Engine line spectrum noise estimation method, vehicle active noise reduction control method

Abstract

The present application relates to an engine line spectrum noise estimation method and a vehicle active noise reduction control method, which includes: calculating the main order frequency of the engine line spectrum noise according to the real-time speed of the engine; obtaining the primary noise through the internal model control framework, and performing spectrum estimation on the primary noise through the frequency estimation algorithm to obtain the main order frequency and harmonic frequency of the engine line spectrum noise; using a pre-trained neural network model as an engine frequency prediction model, combined with engine operation data as input, and outputting the predicted main order frequency and harmonic frequency of the engine line spectrum noise in real time; performing a comprehensive analysis on the above frequency prediction data to generate the final order frequency corresponding to the engine line spectrum noise. The present application combines the comprehensive output results of different estimation schemes to generate an order frequency that is closer to the actual noise, reduce the impact of frequency imbalance on engine noise control, thereby achieving a better order noise control effect and providing users with a better noise reduction experience.

Description

Technical Field

[0001] The present invention relates to the technical field of automobile noise processing, and in particular to a method for estimating engine line spectrum noise, and active noise reduction control based on engine line spectrum noise estimation. Background Art

[0002] In recent years, the rapid development of my country's automotive industry and the extensive construction of high-grade highways have led to increasing average vehicle speeds, resulting in increasingly high levels of vehicle noise. Consequently, the impact of interior noise on ride comfort and customer satisfaction has gradually attracted attention. While providing power, the engines in fuel-powered vehicles also generate unpleasant low-frequency noise. This is particularly true during idling and low-speed driving, where engine noise dominates the noise spectrum, severely impacting driver and passenger comfort and inconsistent with current automotive market demands.

[0003] Currently, passive noise reduction methods based on sound absorption, sound insulation and vibration isolation are generally used to reduce noise in passenger car cabs. The noise reduction mechanism of this type of method is mainly to use porous materials such as sound-absorbing cotton and sound insulation structures such as single-layer thin plates to interact with sound waves to consume noise energy. It has a good noise reduction effect on medium and high-frequency noise such as noise caused by road excitation in the cab and wind noise under high-speed driving. However, it is helpless against the low-frequency order noise of the engine mentioned above, and it is relatively costly.

[0004] To address these challenges, Active Noise Control (ANC) technology has emerged. This technology, also known as active noise reduction, utilizes an additional sound source (typically a speaker) to emit sound waves with the same frequency, amplitude, and 180 degrees phase shift as the primary noise. These sound waves destructively interfere with the primary noise in space, thereby attenuating the primary noise. Currently, the primary method for actively controlling engine line spectrum noise utilizes the multi-channel adaptive notch FxLMS algorithm.

[0005] The current common method for obtaining engine line spectrum noise is to estimate the frequency of the main-order noise using engine speed and the number of cylinders. However, in practice, due to factors such as temperature and humidity fluctuations, component aging, and estimation errors, accurate main-order noise frequencies are often not available. This is known as frequency offset, and frequency offset can severely degrade the control performance of active noise control based on the main-order noise, or even completely eliminate its effectiveness. Summary of the Invention

[0006] The present invention application provides an engine line spectrum noise prediction method and a vehicle active noise reduction control method, which can achieve more accurate calculation of order frequencies, based on which vehicle active noise reduction control can be implemented to achieve better noise reduction effects.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is:

[0008] The present invention provides an engine line spectrum noise prediction method, which includes:

[0009] Calculating the main order frequency of the engine line spectrum noise according to the real-time speed of the engine as the first frequency prediction data;

[0010] Primary noise is calculated and acquired through an internal model control framework, and spectrum estimation is performed on the primary noise through a frequency estimation algorithm to separate and acquire the main order frequency and harmonic frequency of the engine line spectrum noise as second frequency prediction data;

[0011] Pre-training a neural network model to obtain an engine frequency prediction model, acquiring engine operating data in real time, using the engine operating data as input to the engine frequency prediction model, and outputting predicted main-order frequencies and harmonic frequencies of the engine line spectrum noise in real time as third frequency prediction data;

[0012] Comprehensively analyze the first frequency prediction data, the second frequency prediction data, and the third frequency prediction data to generate the final order frequency corresponding to the engine line spectrum noise.

[0013] According to one embodiment of the present invention, a comprehensive analysis is performed on the first frequency prediction data, the second frequency prediction data, and the third frequency prediction data to generate the final order frequency corresponding to the engine line spectrum noise, including the following working steps:

[0014] Extract the candidate fundamental frequencies of the first frequency prediction data, the second frequency prediction data, and the third frequency prediction data respectively to obtain the corresponding first candidate fundamental frequencies. , the second candidate fundamental frequency and the third candidate fundamental frequency , and obtain the first weight corresponding to each candidate fundamental frequency , second weight , the third weight ;

[0015] The fundamental frequency of the final output order frequency is calculated according to the following formula: ,

[0016]

[0017] The number of order frequencies of the final output is calculated according to the following formula:

[0018]

[0019] in, is the number of order frequencies in the third frequency prediction data,

[0020]

[0021] The first in the second frequency prediction data frequencies, is the number of order frequencies in the second frequency prediction data, For the general When sorted in ascending order, the index corresponding to the maximum frequency with a confidence level of 1. If all confidence levels are not 1, then ;

[0022] The final output order frequency is .

[0023] According to one embodiment of the present invention, the first candidate fundamental frequency The frequency is the same as the first frequency prediction data, and the first weight It is between 0.3 and 0.4.

[0024] According to one embodiment of the present invention, the third candidate base frequency The frequency is the same as the third frequency prediction data, and the third weight It is between 0.2 and 0.3.

[0025] According to one embodiment of the present invention, candidate fundamental frequencies are extracted from the second frequency prediction data according to the following formula:

[0026]

[0027]

[0028]

[0029] in, The candidate fundamental frequency is obtained by extracting the second frequency prediction data. The first in the second frequency prediction data frequencies, is the number of main order frequencies and harmonic frequencies in the second frequency prediction data, Set to 2 or 3 or 4, is the weight of the candidate fundamental frequency obtained by extracting the second frequency prediction data, The second frequency prediction data The confidence level of a frequency.

[0030] According to one embodiment of the present invention, when extracting candidate fundamental frequencies from the second frequency prediction data, multiple candidate fundamental frequencies are obtained, and a candidate fundamental frequency list is formed as follows:

[0031]

[0032] The multiple candidate fundamental frequencies each have a corresponding weight, and a corresponding weight list is formed as follows:

[0033]

[0034] Setting thresholds , 0.5≤ ≤4 if ,but ,renew and delete , and are respectively the first and The candidate frequencies of ,renew and delete , and are the first and weights;

[0035] Select the candidate fundamental frequency closest to the first candidate fundamental frequency from the updated candidate fundamental frequency list The frequency of the second candidate fundamental frequency of the second frequency prediction data and take the second candidate fundamental frequency The corresponding weight is used as the second weight of the second frequency prediction data .

[0036] According to one embodiment of the present invention, during the acquisition of the first frequency prediction data,

[0037] The real-time speed of the engine is obtained in real time. After obtaining the real-time speed of the engine, the main order frequency of the engine line spectrum noise is calculated using the following formula:

[0038]

[0039] is the main order frequency of the engine line spectrum noise, nc (number of clinders) is the number of cylinders in the engine, rpm is the engine speed (revolutions per minute), is the stroke coefficient.

[0040] According to one embodiment of the present invention, the process of obtaining the second frequency prediction data includes the following steps:

[0041] Two microphones are installed inside the vehicle carrying the engine. The primary noise at the two microphones is estimated by internal model control and is recorded as and ,right and Do power spectrum estimation, find and obtain Power spectrum and The peak frequency point in the power spectrum that meets the minimum amplitude limit and the minimum distance limit;

[0042] After removing the null values ​​in the peak frequency points, the two power spectrum information are synchronized to separate and obtain the main order frequency and harmonic frequency of the engine line spectrum noise as the second frequency prediction data.

[0043] According to one embodiment of the present invention, the process of pre-training a neural network model to obtain an engine frequency prediction model includes the following steps:

[0044] Collecting several sets of engine data offline as training sets, wherein the engine data includes engine operating data and the main order frequency and harmonic frequency of the corresponding engine line spectrum noise;

[0045] The neural network model is pre-trained using the training set, and the trained neural network model is the engine frequency prediction model.

[0046] According to one embodiment of the present invention, the engine operating speed includes vehicle speed, engine speed, coolant temperature, load torque, and throttle opening.

[0047] In particular, the present invention also provides a vehicle active noise reduction control method, wherein the order frequency of the engine line spectrum noise is obtained, and active control is performed based on the order frequency of the engine line spectrum noise, and the order frequency of the engine line spectrum noise is calculated and obtained using the engine line spectrum noise estimation method described above.

[0048] Compared with the prior art, the advantages and beneficial effects of the engine line spectrum noise prediction method and vehicle active noise reduction control method of the patent application of this invention are:

[0049] This application uses the comprehensive output results of three different estimation schemes through engine speed, spectrum analysis and model prediction based on relevant prior information. After comprehensive evaluation, it can output more accurate engine order noise frequency and order number, reduce the impact of frequency imbalance on engine noise control, and thus achieve better order noise control effect, making the final active noise reduction effect reliable and stable, thereby providing users with a better noise reduction experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Hereinafter, some specific embodiments of the present invention will be described in detail in an exemplary and non-limiting manner with reference to the accompanying drawings. The same reference numerals in the accompanying drawings indicate the same or similar components or parts. It should be understood by those skilled in the art that these drawings are not necessarily drawn to scale. In the accompanying drawings:

[0051] Figure 1 1 is a schematic diagram of the working process of an engine line spectrum noise estimation method according to one embodiment of the present invention;

[0052] Figure 2 2 is a schematic diagram of a process for obtaining second frequency prediction data according to embodiment 1 of the present invention;

[0053] Figure 3 2 is a schematic diagram of a process for obtaining the third frequency prediction data according to Embodiment 1 of the present invention;

[0054] Figure 4 This is a principle block diagram of generating the order frequency corresponding to the final engine line spectrum noise based on the three frequency prediction data in Example 1 of the present invention;

[0055] Figure 5 4 is a principle block diagram of the vehicle active noise reduction control method according to embodiment 1 of the present invention. DETAILED DESCRIPTION

[0056] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0057] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0058] Example 1:

[0059] This embodiment provides a method for estimating engine line spectrum noise, such as Figure 1 As shown, it includes:

[0060] Calculating the main order frequency of the engine line spectrum noise according to the real-time speed of the engine as the first frequency prediction data;

[0061] Primary noise is calculated and acquired through an internal model control framework, and spectrum estimation is performed on the primary noise through a frequency estimation algorithm to separate and acquire the main order frequency and harmonic frequency of the engine line spectrum noise as second frequency prediction data;

[0062] Pre-training a neural network model to obtain an engine frequency prediction model, acquiring engine operating data in real time, using the engine operating data as input to the engine frequency prediction model, and outputting predicted main-order frequencies and harmonic frequencies of the engine line spectrum noise in real time as third frequency prediction data;

[0063] Comprehensively analyze the first frequency prediction data, the second frequency prediction data, and the third frequency prediction data to generate the final order frequency corresponding to the engine line spectrum noise.

[0064] It can be understood that this application uses the comprehensive output results of three different estimation schemes, including engine speed, spectrum analysis, and model prediction based on relevant prior information, to output more accurate engine order noise frequency and order number after comprehensive evaluation, thereby reducing the impact of frequency imbalance on engine noise control, thereby achieving better order noise control effect, making the final active noise reduction effect reliable and stable, and providing users with a better noise reduction experience.

[0065] Specifically, the first frequency prediction data, the second frequency prediction data, and the third frequency prediction data are comprehensively analyzed to generate the order frequency corresponding to the final engine line spectrum noise, such as Figure 4 As shown, the following steps are included:

[0066] Extract the candidate fundamental frequencies of the first frequency prediction data, the second frequency prediction data, and the third frequency prediction data respectively to obtain the corresponding first candidate fundamental frequencies. , the second candidate fundamental frequency and the third candidate fundamental frequency , and obtain the first weight corresponding to each candidate fundamental frequency , second weight , the third weight ;

[0067] The fundamental frequency of the final output order frequency is calculated according to the following formula: ,

[0068]

[0069] The number of order frequencies of the final output is calculated according to the following formula:

[0070]

[0071] in, is the number of order frequencies in the third frequency prediction data,

[0072]

[0073] The first in the second frequency prediction data frequencies, is the number of main-order frequencies in the second frequency prediction data, For the general When sorted in ascending order, the index corresponding to the maximum frequency with a confidence level of 1. If all confidence levels are not 1, then ;

[0074] The final output order frequency is .

[0075] Among them, the first candidate fundamental frequency The frequency is the same as the first frequency prediction data, and the first weight is between 0.3 and 0.4; the third candidate fundamental frequency The frequency is the same as the third frequency prediction data, and the third weight It is between 0.2 and 0.3.

[0076] The main order frequency of the engine line spectrum noise is calculated according to the real-time speed of the engine. In the process of obtaining the first frequency prediction data, the real-time speed of the engine is obtained in real time. After obtaining the real-time speed of the engine, the main order frequency of the engine line spectrum noise is calculated using the following formula:

[0077]

[0078] is the main order frequency of the engine line spectrum noise, nc (number of clinders) is the number of cylinders in the engine, rpm is the engine speed (revolutions per minute), Is the stroke coefficient. The values ​​corresponding to two-stroke and four-stroke engines are 1 and 2 respectively.

[0079] In this embodiment, the first frequency prediction data is subjected to candidate fundamental frequency extraction, and the extraction is performed according to the following formula: , the first weight .

[0080] The process of obtaining the second frequency prediction data includes the following steps:

[0081] Two microphones are installed inside the vehicle carrying the engine. The primary noise at the two microphones is estimated by internal model control and is recorded as and ,right and Do power spectrum estimation, find and obtain Power spectrum and The peak frequency point in the power spectrum that meets the minimum amplitude limit and the minimum distance limit;

[0082] After removing the null values ​​in the peak frequency points, the two power spectrum information are synchronized to separate and obtain the main order frequency and harmonic frequency of the engine line spectrum noise as the second frequency prediction data.

[0083] Specifically, if Figure 2 As shown, the process of obtaining the second frequency prediction data is as follows:

[0084] 1. Use two microphones mic1 and mic2 to extract the noise of the working engine in real time and estimate the primary noise and ,right and Perform power spectrum estimation;

[0085] 2. Find out Power spectrum and The peak frequencies in the power spectrum that meet the minimum amplitude and minimum distance limits include peak_loc_buff1, peak_loc_buff2, and their corresponding peak values ​​peak_value_buff1 and peak_value_buff2.

[0086] in:

[0087]

[0088]

[0089]

[0090]

[0091] in and They are and The number of frequencies in the power spectrum that meet the conditions.

[0092] 3. Process null values.

[0093] 3.1 If peak_loc_buff1 and peak_loc_buff2 are both empty, output the previous output.

[0094] 3.2 If peak_loc_buff1 is empty and peak_loc_buff2 is not empty, then output peak_loc_buff2 and the corresponding frequency weight is set to 0.5, i.e. confidence2: The value is assigned to 0.5, which is recorded as .

[0095] 3.3 If peak_loc_buff1 is not empty and peak_loc_buff2 is not empty, then output peak_loc_buff1 and the corresponding weight is set to 0.5, i.e. confidence1: .

[0096] 3.4 If both peak_loc_buff1 and peak_loc_buff2 are not empty, go to step 4.

[0097] 4. Perform intersection matching on peak_loc_buff1 and peak_loc_buff2 to filter out the common frequency common_peak_loc:

[0098]

[0099] in is the number of common frequencies, and the weight corresponding to the common frequencies is set to 1.

[0100]

[0101] The common frequencies are deleted from peak_loc_buff1 and peak_loc_buff2, and the corresponding peak values ​​are also deleted from peak_value_buff1 and peak_value_buff2.

[0102] 5. Handle the null values ​​of peak_loc_buff1 and peak_loc_buff2 after update.

[0103] 5.1 If at least one of peak_loc_buff1 and peak_loc_buff2 is empty, the output of the null value processing module is combined with the output of the intersection matching module and then output.

[0104] 5.2 If both peak_loc_buff1 and peak_loc_buff2 are not empty, go to step 6.

[0105] 6. Merge and sort peak_value_buff1 and peak_value_buff2 to filter out the corresponding frequencies of the peaks that meet the minimum distance limit.

[0106]

[0107] The corresponding weight is set to 0.5,

[0108]

[0109] in, is the number of frequencies filtered out.

[0110] 7. The output of step 6 and the combination of the intersection matching are output, and the output is the main order frequency and harmonic frequency of the engine line spectrum noise.

[0111] In this embodiment, when extracting candidate fundamental frequencies from the second frequency prediction data, the extraction is performed according to the following formula:

[0112]

[0113]

[0114]

[0115] in, The candidate fundamental frequency is obtained by extracting the second frequency prediction data. The first in the second frequency prediction data frequencies, is the number of main order frequencies and harmonic frequencies in the second frequency prediction data, Set to 2 or 3 or 4, is the weight of the candidate fundamental frequency obtained by extracting the second frequency prediction data, The second frequency prediction data The confidence level of a frequency.

[0116] When extracting candidate fundamental frequencies from the second frequency prediction data, multiple candidate fundamental frequencies are obtained, and a candidate fundamental frequency list is formed as follows:

[0117]

[0118] The multiple candidate fundamental frequencies each have a corresponding weight, and a corresponding weight list is formed as follows:

[0119]

[0120] Setting thresholds , 0.5≤ ≤4 if ,but ,renew and delete , and are respectively the first and The candidate frequencies of ,renew and delete , and are the first and weights;

[0121] Select the candidate fundamental frequency closest to the first candidate fundamental frequency from the updated candidate fundamental frequency list The frequency of the second candidate fundamental frequency of the second frequency prediction data and from the second candidate fundamental frequency The corresponding weight is used as the second weight of the second frequency prediction data .

[0122] In addition, the process of pre-training the neural network model to obtain the engine frequency prediction model, such as Figure 3 As shown, the following steps are included:

[0123] Collecting several sets of engine data offline as training sets, the engine data including engine operating data and the main order frequency and harmonic frequency of the corresponding engine line spectrum noise, the engine operating speed including vehicle speed, engine speed, coolant temperature, load torque, and throttle opening;

[0124] The neural network model is pre-trained using the training set, and the trained neural network model is the engine frequency prediction model.

[0125] Subsequently, by acquiring engine operating data (including vehicle speed, engine speed, coolant temperature, load torque, throttle opening, etc.) in real time, the engine operating data is used as input to the engine frequency prediction model, and the predicted main-order frequency and harmonic frequency of the engine line spectrum noise are output in real time as the third frequency prediction data.

[0126] Example 2:

[0127] This embodiment also describes a vehicle active noise reduction control method, wherein the order frequency of the engine line spectrum noise is obtained, and active control is performed based on the order frequency of the engine line spectrum noise. The order frequency of the engine line spectrum noise is calculated using the engine line spectrum noise estimation method described in Example 1.

[0128] like Figure 5 As shown, the vehicle active noise reduction control method of this embodiment uses a multi-channel adaptive notch FxLMS algorithm to perform active noise reduction processing based on the harmonics of the engine line spectrum noise generated in Example 1, thereby effectively improving the noise reduction effect.

[0129] The above embodiments are only for illustrating the technical concept and features of the present invention. Their purpose is to enable people familiar with this technology to understand the contents of the present invention and implement them accordingly. They are not intended to limit the scope of protection of the present invention. Any equivalent changes or modifications made according to the spirit of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for estimating engine line spectrum noise, characterized in that: include: Calculating the main order frequency of the engine line spectrum noise according to the real-time speed of the engine as the first frequency prediction data; Primary noise is calculated and acquired through an internal model control framework, and spectrum estimation is performed on the primary noise through a frequency estimation algorithm to separate and acquire the main order frequency and harmonic frequency of the engine line spectrum noise as second frequency prediction data; Pre-training a neural network model to obtain an engine frequency prediction model, acquiring engine operating data in real time, using the engine operating data as input to the engine frequency prediction model, and outputting predicted main-order frequencies and harmonic frequencies of the engine line spectrum noise in real time as third frequency prediction data; Extract the candidate fundamental frequencies of the first frequency prediction data, the second frequency prediction data, and the third frequency prediction data respectively to obtain the corresponding first candidate fundamental frequencies. , the second candidate fundamental frequency and the third candidate fundamental frequency , and obtain the first weight corresponding to each candidate fundamental frequency , second weight , the third weight ; The fundamental frequency of the final output order frequency is calculated according to the following formula: , The number of order frequencies of the final output is calculated according to the following formula: in, is the number of order frequencies in the third frequency prediction data, The first in the second frequency prediction data frequencies, is the number of order frequencies in the second frequency prediction data, For the general When sorted in ascending order, the index corresponding to the maximum frequency with a confidence level of 1. If all confidence levels are not 1, then ; The order frequency corresponding to the final engine line spectrum noise is .

2. The engine line spectrum noise estimation method according to claim 1, characterized in that: The first candidate fundamental frequency The frequency is the same as the first frequency prediction data, and the first weight It is between 0.3 and 0.

4.

3. The engine line spectrum noise estimation method according to claim 1, characterized in that: The third candidate fundamental frequency The frequency is the same as the third frequency prediction data, and the third weight It is between 0.2 and 0.

3.

4. The engine line spectrum noise estimation method according to claim 1, characterized in that: The candidate fundamental frequency is extracted from the second frequency prediction data according to the following formula: in, The candidate fundamental frequency is obtained by extracting the second frequency prediction data. The first in the second frequency prediction data frequencies, is the number of main order frequencies and harmonic frequencies in the second frequency prediction data, Set to 2 or 3 or 4, is the weight of the candidate fundamental frequency obtained by extracting the second frequency prediction data, The second frequency prediction data The confidence level of a frequency.

5. The engine line spectrum noise estimation method according to claim 4, characterized in that: When extracting candidate fundamental frequencies from the second frequency prediction data, multiple candidate fundamental frequencies are obtained, and a candidate fundamental frequency list is formed as follows: The multiple candidate fundamental frequencies each have a corresponding weight, and a corresponding weight list is formed as follows: Setting thresholds , 0.5≤ ≤4 if ,but ,renew and delete , and are respectively the first and The candidate frequencies of ,renew and delete , and are the first and weights; Select the candidate fundamental frequency closest to the first candidate fundamental frequency from the updated candidate fundamental frequency list The frequency of the second candidate fundamental frequency of the second frequency prediction data and take the second candidate fundamental frequency The corresponding weight is used as the second weight of the second frequency prediction data .

6. The engine line spectrum noise estimation method according to claim 1, characterized in that: During the acquisition of the first frequency prediction data, The real-time speed of the engine is obtained in real time. After obtaining the real-time speed of the engine, the main order frequency of the engine line spectrum noise is calculated using the following formula: is the main order frequency of the engine line spectrum noise, nc is the number of cylinders of the engine, rpm is the engine speed, is the stroke coefficient.

7. The engine line spectrum noise estimation method according to claim 1, characterized in that: The process of obtaining the second frequency prediction data includes the following steps: Two microphones are installed inside the vehicle carrying the engine. The primary noise at the two microphones is estimated by internal model control and is recorded as and ,right and Do power spectrum estimation, find and obtain Power spectrum and The peak frequency point in the power spectrum that meets the minimum amplitude limit and the minimum distance limit; After removing the null values ​​in the peak frequency points, the two power spectrum information are synchronized to separate and obtain the main order frequency and harmonic frequency of the engine line spectrum noise as the second frequency prediction data.

8. The engine line spectrum noise estimation method according to claim 1, characterized in that: The process of pre-training a neural network model to obtain an engine frequency prediction model includes the following steps: Collecting several sets of engine data offline as training sets, wherein the engine data includes engine operating data and the main order frequency and harmonic frequency of the corresponding engine line spectrum noise; The neural network model is pre-trained using the training set, and the trained neural network model is the engine frequency prediction model.

9. The engine line spectrum noise estimation method according to claim 1 or 8, characterized in that: The engine operating speed includes vehicle speed, engine speed, coolant temperature, load torque, and throttle opening.

10. A vehicle active noise reduction control method, characterized in that: The order frequency of the engine line spectrum noise is obtained, and active control is performed based on the order frequency of the engine line spectrum noise, wherein the order frequency of the engine line spectrum noise is calculated and obtained using the engine line spectrum noise estimation method according to any one of claims 1 to 9.

Images