Active noise reduction method for automobile engine

Abstract

The present invention discloses an active noise reduction method for an automobile engine, comprising: S1, offline parameter acquisition, including: (1) acquiring a secondary channel transfer function from an onboard speaker to a human ear position; (2) acquiring adaptive filter weight coefficients corresponding to different vehicle operating conditions; S2, online noise reduction: based on the current vehicle operating condition, retrieving the adaptive filter weight coefficients under the same or similar operating condition as the current vehicle operating condition in step S1, generating a speaker drive signal to emit a cancellation signal. The active noise reduction method for an automobile engine does not require convergence time during the actual noise reduction process, can improve the stability of the algorithm, can achieve optimal noise reduction effect at the human ear position, and improve the noise reduction effect. Moreover, no error microphone is required during the online noise reduction process, which can reduce the cost of the active noise reduction system.

Description

Technical Field

[0001] The present invention relates to the technical field of active noise control, and in particular to a method for active noise reduction of an automobile engine. Background Art

[0002] The engine is one of the main noise sources of fuel vehicles. To reduce automobile engine noise, OEMs use a variety of passive noise reduction methods, such as adding sound-absorbing cotton, vibration absorbers, damping plates in the transmission path, and using connecting bushings with better vibration isolation capabilities. However, these passive noise reduction methods will increase the overall weight of the vehicle body and even affect the vehicle's handling performance. At the same time, the noise reduction effect is limited. Therefore, engine active noise reduction methods are applied to the field of automobile noise and vibration control.

[0003] Traditional engine active noise reduction systems are based on the FxLMS algorithm and include hardware such as speakers, error microphones, and a controller. The controller obtains real-time engine speed information from the vehicle's CAN bus, generates a reference signal for signal processing, and updates the adaptive filter weights to minimize the mean square error of the engine's primary order noise. This generates a secondary signal that drives the speaker to produce sound, thereby canceling the engine noise inside the vehicle. This method can effectively reduce engine noise at target points within the vehicle, but is limited by the algorithm's inherent limitations. It requires time to converge and obtain the optimal adaptive filter coefficients. When the engine speed changes rapidly, the algorithm struggles to track changes in real time and iterate until complete convergence, impacting control effectiveness. Furthermore, the error microphone's mounting position is affected by the vehicle's structure, resulting in the actual control position not being at the human ear, thus affecting the subjective perception of noise reduction. Summary of the Invention

[0004] The purpose of the present invention is to provide an active noise reduction method for an automobile engine with stable performance, good noise reduction effect and low cost in response to the problems in the prior art.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] A method for active noise reduction of an automobile engine comprises the following steps:

[0007] S1, offline parameter acquisition, including:

[0008] (1) Obtain the secondary channel transfer function from the vehicle speaker to the human ear position;

[0009] (2) Based on the FxLMS algorithm, the engine speed generates two reference signals, sine and cosine, as the input signals of the algorithm. Combined with the secondary channel transfer function, the Filter-x signal is constructed. The minimum mean square value of the noise error signal at the human ear position is used as the optimization goal to obtain the corresponding adaptive filter weight coefficients under different vehicle operating conditions.

[0010] S2, online noise reduction: according to the current vehicle operating condition, the adaptive filter weight coefficients under the same or similar operating conditions as the current vehicle operating condition in step S1 are retrieved to generate a speaker drive signal to send out a cancellation signal.

[0011] Preferably, in step S1, the secondary channel transfer function modeling method is: the vehicle-mounted speaker sends a reference signal x(n), and the response signal e(n) is collected by a microphone arranged at the human ear position, and the weight coefficient W(n) of the modeling filter from the speaker to the human ear position is calculated according to the reference signal x(n) and the response signal e(n) as the secondary channel transfer function.

[0012] Furthermore, the weight coefficient W(n) of the secondary channel transfer function is calculated as follows: W(n+1)=W(n)+2μX(n)e(n), where μ is the filter iterative update step size, and X(n)=[x(n), x(n-1), ..., x(n-L+1)]. T , L is the length of W(n).

[0013] Furthermore, a white noise or a swept frequency signal is used to drive the vehicle speaker to emit a reference signal x(n).

[0014] Preferably, the operating conditions of the vehicle include vehicle gear position and engine speed.

[0015] Furthermore, obtaining the adaptive filter weight coefficients corresponding to different vehicle operating conditions specifically includes the following steps:

[0016] (1) Obtain the main order noise y of the engine at the human ear position in the car at different engine speeds in each gear MiNj ;

[0017] (2) Calculate the FIR filter coefficient W corresponding to the minimum mean square error of the main order noise of the vehicle engine under each operating condition. MiNj ;

[0018] (3) Constructing the adaptive filter weight coefficient W MN(n) database.

[0019] Furthermore, the main order noise of the engine at the human ear position is collected by a microphone arranged at the human ear position. MiNj .

[0020] Furthermore, the method for setting the speed of the car in each gear is as follows: the minimum engine speed N of the car in each gear position is set by a step size k. min and maximum speed N max The speed range between is discretized, speed N j N j∈[N min , N min+k , N min+2k ,......,N max ].

[0021] Furthermore, the FIR filter coefficient W MiNj The calculation formula is: MiNj =-(E[r(n)r T (n)]) -1 E[d(n)r(n)], where E(*) represents the time average of the independent variable, r(n) is the Filter-x signal, which is the convolution of the input signal and the secondary channel transfer function, and d(n) = y MiNj , is the expected signal.

[0022] Furthermore, in step S2, the calculation formula of the speaker driving signal y(n) is: , where x is a vector, the length of x is L, l = 1, 2...L.

[0023] Due to the application of the above-mentioned technical solution, the present invention has the following advantages over existing technologies: The present invention's active noise reduction method for automobile engines obtains the optimal adaptive filter weight coefficients for each vehicle operating condition offline. During the actual noise reduction process, the filter weight coefficients can be directly applied based on the vehicle's actual operating conditions, eliminating the need for convergence time and improving algorithm stability. Furthermore, this active noise reduction method uses the human ear as the noise reduction target point, optimizing the noise reduction effect at the human ear, thereby improving the noise reduction effect. Furthermore, the online noise reduction process eliminates the need for an error microphone, reducing the cost of the active noise reduction system and improving its applicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Attachment Figure 1 This is a flow chart of the implementation of the active noise reduction method for an automobile engine according to the present invention. DETAILED DESCRIPTION

[0025] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0026] The implementation process of the automobile engine active noise reduction method of the present invention is as follows: Figure 1 The specific implementation steps are as follows:

[0027] S1, offline parameter acquisition.

[0028] (1) Obtain the secondary channel transfer function from the vehicle speaker to the human ear position offline.

[0029] In this embodiment, the secondary channel transfer function modeling method is as follows: white noise or a swept frequency signal is used to drive the vehicle speaker to emit a reference signal x(n), a response signal e(n) is collected by a microphone arranged at the human ear position, and the weight coefficient W(n) of the modeling filter from the speaker to the human ear position is calculated based on the reference signal x(n) and the response signal e(n) as the secondary channel transfer function.

[0030] The calculation formula of the weight coefficient W(n) of the secondary channel transfer function is:

[0031] W(n+1)=W(n)+2μX(n)e(n)

[0032] Where μ is the filter iteration update step size, X(n) = [x(n), x(n-1), ..., x(n-L+1)] T , L is the length of W(n).

[0033] (2) Offline acquisition of the adaptive filter weight coefficients corresponding to different vehicle operating conditions.

[0034] Based on the FxLMS algorithm, the engine speed generates two reference signals, sine and cosine, as inputs. The Filter-x signal is constructed using the secondary channel transfer function. Minimizing the mean square value of the noise error signal at the target point is the optimization objective. The corresponding adaptive filter coefficients are then obtained, thereby achieving active control of interior noise. In this embodiment, the target point is the location of the human ear.

[0035] Different vehicle operating conditions produce different corresponding engine noise. Therefore, it is necessary to establish a database of active noise reduction system filter weight coefficients under various operating conditions so that the filter can be called according to the corresponding operating condition type during the actual noise reduction process, thereby accurately and effectively realizing active control of the vehicle's engine noise.

[0036] The operating conditions of the above-mentioned vehicle refer to the gear position when the vehicle is running and the engine speed at the gear position.

[0037] The steps for obtaining the filter weight coefficients of the vehicle under various operating conditions are as follows:

[0038] (1) Obtain the main order noise y of the engine at the human ear position in the car at different engine speeds in each gear MiNj .

[0039] Taking a car in a certain gear position as an example, the car has a minimum speed N corresponding to the gear position. min and maximum speed N max , with a step length of k, the minimum engine speed N min and maximum speed N max The speed interval between is discretized, that is, the speed N after discretizationj N j ∈[N min , N min+k , N min+2k ,......,N max ], thereby obtaining the engine speed of the car at that gear position. By placing a microphone at the human ear position, the main order noise of the engine at each speed can be collected. MiNj The smaller the K value, the better the noise reduction effect, but it will increase the amount of calculation.

[0040] Similarly, obtain the main order noise y of the engine at the human ear position in the car at different engine speeds when the car is in other gear positions MiNj .

[0041] (2) Calculate the FIR filter coefficient W corresponding to the minimum mean square error of the main order noise of the vehicle engine under each operating condition. MiNj .

[0042] The input signal, stage channel transfer function and main order noise of the engine are obtained according to the engine speed. MiNj , calculate the FIR filter coefficient W corresponding to the operating conditions MiNj , the calculation formula is:

[0043] W MiNj =-(E[r(n)r T (n)]) -1 E[d(n)r(n)]

[0044] Where E(*) represents the time average of the independent variable; r(n) is the Filter-x signal, which is the convolution of the input signal and the secondary channel transfer function; d(n) = y MiNj , is the expected signal.

[0045] (3) Constructing the adaptive filter weight coefficient W MN(n) database.

[0046] Through step (2), we can calculate a series of FIR filter coefficients W at different gear positions and speeds of the car. MiNj , construct the adaptive filter weight coefficient W MN(n) database.

[0047] S2, online noise reduction.

[0048] Based on the current vehicle operating conditions, i.e., the vehicle's real-time gear position and engine speed, the adaptive filter weight coefficients for the same or similar engine speed as the current vehicle gear position are retrieved from the filter weight coefficient WMN(n) database to generate a speaker drive signal y(n). A cancellation signal is then emitted through the speaker to reduce the engine noise inside the vehicle, thereby achieving the effect of active noise reduction.

[0049] The loudspeaker drive signal y(n) is calculated as:

[0050]

[0051] Wherein, x is a vector, the length of x is L, l=1,2……L.

[0052] 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 content of the present invention and implement it. 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 active noise reduction of an automobile engine, characterized by: The steps include: S1, offline parameter acquisition, including: (1) Obtain the secondary channel transfer function from the vehicle speaker to the human ear position, The secondary channel transfer function modeling method is as follows: the vehicle speaker emits a reference signal x(n), a response signal e(n) is collected by a microphone arranged at the human ear position, and the weight coefficient W(n) of the modeling filter from the speaker to the human ear position is calculated based on the reference signal x(n) and the response signal e(n) to serve as the secondary channel transfer function; The weight coefficient W(n) of the secondary channel transfer function is calculated as follows: W(n+1)=W(n)+2μX(n)e(n), where μ is the filter iteration update step size, and X(n)=[x(n), x(n-1), ..., x(n-L+1)]. T , L is the length of W(n); (2) Based on the FxLMS algorithm, the engine speed generates two reference signals, sine and cosine, as the input signals of the algorithm. Combined with the secondary channel transfer function, the Filter-x signal is constructed. The minimum mean square value of the noise error signal at the human ear position is used as the optimization goal to obtain the corresponding adaptive filter weight coefficients under different vehicle operating conditions. The vehicle's operating conditions include vehicle gear position and engine speed; Obtaining the adaptive filter weight coefficients corresponding to different vehicle operating conditions specifically includes the following steps: 1) Obtain the main order noise y of the engine at the human ear position in the car at different engine speeds in each gear MiNj ; The main order noise of the engine at the human ear position is collected by a microphone arranged at the human ear position MiNj ; 2) Calculate the FIR filter coefficient W corresponding to the minimum mean square error of the main order noise of the engine in each operating condition of the vehicle MiNj ; The method for setting the speed of the car in each gear is as follows: the minimum engine speed N of the car in each gear position is set by a step length k. min and maximum speed N max The speed range between is discretized, speed N j N j ∈[N min , N min+k , N min+2k ,......,N max ]; FIR filter coefficients W MiNj The calculation formula is: MiNj =-(E[r(n)r T (n)]) -1 E[d(n)r(n)], where E(*) represents the time average of the independent variable, r(n) is the Filter-x signal, which is the convolution of the input signal and the secondary channel transfer function, and d(n) = y MiNj , is the expected signal; 3) Construct adaptive filter weight coefficient W MN(n) database; S2, online noise reduction: according to the current vehicle operating condition, the adaptive filter weight coefficients under the same or similar operating conditions as the current vehicle operating condition in step S1 are retrieved to generate a speaker drive signal to send out a cancellation signal.

2. The method for active noise reduction of an automobile engine according to claim 1, characterized in that: Use white noise or a frequency sweep signal to drive the car speaker to emit a reference signal x(n).

3. The method for active noise reduction of an automobile engine according to claim 1, characterized in that: In step S2, the calculation formula of the speaker driving signal y(n) is: Wherein, x is a vector, the length of x is L, l=1,2......L.

Images