Vehicle interior noise active control method and device combined with virtual sensing and vehicle

Abstract

The application discloses a kind of active control method, device and car of automobile interior noise combined with virtual sensing.The method can include: for the interior noise of vehicle at different speeds, by reference signal, physical monitoring signal and virtual error signal respectively construct multiple groups of auxiliary filter containing optimal control filter information and observation filter containing the transfer function between physical monitoring point and virtual error point;By mean method, construct composite observation filter;At different speeds, based on least mean square estimation error matching mechanism, by reference signal, physical monitoring signal, auxiliary filter, composite observation filter, active noise control is carried out at noise reduction target position.The present application can significantly reduce the noise at ear at different speeds while the arrangement position of error microphone does not affect the normal activities of passengers, improve the comfort of riding.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of active noise control, and more particularly, to an in-vehicle noise active control method and device combining virtual sensing and a vehicle. BACKGROUND

[0002] Active noise control (ANC) is widely used to solve low-frequency noise pollution caused by industrial equipment. This technology uses loudspeakers, known as secondary sound sources, to emit counteracting sound waves to cancel out noise at the current location, achieving the purpose of reducing low-frequency noise. New energy vehicles produce drum noise, cavity noise and booming noise during driving, and these noises are usually low-frequency noises below 250 Hz. Therefore, ANC technology can be used to effectively suppress in-vehicle noise of new energy vehicles. At present, the feedforward ANC scheme is concerned due to its stability and high noise reduction performance.

[0003] Adaptive filtering algorithms play an important role in ANC; and the Filtered-x least mean square (FxLMS) algorithm is the most widely used adaptive filtering method, which ensures a local quiet zone around the error microphone. However, it is sometimes inconvenient to place the error microphone on the target for a long time. Therefore, the virtual sensing method has been proposed to solve this problem. Generally, the virtual sensing method is divided into two stages, namely the training and control stages. In the training stage, a temporary microphone is placed at the target as a virtual error microphone to obtain the transfer function containing information related to the ANC system. In the control stage, the temporary microphone is removed, and the pre-modeled transfer function is used to generate a controller to attenuate the noise in the desired area. The virtual sensing algorithm can significantly reduce the restrictions on the installation location of the physical monitoring microphone, and still form a quiet zone near the human ear in the case where it is inconvenient to place a virtual error microphone near the human ear in the vehicle. Therefore, the study of the virtual sensing method has high practical value.

[0004] The Additional Filter Method (AFM) and the Remote Microphone Method (RMM) are two widely adopted traditional virtual sensing methods. AFM is essentially a model-referenced adaptive control strategy where the virtual error signal is implicitly estimated using a pre-trained additional filter, and information about the optimal controller is already contained within the additional filter. RMM uses pre-modeled observation filters and monitored signals to explicitly compute the virtual error signal of the adaptive controller. However, if the primary noise source changes significantly during the control phase compared to the training phase, the frequency responses of the pre-trained additional and observation filters become unsuitable, requiring remodeling to achieve effective noise reduction at the target location. Using wider-bandwidth primary noise for training during the AFM and RMM phases is a compromise solution; however, as a trade-off, the effectiveness of suppressing all primary noise is limited.

[0005] To overcome this limitation, existing technologies utilize reference signals measured in feedforward ANC systems to calculate a linear combination of auxiliary filters based on a frequency band matching mechanism. However, this method requires high precision in frequency matching, making it unsuitable for in-vehicle noise control in new energy vehicles. Other existing technologies propose a virtual sensing method based on relative paths for feedforward ANC systems, which improves performance in responding to interference variations by estimating interference and anti-noise signals in the target noise reduction region. However, establishing an accurate relative path model is crucial in this method, and its improvement in noise reduction performance is relatively limited.

[0006] In summary, existing virtual sensing methods lack specific design for changes in the primary noise sources of ANC systems within new energy vehicles, and urgently need further improvement. Therefore, it is necessary to develop a method, device, and vehicle for active noise control within new energy vehicles that incorporates virtual sensing.

[0007] The information disclosed in the background section of this invention is intended only to enhance the understanding of the general background of this invention, and should not be construed as an admission or in any way implying that such information constitutes prior art known to those skilled in the art. Summary of the Invention

[0008] This invention proposes a method, device, and vehicle for active control of in-vehicle noise in new energy vehicles that combines virtual sensing. It utilizes a feedforward ANC method to reduce the noise generated at the human ear at different vehicle speeds. It can significantly reduce the noise at the ear at different vehicle speeds without affecting the normal activities of passengers, thereby improving the comfort of the ride.

[0009] In a first aspect, embodiments of this disclosure provide an active method for controlling in-vehicle noise in conjunction with virtual sensing, including:

[0010] To address the in-vehicle noise at different vehicle speeds, multiple sets of auxiliary filters containing optimal control filter information and observation filters containing the transfer function between physical monitoring points and virtual error points are constructed using reference signals, physical monitoring signals, and virtual error signals, respectively.

[0011] A synthetic observation filter was constructed using the mean method;

[0012] Based on the minimum mean square estimation error matching mechanism at different vehicle speeds, active noise control is performed at the noise reduction target location through the reference signal, the physical monitoring signal, the auxiliary filter, and the synthetic observation filter.

[0013] Preferably, for in-vehicle noise at different vehicle speeds, multiple sets of auxiliary filters containing optimal control filter information and observation filters containing the transfer function between physical monitoring points and virtual error points are constructed using reference signals, physical monitoring signals, and virtual error signals, respectively.

[0014] The least mean square algorithm is used to construct physical and virtual secondary channels from the secondary sound source to the physical monitoring microphone and the virtual error microphone, respectively.

[0015] At a certain fixed vehicle speed The reference signal matrix acquired by the accelerometer at any given time is:

[0016]

[0017] in, It represents the past. The first sample Reference vectors, Represents the transpose symbol;

[0018] To address interference noise at different vehicle speeds, an auxiliary filter incorporating optimal controller information and an observation filter containing the transfer function between the physical monitoring microphone and the virtual error microphone are trained for the corresponding vehicle speeds. Auxiliary filter at vehicle speed for:

[0019]

[0020] Corresponding to the Observation filter at vehicle speed for:

[0021]

[0022] in, Representing the The reference signal to the first The order of the physical monitoring signals is The auxiliary filter coefficient vector, Representing the The physical monitoring signal to the first The order of the virtual error signals is The observation filter coefficient vector, Represents the transpose symbol.

[0023] Preferably, training the auxiliary filter containing optimal controller information at the corresponding vehicle speed includes:

[0024] The control filter matrix is:

[0025]

[0026] in, Representing the The reference signal to the first The order of the output signals of each secondary source is The control filter coefficient vector;

[0027] Based on the traditional filter reference mean square adaptive algorithm The update formula is:

[0028]

[0029] in, Indicates the first The residual error signal at each virtual error microphone, and the filtered reference signal. Indicates the first One reference signal and virtual secondary channel convolution, Representing the The secondary source outputs a signal to the first... The order of the virtual error signals is The virtual secondary channel vector;

[0030] After the control filter converges, the auxiliary filter begins training, and the LMS algorithm is used to model and obtain the first... The reference signal to the first The order of the physical monitoring signals is The auxiliary filter coefficient vector is Its update formula is:

[0031]

[0032] in, The auxiliary filter is in the first... an error signal output at the physical monitoring microphone, wherein is an error signal collected at the physical monitoring microphone.

[0033] Preferably, training the observation filter containing the transfer function between the physical monitoring microphone and the virtual error microphone at the corresponding vehicle speed comprises:

[0034] The observation filter is trained by the LMS algorithm, and the observation filter is:

[0035]

[0036] wherein, represents a vector of order L composed of the interference signals at the th virtual error microphone, represents a vector of order L composed of the interference signals at the th physical monitoring microphone; The update formula of the observation filter is obtained by the LMS algorithm:

[0037]

[0038]

[0039] wherein, represents the relative error signal for updating the observation filter.

[0040] Preferably, constructing the synthesized observation filter by the mean method comprises:

[0041] The measured observation filter is averaged, and the corresponding synthesized observation filter coefficient vector of order L from the th physical monitoring signal to the th virtual error signal is:

[0042]

[0043] wherein, represents a preset vehicle speed category considered.

[0044] Preferably, based on the minimum mean square error matching mechanism at different vehicle speeds, active noise control is performed at the noise reduction target position by the reference signal, the physical monitoring signal, the auxiliary filter, and the synthesized observation filter, comprising:

[0045] The interference signal at the virtual point is estimated based on the physical monitoring signal, the physical secondary channel, and the synthesized observation filter.​​​​​

[0046] ;

[0047] By constructing a parallel structure for multiple sets of auxiliary filters corresponding to different vehicle speeds, the first set of auxiliary filters corresponding parallel control filter is:

[0048]

[0049] wherein, represents the order of the first reference signal to the first secondary source output signal is the parallel control filter coefficient vector of the order of

[0050] By the reference signal, the first set of control filters corresponding parallel output signal is:

[0051]

[0052] wherein, represents the first output signal vector of the order of

[0053] By the parallel output signal and the virtual secondary channel, estimate the parallel virtual error signal at the virtual point under the action of the parallel control filter.

[0054] Repeat the above steps until a pre-set frame length , filter the one with the smallest cumulative energy in the parallel virtual error signal, and use the corresponding auxiliary filter for actual control filter update and drive the secondary source to emit a signal to cancel the interference at the virtual point.

[0055] Preferably, by the parallel output signal and the virtual secondary channel, estimate the parallel virtual error signal at the virtual point under the action of the parallel control filter includes:

[0056] For the first set of parallel filters, the corresponding output signal is , the error signal output by the auxiliary filter at the first physical monitoring microphone is:

[0057] ;

[0058] By FxLMS algorithm, ​The update formula of the actual control filter is:

[0059]

[0060] wherein, represents the convolution of the first reference signal and the virtual secondary channel , represents the order of the first secondary source output signal to the first physical monitoring signal, and the physical secondary channel vector of the first ;

[0061] The corresponding parallel virtual error signal is:

[0062] ;

[0063] For the first group of parallel filters, the parallel virtual error cumulative energy obtained at the time of is .

[0064] Preferably, the one with the minimum energy in the parallel virtual error signal is screened, and the corresponding auxiliary filter is used for actual control filter updating and driving the secondary source to emit a signal to offset the interference at the virtual point, which comprises:

[0065] Determining the auxiliary filter based on a minimum mean square error matching mechanism;

[0066] Selecting the one with the minimum cumulative energy in all groups of parallel auxiliary filters, and assigning the corresponding auxiliary filter to the auxiliary filter used for updating the actual controller, that is, ;

[0067] The output signal driving the secondary source is , and the error signal output by the auxiliary filter at the first physical monitoring microphone is:

[0068]

[0069] The update formula of the actual control filter using the FxLMS algorithm is:

[0070]

[0071] Then, initialization is performed, that is, the virtual error cumulative energy is set to 0, and the initial values of the parallel auxiliary filters are all set to , and the corresponding parallel control filters are all set to .

[0072] In a second aspect, the embodiments of the present disclosure further provide an automobile in-cabin noise active control device combined with virtual sensing, comprising a reference accelerometer, an adaptive control system, a secondary sound source, a physical monitoring microphone and a virtual error microphone, wherein:

[0073] The plurality of reference accelerometers are arranged on the chassis of the vehicle to collect road noise transmitted into the vehicle cabin during driving of the vehicle;

[0074] The adaptive control system is used to send the reference signal collected by the reference accelerometer to the secondary sound source after adaptive filtering and calculation, to drive the secondary sound source to emit interference-cancelling sound;

[0075] The plurality of physical monitoring microphones are arranged at the roof of the vehicle to obtain physical monitoring signals;

[0076] The virtual error microphone is arranged at the binaural position of the vehicle occupant in the training stage to obtain a virtual error signal;

[0077] The adaptive control system comprises an analog / digital converter, a digital signal processor, a digital / analog converter and a power amplifier circuit, wherein:

[0078] The analog / digital converter is used to convert the analog signal collected by the reference accelerometer into a digital signal to be processed and input into the digital signal processor;

[0079] The digital signal processor is used to filter and calculate the digital signal, and the calculated result is transmitted to the digital / analog converter in the form of a digital signal;

[0080] The digital / analog converter is used to convert the output digital signal into an analog signal;

[0081] The power amplifier circuit is used to moderately amplify the analog signal output by the digital / analog converter and send it to the secondary sound source to drive the secondary sound source to emit interference-cancelling sound.

[0082] In a third aspect, the embodiments of the present disclosure further provide an automobile comprising an automobile in-cabin noise active control device combined with virtual sensing.

[0083] The beneficial effects are as follows:

[0084] 1. The method has robustness to the change of the primary noise source caused by the change of the speed of the new energy vehicle, and can be used in a multi-channel adaptive feedforward ANC system; the method combines the advantages of AFM and RMM, and reduces the precision requirement for modeling of the observation filter.

[0085] 2. The method of the present application uses the virtual error signal estimated by the synthetic observer filter not directly for updating the adaptive control filter, which reduces the precision required for modeling the observer filter.

[0086] 3. The method of the present application is divided into two stages; in the training stage, the underlying dominant noise is used to train a bank of auxiliary filters and a synthetic observer filter; in the control stage, the best parallel auxiliary filter is selected according to the minimum mean square error matching mechanism to minimize the energy of the estimated virtual error signal in a frame. The matching and selection of the auxiliary filters are performed frame by frame, so that the most suitable auxiliary filter is used for each frame of noise.

[0087] The method and apparatus of the present application have other features and advantages which will be apparent from, or that will be elucidated with regard to, the drawings accompanying and the written description of specific embodiments herein, which are jointly used to explain the specific principles of the application. BRIEF DESCRIPTION OF DRAWINGS

[0088] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings, in which like reference characters designate like elements in the several views.

[0089] Figure 1 A flow chart showing the steps of the active control method for in-vehicle noise of an automobile incorporating virtual sensing according to the present application is shown.

[0090] Figure 2 A schematic diagram showing the training stage of the control method according to one embodiment of the present application is shown.

[0091] Figure 3 A schematic diagram showing the control stage of the control method according to one embodiment of the present application is shown.

[0092] Figure 4 A flow chart showing the steps of the active control method for in-vehicle noise of an automobile incorporating virtual sensing according to one embodiment of the present application is shown. DETAILED DESCRIPTION

[0093] Preferred embodiments of the present application will be described in more detail below. Although the preferred embodiments of the present application are described below, it is understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein.

[0094] To facilitate understanding of the solutions and effects of the embodiments of the present invention, three specific application examples are given below. Those skilled in the art should understand that these examples are merely for the purpose of understanding the present invention, and any specific details therein are not intended to limit the present invention in any way.

[0095] Example 1

[0096] Figure 1 A flowchart illustrating the steps of an active vehicle in-vehicle noise control method incorporating virtual sensing according to the present invention is shown.

[0097] like Figure 1 As shown, the active noise control method for vehicle interiors combined with virtual sensing includes: Step 101, for vehicle interior noise at different vehicle speeds, constructing multiple sets of auxiliary filters containing optimal control filter information and observation filters containing the transfer function between physical monitoring points and virtual error points using reference signals, physical monitoring signals, and virtual error signals respectively; Step 102, constructing a synthetic observation filter using the mean method; Step 103, at different vehicle speeds, based on the minimum mean square estimation error matching mechanism, performing active noise control at the noise reduction target location using reference signals, physical monitoring signals, auxiliary filters, and synthetic observation filters.

[0098] In one example, for the in-vehicle noise at different vehicle speeds, multiple sets of auxiliary filters containing optimal control filter information and observation filters containing the transfer function between physical monitoring points and virtual error points are constructed using reference signals, physical monitoring signals, and virtual error signals, respectively.

[0099] The least mean square algorithm is used to construct physical and virtual secondary channels from the secondary sound source to the physical monitoring microphone and the virtual error microphone, respectively.

[0100] At a certain fixed vehicle speed The reference signal matrix acquired by the accelerometer at any given time is:

[0101]

[0102] in, It represents the past. The first sample Reference vectors, Represents the transpose symbol;

[0103] To address interference noise at different vehicle speeds, an auxiliary filter incorporating optimal controller information and an observation filter containing the transfer function between the physical monitoring microphone and the virtual error microphone are trained for the corresponding vehicle speeds. Auxiliary filter at vehicle speed for:

[0104]

[0105] Corresponding to the Observation filter at vehicle speed for:

[0106]

[0107] in, Representing the The reference signal to the first The order of the physical monitoring signals is The auxiliary filter coefficient vector, Representing the The physical monitoring signal to the first The order of the virtual error signals is The observation filter coefficient vector, Represents the transpose symbol.

[0108] In one example, training an auxiliary filter containing optimal controller information at the corresponding vehicle speed includes:

[0109] The control filter matrix is:

[0110]

[0111] in, Representing the The reference signal to the first The order of the output signals of each secondary source is The control filter coefficient vector;

[0112] Based on the traditional filter reference mean square adaptive algorithm The update formula is:

[0113]

[0114] in, Indicates the first The residual error signal at each virtual error microphone, and the filtered reference signal. Indicates the first One reference signal and virtual secondary channel convolution, Representing the The secondary source outputs a signal to the first... The order of the virtual error signals is The virtual secondary channel vector;

[0115] After the control filter converges, the auxiliary filter begins training, and the LMS algorithm is used to model and obtain the first... The reference signal to the first The order of the physical monitoring signals is The auxiliary filter coefficient vector is Its update formula is:

[0116]

[0117] in, The auxiliary filter is in the first... The error signal output from a physical monitoring microphone. ,in It is the first Error signals collected from a physical monitoring microphone.

[0118] In one example, training an observation filter containing the transfer function between the physical monitoring microphone and the virtual error microphone at the corresponding vehicle speed includes:

[0119] The observation filter is trained using the LMS algorithm, and the observation filter is as follows:

[0120]

[0121] in, Representing the The order of the interference signals at each virtual error microphone is: The vector, Representing the A vector of order L composed of interference signals from each physical monitoring microphone;

[0122] Obtained through the LMS algorithm The update formula is:

[0123]

[0124] in, This represents the relative error signal for updating the observation filter.

[0125] In one example, constructing a synthetic observation filter using the mean method includes:

[0126] The measured observation filters are averaged, and the corresponding first... The physical monitoring signal to the first The order of the virtual error signals is Synthetic observation filter coefficient vector for:

[0127]

[0128] in, This represents the preset speed range.

[0129] In one example, the active noise control at the target position of noise reduction based on the minimum mean square error matching mechanism at different vehicle speeds, by reference signals, physical monitoring signals, auxiliary filters, synthetic observation filters, includes:

[0130] The interference signal at the virtual point is estimated based on the physical monitoring signal, the physical secondary channel and the synthetic observation filter as:

[0131]

[0132] By constructing a parallel structure for multiple groups of auxiliary filters for different vehicle speeds, the first group of auxiliary filters corresponding to the parallel control filter is:

[0133]

[0134] wherein, represents the order of the th reference signal to the th secondary source output signal is the parallel control filter coefficient vector of the

[0135] By the reference signal, the th control filter corresponding to the parallel output signal is:

[0136]

[0137] wherein, represents the th output signal vector with an order of

[0138] By the parallel output signal and the virtual secondary channel, the parallel virtual error signal at the virtual point under the action of the parallel control filter is estimated.

[0139] Repeat the above steps until a pre-set frame length is reached, select the one with the minimum cumulative energy in the parallel virtual error signal, and use the corresponding auxiliary filter for actual control filter update and drive the secondary source to emit a signal to cancel the interference at the virtual point.

[0140] In one example, by the parallel output signal and the virtual secondary channel, the parallel virtual error signal at the virtual point under the action of the parallel control filter includes:

[0141] For the th parallel filter, its corresponding output signal is​​ , the error signal output by the auxiliary filter at the mth physical monitoring microphone is:

[0142]

[0143] Through the FxLMS algorithm, the update formula is:

[0144]

[0145] wherein, represents the convolution of the mth reference signal and the virtual secondary channel , represents the mth secondary source output signal to the mth physical monitoring signal, and the order of the physical secondary channel vector The corresponding parallel virtual error signal is:

[0146] For the mth group of parallel filters, the parallel virtual error cumulative energy obtained at the time t is

[0147]

[0148] .

[0149] In one example, the one with the minimum energy in the parallel virtual error signal is screened, and the corresponding auxiliary filter is used for actual controller update and driving the secondary source to emit a signal that cancels the interference at the virtual point, comprising:

[0150] Determining the auxiliary filter based on the minimum mean square error matching mechanism;

[0151] Selecting the group corresponding to the minimum cumulative energy among all groups of parallel auxiliary filters, and assigning the corresponding auxiliary filter to the auxiliary filter used for updating the actual controller, that is, ;

[0152] The output signal driving the secondary source is , and the error signal output by the auxiliary filter at the mth physical monitoring microphone is:

[0153]

[0154] The actual control filter is obtained by using the FxLMS algorithm ​​​​​​​​​​​The update formula is:

[0155]

[0156] Then, an initialization operation is performed, which involves accumulating virtual error energy. The initial values ​​of the parallel auxiliary filters are all set to 0. The corresponding parallel control filters are all set to .

[0157] Figure 2 A schematic diagram of a control method according to an embodiment of the present invention during the training phase is shown.

[0158] Figure 3 A schematic diagram of a control method according to an embodiment of the present invention is shown in the control phase.

[0159] Figure 4 A flowchart illustrating the steps of an active vehicle in-vehicle noise control method incorporating virtual sensing according to an embodiment of the present invention is shown.

[0160] Specifically, such as Figures 2-4 As shown, during the training phase, multiple sets of auxiliary filters containing optimal control filter information and observation filters containing the transfer function between physical monitoring points and virtual error points are trained using reference signals, physical monitoring signals, and virtual error signals to address the in-vehicle noise of new energy vehicles at different speeds. During the control phase, the virtual error microphone is removed, and the new energy vehicle achieves active noise control at both ear positions based on the minimum mean square estimation error matching mechanism using reference signals, physical monitoring signals, auxiliary filters, and synthetic observation filters when driving at different speeds.

[0161] Specifically, the training phase is implemented as follows:

[0162] 1. Secondary channel identification

[0163] With the new energy vehicle stationary, secondary sound sources at the four doors sequentially emit band-limited broadband white noise ranging from 20 to 1000 Hz. Noise signals are collected by four physical monitoring microphones in the vehicle's ceiling and eight virtual error microphones at the driver's ear. Using this data, the physical secondary paths are modeled using the LMS algorithm. and virtual secondary pathways .

[0164] 2. Parallel Auxiliary Filter Design

[0165] Based on the characteristics of road noise generated by new energy vehicles, this example mainly considers the noise suppression effect when new energy vehicles switch between speeds of 40 km / h (kph), 50 kph, 60 kph, 70 kph and 80 kph.

[0166] All secondary sound sources were silenced. The new energy vehicle was then driven at 40 kph, 50 kph, 60 kph, 70 kph, and 80 kph sequentially to collect signals and train the auxiliary filter. Figure 2 As shown in the middle left figure. At a certain fixed vehicle speed, The reference signal matrix acquired by the accelerometer at any given time can be represented as:

[0167]

[0168] in, It represents the past. The first sample Reference vectors, Represents the transpose sign; the control filter matrix can be represented as

[0169]

[0170] in Representing the The reference signal to the first The order of the output signals of each secondary source is The control filter coefficient vector; based on the traditional FxLMS adaptive algorithm. The update formula can be written as

[0171]

[0172] in Indicates the first The residual error signal at each virtual error microphone, and the filtered reference signal. Indicates the first One reference signal and virtual secondary channel convolution, Representing the The secondary source outputs a signal to the first... The order of the virtual error signals is The virtual secondary channel vector; after the control filter fully converges, the auxiliary filter corresponding to the vehicle speed begins training; the LMS algorithm is used to model and obtain the first... The reference signal to the first The order of the physical monitoring signals is The auxiliary filter coefficient vector is its update formula is

[0173]

[0174] wherein is the error signal output by the auxiliary filter at the th physical monitoring microphone, which can be expressed as

[0175]

[0176] wherein is the error signal collected at the th physical monitoring microphone.

[0177] When the auxiliary filter is fully converged, the corresponding auxiliary filter at this vehicle speed is designed. According to the above method, the new energy vehicle is sequentially placed at five vehicle speeds to construct auxiliary filters, and the parallel auxiliary filter group finally obtained is .

[0178] 3. Synthesis observation filter design

[0179] Set all secondary sound sources to be silent, and let the new energy vehicle drive at 40 kph, 50 kph, 60 kph, 70 kph and 80 kph to collect signals and train auxiliary filters, as shown in the right graph in FIG. 2. At a certain fixed vehicle speed, interference signals are collected through the microphones at the physical monitoring points and the virtual error points, and the current vehicle speed corresponding observation filter is trained by using LMS algorithm combined with these signals. The relationship of the observation filter can be expressed as Figure 2

[0180]

[0181] wherein represents the interference signal at the th virtual error microphone, which forms a vector of order represents the interference signal at the th physical monitoring microphone, which forms a vector of order L, and the update formula of

[0182]

[0183] wherein represents the relative error signal for updating the observation filter. When the auxiliary filter is fully converged, the corresponding observation filter at this vehicle speed is designed. According to the above method, the new energy vehicle is sequentially placed at five vehicle speeds to construct observation filters, and the observation filter group finally obtained is ​​​The measured observation filter is averaged to reduce the total deviation of the estimated virtual error signal of the new energy vehicle at different vehicle speeds, and the order of the corresponding first physical monitoring signal to the first virtual error signal is The synthetic observation filter coefficient vector can be represented as

[0184]

[0185] The synthetic observation filter designed is

[0186] The implementation process of the control stage is as follows:

[0187] 1. First, remove all virtual error microphones connected to the ANC controller. Based on the physical monitoring signal, the physical secondary channel and the synthetic observation filter, the disturbance signal at the virtual point can be estimated. The new energy vehicle normally drives at a speed of 40-80kph. A plurality of auxiliary filters for different vehicle speeds are used to construct a parallel structure, and the first auxiliary filter corresponds to the corresponding parallel control filter , which can be represented as

[0188]

[0189] Among them represents the order of the first reference signal to the first secondary source output signal The parallel control filter coefficient vector is The reference signal obtained by using the accelerometer can obtain the first control filter The corresponding parallel output signal is , which can be represented as

[0190]

[0191] Among them represents the first output signal vector with an order of Then, using the parallel output signal and the virtual secondary channel, the parallel virtual error signal at the virtual point under the action of the parallel control filter can be estimated. The disturbance signal at the first virtual error microphone can be estimated as

[0192]

[0193] For the first parallel filter, the corresponding output signal can be represented as , the error signal outputted by the auxiliary filter at the mth physical monitoring microphone may be expressed as

[0194]

[0195] The update formula of the auxiliary filter can be written as

[0196]

[0197] wherein the convolution of the mth reference signal and the virtual secondary channel , the order of the mth secondary source output signal to the mth physical monitoring signal is ; thus the corresponding parallel virtual error signal is

[0198]

[0199] From the above, for the mth group of parallel filters, the parallel virtual error cumulative energy obtained at the time of t is . 2. When the sampling point cumulative reaches a pre-set frame length , the one with the minimum energy in the parallel virtual error signal is screened, and the corresponding auxiliary filter is used for actual controller update and driving the secondary source to emit a signal to offset the interference at the virtual point; then initialization is performed, and the method proposed in the application is used to continue the calculation of the next frame. The one with the minimum cumulative energy in all groups of parallel auxiliary filters is selected, and the corresponding parallel auxiliary filter

[0200] is assigned to the auxiliary filter used for updating the actual controller, that is ; the output signal driving the secondary source is , and the error signal outputted by the auxiliary filter at the mth physical monitoring microphone may be expressed as

[0201]

[0202] The update formula of the actual control filter can be written as

[0203] ​​​​​​​​​​

[0204] After the initialization operation, that is, the virtual error cumulative energy is set to 0, and the initial values of the parallel auxiliary filters are set to , and the corresponding parallel control filters are set to .

[0205] Example 2

[0206] The active control device for vehicle interior noise combined with virtual sensors comprises a reference accelerometer, an adaptive control system, a secondary sound source, a physical monitoring microphone and a virtual error microphone, wherein:

[0207] A plurality of reference accelerometers are arranged on the chassis of the vehicle to collect road noise transmitted into the vehicle cabin during vehicle driving;

[0208] The adaptive control system is used to send the reference signal collected by the reference accelerometer to the secondary sound source after adaptive filtering and calculation, to drive the secondary sound source to emit interference-cancelling sound;

[0209] A plurality of physical monitoring microphones are arranged at the roof of the vehicle to obtain physical monitoring signals;

[0210] The virtual error microphone is arranged at the binaural position of the vehicle occupant in the training stage to obtain a virtual error signal;

[0211] The adaptive control system comprises an analog / digital converter, a digital signal processor, a digital / analog converter and a power amplifier circuit, wherein:

[0212] The analog / digital converter is used to convert the analog signal collected by the reference accelerometer into a digital signal to be processed and input into the digital signal processor;

[0213] The digital signal processor is used to filter and calculate the digital signal, and the calculated result is transmitted to the digital / analog converter in the form of a digital signal;

[0214] The digital / analog converter is used to convert the output digital signal into an analog signal;

[0215] The power amplifier circuit is used to moderately amplify the analog signal output by the digital / analog converter and send it to the secondary sound source to drive the secondary sound source to emit interference-cancelling sound.

[0216] Specifically, a plurality of reference accelerometers are arranged on the chassis of the new energy vehicle to collect road noise entering the vehicle cabin during driving. A plurality of built-in speakers in the doors of the vehicle are used as secondary sound sources to generate secondary sound signals to offset the noise in the vehicle. A plurality of physical monitoring microphones are arranged on the roof of the vehicle to obtain physical monitoring signals. In the training stage, a plurality of virtual error microphones are respectively placed at the binaural positions of the passengers in the vehicle to obtain virtual error signals, and in the real-time control stage, all the virtual error microphones are removed.

[0217] An adaptive control system is used to adaptively filter and calculate the reference signals collected by the accelerometers and send them to the secondary sound sources to drive the secondary sound sources to emit interference-cancelling sound, thereby achieving active noise control. The adaptive control system includes an analog / digital converter, a digital signal processor, a digital / analog converter, and a power amplifier circuit. The analog / digital converter is used to convert the analog signals collected by the reference accelerometers into digital signals to be processed and input into the digital signal processor. The digital signal processor is used to filter and calculate the digital signals and deliver the calculated results in the form of digital signals to the digital / analog converter. The digital / analog converter is used to convert the output digital signals into analog signals. The power amplifier circuit is used to moderately amplify the analog signals output by the digital / analog converter and send them to the secondary sound sources to drive the secondary sound sources to emit interference-cancelling sound.

[0218] In an embodiment, the number of reference accelerometers is 8, which are respectively located at the auxiliary frame and the passive end of the shock absorber spring near the four wheel hubs of the vehicle. The number of secondary sound sources is 4, which are respectively located at the bottom of the four doors. The number of physical monitoring microphones is 4, which are respectively located at the roof positions opposite the four seats. The number of temporarily placed virtual error microphones in the training stage is 8, which are respectively located at the binaural positions of the passengers corresponding to the four seats.

[0219] Example 3

[0220] The present disclosure provides a car comprising the above-mentioned active control device for in-vehicle noise of a car combined with virtual sensors.

[0221] Those skilled in the art will understand that the purpose of the above description of the embodiments of the present application is only to exemplarily illustrate the beneficial effects of the embodiments of the present application, and is not intended to limit the embodiments of the present application to any examples given.

[0222] The above has described the embodiments of the present application, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A method for active control of in-vehicle noise combining virtual sensing, characterized in that, include: To address in-vehicle noise at different vehicle speeds, multiple sets of auxiliary filters containing optimal control filter information and observation filters containing the transfer function between physical monitoring points and virtual error points are constructed using reference signals, physical monitoring signals, and virtual error signals. This includes constructing physical secondary channels and virtual secondary channels from secondary sound sources to physical monitoring microphones and virtual error microphones using the least mean square algorithm. A synthetic observation filter is constructed using the mean method; Based on the minimum mean square estimation error matching mechanism at different vehicle speeds, active noise control is performed at the noise reduction target location through the reference signal, the physical monitoring signal, the auxiliary filter, and the synthetic observation filter. The synthetic observation filter constructed using the mean method includes: The measured observation filters are averaged, and the corresponding first... The physical monitoring signal to the first The order of the virtual error signals is Synthetic observation filter coefficient vector for: in, This represents the preset types of vehicle speeds considered. Representing the The physical monitoring signal to the first The order of the virtual error signals is The observation filter coefficient vector; Specifically, the active noise control at the noise reduction target location, based on the minimum mean square estimation error matching mechanism at different vehicle speeds and utilizing the reference signal, the physical monitoring signal, the auxiliary filter, and the synthetic observation filter, includes: Based on the physical monitoring signal, the physical secondary channel, and the synthetic observation filter, the interference signal at the virtual point is estimated as follows: ； By constructing a parallel structure with multiple sets of auxiliary filters for different vehicle speeds, the first... Group Auxiliary Filter Corresponding parallel control filter for: in, Representing the The reference signal to the first The order of the output signals of each secondary source is The parallel control filter coefficient vector; Through the reference signal, or the first Group control filter Corresponding parallel output signal for: in, Representing the The order is The output signal vector; The parallel virtual error signal at the virtual point is estimated using the parallel output signal and the virtual secondary channel under the action of the parallel control filter. Repeat the above steps until the preset frame length is reached. The parallel virtual error signal with the smallest cumulative energy is selected, and its corresponding auxiliary filter is used to update the actual control filter, and the secondary source is driven to emit a signal to cancel the interference at the virtual point.

2. The active vehicle in-vehicle noise control method combining virtual sensing according to claim 1, wherein, To address in-vehicle noise at different speeds, multiple sets of auxiliary filters containing optimal control filter information and observation filters containing the transfer function between physical monitoring points and virtual error points are constructed using reference signals, physical monitoring signals, and virtual error signals. At a certain fixed vehicle speed The reference signal matrix acquired by the accelerometer at any given time is: in, It represents the past. The first sample Reference vectors, Represents the transpose symbol; To address interference noise at different vehicle speeds, an auxiliary filter incorporating optimal controller information and an observation filter containing the transfer function between the physical monitoring microphone and the virtual error microphone are trained for the corresponding vehicle speeds. Auxiliary filter at vehicle speed for: Corresponding to the Observation filter at vehicle speed for: in, Representing the The reference signal to the first The order of the physical monitoring signals is The auxiliary filter coefficient vector, Represents the transpose symbol.

3. The active vehicle in-vehicle noise control method combining virtual sensing according to claim 2, wherein, Training the auxiliary filter containing optimal controller information at the corresponding vehicle speed includes: The control filter matrix is: in, Representing the The reference signal to the first The order of the output signals of each secondary source is The control filter coefficient vector; Based on the traditional filter reference mean square adaptive algorithm The update formula is: in, Indicates the first The residual error signal at each virtual error microphone, and the filtered reference signal. Indicates the first One reference signal and virtual secondary channel convolution, Representing the The secondary source outputs a signal to the first... The order of the virtual error signals is The virtual secondary channel vector; After the control filter converges, the auxiliary filter begins training, and the LMS algorithm is used to model and obtain the first... The reference signal to the first The order of the physical monitoring signals is The auxiliary filter coefficient vector is Its update formula is: in, The auxiliary filter is in the first... The error signal output from a physical monitoring microphone. ,in It is the first Error signals collected from a physical monitoring microphone.

4. The active vehicle in-vehicle noise control method combining virtual sensing according to claim 3, wherein, The observation filter, which incorporates the transfer function between the physical monitoring microphone and the virtual error microphone, is trained at the corresponding vehicle speed. The observation filter is trained using the LMS algorithm, and the observation filter is as follows: in, Representing the The order of the interference signals at each virtual error microphone is: The vector, Representing the A vector of order L composed of interference signals from each physical monitoring microphone; Obtained through the LMS algorithm The update formula is: in, This represents the relative error signal for updating the observation filter.

5. The active vehicle in-vehicle noise control method combining virtual sensing according to claim 1, wherein, The parallel virtual error signal at the virtual point, estimated using the parallel output signal and the virtual secondary channel, under the action of the parallel control filter, includes: For the A group of parallel filters, whose corresponding output signal is The auxiliary filter in the first Error signal output from each physical monitoring microphone for: ； Using the FxLMS algorithm The update formula is: in, Indicates the first One reference signal and virtual secondary channel convolution, Representing the The secondary source outputs a signal to the first... The order of the physical monitoring signals is The physical secondary channel vector; The corresponding parallel virtual error signal is: ； For the A group of parallel filters, which in The cumulative energy of the parallel virtual error obtained at time step is , Representing the The order of the interference signals at each virtual error microphone is: The vector, Representing the The secondary source outputs a signal to the first... The order of the virtual error signals is The virtual secondary channel vector.

6. The active vehicle in-vehicle noise control method combining virtual sensing according to claim 1, wherein, The process involves selecting the virtual error signal with the lowest energy from the parallel signals, using its corresponding auxiliary filter to update the actual control filter, and driving the secondary source to emit a signal that cancels out the interference at the virtual point. The auxiliary filter is determined based on the minimum mean square estimation error matching mechanism; Select the group with the smallest cumulative energy among all groups of parallel auxiliary filters, and then set the corresponding auxiliary filter... The value is assigned to the auxiliary filter used to update the actual controller, i.e. ; The output signal of the driving secondary source is obtained as The auxiliary filter in the first Error signal output from each physical monitoring microphone for: It is the first Error signals collected from a physical monitoring microphone; Using the FxLMS algorithm to obtain the actual control filter The update formula is: Then, an initialization operation is performed, which involves accumulating virtual error energy. The initial values ​​of the parallel auxiliary filters are all set to 0. The corresponding parallel control filters are all set to , Indicates the first One reference signal and virtual secondary channel The convolution.

7. A vehicle in-vehicle noise active control device incorporating virtual sensing, utilizing the vehicle in-vehicle noise active control method incorporating virtual sensing as described in any one of claims 1-6, characterized in that, It includes a reference accelerometer, an adaptive control system, a secondary sound source, a physical monitoring microphone, and a virtual error microphone, among which: Multiple reference accelerometers are installed in the vehicle's chassis to collect road noise transmitted into the passenger compartment while the vehicle is in motion; An adaptive control system is used to adaptively filter and calculate the reference signal collected by the reference accelerometer and then send it to the secondary sound source to drive the secondary sound source to emit sound that cancels out interference. Multiple physical monitoring microphones are installed on the roof of the vehicle to acquire physical monitoring signals; During the training phase, virtual error microphones are placed at the ears of the occupants inside the vehicle to acquire virtual error signals. The adaptive control system includes an analog-to-digital converter, a digital signal processor, a digital-to-analog converter, and a power amplifier circuit, wherein: An analog-to-digital converter is used to convert the analog signal acquired by the reference accelerometer into a digital signal to be processed and input into a digital signal processor; A digital signal processor is used to perform filtering and calculation operations on digital signals and transmit the calculated results to a digital-to-analog converter in the form of digital signals. A digital-to-analog converter is used to convert an output digital signal into an analog signal. The power amplifier circuit is used to amplify the analog signal output from the digital-to-analog converter to a suitable extent and send it to the secondary sound source, driving the secondary sound source to emit sound that cancels out interference.

8. A car, characterized in that, Including the active vehicle in-vehicle noise control device combining virtual sensing as described in claim 7.

Images