Active noise reduction device and method for reactor chamber

Abstract

The invention particularly relates to an active noise reduction device and method for a reactor chamber, and belongs to the technical field of active noise reduction. The active noise reduction device can be produced in an industrial prefabrication mode according to the actual size on site, rapid assembly is achieved through bolted joints, and the integrity of an existing building structure is effectively kept; an independent waveguide sound channel is adopted for noise reduction, and the noise propagation area is reduced based on the acoustic waveguide theory. According to the active noise reduction method disclosed by the invention, effective noise reduction in a long channel and a small section is realized by introducing acoustic mode decomposition, multi-mode decoupling control, a time delay compensation mechanism and a stability enhancement measure. And multi-mode independent control and differentiated step length are realized through dynamic global step length. Through modal step size attenuation and dynamic adjustment, stability, low-order modal rapid convergence and high-order modal stable suppression are remarkably improved, the method is suitable for a complex sound field, and the algorithm has rapid convergence, high-precision noise reduction and high robustness in the complex sound field through combination of the two modes.

Description

Technical Field

[0001] The present invention specifically relates to an active noise reduction device and method for a reactor chamber, and belongs to the technical field of active noise reduction. Background Art

[0002] As residential electricity demand increases, the demand for reactor rooms also grows. Heat generated by reactors during operation must be dissipated through ventilation systems, resulting in a noticeable low-frequency hum or airflow noise from the ventilation equipment and the reactors themselves. As the number of reactor rooms increases or their operation time increases, this persistent noise is emitted through the exterior windows of the reactor rooms, significantly impacting the lives of surrounding residents and even triggering complaints. While traditional noise reduction methods (such as soundproof walls and sound-absorbing barriers) are somewhat effective, they are costly to construct, require significant space, and are less effective against low-frequency noise. Active noise reduction technology has a good effect on low-frequency noise reduction, but when applied in the reactor room, the following problems were found: The following problems were found: 1. Active noise reduction technology has a good effect on low-frequency noise reduction, and performs well in closed small spaces (such as noise-canceling headphones and cockpits), but it is rarely used in large spaces such as reactor rooms; 2. Conventional active noise reduction algorithms (such as FXLMS) have a dynamic response bottleneck under reactor room conditions. Its fixed step size mechanism is difficult to adapt to sudden changes in noise amplitude (such as sound pressure level jumps caused by reactor start-up and shutdown or sudden load changes), resulting in delayed convergence of weight coefficients and transient deterioration of noise reduction effects; 3. In the active control system of low-frequency noise, since the formation of standing waves needs to meet the matching relationship between the wavelength of the sound wave and the propagation distance, the short channel cannot provide sufficient phase accumulation space for low-frequency components below 100 Hz; 4. In the long-channel active noise reduction system, the sound wave propagation delay will cause phase mismatch between the secondary sound source signal and the original noise. Summary of the Invention

[0003] The technical problem to be solved by the present invention is: how to reduce the noise transmitted from the outer window of the reactor room.

[0004] To solve the above technical problems, the present invention proposes a first technical solution: an active noise reduction device for a reactor chamber, comprising a mounting frame mounted at a window of the reactor chamber and n waveguide acoustic noise reduction assemblies mounted on the mounting frame; n is a natural number greater than or equal to 1; the n waveguide acoustic noise reduction assemblies fill the window of the reactor chamber;

[0005] The waveguide acoustic noise reduction component includes a waveguide acoustic noise reduction channel, a first acoustic wave sensor, a second acoustic wave sensor and a noise reduction control unit; one side of the waveguide acoustic noise reduction channel passes through the window and is connected to the outside world, and the other side passes through the window and is connected to the inside of the reactor chamber; the first acoustic wave sensor is installed inside the side of the waveguide acoustic noise reduction channel that is connected to the inside of the reactor chamber; the second acoustic wave sensor is installed inside the side of the waveguide acoustic noise reduction channel that is connected to the outside world; the noise reduction control unit is installed at the inner center of the waveguide acoustic noise reduction channel and is connected to the sound generating unit;

[0006] When the first acoustic wave sensor receives a noise signal, the collected noise signal is transmitted to the noise reduction control unit for calculation, and the sound generating unit generates a noise reduction wave having the same amplitude and opposite phase as the noise signal received by the first acoustic wave sensor according to the calculation result, thereby eliminating the noise signal;

[0007] When the second acoustic wave sensor receives the residual noise signal, the collected residual noise signal is transmitted to the noise reduction control unit for updating, thereby updating the calculation parameters of the noise reduction control unit for calculating the noise reduction wave.

[0008] Furthermore, an air valve is arranged between the second acoustic wave sensor and the noise reduction control unit, and the air valve includes a mounting frame, an electric actuator, a rotating shaft and fan blades arranged on the waveguide noise reduction channel; the electric actuator is installed on the mounting frame, one end of the rotating shaft is connected to the electric actuator, and the other end is inserted into the waveguide noise reduction channel, and the fan blades are installed on the rotating shaft.

[0009] To solve the above-mentioned first technical problem, the second technical solution proposed by the present invention is: an active noise reduction method applied to the above-mentioned active noise reduction device, comprising the following steps:

[0010] Step 1: Set the noise reduction period T of the active noise reduction device, define the first noise reduction period when the waveguide acoustic noise reduction component on the active noise reduction device detects a noise signal as the first noise reduction period T1, and define the second noise reduction period to the Nth noise reduction period when the waveguide acoustic noise reduction component continuously detects the noise signal as the second noise reduction period T2 to the Nth noise reduction period T N ;

[0011] Step 2: In the first noise reduction cycle T1 of the active noise reduction device, m noise data are collected in real time and collected to form a first reference noise signal sequence P 1 ; As shown in the following formula (1),

[0012] P 1 ={p 1 (1) p1 (2), ..., p 1 (m)} (1)

[0013] In formula (1), p 1 (1) p 1 (2) ... to p 1 (m) are respectively the first noise data, the second noise data to the mth noise data collected in real time within the first noise reduction period T1;

[0014] According to the first reference noise signal sequence P 1 All data in the first noise reduction cycle T1 are used to determine the first mode M of the noise collected by the active noise reduction device. 1 ;

[0015] The first reference noise signal sequence P 1 All the data in are substituted into the following formula (2) for channel convolution processing to obtain m primary signals and secondary signals respectively.

[0016]

[0017] In formula (2), p 1 (i) is the first reference noise signal sequence P 1 The i-th noise data in h r (i) is the i-th primary convolution coefficient of the channel convolution; h p (i) is the i-th secondary convolution coefficient of the channel convolution; x 1 (i) is the p 1 (i) After the h r (i) The i-th primary signal formed after convolution; d 1 (i) is the p 1 (i) After the h p (i) The i-th secondary signal formed after convolution;

[0018] Collect m primary signals and secondary signals to form the first primary signal sequence X 1 and the first secondary signal sequence D 1 ;

[0019] Step 3: Initialize the filter weight coefficient of the transversal filter, and use the initialized filter weight coefficient as the first filter weight coefficient W of the transversal filter in the first noise reduction period T1. 1 ; The first primary signal sequence X 1 Input into the transversal filter and obtain the first reverse cancellation signal sequence S of the active noise reduction device in the first noise reduction period T1 according to the following formula (3): 1 ,

[0020] S 1 =X 1 *W 1 *H s (3)

[0021] In formula (3), H s is the expected convolution coefficient sequence of channel convolution;

[0022] The first reverse cancellation signal sequence S 1 Sending the corresponding sound waves to the noise reduction control unit in the active noise reduction device in the waveguide acoustic noise reduction channel to complete the noise reduction work of the active noise reduction device in the first noise reduction period T1;

[0023] Step 4: In the second noise reduction cycle T2 of the active noise reduction device, the first reverse cancellation signal sequence S 1 and the first secondary signal sequence D 1 Substitute into the following formula (4) to make a judgment:

[0024] ||(S 1 +D 1 )||≤ΔE (4)

[0025] In formula (4), ||(S i +D i )|| is the first reverse cancellation signal sequence S 1 and the first secondary signal sequence D 1 The modulus value after superposition; ΔE is the noise reduction judgment threshold, which is an empirical value;

[0026] If formula (4) is satisfied, the first filter weight coefficient W 1 , as the second filtering weight coefficient W of the transversal filter in the second noise reduction period T2 2 , and continue to perform step 5 below;

[0027] If the formula (4) is not satisfied, the first filter weight coefficient W is calculated by the following formula (5): 1 Update to obtain the second filtering weight coefficient W of the transversal filter in the second noise reduction period T2 2 , and continue to perform step 5 below;

[0028]

[0029] In formula (4), W i+1 and W i are the wave weight coefficients of the transverse filter in the (i+1)th noise reduction cycle and the (i)th noise reduction cycle, respectively; is the adaptive step size of the transversal filter in the i-th denoising cycle; S i and D i are respectively the reverse cancellation signal sequence and the secondary signal sequence of the active noise reduction device in the i-th noise reduction cycle; X i is the primary signal sequence of the active noise reduction device in the i-th noise reduction cycle; H c is the transfer function; μ max is the maximum step size, which is an empirical value; α is the attenuation factor, which is an empirical value; M i is the i-th mode of the noise data collected by the active noise reduction device in the i-th noise reduction cycle; γ is the regularization factor, which is an empirical value; z is a complex frequency domain variable; τ is the sound wave propagation delay; T is the length of the noise reduction cycle; K is the filter order; h k is the filter coefficient of the filter at the kth order; G b is the oversampling multiple, which is the empirical value;

[0030] Step 5: Repeat the principle of step 2 to obtain the second primary signal sequence X in the second noise reduction period T2 of the active noise reduction device. 2 and the second secondary signal sequence D 2 The second primary signal sequence X 2 and the second filtering weight coefficient W obtained in step 4 2 Substitute into formula (3) to replace the first primary signal sequence X 1 and the first filtering weight coefficient W 1 The second reverse cancellation signal sequence S of the active noise reduction device in the second noise reduction period T2 is calculated 2 ;

[0031] The second reverse cancellation signal sequence S 2 Sending the corresponding sound waves to the noise reduction control unit in the active noise reduction device in the waveguide acoustic noise reduction channel to complete the noise reduction work of the active noise reduction device in the second noise reduction period T2;

[0032] Step 6: Repeat the principles of steps 4 to 5 to complete the active noise reduction device in the third noise reduction cycle T3 to the Nth noise reduction cycle T N Noise reduction work inside.

[0033] Beneficial effects of the present invention: 1. The active noise reduction device in the present invention can be produced by industrial prefabrication according to the actual size of the site, and can be quickly assembled through bolted nodes, which increases the on-site installation efficiency by more than 60%, and realizes zero-wet and zero-fire operations, effectively maintaining the integrity of the existing building structure; adopts independent waveguide sound channel noise reduction, based on the acoustic waveguide theory, reduces the noise propagation area, and converts the three-dimensional diffuse sound field into one-dimensional plane wave propagation through the boundary constraint effect, converting the large space noise reduction problem into noise reduction in a small space channel, thereby improving the noise reduction effect. 2. The present invention proposes an active noise reduction method for the noise reduction problem, which realizes effective noise reduction in long channels and small cross-sections by introducing acoustic mode decomposition, multi-modal decoupling control, time delay compensation mechanism and stability enhancement measures. Multi-modal independent control and differentiated step sizes are achieved through dynamic global step sizes. Through modal step attenuation and dynamic adjustment, stability is significantly improved, low-order modes converge quickly, and high-order modes are stably suppressed. It is suitable for complex sound fields (such as reactor rooms and long channels). The combination of the two enables the algorithm to have fast convergence, high-precision noise reduction and strong robustness in complex sound fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a rendering of the active noise reduction device of the present invention installed in the window opening of the reactor room.

[0035] Figure 2 It is a structural schematic diagram of the waveguide acoustic noise reduction component in the active noise reduction device of the present invention.

[0036] Figure 3 4 is a flow chart of the active noise reduction method of the present invention. DETAILED DESCRIPTION

[0037] The following further describes an active noise reduction device and method for a reactor chamber according to the present invention in conjunction with the accompanying drawings and specific embodiments.

[0038] Example

[0039] The active noise reduction device of the reactor chamber in this embodiment, such as Figure 1 and Figure 2 As shown, it includes a mounting frame installed at the window of the reactor room and 9 waveguide noise reduction components installed on the mounting frame;

[0040] The waveguide acoustic noise reduction component includes a waveguide acoustic noise reduction channel 1, a first acoustic wave sensor 5, a second acoustic wave sensor 10 and a noise reduction control unit 7; one side of the waveguide acoustic noise reduction channel 1 passes through a window and is connected to the outside world, and the other side passes through a window and is connected to the inside of the reactor chamber; the first acoustic wave sensor 5 is installed inside the side of the waveguide acoustic noise reduction channel 1 connected to the inside of the reactor chamber; the second acoustic wave sensor 10 is installed inside the side of the waveguide acoustic noise reduction channel 1 connected to the outside world; the noise reduction control unit 7 is installed at the center of the interior of the waveguide acoustic noise reduction channel and is connected to the sound generating unit 6; a wind valve 4 is arranged between the second acoustic wave sensor 10 and the noise reduction control unit 7, and the wind valve 4 includes a mounting frame 3 arranged on the waveguide noise reduction channel 1, an electric actuator 8, a rotating shaft 11 and fan blades; the electric actuator 8 is installed on the mounting frame 3, one end of the rotating shaft 11 is connected to the electric actuator 8, and the other end is inserted into the waveguide noise reduction channel 1, and the fan blades are installed on the rotating shaft 11.

[0041] When the first acoustic wave sensor 5 receives a noise signal, the collected noise signal is transmitted to the noise reduction control unit 7 for calculation. The sound generating unit 6 generates a noise reduction wave with the same amplitude and opposite phase as the noise signal received by the first acoustic wave sensor 5 according to the calculation result, thereby eliminating the noise signal.

[0042] When the second acoustic wave sensor 10 receives the residual noise signal, it transmits the collected residual noise signal to the noise reduction control unit 7 for updating, thereby updating the calculation parameters of the noise reduction control unit 7 for calculating the noise reduction wave.

[0043] The active noise reduction method applied to the above active noise reduction device in this embodiment is as follows: Figure 3 As shown, the following steps are included:

[0044] Step 1: Set the noise reduction cycle T of the active noise reduction device. Define the first noise reduction cycle when the waveguide acoustic noise reduction component on the active noise reduction device detects a noise signal as the first noise reduction cycle T1, and define the second noise reduction cycle to the Nth noise reduction cycle when the waveguide acoustic noise reduction component continuously detects the noise signal as the second noise reduction cycle T2 to the Nth noise reduction cycle T N ;

[0045] Step 2: In the first noise reduction cycle T1 of the active noise reduction device, m noise data are collected in real time and collected to form a first reference noise signal sequence P 1 ; As shown in the following formula (1),

[0046] P 1 ={p 1 (1) p 1 (2), ..., p 1 (m)} (1)

[0047] In formula (1), p1 (1) p 1 (2) ... to p 1 (m) are respectively the first noise data, the second noise data to the mth noise data collected in real time during the first noise reduction period T1;

[0048] According to the first reference noise signal sequence P 1 All data in the active noise reduction device determine the first mode M of the noise collected in the first noise reduction cycle T1 1 ;

[0049] The first reference noise signal sequence P 1 All the data in are substituted into the following formula (2) for channel convolution processing to obtain m primary signals and secondary signals respectively.

[0050]

[0051] In formula (2), p 1 (i) is the first reference noise signal sequence P 1 The i-th noise data in h r (i) is the i-th primary convolution coefficient of channel convolution; h p (i) is the i-th subconvolution coefficient of the channel convolution; x 1 (i) is p 1 (i) After h r (i) The i-th primary signal formed after convolution; d 1 (i) is p 1 (i) After h p (i) The i-th secondary signal formed after convolution;

[0052] Collect m primary signals and secondary signals to form the first primary signal sequence X 1 and the first secondary signal sequence D 1 ;

[0053] Step 3: Initialize the filter weight coefficient of the transversal filter, and use the initialized filter weight coefficient as the first filter weight coefficient W of the transversal filter in the first noise reduction cycle T1 1 ; The first primary signal sequence X 1 Input into the transversal filter and obtain the first reverse cancellation signal sequence S of the active noise reduction device in the first noise reduction period T1 according to the following formula (3): 1 ,

[0054] S 1 =X 1 *W 1 *H s (3)

[0055] In formula (3), H s is the expected convolution coefficient sequence of channel convolution;

[0056] The first reverse offset signal sequence S 1 The signal is sent to the noise reduction control unit in the active noise reduction device to emit corresponding sound waves in the waveguide acoustic noise reduction channel, completing the noise reduction work of the active noise reduction device in the first noise reduction cycle T1;

[0057] Step 4: In the second noise reduction cycle T2 of the active noise reduction device, the first reverse cancellation signal sequence S 1 and the first secondary signal sequence D 1 Substitute into the following formula (4) to make a judgment:

[0058] ||(S 1 +D 1 )||≤ΔE (4)

[0059] In formula (4), ||(S i +D i )|| is the first reverse cancellation signal sequence S 1 and the first secondary signal sequence D 1 The modulus value after superposition; ΔE is the noise reduction judgment threshold, which is an empirical value;

[0060] If formula (4) is satisfied, the first filter weight coefficient W 1 , as the second filter weight coefficient W of the transversal filter in the second noise reduction period T2 2 , and continue to perform step 5 below;

[0061] If the formula (4) is not satisfied, the first filter weight coefficient W is calculated by the following formula (5): 1 Update to obtain the second filtering weight coefficient W of the transverse filter in the second noise reduction period T2 2 , and continue to perform step 5 below;

[0062]

[0063] In formula (4), W i+1 and W i are the wave weight coefficients of the transverse filter in the i+1th denoising cycle and the ith denoising cycle, respectively; is the adaptive step size of the transversal filter in the i-th denoising cycle; S i and D i are the reverse cancellation signal sequence and secondary signal sequence of the active noise reduction device in the i-th noise reduction cycle; X i is the primary signal sequence of the active noise reduction device in the i-th noise reduction cycle; H c is the transfer function; μ maxis the maximum step size, which is an empirical value; α is the attenuation factor, which is an empirical value; M i is the i-th mode of the noise data collected by the active noise reduction device in the i-th noise reduction cycle; γ is the regularization factor, which is an empirical value; z is a complex frequency domain variable; τ is the sound wave propagation delay; T is the length of the noise reduction cycle; K is the filter order; h k is the filter coefficient of the filter at the kth order; G b is the oversampling multiple, which is the empirical value;

[0064] Step 5: Repeat the principle of step 2 to obtain the second primary signal sequence X in the second noise reduction cycle T2 of the active noise reduction device. 2 and the second secondary signal sequence D 2 ; The second primary signal sequence X 2 and the second filter weight coefficient W obtained in step 4 2 Substitute into formula (3) to replace the first primary signal sequence X 1 and the first filter weight coefficient W 1 The second reverse cancellation signal sequence S of the active noise reduction device in the second noise reduction period T2 is calculated 2 ;

[0065] The second reverse offset signal sequence S 2 The signal is sent to the noise reduction control unit in the active noise reduction device to emit corresponding sound waves in the waveguide acoustic noise reduction channel, thereby completing the noise reduction work of the active noise reduction device in the second noise reduction cycle T2;

[0066] Step 6: Repeat the principles of steps 4 to 5 to complete the active noise reduction device from the third noise reduction cycle T3 to the Nth noise reduction cycle T N Noise reduction work inside.

[0067] The present invention is described above by way of example in conjunction with the accompanying drawings. It is obvious that the specific implementation of the present invention is not limited to the above-mentioned method. As long as various non-substantial improvements are made using the method concept and technical solution of the present invention, or the concept and technical solution of the present invention are directly applied to other occasions without improvement, they are all within the scope of protection of the present invention.

Claims

1. An active noise reduction device for a reactor chamber, comprising a mounting frame mounted at a reactor chamber window and n waveguide acoustic noise reduction assemblies mounted on the mounting frame; n is a natural number greater than or equal to 1; the n waveguide acoustic noise reduction assemblies completely fill the reactor chamber window, characterized in that: The waveguide acoustic noise reduction assembly includes a waveguide acoustic noise reduction channel, a first acoustic wave sensor, a second acoustic wave sensor, and a noise reduction control unit; one side of the waveguide acoustic noise reduction channel passes through the window and communicates with the outside world, and the other side passes through the window and communicates with the interior of the reactor chamber; the first acoustic wave sensor is installed inside the side of the waveguide acoustic noise reduction channel that communicates with the interior of the reactor chamber; The second acoustic wave sensor is installed inside the side of the waveguide acoustic noise reduction channel connected to the outside world; the noise reduction control unit is installed at the center of the waveguide acoustic noise reduction channel and is connected to the sound generating unit; When the first acoustic wave sensor receives a noise signal, the collected noise signal is transmitted to the noise reduction control unit for calculation, and the sound generating unit generates a noise reduction wave having the same amplitude and opposite phase as the noise signal received by the first acoustic wave sensor according to the calculation result, thereby eliminating the noise signal; When the second acoustic wave sensor receives the residual noise signal, the collected residual noise signal is transmitted to the noise reduction control unit for updating, thereby updating the calculation parameters of the noise reduction control unit for calculating the noise reduction wave.

2. The active noise reduction device according to claim 1, characterized in that: An air valve is arranged between the second acoustic wave sensor and the noise reduction control unit, and the air valve includes a mounting frame, an electric actuator, a rotating shaft and fan blades arranged on the waveguide noise reduction channel; the electric actuator is installed on the mounting frame, one end of the rotating shaft is connected to the electric actuator, and the other end is inserted into the waveguide noise reduction channel, and the fan blades are installed on the rotating shaft.

3. An active noise reduction method applied to the active noise reduction device according to claim 1, characterized in that: The following steps are involved: Step 1: Set the noise reduction period T of the active noise reduction device, define the first noise reduction period when the waveguide acoustic noise reduction component on the active noise reduction device detects a noise signal as the first noise reduction period T1, and define the second noise reduction period to the Nth noise reduction period when the waveguide acoustic noise reduction component continuously detects the noise signal as the second noise reduction period T2 to the Nth noise reduction period T N ; Step 2: In the first noise reduction cycle T1 of the active noise reduction device, m noise data are collected in real time and collected to form a first reference noise signal sequence P 1 ; As shown in the following formula (1), P 1 ={p 1 (1)、p 1 (2)、...、p 1 (m)} (1) In formula (1), p 1 (1) p 1 (2) ... to p 1 (m) are respectively the first noise data, the second noise data to the mth noise data collected in real time within the first noise reduction period T1; According to the first reference noise signal sequence P 1 All data in the first noise reduction cycle T1 are used to determine the first mode M of the noise collected by the active noise reduction device. 1 ; The first reference noise signal sequence P 1 All the data in are substituted into the following formula (2) for channel convolution processing to obtain m primary signals and secondary signals respectively. In formula (2), p 1 (i) is the first reference noise signal sequence P 1 The i-th noise data in h r (i) is the i-th primary convolution coefficient of the channel convolution; h p (i) is the i-th secondary convolution coefficient of the channel convolution; x 1 (i) is the p 1 (i) After the h r (i) The i-th primary signal formed after convolution; d 1 (i) is the p 1 (i) After the h p (i) The i-th secondary signal formed after convolution; Collect m primary signals and secondary signals to form the first primary signal sequence X 1 and the first secondary signal sequence D 1 ; Step 3: Initialize the filter weight coefficient of the transversal filter, and use the initialized filter weight coefficient as the first filter weight coefficient W of the transversal filter in the first noise reduction period T1. 1 ; The first primary signal sequence X 1 Input into the transversal filter and obtain the first reverse cancellation signal sequence S of the active noise reduction device in the first noise reduction period T1 according to the following formula (3): 1 , S 1 =X 1 *W 1 *H s (3) In formula (3), H s is the expected convolution coefficient sequence of channel convolution; The first reverse cancellation signal sequence S 1 Sending the corresponding sound waves to the noise reduction control unit in the active noise reduction device in the waveguide acoustic noise reduction channel to complete the noise reduction work of the active noise reduction device in the first noise reduction period T1; Step 4: In the second noise reduction cycle T2 of the active noise reduction device, the first reverse cancellation signal sequence S 1 and the first secondary signal sequence D 1 Substitute into the following formula (4) to make a judgment: ||(S 1 +D 1 )||≤ΔE (4) In formula (4), ||(S i +D i )|| is the first reverse cancellation signal sequence S 1 and the first secondary signal sequence D 1 The modulus value after superposition; ΔE is the noise reduction judgment threshold, which is an empirical value; If formula (4) is satisfied, the first filter weight coefficient W 1 , as the second filtering weight coefficient W of the transversal filter in the second noise reduction period T2 2 , and continue to perform step 5 below; If the formula (4) is not satisfied, the first filter weight coefficient W is calculated by the following formula (5): 1 Update to obtain the second filtering weight coefficient W of the transversal filter in the second noise reduction period T2 2 , and continue to perform step 5 below; In formula (4), W i+1 and W i are the wave weight coefficients of the transverse filter in the (i+1)th noise reduction cycle and the (i)th noise reduction cycle, respectively; is the adaptive step size of the transversal filter in the i-th denoising cycle; S i and D i are respectively the reverse cancellation signal sequence and the secondary signal sequence of the active noise reduction device in the i-th noise reduction cycle; X i is the primary signal sequence of the active noise reduction device in the i-th noise reduction cycle; H c is the transfer function; μ max is the maximum step length, is the experience value; α is the attenuation factor, which is an empirical value; M i is the i-th mode of the noise data collected by the active noise reduction device in the i-th noise reduction cycle; γ is the regularization factor, which is an empirical value; z is a complex frequency domain variable; τ is the sound wave propagation delay; T is the length of the noise reduction cycle; K is the filter order; h k is the filter coefficient of the filter at the kth order; G b is the oversampling multiple, which is the empirical value; Step 5: Repeat the principle of step 2 to obtain the second primary signal sequence X in the second noise reduction period T2 of the active noise reduction device. 2 and the second secondary signal sequence D 2 The second primary signal sequence X 2 and the second filtering weight coefficient W obtained in step 4 2 Substitute into formula (3) to replace the first primary signal sequence X 1 and the first filtering weight coefficient W 1 The second reverse cancellation signal sequence S of the active noise reduction device in the second noise reduction period T2 is calculated 2 ; The second reverse cancellation signal sequence S 2 Sending the corresponding sound waves to the noise reduction control unit in the active noise reduction device in the waveguide acoustic noise reduction channel to complete the noise reduction work of the active noise reduction device in the second noise reduction period T2; Step 6: Repeat the principles of steps 4 to 5 to complete the active noise reduction device in the third noise reduction cycle T3 to the Nth noise reduction cycle T N Noise reduction work inside.

Images