amplified to a standard range. For the reverse sound wave signal, in some cases, it is necessary to adjust its amplitude through gain control to match the intensity of the control target. After the amplitude calibration, it is also necessary to verify whether the calibrated signal amplitude meets the input requirements of the system. The phase relationship between the reverse sound wave signal and the target noise source is ensured by phase adjustment. Phase calibration is usually performed based on time alignment to ensure that the waveforms of the reverse sound wave and the noise source match in space and time. According to the calculated phase difference, the phase of the reverse sound wave is adjusted to make it reverse to the phase of the noise source. In some cases, it is necessary to adjust the signal by adding or subtracting a specific phase compensation amount. In some actual systems, the reverse sound wave signal generates a time delay due to different propagation media. Therefore, it is also necessary to perform delay correction to ensure that the time and space relationship between the signal and the target noise source matches by adjusting the start time of the signal. After the amplitude and phase calibration is completed, the signal needs to be further tuned to meet the specific control accuracy and stability requirements. This can be achieved by iteratively adjusting the calibrated signal and continuously optimizing it based on the actual feedback and control effect of the noise source. In actual applications, multiple rounds of feedback correction will be performed to gradually optimize the interference effect of the reverse sound wave by adjusting the amplitude and phase of the signal until the ideal noise suppression effect is achieved. The reverse sound wave signal data is obtained after amplitude and phase calibration.

[0108] The present invention converts the frequency domain signal into the time domain signal through inverse Fourier transform, so that the frequency and phase characteristics of the reverse sound wave signal are accurately presented in the time domain. The generation of the initial time domain signal of the reverse sound wave provides a basis for subsequent smoothing and calibration, ensuring that the reverse sound wave can better match the characteristics of the original noise in the time domain, and improving the time domain accuracy of the noise reduction effect. The smoothing process removes the sharp discontinuities and abnormal peaks in the signal, making the reverse sound wave time domain signal more stable and reducing unnecessary distortion in the signal waveform. The generation of this smooth reverse sound wave time domain signal helps to avoid the reverse sound wave from generating secondary noise in spatial propagation, and improves the stability and silencing effect of the noise reduction system. Amplitude and phase calibration ensure that the smoothed reverse sound wave signal accurately meets the characteristics of the original noise signal in amplitude and phase. The obtained reverse sound wave signal data can ensure that the reverse sound wave and the original noise phase are accurately canceled, thereby achieving the best noise reduction effect. The calibrated signal data enables the noise reduction system to achieve comprehensive matching in frequency, amplitude and phase, greatly improving the accuracy of noise control.

[0109] Preferably, step S3 comprises the following steps:

[0110] Step S31: converting the reverse sound wave signal data into an analog sound wave signal recognizable by a wide-band speaker, and performing segment processing and gain control on the signal to obtain a reverse sound wave driving signal;

[0111] In the embodiment of the present invention, the reverse sound wave frequency domain signal is subjected to inverse Fourier transformation to generate a continuous analog signal. The signal is consistent with the current or voltage signal format that can be received by the wideband speaker, and can drive the speaker to output sound waves. The converted analog signal needs to ensure that the amplitude and phase of the reverse sound wave are consistent to ensure that it can interfere with the noise source in the opposite direction. In order to effectively process and optimize the reverse sound wave driving signal, the signal is usually processed in segments. This is because a long-term continuous sound wave signal is difficult to control or will produce discontinuities in the time domain. Therefore, the signal is divided into multiple small segments for better refined control. The signal is cut through a preset time window, and each signal segment can be processed and adjusted independently to improve the accuracy of the signal. The segmented signal can be adaptively adjusted according to the changes in the actual noise source, especially under different working conditions, the length and amplitude of the signal segment can be dynamically adjusted to ensure the quality of the signal. The purpose of gain control is to adjust the amplitude of the reverse sound wave signal to adapt it to the performance requirements of the wideband speaker and ensure that the signal is strong enough to counter the original noise. Gain control needs to be set according to the output characteristics of the speaker and the intensity of the target noise. The gain control algorithm adjusts the gain of the signal based on real-time monitoring data or theoretical calculations. An adaptive gain control method can be used to dynamically adjust the gain value based on monitoring feedback or changes in the noise environment to achieve the best noise elimination effect. After the above signal conversion and gain control, the final reverse sound wave drive signal is obtained, which can be directly used as the input signal of the wide-band speaker.

[0112] Step S32: inputting the reverse sound wave driving signal into the wide-band speaker for output, superimposing the reverse sound wave output by the wide-band speaker and the original noise generated by the generator device in space to obtain superimposed spatial sound field data;

[0113] The embodiment of the present invention inputs the reverse sound wave driving signal that has been gain adjusted and optimized into the wideband speaker. After receiving this signal, the speaker will generate a corresponding sound wave output according to the frequency, amplitude and phase of the signal. The wideband speaker is a speaker that can cover a wide frequency range and is usually designed to process low-frequency to high-frequency signals. Through the output of the wideband speaker, a reverse sound wave including a noise interference component can be generated, and the sound wave outputted by the wideband speaker should interfere with the noise generated by the generator device. When the speaker outputs the reverse sound wave, the sound wave will propagate in the air and superimpose on the original noise generated by the generator device. Since the two sound waves have opposite phases and the frequency characteristics of the reverse sound wave correspond to the frequency components of the noise, the superimposed sound wave can reduce or eliminate the original noise through interference. According to the phase and amplitude of the reverse sound wave, the reverse sound wave will interfere with the original noise in the propagation area of ​​the noise source. Ideally, the superposition of the reverse sound wave and the noise will significantly reduce the noise intensity. The effect of spatial superposition depends on the phase difference between the reverse sound wave and the original noise, the degree of frequency matching and the spatial characteristics of sound wave propagation. By tracing the superposition effect of the reverse sound wave and the original noise through simulation or actual measurement, the spatial sound field data after superposition can be obtained. The spatial sound field data includes the intensity and frequency components of the superimposed sound waves and the changes in sound pressure at different locations in space. In order to evaluate and optimize the superposition effect, it is necessary to monitor the changes in the sound field in real time to obtain the superimposed spatial sound field data.

[0114] Step S33: Arrange noise monitoring sensors in the preset noise reduction area based on the superimposed spatial sound field data, monitor the superimposed spatial sound field in real time, and perform effect analysis on the monitoring data to obtain noise reduction effect feedback data.

[0115] The embodiment of the present invention presets a noise reduction area around the generator device or within the noise propagation range, and selects appropriate monitoring points according to the location of the noise source, the environmental structure and the target noise reduction effect. Noise monitoring sensors should be arranged in different spatial positions to cover the entire noise reduction area to ensure that the noise changes at different positions can be fully reflected. Highly sensitive noise sensors or sound pressure sensors are selected. These devices should have strong anti-interference capabilities and can accurately detect changes in the sound field in the complex environment of the generator device. The sensor needs to be able to measure noise signals with a wide frequency range and have a high sampling rate to ensure real-time acquisition of sound pressure data in the space. The sensor collects the spatial sound field data after superposition in real time, including information such as sound pressure level, frequency component, and spatial position. These data can reflect the overall sound field situation after the reverse sound wave is superimposed with the original noise. The collected sound field data is transmitted to the central control system in real time through the data transmission network for immediate processing. The control system performs synchronous analysis of signals from different sensors to ensure the integrity and accuracy of the monitoring data. According to the data provided by the sensor, the difference between the superimposed sound field and the original noise is compared to analyze the noise reduction effect. Commonly used analysis methods include sound pressure level difference analysis, noise spectrum change analysis, and signal-to-noise ratio improvement evaluation. The goal of the analysis is to evaluate the noise reduction effect of the reverse sound wave and determine whether the expected noise reduction level has been achieved. The noise changes at different locations need to be considered during the analysis to evaluate the effect of the reverse sound wave in the entire noise reduction area. Based on the analysis results of the real-time monitoring data, noise reduction effect feedback data is generated.

[0116] The present invention converts a digital reverse sound wave signal into an analog signal that can be recognized by a wide-band speaker, so that the signal is suitable for physical output, and performs segmented processing and gain control on the signal, thereby optimizing the signal strength in different frequency bands. The generated reverse sound wave driving signal not only ensures the signal accuracy, but also improves the output quality of the sound, ensures that the speaker can output a more accurate noise reduction signal, and lays the foundation for subsequent spatial noise reduction. The reverse sound wave is output by a wide-band speaker, and is physically superimposed with the original noise in space. The reverse sound wave can effectively cancel the noise and reduce the overall noise. The generation of superimposed spatial sound field data reflects the sound field distribution of the noise reduction area, and provides a reliable data basis for subsequent noise reduction effect monitoring and adjustment. The noise level is monitored in real time in the superimposed sound field, and the noise reduction effect can be dynamically evaluated and adjusted. The generated noise reduction effect feedback data provides a basis for the adaptive adjustment of the system, so that the noise reduction system can timely optimize the reverse sound wave signal, improve the noise reduction performance and realize closed-loop control.

[0117] Preferably, step S4 comprises the following steps:

[0118] Step S41: analyzing the noise reduction amount of each frequency band on the noise reduction effect feedback data to obtain noise reduction effect analysis data;

[0119] The embodiment of the present invention divides the spectral components in the noise reduction effect feedback data into different frequency bands. Generally, the noise signal has different effects in different frequency bands such as low frequency, medium frequency and high frequency, so different frequency bands need to be evaluated separately during analysis. For example, the impact of low-frequency noise is greater, while high-frequency noise is more difficult to be effectively suppressed. The noise in each frequency band is quantified, and the noise reduction amount in each frequency band is calculated. The noise reduction amount can be obtained by calculating the difference in sound pressure level after the noise source signal and the reverse sound wave signal are superimposed. For example, the noise reduction amount of a certain frequency band can be obtained by comparing the original noise signal of the frequency band with the change in the signal after reverse sound wave processing. Common noise reduction analysis methods include sound pressure level comparison method, spectrum comparison method, signal-to-noise ratio improvement method, etc. These methods are helpful to evaluate the noise control effect of reverse sound waves on different frequency bands. By analyzing the noise reduction effect of each frequency band, the noise reduction amount of each frequency band is obtained. These noise reduction amounts will reflect the suppression effect of reverse sound waves on the noise of each frequency band. Noise reduction effect analysis data usually includes key information such as frequency band noise reduction amount, spectrum changes, and noise intensity changes.

[0120] Step S42: adjusting the frequency, phase and gain parameters of the filter using an adaptive filtering algorithm based on the noise reduction effect analysis data to obtain filter adjustment parameter data;

[0121] In the embodiment of the present invention, before the adaptive filtering is adjusted, the noise reduction effect analysis data is processed as necessary to ensure the accuracy and consistency of the data. It is necessary to denoise, smooth or filter the noise data to remove potential interference and maintain the quality of the signal. According to the noise reduction effect analysis data, a suitable adaptive filtering algorithm is selected. Initialize the filter parameters and set the preliminary conditions for adaptive filtering. For example, an initial frequency range or phase value is set as the starting point for filter adjustment. According to the frequency band noise reduction amount in the noise reduction effect analysis data, the filter will automatically adjust the frequency parameters according to the poor noise reduction effect of certain frequency bands to strengthen the suppression of these frequency bands. For example, if the noise reduction effect of some low-frequency or high-frequency components is not obvious, the adaptive filter will adjust its frequency response to enhance the suppression of these frequency bands. According to the feedback data, especially the comparison of the phase between the noise source and the reverse sound wave, the phase response of the filter will be corrected accordingly. Adjusting the phase parameter helps to ensure that the interference effect between the reverse sound wave and the original noise is optimal, thereby achieving the maximum noise reduction effect. The filter will adjust the gain value according to the noise reduction effect of each frequency band to optimize the amplitude of the sound signal and avoid the signal in some frequency bands being too strong or too weak, thereby ensuring the balance of each frequency band and the best sound wave interference effect. During the adaptive filtering process, the filter will continuously adjust its parameters according to the real-time input noise reduction effect data and the changes in the noise environment. Each parameter adjustment of the filter will be based on the previously fed back data, forming a dynamic adjustment process. In this way, the filter can be continuously optimized according to real-time changes in different working environments. The goal of adaptive filtering is to minimize the error between the input signal and the desired output, so the filter will continue to adjust the frequency, phase and gain until the system achieves the optimal noise reduction effect. Finally, after a series of adaptive adjustments, the filter parameters will be optimized and output as filter adjustment parameter data.

[0122] Step S43: optimizing and calculating the reverse acoustic wave characteristic data according to the filter adjustment parameter data to obtain adjusted reverse acoustic wave characteristic data.

[0123] According to the frequency adjustment parameters of the filter in the embodiment of the present invention, the frequency characteristics of the reverse sound wave will be recalculated. Specifically, according to the optimized frequency response, the frequency component of the reverse sound wave is adjusted to better interfere with and eliminate the target noise frequency band. If the noise in certain frequency bands is not effectively suppressed, the system will make additional adjustments to enhance the reverse signals in these frequency bands. The adjusted phase parameters will be used to correct the phase characteristics of the reverse sound wave. The system will compensate according to the adjusted phase difference to maximize the interference effect between the phase of the reverse sound wave and the original noise. This process can eliminate the phenomenon of insufficient noise reduction caused by the initial phase mismatch. The gain adjustment parameter will be used to adjust the amplitude of the reverse sound wave. The system will adjust the energy of the reverse sound wave according to the optimized gain value so that it reaches the optimal sound pressure in the required frequency band, thereby achieving the maximum noise suppression effect. The reverse sound wave characteristic data will be further calculated based on the adjusted frequency, phase and gain parameters. The optimization calculation can accurately calculate the adjusted reverse sound wave characteristics based on numerical models such as Fourier transform, time-frequency analysis, amplitude modulation, etc. These calculations can be verified by simulation models to ensure that each frequency band and phase adjustment can be accurately reflected in the reverse acoustic wave characteristic data. A new reverse acoustic wave signal is generated through the adjusted reverse acoustic wave characteristic data.

[0124] The present invention generates noise reduction effect analysis data by performing a detailed analysis of the noise reduction amount of each frequency band, so that the system can identify the frequency band differences of the noise reduction effect. This analysis can provide accurate feedback for the optimization of subsequent filters, ensuring that effective noise reduction can be achieved in different frequency bands, thereby improving the overall noise reduction performance. The key parameters of the filter are dynamically adjusted using an adaptive filtering algorithm, and the frequency, phase and gain are adjusted according to the noise reduction requirements of different frequency bands to achieve more refined frequency band noise reduction control. The generated filter adjustment parameter data provides a dynamic parameter adjustment scheme, so that the system can be adaptively adjusted in different noise environments, effectively enhancing the stability and flexibility of noise reduction. The reverse acoustic wave characteristic data is optimized and calculated to ensure that the reverse acoustic wave signal can match the original noise in terms of frequency, phase and gain, thereby achieving more efficient noise cancellation. The generated adjusted reverse acoustic wave characteristic data is the key to optimizing the reverse acoustic wave output, further improving the noise cancellation effect and reducing the noise residual of the system, and providing a solid foundation for closed-loop adaptive control.

[0125] Preferably, step S42 includes the following steps:

[0126] Step S421: Calculate the residual noise energy of each frequency band on the noise reduction effect analysis data to identify the key frequency band and generate frequency band noise energy distribution data;

[0127] The embodiment of the present invention decomposes the noise reduction effect analysis data into energy components of different frequency bands. The noise energy of each frequency band is obtained by integrating the power spectrum of the signal in the frequency band, which usually involves Fourier transform or wavelet transform of the time domain signal or the frequency domain signal. The residual noise energy refers to the noise component that still exists after the reverse sound wave is applied. An energy spectrum calculation method, such as an integral method, is used to quantify the residual noise energy of each frequency band. Specifically, frequency analysis is used to extract the spectrum of the signal, and then the energy of each frequency band is calculated. On this basis, the frequency bands with the most significant residual noise are determined according to the energy value. Usually, these frequency bands have a greater impact on the overall noise control effect, so they need to be optimized. Key frequency bands refer to those frequency bands with higher residual noise energy and have a significant impact on the noise reduction effect. These frequency bands contain the main frequency components of the equipment noise, or frequency areas that the reverse sound wave fails to effectively cover. By analyzing the noise energy of each frequency band, those frequency bands that need special attention and optimization can be identified. By setting an energy threshold, the frequency bands with higher residual noise energy are marked. Based on the result of the residual noise energy calculation, the band noise energy distribution data is generated.

[0128] Step S422: Optimize and adjust the key frequency bands in the frequency band noise energy distribution data using an adaptive filtering algorithm to obtain filter adjustment parameter data.

[0129] The embodiment of the present invention selects frequency bands with higher residual noise energy as "key frequency bands" by analyzing the frequency band noise energy distribution data. These frequency bands usually have a greater impact on the noise reduction effect. The selection of key frequency bands is based on a preset threshold, for example, selecting frequency bands with noise energy values ​​greater than a certain specific value. The key frequency band covers a certain frequency range, such as low frequency (20Hz-200Hz), medium frequency (200Hz-2kHz) or high frequency (above 2kHz), and the specific range is set according to the characteristics of the noise and the working conditions of the generator device. An adaptive filter is a filter that can automatically adjust the filter parameters according to the characteristics of the input signal. Its purpose is to make the output signal gradually approach the ideal noise reduction target by continuously adjusting the characteristics of the filter. The selection of a suitable adaptive filtering algorithm depends on the noise characteristics, real-time computing power and noise reduction accuracy requirements. The adaptive filter dynamically adjusts its frequency, gain and phase parameters according to the changes in the input signal, so that the signal output by the filter optimizes the noise reduction effect. Specifically, the filter will enhance or attenuate the key frequency band to reduce the residual noise of the frequency band. According to the noise energy characteristics of the key frequency band, the operating frequency of the filter is adjusted so that it is concentrated in the noise frequency range. By adjusting the gain, the energy of a specific frequency band is enhanced or attenuated to reduce the noise energy. The phase relationship between the reverse sound wave and the noise signal is optimized so that the two signals can be effectively superimposed to further improve the noise reduction effect. The adaptive filter outputs the filter adjustment parameter data based on the optimization results of the algorithm.

[0130] The present invention can identify the high-energy noise bands that still exist after noise reduction by calculating the residual noise energy of each frequency band, thereby determining the key frequency bands for noise reduction. The generated frequency band noise energy distribution data provides a refined distribution analysis of the residual noise, which helps to accurately lock the frequency bands that need to be further optimized and improve the pertinence and accuracy of noise reduction. By optimizing and adjusting the key frequency bands through the adaptive filtering algorithm, the filter parameters can be adjusted in a targeted manner to effectively reduce the impact of residual noise. The generated filter adjustment parameter data enables the system to suppress noise more efficiently on the key frequency bands, thereby achieving better noise reduction effects on the most needed frequency bands.

[0131] Preferably, the present invention further provides a generator device for executing the control method of the generator device as described above, the generator device comprising:

[0132] The noise characteristic monitoring module is used to monitor the noise frequency, amplitude and phase of the generator device in real time during operation and generate a noise characteristic data set;

[0133] The noise characteristic analysis and reverse signal generation module is used to extract the main frequency and phase characteristics of the noise source from the noise characteristic data set, and reversely analyze the data to generate reverse sound wave signal data with the opposite phase to the noise source;

[0134] The reverse sound wave transmission and superposition module is used to transmit the reverse sound wave signal data to the reverse sound wave generating device of the generator device, output the reverse sound wave through the wide-band speaker and superimpose it with the original noise in space, and monitor the noise after noise reduction in real time to obtain noise reduction effect feedback data;

[0135] The reverse signal adaptive adjustment module is used to dynamically adjust the frequency, amplitude and phase of the reverse sound wave through an adaptive filtering algorithm based on the noise reduction effect feedback data to obtain the adjusted reverse sound wave characteristic data;

[0136] The closed-loop control and adaptive optimization module is used to apply the adjusted reverse acoustic wave characteristic data to step S3 to dynamically optimize the reverse acoustic wave and realize closed-loop adaptive control.

[0137] The present invention can monitor the noise frequency, amplitude and phase of the generator device in real time during operation, and generate a noise characteristic data set. By real-time monitoring of the main characteristics of the noise, timely and accurate acquisition of the noise source information is ensured, and reliable data support is provided for subsequent reverse signal analysis and generation, thereby improving the accuracy of noise reduction control. By analyzing the noise characteristic data set, the main frequency and phase characteristics of the noise source are extracted, and reverse sound wave signal data with a phase opposite to the noise source is generated. Through reverse analysis, a sound wave signal with a reverse phase can be accurately generated, thereby effectively offsetting the main components of the noise source, providing a basis for achieving significant noise reduction. The reverse sound wave signal data is transmitted to the reverse sound wave generating device of the generator device, and is output through a wide-band speaker and superimposed in space with the original noise. The real-time monitoring function of this module can capture the noise reduction effect after superposition, and generate feedback data, which provides a real-time reference for the adjustment of the reverse signal, and ensures the continuous optimization and verification of the noise reduction effect. Using an adaptive filtering algorithm, this module dynamically adjusts the frequency, amplitude and phase of the reverse sound wave according to the feedback data to generate optimized reverse sound wave characteristic data. Through adaptive adjustment, the reverse acoustic wave signal can be flexibly adjusted and precisely controlled, which can better cope with the noise changes in the operation of the generator device, and further enhance the noise reduction effect and stability of the system. Closed-loop control is realized, and the adjusted reverse acoustic wave characteristic data is applied to the noise superposition process to achieve dynamic optimization of the reverse acoustic wave. Closed-loop adaptive control can ensure that the entire system automatically adjusts to the optimal noise reduction state during operation, achieving long-term and continuous noise reduction effects. The adaptive optimization characteristics of this module help the system maintain efficient noise reduction performance in a changing noise environment.

[0138] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is not limited by the above description, so it is intended to include all changes within the meaning and scope of the equivalent elements of the application documents in the present invention.

[0139] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for controlling a generator device, characterized in that: The following steps are involved: Step S1: Real-time monitoring of the noise frequency, amplitude and phase of the generator device during operation to generate a noise characteristic data set; Step S2: extracting the main frequency and phase characteristics of the noise source from the noise characteristic data set, and performing reverse analysis on the data to generate reverse sound wave signal data with a phase opposite to that of the noise source; Step S3: transmitting the reverse sound wave signal data to the reverse sound wave generating device of the generator device, outputting the reverse sound wave through the broadband speaker to superimpose the reverse sound wave with the original noise in space, and monitoring the noise after noise reduction in real time to obtain noise reduction effect feedback data; Step S4: dynamically adjusting the frequency, amplitude and phase of the reverse sound wave through an adaptive filtering algorithm based on the noise reduction effect feedback data to obtain adjusted reverse sound wave characteristic data; Step S5: Apply the adjusted reverse acoustic wave characteristic data to step S3 to dynamically optimize the reverse acoustic wave and achieve closed-loop adaptive control.

2. The control method of the generator device according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: Arranging an acoustic sensor array in the noise generating area of ​​the generator device and establishing sensor coordinates to record the spatial position information of each monitoring point and obtain sensor position coordinate data; Step S12: Based on the sensor position coordinate data, the ambient background noise at each wind speed is collected when the generator device is stopped to obtain background noise spectrum data; Step S13: Based on the sensor position coordinate data, the sound pressure signal under the operation state of the generator device is collected, and analog-to-digital conversion and digital processing are performed to obtain original time domain sound pressure data; Step S14: performing a fast Fourier transform on the original time-domain sound pressure data and calculating the power spectrum density to obtain the runtime noise spectrum data; Step S15: Calculate the phase information of each frequency band signal on the original time-domain sound pressure data, and perform phase unwrapping processing to obtain the runtime noise phase spectrum data; Step S16: preprocessing the runtime noise spectrum data and the runtime noise phase spectrum data, and integrating the preprocessed data to obtain runtime noise frequency domain data; Step S17: Integrate the sensor position coordinate data, background noise spectrum data, original time domain sound pressure data, and runtime noise frequency domain data, and timestamp the data to generate a noise characteristic data set.

3. The control method of the generator device according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: performing background noise elimination on the noise characteristic data set, and using a digital filter to eliminate high-frequency interference to obtain a pure noise characteristic data set; Step S22: performing time-frequency analysis on the pure noise characteristic data set, and calculating the spectrum peak and energy distribution to identify the main frequency components, and obtain the main frequency component list data, wherein the main frequency component list data includes the energy contribution of each frequency and the frequency time-varying characteristic data; Step S23: calculating the phase characteristics of the main frequency components of the pure noise characteristic data set based on the main frequency component list data, and performing phase continuity correction to analyze the phase change law and obtain the phase characteristic data of the main frequency; Step S24: calculating the reverse phase value of each frequency component based on the main frequency component list and the phase characteristic data of the main frequency, and determining the amplitude ratio according to the energy contribution of each frequency to obtain the reverse sound wave frequency characteristic data and the reverse sound wave phase characteristic data; Step S25: Using inverse Fourier transform to synthesize the reverse acoustic wave frequency characteristic data and the reverse acoustic wave phase characteristic data into a time domain signal, and performing signal smoothing processing to obtain reverse acoustic wave signal data.

4. The control method of the generator device according to claim 3, characterized in that: Step S22 includes the following steps: Step S221: performing feature analysis on the noise signal in the pure noise characteristic data set, and selecting a wavelet basis function based on the analysis result to generate a wavelet decomposition layer number; Step S222: performing multi-level wavelet transform on the noise signal in the pure noise characteristic data set based on the wavelet decomposition layer number, extracting the approximate coefficients and detail coefficients of each frequency band to generate a reconstructed signal of each frequency band; Step S223: analyzing the energy variation of the reconstructed signals of each frequency band to generate main frequency component list data.

5. The control method of the generator device according to claim 3, characterized in that: Step S23 includes the following steps: Step S231: extracting the reference phase value of the main frequency component of the pure noise characteristic data set based on the main frequency component list data, and calculating the phase stability index of each frequency to obtain the reference phase data; Step S232: performing phase change analysis on the reference phase data based on the frequency time-varying characteristic data to identify the phase modulation law, and calculating the phase change rate to obtain dynamic phase characteristic data; Step S233: integrating the reference phase data with the dynamic phase characteristic data to obtain the phase characteristic data of the main frequency.

6. The control method of the generator device according to claim 3, characterized in that: Step S25 includes the following steps: Step S251: based on the reverse acoustic wave frequency characteristic data and the reverse acoustic wave phase characteristic data, the frequency domain signal is converted into the time domain by using an inverse Fourier transform to generate an initial time domain signal of the reverse acoustic wave; Step S252: smoothing the initial time domain signal of the reverse acoustic wave, and eliminating the sharp discontinuities and abnormal peaks generated in the inverse Fourier transform process, to obtain a smoothed time domain signal of the reverse acoustic wave; Step S253: performing amplitude and phase calibration on the smoothed reverse acoustic wave time domain signal to obtain reverse acoustic wave signal data.

7. The control method of the generator device according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: converting the reverse sound wave signal data into an analog sound wave signal recognizable by a wide-band speaker, and performing segment processing and gain control on the signal to obtain a reverse sound wave driving signal; Step S32: inputting the reverse sound wave driving signal into the wide-band speaker for output, superimposing the reverse sound wave output by the wide-band speaker and the original noise generated by the generator device in space to obtain superimposed spatial sound field data; Step S33: Arrange noise monitoring sensors in the preset noise reduction area based on the superimposed spatial sound field data, monitor the superimposed spatial sound field in real time, and perform effect analysis on the monitoring data to obtain noise reduction effect feedback data.

8. The control method of the generator device according to claim 1, characterized in that: Step S4 includes the following steps: Step S41: analyzing the noise reduction amount of each frequency band on the noise reduction effect feedback data to obtain noise reduction effect analysis data; Step S42: adjusting the frequency, phase and gain parameters of the filter using an adaptive filtering algorithm based on the noise reduction effect analysis data to obtain filter adjustment parameter data; Step S43: optimizing and calculating the reverse acoustic wave characteristic data according to the filter adjustment parameter data to obtain adjusted reverse acoustic wave characteristic data.

9. The control method of the generator device according to claim 1, characterized in that: Step S42 includes the following steps: Step S421: Calculate the residual noise energy of each frequency band on the noise reduction effect analysis data to identify the key frequency band and generate frequency band noise energy distribution data; Step S422: Optimize and adjust the key frequency bands in the frequency band noise energy distribution data using an adaptive filtering algorithm to obtain filter adjustment parameter data.

10. A generator device, characterized in that: The invention comprises a reverse sound wave generating device, a wide-band loudspeaker and a control system, wherein the reverse sound wave generating device is disposed in the generator device, the wide-band loudspeaker is disposed in the reverse sound wave generating device, the control system is installed in the generator device, the control system is used to execute the control method of the generator device according to claim 1, and the control system comprises: The noise characteristic monitoring module is used to monitor the noise frequency, amplitude and phase of the generator device in real time during operation and generate a noise characteristic data set; The noise characteristic analysis and reverse signal generation module is used to extract the main frequency and phase characteristics of the noise source from the noise characteristic data set, and reversely analyze the data to generate reverse sound wave signal data with the opposite phase to the noise source; The reverse sound wave transmission and superposition module is used to transmit the reverse sound wave signal data to the reverse sound wave generating device of the generator device, output the reverse sound wave through the wide-band speaker and superimpose it with the original noise in space, and monitor the noise after noise reduction in real time to obtain noise reduction effect feedback data; The reverse signal adaptive adjustment module is used to dynamically adjust the frequency, amplitude and phase of the reverse sound wave through an adaptive filtering algorithm based on the noise reduction effect feedback data to obtain the adjusted reverse sound wave characteristic data; The closed-loop control and adaptive optimization module is used to apply the adjusted reverse acoustic wave characteristic data to step S3 to dynamically optimize the reverse acoustic wave and realize closed-loop adaptive control.

Images