Active control system and method for vibration of exhaust pipeline of air compressor

Abstract

The invention discloses an active control system and method for vibration of an exhaust pipeline of an air compressor, the system comprises the exhaust pipeline with two rigidly fixed ends, and a signal generator, a power amplifier and a vibration exciter which are sequentially connected along a signal transmission direction, and a vibration excitation head of the vibration exciter acts on a pre-set vibration excitation point on the exhaust pipeline; the laser vibration meter, the data acquisition board and the digital signal processing module are sequentially connected in the signal transmission direction; the digital signal processing module generates a driving control signal for the piezoelectric actuator in real time by adopting a self-adaptive active vibration control strategy based on the collected vibration signal; a laser beam of the laser vibration meter vertically irradiates a measurement point on the surface of the exhaust pipeline, the piezoelectric actuator is arranged at a preset position on the surface of the exhaust pipeline, pipeline vibration is simulated through the vibration exciter, and a control signal is generated in real time by adopting a parallel FxLMS self-adaptive control strategy based on frequency band decomposition of a reference signal; and after power amplification, the piezoelectric actuator adhered to the surface of the pipeline is driven to generate deformation to counteract vibration.

Description

Technical Field

[0001] This invention belongs to the fields of active vibration control and compressor technology, specifically relating to an active vibration control system and method for air compressor exhaust pipes. Background Technology

[0002] In reciprocating compressor piping systems, severe airflow pulsations are often generated due to the intermittent pressurization of the fluid by the compressor and the turbulent flow of gas within the pipeline. Excessive airflow pulsations not only significantly increase the operating power consumption of the compressor but also induce strong vibrations in the piping system. Over time, this can lead to loosening of pipe joints, failure of supporting structures, and even fatigue cracking of the pipes, seriously threatening the safe and stable operation of the equipment.

[0003] Existing vibration reduction technologies mainly focus on passive vibration isolation or offline simulation analysis. Passive methods lack adaptability to different operating conditions and struggle to cope with wide-frequency vibrations under complex conditions such as variable compressor speed and load; furthermore, their reliability and maintainability are insufficient in high-temperature, high-pressure, and highly corrosive environments. Simulation analysis, on the other hand, cannot achieve real-time monitoring and dynamic intervention of the operating status. In the field of active control, standard FxLMS often suffers from decreased control performance when processing wide-frequency vibration signals due to differences in the convergence characteristics of different frequency components, making it difficult to meet the requirements for coordinated suppression of multi-frequency composite vibrations.

[0004] Patent CN110486562A discloses an active vibration control device for pipelines, relating to the field of vibration control. It includes a pipeline, a signal conversion unit, a signal acquisition unit, a control unit, and an execution unit. The receiving end of the signal acquisition unit is connected to the output end of the signal conversion unit, and the input end of the control unit is connected to the output end of the signal acquisition unit. The control unit is connected to the execution unit. The signal conversion unit is fixed at the first-order vibration mode of the pipeline. Based on a power-law approaching sliding mode variable structure control algorithm, it can control vibration more quickly, thereby actively controlling the pipeline's vibration and ensuring that the normal operation of the pipeline is not affected by vibration. Patent CN118729082A discloses a ring-shaped active vibration damping device and its control system in a pipeline system. The ring-shaped active vibration damping device includes a ring-shaped support structure, four piezoelectric actuators, and four accelerometers. Piezoelectric actuators are fixed on the inner surface of the ring-shaped support structure at 45°, 135°, 225°, and 315° directions, and the arrangement direction of the piezoelectric actuators is along the radial direction of the ring-shaped support structure. These four piezoelectric actuators can output driving force in any radial direction along the piping system, thereby achieving active vibration suppression. An accelerometer is also fixed near the surface of each piezoelectric actuator to measure vibration signals. This ring-shaped active vibration damping device can effectively suppress pipeline vibration and reduce or eliminate its adverse effects. However, it does not perform vibration signal acquisition and control at the same cross-section, nor does it provide a specific active control method. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides an active control system and method for vibration of air compressor exhaust pipes. It innovatively introduces an adaptive control strategy based on reference signal band decomposition and parallel FxLMS algorithm. This adaptive control strategy achieves independent optimized control of different frequency components by performing band decomposition and parallel processing on the vibration signal, significantly improving the broadband vibration suppression effect.

[0006] To achieve the above objectives, in a first aspect, the present invention provides an active control system for vibration of an air compressor exhaust pipe, comprising: The exhaust pipe is rigidly fixed at both ends, and a signal generator, a power amplifier and an exciter are connected in sequence along the signal transmission direction. The excitation head of the exciter acts on a pre-set excitation point on the exhaust pipe. It also includes a laser vibration meter, a data acquisition board, and a digital signal processing module connected sequentially along the signal transmission direction; the digital signal processing module generates drive control signals for the piezoelectric actuator in real time based on the acquired vibration signals and adopts an adaptive active vibration control strategy. The laser beam of the laser vibrometer is vertically irradiated onto the measurement point on the surface of the exhaust pipe, and the piezoelectric actuator is set at a preset position on the surface of the exhaust pipe.

[0007] Furthermore, two laser vibrometers are installed, and the two laser vibrometers are fixedly mounted by brackets.

[0008] Furthermore, four protrusions are evenly distributed around the outer wall of the middle section of the exhaust pipe. The four protrusions are located at the same cross section of the exhaust pipe. Two laser vibration meters irradiate the protrusions directly in front and directly below, respectively. The exciter acts on the outer wall of the exhaust pipe between the irradiation points of the two laser vibration meters.

[0009] Furthermore, on the radial cross-section of the pipe, the installation positions of the piezoelectric actuators and the measuring points are distributed at 90° intervals, and the measuring points are the positions directly opposite the laser vibration meter.

[0010] Secondly, this invention provides an active vibration control method for exhaust pipes based on the above-mentioned system, comprising: simulating target vibration on the exhaust pipe using a vibrator; non-contactly acquiring vibration displacement signals using a laser vibration meter; generating control signals in real time using a parallel FxLMS adaptive control algorithm in a digital signal processing module; amplifying the control signals to drive a piezoelectric actuator to generate deformation-canceling vibration; and ensuring continuous vibration mean square values. N Stop when the value falls below the threshold.

[0011] Furthermore, the parallel FxLMS adaptive control algorithm is used to generate control signals in real time, including: S1: Design an adaptive bandpass filter bank; S2: Initialize the adaptive bandpass filter bank parameters; S3: Obtain reference signal x ( n ) and error signal e ( n Reference signal x ( n The original sine wave signal output from the signal generator connected to the exciter is used directly, along with the error signal. e ( n The laser vibration meter, which uses a data acquisition board, collects the vibration displacement signal of the exhaust pipe in real time. S4: Combine the adaptive bandpass filter to decompose the reference signal frequency band and generate the filtered reference signal; S5: Generate a control signal at a specific moment. y ( n ); S6: Control signal y ( n Amplified to the effective operating voltage of the piezoelectric actuator; S7: The piezoelectric actuator deforms under the amplified control signal to counteract the vibration of the exhaust pipe at the specified moment; S8: Update the filter weights for the next time step based on the instantaneous gradient descent method; S9: At each sampling time n Repeat S3-S8, when e ( n The iteration stops when the mean square value of the signal is continuously lower than the threshold.

[0012] Furthermore, the filter in S1 is a wavelet packet, and has K A bandpass filter.

[0013] Furthermore, the parameters for S2 initialization include K The order of the adaptive filter L k , K a convergence step size factor μ k , K Initialized weight coefficient vector w k (0), K Sub-band secondary channel estimation model .

[0014] Furthermore, the reference signal mentioned in step 3 x ( n The signal is from the exciter drive signal, and the error signal is from the exciter drive signal. e ( n (Source: Laser Vibration Meter)

[0015] Furthermore, by combining an adaptive bandpass filter, the reference signal frequency band is decomposed to generate a filtered reference signal including: S41, the reference signal at the current time x ( n ) and its historical values ​​are input into S1 as designed K An adaptive bandpass filter is obtained. K Reference signal samples of each sub-band ; S42, for each sub-band k Construct sub-band reference signal vector :

[0016] S43, for each sub-band k The construction length is M k Secondary channel input vector :

[0017] S44, For each sub-band k Use the corresponding sub-band secondary channel model (coefficient For the secondary channel input vector Perform filtering to generate a filtered reference signal. :

[0018] S45, based on the filtered reference signal For each sub-band k Construct its filtered reference signal vector : .

[0019] Compared with existing technologies, this invention has at least the following advantages: The system achieves active vibration excitation by rigidly fixing the exhaust pipe and configuring a vibrator and signal generation link. The excitation head can accurately excite specific modal vibrations when applied to a preset point, providing a basis for verifying control strategies; This invention achieves real-time online monitoring and dynamic suppression of exhaust pipe vibration through the synergy of non-contact measurement with a laser vibration meter and active control with a piezoelectric actuator, overcoming the shortcomings of passive vibration reduction, such as poor flexibility and inability to dynamically respond to changes in operating conditions; The laser sensor and piezoelectric actuator used in this invention have convenient installation methods, intuitive and accurate vibration suppression effects, and a wide range of applications, which can be extended to various tubular and plate / shell structures.

[0020] This invention presents an adaptive control strategy based on the parallel FxLMS algorithm of reference signal frequency band decomposition. By performing sub-band decomposition and independent optimization on broadband vibration signals, it achieves decoupled control of different frequency components, fundamentally solving the problem of reduced control performance caused by frequency-dependent convergence characteristics in broadband vibration suppression.

[0021] Furthermore, the laser vibrometer adopts non-contact measurement, and the vertical illumination design ensures the accuracy of vibration signal acquisition. The dual measurement point configuration is fixed by a bracket to enhance system stability.

[0022] Furthermore, the circumferentially distributed boss structure forms a standard measurement reference surface, and the dual measuring point layout at the front and bottom, combined with the centrally installed exciter, effectively captures the bending vibration mode of the pipeline.

[0023] Furthermore, the design of distributing the piezoelectric actuators and measurement points at 90° intervals conforms to the principle of vibration wave interference, maximizing the cancellation of vibration energy. The digital signal processing module employs an adaptive control strategy, achieving dynamic optimization of vibration suppression through real-time feedback adjustment.

[0024] Furthermore, the collaborative layout of multiple sensors and the spatial phase difference configuration of actuators form a spatial sampling matrix for closed-loop control, enhancing the system's ability to suppress complex vibration modes. Attached Figure Description

[0025] Figure 1A block diagram of the active vibration control system for exhaust pipes provided by the present invention.

[0026] Figure 2 The flowchart of the active vibration control method for exhaust pipes provided by the present invention is shown.

[0027] Figure 3 The system block diagram for reference signal generation after reference signal frequency band decomposition and filtering provided by the present invention.

[0028] Figure 4 The arrangement of the exhaust pipe boss, laser vibration meter and piezoelectric actuator provided by the present invention.

[0029] Figure 5 The arrangement of the active vibration control system for exhaust pipes provided by the present invention.

[0030] Explanation of reference numerals in the attached diagram: 1-Exhaust pipe; 2-Signal generator; 3-Power amplifier; 4-Vibrator; 5-Laser vibrometer; 6-Data acquisition board; 7-Digital signal processing module; 8-Piezoelectric ceramic drive power supply; 9-Piezoelectric actuator; 10-Boss. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. It should be noted that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the invention.

[0032] In the description of this invention, it should be understood that the terms "center," "upper," "lower," "front," and "rear" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0033] refer to Figure 1 and Figure 4The present invention discloses an active control system for the vibration of an air compressor exhaust pipe, comprising an exhaust pipe 1, a signal generator 2, a power amplifier 3, a vibrator 4, a laser vibration meter 5, a data acquisition board 6, a signal processing module 7, a piezoelectric ceramic drive power supply 8, and a piezoelectric actuator 9. The exhaust pipe 1 adopts a typical exhaust pipe section that matches the target air compressor. Both ends of the exhaust pipe 1 are rigidly fixed to a stable experimental support frame by clamps to ensure that the boundary condition of the exhaust pipe 1 is that both ends are fixedly supported. The middle section of the exhaust pipe 1 is the main vibration point. To ensure the stable installation of the piezoelectric actuator 9, the laser vibration meter 5 can accurately sense the vibration that needs to be suppressed. Four bosses are installed in the middle section of the exhaust pipe 1. The four bosses are in the same axial position and are evenly distributed at 90° intervals on the outer wall of the exhaust pipe 1 directly above, directly in front of, directly below, and directly behind.

[0034] refer to Figure 4 and Figure 5 The signal generated by the signal generator 2 is used to simulate the frequency characteristics of the target vibration source; the output signal of the signal generator 2 is connected to the input terminal of the power amplifier 3 for power amplification; the output terminal of the power amplifier 3 is connected to the exciter 4, and the excitation head of the exciter 4 acts on the pre-selected excitation point on the middle section of the exhaust pipe 1 to simulate actual vibration. The exciter 4 generates a mechanical excitation force of the same frequency according to the input amplified signal, driving the exhaust pipe 1 to generate forced vibration.

[0035] The input terminal of the data acquisition board 6 is connected to the output terminal of the laser vibration meter 5; the input terminal of the digital signal processing module 7 is connected to the output terminal of the data acquisition board 6; the input terminal of the piezoelectric ceramic driving power supply 8 is connected to the output terminal of the digital signal processing module 7; the output terminal of the digital signal processing module 7 is connected to the input terminal of the piezoelectric ceramic driving power supply 8 via the signal generator 2 and the power amplifier 3 in sequence. Two piezoelectric actuators 9 are attached to preset positions on the surface of the exhaust pipe 1, and their input ends are connected to the output end of the piezoelectric ceramic drive power supply 8; two laser vibration meters 5 are fixed by brackets, and the laser beams are vertically irradiated to the measurement points on the surface of the exhaust pipe 1 to collect vibration displacement signals in real time.

[0036] The digital signal processing module 7 is configured as a controller that executes a parallel FxLMS adaptive control algorithm based on reference signal frequency band decomposition, and generates a piezoelectric actuator 9 drive signal in real time.

[0037] The laser probes of the laser vibration meter 5 are fixed by a bracket. Their positions and angles are adjusted so that the laser beam from one laser vibration meter 5 is perpendicularly irradiated onto the front protrusion on the outer wall of the middle section of the exhaust pipe 1, and the laser beam from the other laser vibration meter is perpendicularly irradiated onto the lower protrusion on the outer wall of the middle section of the exhaust pipe 1. Furthermore, special reflective patches are attached to the front and lower protrusions of the pipe to improve the signal-to-noise ratio and accurately detect the vibrations that need to be suppressed.

[0038] The data acquisition board 6 inputs the vibration signal measured by the laser vibrometer 5 into its analog input channel. Simultaneously, the drive signal of the exciter 4 is input into another analog input channel as a reference signal.

[0039] The input terminal of the piezoelectric ceramic drive power supply 8 receives control signals from the digital signal processing module 7, and the output terminal of the piezoelectric ceramic drive power supply 8 is connected to the piezoelectric actuator 9 to provide the required drive control signals.

[0040] Two piezoelectric actuators 9 are provided, each attached to the middle section of the outer wall of the exhaust pipe 1. One piezoelectric actuator 9 is installed on the boss 10 directly above the pipe, and the other piezoelectric actuator 9 is installed on the boss 10 directly behind the pipe. The piezoelectric actuator 9 is connected to the piezoelectric ceramic drive power supply 8, and after being supplied with a control voltage signal generated by the digital signal processing module 7, it generates expansion and contraction deformation to actively cancel vibration.

[0041] refer to Figure 2 The present invention provides an active vibration control method for air compressor exhaust pipes, which specifically includes the following steps: S1. Designing filter banks Include K An adaptive bandpass filter, the adaptive bandpass filter will filter the reference signal x ( n Decomposed into K Different subbands, adaptive bandpass filter type selection Daubechies wavelet filter.

[0042] S2, Initialization parameters initialization K The weight vector of an adaptive bandpass filter w k (0), k =1,2,…, K The length of each weight vector L k Initialize the secondary channel estimation model ,initialization K Sub-band secondary channel model Its model length is M k That is, coefficient vector The length is M k ;initialization K Step size factor A larger step size is used to accelerate convergence in the main vibration harmonic frequency band, while a smaller step size is used to improve stability in the secondary or noise frequency band; initial threshold is used. ϵ .

[0043] S3. Obtain the reference signal x ( n ) and error signal e ( n ) Reference signal x ( n The original sine wave signal output from the signal generator 2 connected to the exciter 4 is used directly. This original sine wave signal contains the target vibration frequency information to be suppressed and the error signal. e ( n The laser vibration meter 5, which is read by the data acquisition board 6, collects the vibration displacement signal of the exhaust pipe 1 in real time.

[0044] S4. Reference signal frequency band decomposition and filtering followed by reference signal generation. Figure 3 .

[0045] S41, Set the reference signal at the current time. x ( n ) and its historical values ​​are input into S1 as designed K An adaptive bandpass filter is obtained. K Reference signal samples of each sub-band .

[0046] S42. Constructing the sub-band reference signal vector For each sub-band k Construct sub-band reference signal vector :

[0047] S43, For each sub-band k The construction length is M k Secondary channel input vector :

[0048] S44. Generate the filtered reference signal For each sub-band k Use the corresponding sub-band secondary channel model (coefficient ) Sub-band secondary channel signal vector Perform filtering to generate a filtered reference signal. :

[0049] S45. Based on the filtered reference signal For each sub-band k Construct its filtered reference signal vector :

[0050] S5. Generate a control signal for a specific moment.

[0051] For each sub-band k Using its adaptive bandpass filter weights and its sub-band reference signal vector Calculate the sub-band control signal:

[0052] All k Each sub-band control signal Adding them together, we obtain the final full-band control signal:

[0053] S6, Power Amplifier 3 amplifies the control signal. The synthesized full-band control signal The digital signal is converted into an analog signal by the DAC, and then amplified by the power amplifier 3 to the voltage range required by the piezoelectric actuator 9. .

[0054] S7, piezoelectric actuator 9 generates deformation to counteract vibration. Amplified signal Drive the piezoelectric actuator 9 to generate deformation To counteract the vibration of exhaust pipe 1.

[0055]

[0056] in, S ( z ) is a real secondary channel.

[0057] S8. Update the filter weights for the next time step based on the instantaneous gradient descent method. For each sub-band k Independent weight updates: using the same full-band error signal e ( n ) and the filtered reference signal vector of that sub-band The weight vector for the next time step of the subband is calculated using the FxLMS update formula. :

[0058] S9, at each sampling time n Repeat steps S3-S8, when e ( n The iteration stops when the mean square value of the signal is continuously below the threshold. At each sampling time n Repeat steps 3-8, using the full-band error signal. e ( n Calculate the mean square value to monitor whether the mean square value is continuous. N The second time below the threshold ϵ If the condition is met, the iteration stops; otherwise, the iteration continues to the next sampling time until the loop ends.

[0059] This invention provides an active control system and method for vibration of air compressor exhaust pipes. Addressing the problems of existing passive vibration reduction technologies being unable to adapt to complex operating conditions and traditional active control algorithms struggling to coordinate and suppress broadband vibrations, the system includes an exhaust pipe rigidly fixed at both ends, a signal generator, a power amplifier, and a vibrator connected sequentially along the signal transmission direction. The vibrator's excitation head acts on a pre-set excitation point on the exhaust pipe. It also includes a laser vibrometer, a data acquisition board, and a digital signal processing module connected sequentially along the signal transmission direction. Based on the acquired vibration signal, the digital signal processing module uses an adaptive active vibration control strategy to generate a drive control signal for the piezoelectric actuator in real time. The laser beam from the laser vibrometer vertically illuminates the measurement point on the exhaust pipe surface, and the piezoelectric actuator is positioned at a preset location on the exhaust pipe surface. The method simulates pipe vibration through the vibrator, collects vibration signals from the laser vibrometer, and transmits them to the digital signal processing module via the data acquisition board. The core employs a parallel FxLMS adaptive control strategy based on reference signal frequency band decomposition to generate a control signal in real time. After power amplification, the signal drives the piezoelectric actuator adhered to the pipe surface to generate deformation to counteract the vibration. The present invention provides an active control system for the vibration of an air compressor exhaust pipe. This system utilizes the inverse piezoelectric effect of the piezoelectric ceramic material in the piezoelectric actuator to convert electrical energy into mechanical energy. The digital signal processing module uses an adaptive control strategy based on the parallel FxLMS algorithm with reference signal frequency band decomposition to calculate the control signal required for active control. This causes the piezoelectric actuator to produce a certain deformation to suppress the vibration of the exhaust pipe, thereby significantly reducing the airflow pulsation effect in the exhaust pipe of the reciprocating air compressor.

[0060] The above content is only for illustrating the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solution based on the technical concept proposed in this invention shall fall within the scope of protection of the claims of this invention.

Claims

1. An active control system for vibration of an air compressor exhaust pipe, characterized in that, It includes an exhaust pipe (1) with rigid fixed ends, a signal generator (2), a power amplifier (3) and an exciter (4) connected in sequence along the signal transmission direction, and the excitation head of the exciter (4) acts on a pre-set excitation point on the exhaust pipe (1). It also includes a laser vibration meter (5), a data acquisition board (6), and a digital signal processing module (7) connected sequentially along the signal transmission direction; the digital signal processing module (7) generates a drive control signal for the piezoelectric actuator (9) in real time based on the acquired vibration signal and adopts an adaptive active vibration control strategy. The laser beam of the laser vibrometer (5) vertically irradiates the measurement point on the surface of the exhaust pipe (1), and the piezoelectric actuator (9) is set at a preset position on the surface of the exhaust pipe (1).

2. The active vibration control system for air compressor exhaust pipes according to claim 1, characterized in that, Two laser vibration meters (5) are set up, and the two laser vibration meters (5) are fixedly installed by brackets.

3. The active vibration control system for air compressor exhaust pipe according to claim 2, characterized in that, Four protrusions (10) are evenly distributed around the outer wall of the middle section of the exhaust pipe (1). The four protrusions (10) are located at the same cross section of the exhaust pipe (1). Two laser vibration meters (5) irradiate the protrusions directly in front and directly below, respectively. The exciter (4) acts on the outer wall of the exhaust pipe (1) between the irradiation points of the two laser vibration meters (5).

4. The active vibration control system for air compressor exhaust pipe according to claim 1, characterized in that, On the radial section of the pipe, the installation positions of the piezoelectric actuator (9) and the measuring points are distributed at 90° intervals, and the measuring points are the positions directly opposite the laser vibration meter (5).

5. A method for active vibration control of an exhaust pipe based on the system of any one of claims 1-4, characterized in that, include: The target vibration is simulated on the exhaust pipe (1) by the exciter (4); Vibration displacement signals are collected non-contactly using a laser vibrometer (5); the digital signal processing module (7) generates control signals in real time using a parallel FxLMS adaptive control algorithm; the amplified control signals drive the piezoelectric actuator (9) to generate deformation-canceling vibrations; the vibration mean square value is continuous. N Stop when the value falls below the threshold.

6. The active vibration control method for exhaust pipes according to claim 5, characterized in that, The parallel FxLMS adaptive control algorithm is used to generate control signals in real time, including: S1: Design an adaptive bandpass filter bank; S2: Initialize the adaptive bandpass filter bank parameters; S3: Obtain reference signal x ( n ) and error signal e ( n Reference signal x ( n The original sine wave signal output from the signal generator (2) connected to the exciter (4) is used directly, and the error signal is used. e ( n The vibration displacement signal of the exhaust pipe (1) is collected in real time by the laser vibration meter (5) read by the data acquisition board (6); S4: Combine the adaptive bandpass filter to decompose the reference signal frequency band and generate the filtered reference signal; S5: Generate a control signal at a specific moment. y ( n ); S6: Control signal y ( n Amplified to the effective operating voltage of the piezoelectric actuator (9); S7: The piezoelectric actuator (9) deforms under the amplified control signal to counteract the vibration of the exhaust pipe at the said moment; S8: Update the filter weights for the next time step based on the instantaneous gradient descent method; S9: At each sampling time n Repeat S3-S8, when e ( n The iteration stops when the mean square value of the signal is continuously lower than the threshold.

7. The active vibration control method for exhaust pipes according to claim 6, characterized in that, The filter in S1 is a wavelet packet, and has K A bandpass filter.

8. The active vibration control method for exhaust pipes according to claim 6, characterized in that, The parameters for S2 initialization include K The order of the adaptive filter L k , K a convergence step size factor μ k , K Initialized weight coefficient vector w k (0), K Sub-band secondary channel estimation model .

9. The active vibration control method for exhaust pipes according to claim 6, characterized in that, The reference signal mentioned in step 3 x ( n The signal is from the exciter drive signal, and the error signal is from the exciter drive signal. e ( n (Source: Laser Vibration Meter) 10. The active vibration control method for exhaust pipes according to claim 6, characterized in that, By combining an adaptive bandpass filter, the reference signal frequency band is decomposed, and the filtered reference signal includes: S41, the reference signal at the current time x ( n ) and its historical values ​​are input into S1 as designed K An adaptive bandpass filter is obtained. K Reference signal samples of each sub-band ; S42, for each sub-band k Construct sub-band reference signal vector : S43, for each sub-band k The construction length is M k Secondary channel input vector : S44, For each sub-band k Use the corresponding sub-band secondary channel model (coefficient Sub-band reference signal vector Perform filtering to generate a filtered reference signal. : S45, based on the filtered reference signal For each sub-band k Construct its filtered reference signal vector : 。

Images