Online identification method and system for vibration noise active control based on fusion operation

Abstract

The application provides an online identification method and system for active vibration noise control based on fusion operation, which adopts an active vibration noise control system algorithm to realize active control based on a secondary channel, and performs online identification on the secondary channel based on an additional excitation signal; a filtering algorithm is used to denoise and optimize the online identification result; then, a fusion operation is used to determine a fusion identification result based on the optimized new and old identification results, as a final identification result; and active vibration noise control is realized based on the determined online identification result of the secondary channel. The scheme overcomes the defects of the prior art that the accuracy of the online identification result of the secondary channel and the quality of the active control are limited, denoises and corrects the online identification result, reduces interference, and fuses and iterates the new and old secondary channels, so that the result of the secondary channel slowly changes on the basis of the original result of the secondary channel, reduces the instability probability of the system, and improves the active control effect.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of active vibration noise control, and in particular to an online identification method and system for active vibration noise control based on fusion operation. BACKGROUND

[0002] In the filter-x least mean square (FxLMS) algorithm, the identification of the secondary path is a key link of active vibration noise control. The system characteristics of the secondary path remain basically unchanged or change slowly during active control, so the existing technology mostly adopts the offline identification mode of the secondary path, which can also simplify the control algorithm. However, if the system characteristics change significantly during active control, the offline identification result will still be used when the secondary path changes greatly, which will lead to the fact that the vibration noise suppression effect is not obvious, and even the vibration noise situation is deteriorated, because the difference between the offline identification result and the actual secondary path is too large. Therefore, in this case, the online identification of the secondary path should be used to ensure that the stability and control accuracy of the control system are met. The quality of the online identification result of the secondary path is closely related to the effect of vibration noise control.

[0003] Some existing technologies propose to use white noise to perform online identification of the secondary path and participate in active vibration noise control. For example, patent document CN109448686A provides a new algorithm for cross updating of an active noise control system based on online identification of the secondary path, proposes to use a momentum FxLMS algorithm to update the weight values of the control filter, use a variable step size LMS algorithm to update the weight values of the modeling filter, and use a third adaptive filter of the proposed Newton LMS algorithm to eliminate the signal related to the error signal and the reference input signal, thereby improving the modeling accuracy of the modeling filter and the convergence speed of the entire ANC system. The weight values of the modeling filter are directly updated based on the calculation results of the filter algorithm, without considering the correlation factors between the calculation results at different time nodes, so the accuracy and reliability of the calculation results cannot be guaranteed, and it is difficult to achieve high-quality active noise control.

[0004] The information disclosed in the background section of the present application is only intended to deepen the understanding of the general background of the present application, and should not be regarded as acknowledging or implying in any form that the information constitutes prior art known to those skilled in the art. SUMMARY

[0005] To solve the above problems, the application provides an online identification method and system for active vibration noise control based on fusion operation, which uses an active vibration noise control system algorithm to realize active control based on a secondary channel, and performs online identification on the secondary channel based on an additional excitation signal; a filtering algorithm is used to denoise and optimize the online identification result; then a fusion operation is used to determine a fusion identification result based on the optimized new and old identification results, which is used as the final identification result; and the active vibration noise control is realized based on the determined online identification result of the secondary channel. The scheme overcomes the defects of the prior art that the accuracy of the online identification result of the secondary channel and the quality of the active control are limited, denoises and corrects the online identification result, reduces interference, and fuses and iterates the new and old secondary channels, so that the result of the secondary channel changes slowly on the basis of the original result of the secondary channel, reduces the probability of system instability, and improves the effect of active control; preferably, in one embodiment, the method comprises:

[0006] Step S1: using an active vibration noise control system algorithm to obtain an active control signal based on a secondary channel, combining an additional excitation signal as an actuator driving signal, and performing online identification on the secondary channel around a frequency point to be measured for a set time based on the additional excitation signal;

[0007] Step S2: using a filtering algorithm to denoise and optimize the online identification result;

[0008] Step S3: determining a fusion identification result based on the optimized new and old identification results through a fusion operation, which is used as the final identification result;

[0009] Step S4: processing the active control signal of the corresponding frequency point based on the determined online identification result of the secondary channel, and realizing active vibration noise control.

[0010] Further, in one embodiment, in the step S2, the process of denoising and optimizing the online identification result using the filtering algorithm comprises:

[0011] Based on the online identification result of the secondary channel at the previous moment, a state matrix of the online identification result of the secondary channel is used for prior prediction to obtain a prior estimation value of the identification result of the secondary channel at the current moment corresponding to the frequency point;

[0012] The prior estimation value of the system state error covariance matrix corresponding to the frequency point at the current moment is predicted, and then a Kalman gain is calculated in combination with an observation matrix of the corresponding frequency point and a measurement noise covariance matrix of the online identification of the secondary channel;

[0013] The identification result obtained by this identification is filtered using the Kalman gain in combination with the prior estimation value of the secondary channel identification result to obtain a correction value of the identification result, which is used as the optimized online identification result of the secondary channel.

[0014] Optionally, in one embodiment, the step S3, the fusion algorithm shown in the following formula is used to determine the fusion identification result based on the optimized new and old identification results:

[0015]

[0016] wherein w ja (k) is the target secondary path identification result of the jth frequency point at the kth moment in the active control process, M f is the fusion factor of the new and old secondary paths, is the correction value of the secondary path online identification result of the jth frequency point at the kth moment.

[0017] Specifically, in one optional embodiment, the step S2, the Kalman gain is calculated according to the following formula:

[0018]

[0019] wherein R j (k) is the measurement noise covariance matrix of the secondary path online identification of the jth frequency point at the kth moment, K j (k) is the Kalman gain of the secondary path online identification process of the jth frequency point at the kth moment, H j (k) is the observation matrix of the jth frequency point at the kth moment, is the prior estimation value of the system state error covariance matrix of the jth frequency point at the kth moment, and T represents the transposed matrix.

[0020] Further, in one embodiment, the step S2, the identification result correction value is calculated according to the following formula:

[0021]

[0022] wherein, is the correction value of the secondary path online identification result of the jth frequency point at the kth moment, w j (k) is the secondary path online identification result of the jth frequency point at the kth moment obtained by online identification, K j (k) is the Kalman gain of the secondary path online identification process of the jth frequency point at the kth moment, H j (k) is the observation matrix of the jth frequency point at the kth moment, is the prior estimation value of the secondary path identification result of the jth frequency point at the kth moment.

[0023] Preferably, in one embodiment, the method further comprises the step S3-1: after the step S3, an interpolation algorithm is used to analyze the secondary path online identification result corresponding to the to-be-predicted frequency point based on the fusion identification result, the to-be-predicted frequency point includes a frequency point which does not perform secondary path online identification but has identification requirement.

[0024] Optionally, in one embodiment, the method further comprises step S1-1: before the online identification of the secondary channel, setting the identification time of the online identification of the secondary channel in stages for the frequency points to be measured, setting identification time coefficients for the identification time of different stages, respectively calculating the identification step of different stages matched based on the identification time coefficients, and using the identification step for the online identification of the secondary channel.

[0025] Further, in one embodiment, the method further comprises step 1-2: after step S1, obtaining the online identification result and performing filtering processing, selecting the identification result sample set based on the filtered online identification result; using the weight band method to perform identification quality evaluation according to the identification result sample set, selecting the identification result whose quality meets the set requirement, and further performing step S2.

[0026] Specifically, in one preferred embodiment, step S1-2 further comprises:

[0027] After the identification quality evaluation using the weight band method based on the identification result sample set, if the result is passed, further comprising: transforming the identification result into an amplitude-phase form, performing secondary evaluation on the transformed identification result; selecting the sine and cosine weight extreme values of the online identification result of the secondary channel, respectively calculating the amplitude attenuation coefficient extreme value and the phase lag angle extreme value based on the sine weight extreme value and the cosine weight extreme value, and then calculating the secondary evaluation parameter determined by the combination of the amplitude attenuation coefficient evaluation parameter and the phase lag angle evaluation parameter to determine the final evaluation result, and selecting the identification result meeting the evaluation requirement.

[0028] Based on other aspects of the method according to any one or more of the above embodiments, the application further provides a storage medium having a program code stored thereon, which can implement the method according to any one or more of the above embodiments.

[0029] Based on the application aspect of the method according to any one or more of the above embodiments, the application further provides a vibration noise active control online identification system based on fusion operation, which comprises:

[0030] a memory for storing a computer program and a secondary channel model;

[0031] a processor for implementing the steps of the vibration noise active control online identification method based on fusion operation according to any one or more of the above embodiments when executing the computer program.

[0032] Based on the application aspect of the vibration noise active control online identification system according to the above embodiments, the application further provides a vibration noise active control system, which comprises the vibration noise active control online identification system based on fusion operation according to the above embodiments.

[0033] Compared with the closest prior art, the present application also has the following beneficial effects:

[0034] The present application provides an online identification method and system for active vibration noise control based on fusion operation, which adopts an active vibration noise control system algorithm to realize active control based on a secondary channel, and performs online identification on the secondary channel based on an additional excitation signal; a filtering algorithm is used to denoise and optimize the online identification result; a filtering algorithm is used to denoise and correct the online identification result, reduce interference components, and improve the accuracy of the online identification result of the secondary channel;

[0035] Further, the fusion operation is used to determine a fusion identification result based on the optimized new and old identification results as the final identification result; the active vibration noise control is realized based on the determined online identification result of the secondary channel, and the fusion algorithm is used to fuse and iterate the new and old secondary channels; the proportion of the online identification result of the secondary channel is reduced, the secondary channel result changes slowly based on the original secondary channel result, the possibility of system instability is reduced, the adverse effects of the inaccurate online identification result of the secondary channel on the control effect are avoided, and the active control effect is improved

[0036] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be realized and achieved by the structure particularly pointed out in the specification, claims and drawings. BRIEF DESCRIPTION OF DRAWINGS

[0037] The accompanying drawings are included to provide a further understanding of the present application, and constitute a part of the specification, which together with the embodiments of the present application, serve to explain the present application, and do not constitute a limitation on the present application. In the drawings:

[0038] Figure 1 is a flowchart of the online identification method for active vibration noise control based on fusion operation provided by an embodiment of the present application;

[0039] Figure 2 is a schematic diagram of the noise active control connection principle applied to a vibration source in the online identification method for active vibration noise control based on fusion operation provided by the embodiment of the present application;

[0040] Figure 3 is a schematic diagram of the identification time and step design principle of the online identification method for active vibration noise control based on fusion operation provided by another embodiment of the present application;

[0041] Figure 4 is a flowchart of the online identification method for active vibration noise control based on fusion operation provided by an embodiment of the present application;

[0042] Figure 5 is a flowchart of the online identification method of the fusion operation-based active control of vibration and noise according to another embodiment of the present application;

[0043] Figure 6 is a distribution diagram of the equivalent extreme value range of the sine component weight of the online identification method of the fusion operation-based active control of vibration and noise according to an embodiment of the present application;

[0044] Figure 7 is a diagram showing the change of the engine speed during the identification process in the online identification method of the fusion operation-based active control of vibration and noise according to an embodiment of the present application;

[0045] Figure 8 is a diagram showing the time-domain change of the error sensor signal during the identification process in the online identification method of the fusion operation-based active control of vibration and noise according to an embodiment of the present application;

[0046] Figure 9 (a) is a distribution diagram of the sine component weight of the online secondary channel identification result in the frequency range of 100-120 Hz in the online identification method of the fusion operation-based active control of vibration and noise according to an embodiment of the present application;

[0047] Figure 9 (b) is a distribution diagram of the cosine component weight of the online secondary channel identification result in the frequency range of 100-120 Hz in the online identification method of the fusion operation-based active control of vibration and noise according to an embodiment of the present application;

[0048] Figure 10 is a distribution diagram of the filtered online secondary channel identification result in the online identification method of the fusion operation-based active control of vibration and noise according to an embodiment of the present application;

[0049] Figure 11 is a structural diagram of the online identification system of the fusion operation-based active control of vibration and noise according to another embodiment of the present application. DETAILED DESCRIPTION

[0050] The embodiments of the present application will be described in detail below with reference to the accompanying drawings and embodiments, so that the persons skilled in the art can fully understand how the present application applies technical means to solve technical problems and achieve the implementation process of technical effects, and can implement the present application according to the above implementation process. It should be noted that, as long as there is no conflict, each embodiment in the present application and each feature of each embodiment can be combined with each other, and the technical solutions formed thereby are all within the protection scope of the present application.

[0051] Although the process diagram describes the operations as sequential processes, many of the operations can be performed in parallel, concurrently, or at the same time. The order of the operations can be re-arranged. The process can be terminated when its operations are completed, but could also have additional steps not included in the figure. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0052] The computer device includes user devices and network devices. The user devices or clients include, but are not limited to, computers, smart phones, PDAs (Personal Digital Assistants), etc.; the network devices include, but are not limited to, single network servers, server groups composed of multiple network servers, or clouds composed of a large number of computers or network servers based on cloud computing. The computer device can be independently operated to implement the present application, or can be connected to a network and implement the present application by interacting with other computer devices in the network. The network in which the computer device is located includes, but is not limited to, the Internet, a wide area network, a metropolitan area network, a local area network, a VPN network, etc.

[0053] The terms "first", "second", and the like, can be used herein to describe various elements, but the elements should not be limited by these terms. The terms are only used to distinguish one element from another. The term "and / or" as used herein includes any and all combinations of one or more of the associated associated items. When an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element, or intervening elements can be present.

[0054] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and / or "comprising", when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0055] The filter-x least mean square (FxLMS) algorithm is one of the algorithms widely used in active vibration noise control. In the FxLMS algorithm, the identification of the secondary path is a key link of the active vibration noise control. There are two ways of secondary path identification, offline identification and online identification. At present, it is believed that the system characteristics of the secondary path remain basically unchanged or change slowly during the active control, so the existing technology adopts the offline identification of the secondary path, which can also simplify the control algorithm. However, if the system characteristics change significantly during the active control, and the secondary path changes greatly, the offline identification result will be too different from the actual secondary path (more than 90 degrees), which will lead to the fact that the vibration noise suppression effect is not obvious, and even the vibration noise situation is deteriorated. Therefore, in this case, the online identification of the secondary path should be adopted to ensure the stability and control accuracy of the control system.

[0056] Some existing technologies propose to use white noise for online identification of the secondary path and participate in active vibration noise control. For example, patent document CN109448686A provides a new algorithm for cross updating of an active noise control system based on online identification of the secondary path, proposes to use the momentum FxLMS algorithm to update the weight value of the control filter, use the variable step LMS algorithm to update the weight value of the modeling filter, and use the third adaptive filter of the proposed Newton LMS algorithm to eliminate the signal related to the error signal and the reference input signal, so as to improve the modeling accuracy of the modeling filter and the convergence speed of the entire ANC system. During the online identification of the secondary path, the signal has many interference components, which leads to low quality of the identification result. The above scheme directly updates the weight value of the modeling filter based on the calculation result of the filtering algorithm, does not consider the correlation factors between the calculation results at different time nodes, cannot play the positive role of the comprehensive and iterative combination of the new online identification result and the old secondary path identification result, and the accuracy and reliability of the calculation result cannot be guaranteed, so the quality of the active noise control is limited.

[0057] To solve the above problems, the present application provides an online identification method for active vibration noise control based on fusion operation. The present application uses the sweep frequency method to perform online identification of the secondary path. On the basis of the online identification result of the secondary path, the Kalman filter is used to optimize the identification result, the online identification result is denoised and corrected, the interference components are reduced, and the accuracy of the online identification result of the secondary path is improved. At the same time, the fusion algorithm of the new and old secondary path identification results is considered, the new and old secondary paths are fused and iterated, the proportion of the online identification result of the secondary path is reduced, the secondary path result changes slowly on the basis of the original secondary path result, and the possibility of system instability is effectively reduced.

[0058] Next, the detailed flow of the method of the embodiment of the application is described in detail based on the accompanying drawings. The steps shown in the flowchart of the accompanying drawings can be executed in a computer system comprising, for example, a set of computer-executable instructions. Although the logical order of the steps is shown in the flowchart, in some cases, the steps shown or described herein can be executed in a different order.

[0059] Embodiment one:

[0060] Figure 1 The flowchart of the fusion operation-based online identification method for active vibration noise control provided by the first embodiment of the application is shown, referring to Figure 1 It can be seen that the method comprises the following steps.

[0061] Step S1: obtaining an active control signal based on a secondary channel using an active vibration noise control system algorithm, combining an additional excitation signal as an actuator driving signal, and performing online identification of the secondary channel around a frequency point to be measured for a set time based on the additional excitation signal;

[0062] Step S2: denoising and optimizing the online identification result using a filtering algorithm;

[0063] Step S3: determining a fusion identification result based on the optimized new and old identification results through fusion operation, as the final identification result;

[0064] Step S4: processing the active control signal of the corresponding frequency point based on the determined online identification result of the secondary channel, to realize active vibration noise control.

[0065] The embodiment of the application proposes a fusion operation-based online identification method for active vibration noise control. The online identification result of the secondary channel is denoised and corrected by Kalman filtering, and the online identification result of the secondary channel after interference removal and the original online identification result of the secondary channel are iteratively fused by a fusion algorithm. The finally obtained new online identification result of the secondary channel is applied to active vibration noise control, thereby ensuring the control effect.

[0066] In actual application, in the process of realizing vibration noise control by using an active vibration noise control system in the optional embodiment, the following operations are included:

[0067] The constructed notch FxLMS active vibration noise control system generates a reference signal by using a phase accumulation method based on the current vibration or noise frequency obtained;

[0068] The reference signal is filtered by the online identification result of the secondary channel and then input into the LMS algorithm of the active control part (active vibration noise control system), and an error sensor signal is input at the same time, so as to obtain the LMS weight value;

[0069] The acquired weight value is multiplied with the reference signal to obtain y(n), and then the additional excitation signal X is combined to obtain the output signal of the error sensor add The driving signal output to the actuator is obtained, and the vibration noise control is realized to reduce the vibration or noise. In the figure, the oblique arrows above or below the active vibration control algorithm weight block and the secondary path identification result weight block represent the iterative update of the results.

[0070] It should be noted that the reference signal in the secondary path online identification method of the vibration noise active control system is not limited to the phase accumulation method proposed in the above embodiment, but also includes other commonly used reference signal acquisition or generation methods.

[0071] The secondary path identification result of the notch FxLMS algorithm adopted in the embodiment of the application is in the form of amplitude attenuation and phase lag, so the filtering process through the secondary path identification result can be understood as the following logic:

[0072] The amplitude of the obtained reference signal x(n) is multiplied by the corresponding attenuation coefficient, and a lag angle is added to the phase to obtain the filtered reference signal x'(n). In actual application, the reference signal is input to the secondary path identification result as an operation parameter to obtain the filtered reference signal x'(n) after operation.

[0073] An additional excitation signal x add The secondary path is online identified, and the excitation signal is a sweep signal, that is, a sine excitation with a fixed frequency at a certain time. The secondary path online identification result of the jth frequency point at the kth time in the sweep frequency range can be obtained by the following formula:

[0074] w j (k)=w j (k-1)-2μ OID e(k-1)x add (k-1)

[0075] wherein, w j (k)=[w s,j (k)w c,j (k)] T ;

[0076] In the formula, w j (k) is the jth frequency point secondary path identification result at the kth time, μ OID is the secondary path online identification step, e(k) is the error sensor signal at the kth time, x add (k) is the additional sweep excitation signal at the kth time, w s,j (k) is the sine component weight of the jth frequency point secondary path identification result at the kth time, is the weight corresponding to the sine component of the secondary path identification result, and w c,j(k) is the cosine component weight value of the secondary path identification result of the jth frequency point at the kth moment, and T represents a transposed matrix.

[0077] For a single frequency vibration, the vibration of a certain point in the vibration propagation process can be decomposed into a linear combination of a sine component and a cosine component of the same frequency. The weight value of the secondary path online identification result obtained in the application is similar to the weight value corresponding to the sine component or the cosine component, which can be referred to as the sine component weight value or the cosine component weight value.

[0078] The initial value of the secondary path identification result can be set by the researcher, and in actual application, the initial value w j (0) = 0, but it can also be set to other values. With the initial value, further calculation can be performed through the iteration formula, and after a series of iteration calculations, w j (k-1) is obtained, and after further iteration, w j (k) is obtained.

[0079] In the process of realizing vibration noise control by using an active vibration noise control system, the secondary path is identified online based on the additional excitation signal around the frequency point to be measured; the secondary path identification result is used for filtering of a reference signal in the active vibration control process, and the filtered reference signal is used for an active vibration control LMS algorithm in the active vibration control process and obtains an active vibration control algorithm weight value;

[0080] Figure 2 Fig. 1 is a schematic diagram of a noise active control connection principle applied to a vibration source in the online identification method for vibration noise active control based on fusion operation provided by the embodiment of the application. The vibration source takes an automobile active suspension system as an example. The better the secondary path identification result is, the more accurate the active vibration control algorithm weight value obtained is. The more accurate the active vibration control algorithm weight value is, the more accurate the control signal y(n) used in the active vibration control algorithm is, and the better the active vibration control effect is, which is reflected in that the error sensor e(n) signal is smaller.

[0081] In the active vibration control process, the offline identification result of the secondary path or the result after the last online identification is used as the initial value at the beginning of control. After the control starts, the identification result of the secondary path is usually obtained through online identification. In the offline identification process of the secondary path, the active vibration control system is turned off, and only the additional excitation signal x addThe process of inputting excitation to the actuator and obtaining the secondary path identification result (essentially considered as a transfer function) through the response at this time is the process of online secondary path identification. In the active vibration control system, the excitation input to the actuator contains both the excitation signal of the active vibration control system and the additional excitation signal, so the error sensor signal at this time will be used for the calculation of the LMS (The Least Mean Square) weight value in the active vibration control system and the calculation of the weight value in the online secondary path identification system. The calculation methods of the two kinds of weight values are both the LMS method, but the parameters in the two kinds of weight value calculation processes are different. The online secondary path identification system is not always on, and has two states of opening and closing.

[0082] In actual application, whether the secondary path has a large change can be determined through the following operations, so as to determine whether the online identification system at the current frequency point needs to be opened or closed:

[0083] (1) First, the secondary path offline identification is performed for a to-be-judged identification frequency point;

[0084] (2) The offline identification result is applied to the active vibration control system without the online secondary path identification system (in actual execution, the online secondary path identification system is always in the closed state);

[0085] (3) The active vibration control system is opened, and the error sensor signal e(n) in a set time period is obtained, and the root mean square value thereof is calculated as a threshold value;

[0086] (4) The online secondary path identification system is opened, that is, the active vibration control system with the online secondary path identification system is started. Since the secondary path has a certain time-varying property, the secondary path identification result is an estimation of the actual secondary path, and after a certain period of time, the initial secondary path identification result will have a certain gap with the actual secondary path. When the gap is too large, the accuracy of the filtered reference signal will be reduced, thereby reducing the effect of the active vibration control.

[0087] Specifically, the amplitude of the error sensor signal e(n) becomes large, and after a set test time, if the change amplitude of the error sensor signal e(n) is greater than the threshold value obtained in step (3), it is indicated that the gap between the secondary path identification result and the actual secondary path is too large, and the secondary path needs to be re-identified, so the online secondary path identification system needs to be immediately opened for online secondary path identification.

[0088] After re-identification, due to the complex interference factors in the process of online identification of the secondary channel, the identification result is not good; then the accuracy of the reference signal filtered based on the identification result is still not high, and the satisfactory active control effect of vibration noise cannot be obtained.

[0089] Based on this, the application proposes to correct and denoise the obtained online identification result of the secondary channel based on the Kalman filtering method, and combine the identification results of different rounds to perform fusion operation, so as to determine the identification result of the secondary channel with high accuracy and comprehensive distribution, and participate in the active control process of vibration noise, thereby improving the control effect.

[0090] Preferably, in one embodiment, in the step S3, the process of denoising and optimizing the online identification result by using the filtering algorithm comprises:

[0091] Based on the online identification result of the secondary channel at the previous moment, the state matrix of the online identification result of the secondary channel is used for prior prediction to obtain the prior estimation value of the identification result of the secondary channel at the current moment corresponding to the frequency point

[0092] The prior estimation value of the system state error covariance matrix corresponding to the frequency point at the current moment is predicted Then, the Kalman gain is calculated by combining the observation matrix of the corresponding frequency point and the measurement noise covariance matrix of the online identification of the secondary channel.

[0093] The identification result obtained by the current identification is filtered by using the Kalman gain and the prior estimation value of the secondary channel identification result, to obtain the identification result correction value as the optimized online identification result of the secondary channel.

[0094] In a preferred embodiment, the prior estimation value of the identification result of the secondary channel at the current moment corresponding to the frequency point is predicted as follows

[0095]

[0096] In the formula, w j (k-1) is the secondary channel identification result of the jth frequency point at the previous moment, and k is greater than 1; A j is the state matrix of the online identification result of the jth frequency point, is the prior estimation value of the identification result of the jth frequency point at the kth moment.

[0097] Further, in one embodiment, the prior estimation value of the system state error covariance matrix corresponding to the frequency point at the current moment is predicted by the following formula

[0098]

[0099] wherein A j is the state matrix of the online identification result of the secondary channel at the jth frequency point, T represents the transposed matrix, Q j (k) is the process noise covariance matrix of the online identification of the secondary channel at the jth frequency point at the kth moment, is the prior estimation value of the system state error covariance matrix at the jth frequency point at the kth moment, T represents the transposed matrix. j (k-1) is the system state error covariance matrix at the jth frequency point at the k-1th moment.

[0100] In an optional embodiment, the Kalman gain is calculated according to the following formula:

[0101]

[0102] wherein R j (k) is the measurement noise covariance matrix of the online identification of the secondary channel at the jth frequency point at the kth moment, K j (k) is the Kalman gain of the online identification process of the secondary channel at the jth frequency point at the kth moment, H j (k) is the observation matrix of the secondary channel at the jth frequency point at the kth moment. is the prior estimation value of the system state error covariance matrix at the jth frequency point at the kth moment, T represents the transposed matrix.

[0103] Further, the identification result obtained by the present identification is filtered according to the following formula to obtain an identification result correction value:

[0104]

[0105] wherein, is the correction value of the online identification result of the secondary channel at the jth frequency point at the kth moment, w j (k) is the online identification result of the secondary channel at the jth frequency point at the kth moment obtained by online identification, K j (k) is the Kalman gain of the online identification process of the secondary channel at the jth frequency point at the kth moment, H j (k) is the observation matrix of the secondary channel at the jth frequency point at the kth moment. is the prior estimation value of the secondary channel identification result at the jth frequency point at the kth moment.

[0106] In actual application, the system state error covariance matrix correction value at the current moment is also obtained by the following formula, which is used to provide operation support for obtaining the prior estimation value of the system state error covariance matrix at the next moment and calculating the online identification result correction value at the next moment:

[0107]

[0108] wherein Pj (k) is the modified value of the system state error covariance matrix of the jth frequency point at time k, and I is the unit matrix.

[0109] Further, the secondary path results after Kalman filtering correction and the original secondary path are further fused step by step through a fusion factor. In step S3, the fusion recognition result is determined based on the optimized new and old recognition results through a fusion algorithm shown in the following formula:

[0110]

[0111] wherein w ja (k) is the target secondary path recognition result of the jth frequency point at time k actually used in the active control process, M f is a fusion factor of the new and old secondary paths, is the modified value of the secondary path online recognition result of the jth frequency point at time k.

[0112] The process of obtaining the final fusion recognition result based on the new and old secondary path recognition result fusion method in the embodiments of the application is not limited to the method using the fusion factor described in the above embodiments, and other methods easily thought of in the field also belong to the protection scope of the application.

[0113] Therefore, the fusion recognition result obtained through the fusion operation is used to filter the reference signal and participate in the active vibration noise control, and the vibration noise suppression ability of the active vibration noise control system can be effectively improved based on the more accurate secondary path online recognition result.

[0114] Embodiment two:

[0115] The online recognition method for vibration noise active control based on fusion operation provided in the second embodiment of the application comprises the following operations.

[0116] Step S1: An active vibration noise control system algorithm is used to obtain an active control signal based on a secondary path, an additional excitation signal is combined as an actuator driving signal, and the secondary path is identified online for a set time around a to-be-measured frequency point based on the additional excitation signal;

[0117] Step S2: A filtering algorithm is used to optimize and denoise the online recognition result;

[0118] Step S3: A fusion operation is used to determine a fusion recognition result based on the optimized new and old recognition results, which is used as a final recognition result;

[0119] Step S4: processing the active control signal of the corresponding frequency point based on the determined secondary path online identification result, to realize active vibration noise control. Embodiment two is a variant of embodiment one, and the same or corresponding technical features in the above embodiments will not be described again. The following will only describe the different technical features of this embodiment.

[0120] Since the secondary path online identification process is based on the set sweep frequency step to apply the additional sweep frequency excitation signal, for example, the sweep frequency step is △f, △f≥1, there will inevitably be some frequency points that are not identified. For these frequency points, the embodiment of the present application further includes step S3-1 after step S3: analyzing the secondary path online identification result corresponding to the to-be-predicted frequency point based on the fusion identification result; for vibration noise control of the to-be-measured frequency point. Wherein, the to-be-predicted frequency point includes frequency points that have not performed secondary path online identification but have identification requirements.

[0121] In an optional embodiment, the secondary path online identification result of the to-be-predicted frequency point can be obtained by interpolation operation as follows:

[0122]

[0123] Wherein N a =△f / 2, w ja(i) (k) is the actual application secondary path identification result of the to-be-predicted frequency point that is different from the jth frequency point by i Hz at k time, w ja (k) is the target secondary path identification result of the jth frequency point actually used at k time in the active control process.

[0124] It should be noted that the acquisition of the secondary path identification result of the un-identified frequency point is not limited to the operation logic described in the above embodiments of the present application. The protection scope of the present application should also include other easily thought interpolation operation methods.

[0125] The scheme of this embodiment overcomes the deficiency of the prior art online identification that can only identify the secondary path of a limited number of frequency points and cannot identify all frequency points. The fusion operation result of the new and old secondary path results is combined, and the interpolation method is used to update the secondary path of the frequency points that are not directly identified, so that as many secondary path identification results of frequency points as possible are obtained under the premise of ensuring identification efficiency.

[0126] Embodiment three:

[0127] The vibration noise active control online identification method based on fusion operation provided in embodiment three of the present application includes the following operations.

[0128] Step S1: obtaining an active control signal based on a secondary channel by using an active vibration noise control system algorithm, combining an additional excitation signal as an actuator driving signal, and performing online identification of the secondary channel around a frequency point to be measured based on the additional excitation signal for a set time;

[0129] Step S2: denoising and optimizing the online identification result by using a filtering algorithm;

[0130] Step S3: determining a fusion identification result based on the optimized new and old identification results by fusion operation, as a final identification result;

[0131] Step S4: processing the active control signal of the corresponding frequency point based on the determined online identification result of the secondary channel, to realize active vibration noise control. Embodiment three is a variant of the foregoing embodiment, and the same or corresponding technical features in the above embodiments will not be described again. Only the different technical features of this embodiment will be described below.

[0132] The researchers of the present application consider that the process of implementing online identification of the secondary channel is closely related to the accuracy of the identification result. Although some existing technologies in the field provide corresponding limitations on the type of sweep signal for online identification of the secondary channel of the active vibration suppression system, they lack reasonable signal and step parameter quantization design operations, and cannot guarantee reliable optimization of the performance of online identification of the secondary channel.

[0133] Preferably, in one embodiment, the method further comprises:

[0134] Step S1-1: before online identification of the secondary channel, setting the identification time of the online identification of the secondary channel in stages for the frequency point to be measured, setting an identification time coefficient for the identification time of different stages, calculating the identification step matched in different stages based on the identification time coefficient, and performing online identification of the secondary channel.

[0135] For each frequency point object, the corresponding identification time is set in stages. For example, for a certain frequency point f Ident , the corresponding identification time is set by the following formula:

[0136] T=3T1+3T2+3T3

[0137] wherein,

[0138] T1=l·1 / f Ident , T2=4T1, T3=4T2

[0139] In the formula, T represents the identification time of the current frequency point, T1, T2, and T3 represent the time period unit of the first time period, the time period unit of the second time period, and the time period unit of the third time period into which the identification time of the current frequency is divided, respectively; and I represents the number of sine signal periods of identification, and the signal unit period identification time length is 1 / f Ident .

[0140] In the embodiment of the application, the identification time of the current frequency is divided into three time periods, and for different time periods, different identification time coefficients are set, and then the identification step length matched with the identification time coefficient is calculated by using the set step length calculation model.

[0141] In one embodiment, the identification time coefficient is set according to the following logic for different stages of identification time:

[0142] For the first time period 0-3T1, the identification time coefficient τ is set to 25 full periods of the signal unit period identification time length corresponding to the current frequency, and in actual application, the identification time coefficient τ is a constant;

[0143] For the second time period 3T1-3T2, the identification time coefficient τ is set to 4×25 full periods of the signal unit period identification time length corresponding to the current frequency;

[0144] For the second time period 3T2-3T3, the identification time coefficient τ is set to 16×25 full periods of the signal unit period identification time length corresponding to the current frequency.

[0145] In an optional embodiment, the step length calculation model described by the following formula is used to calculate the identification step length of the frequency point to be measured:

[0146]

[0147] In the formula, μ OID is the secondary channel online identification step length of the current frequency point, t p is the sampling period of the secondary channel identification result, τ is the corresponding identification time coefficient, and k is a constant.

[0148] Based on the above embodiment, in actual application, different identification time coefficients are set for different time periods according to the following logic, and then the identification step length matched with the identification time coefficient is calculated by using the step length calculation model:

[0149] (1) For the time period 0-3T1, the identification time coefficient τ is set to 25 full periods of the signal unit period identification time length corresponding to the current frequency, and then the matched identification step length is calculated by substituting the corresponding identification time coefficient into the step length calculation model, as follows:

[0150] (2) for the time period 3T1~3T2, the recognition time coefficient τ is set as: 4*25 full periods of the current frequency corresponding to the signal unit period recognition time length, and then the corresponding recognition time coefficient is substituted into the step length operation model to calculate the matched recognition step length, as follows:

[0151] (3) for the time period 3T2~3T3, the recognition time coefficient τ is set as: 16*25 full periods of the current recognition frequency corresponding to the signal unit period recognition time length, and then the corresponding recognition time coefficient is substituted into the step length operation model to calculate the matched recognition step length, as follows:

[0152] Based on the above operation, the present application sets the step length operation factor of the online recognition algorithm based on 3 different stages, each stage is a matched constant value; and at the beginning of recognition, a larger step length is used, which can quickly converge, and then gradually reduces the step length, as shown in Figure 3 Based on this, the secondary channel is identified online to ensure the robustness of the recognition.

[0153] The means in the above embodiments of the present application determine the recognition time and the recognition step length, and the secondary channel is identified online based on the stages, and then the vibration noise active control is realized based on the recognition result.

[0154] In the control process, after completing the online identification of the secondary channel at the frequency point, the algorithm makes a decision through the preset online identification logic of the secondary channel to determine the frequency point of the next online identification of the secondary channel. The present application does not particularly limit the online identification logic of the secondary channel, and any reasonable logic available in the field can be used.

[0155] The online identification of the secondary channel is realized by using the logic of the above embodiments of the present application, which can improve the signal-to-noise ratio of the signal during the online identification of the secondary channel, so as to achieve the purpose of fully exciting the secondary channel, and at the same time, the convergence speed and the robustness of the recognition algorithm are ensured during the online identification of the secondary channel, so as to ensure the accuracy of the online identification of the secondary channel of the active vibration noise control system and improve the control effect.

[0156] Embodiment Four

[0157] The online identification method of the vibration noise active control based on the fusion operation provided in the fourth embodiment of the present application includes the following operations.

[0158] Step S1: using the active vibration noise control system algorithm to obtain an active control signal based on a secondary channel, combining an additional excitation signal as an actuator driving signal, and performing online identification of the secondary channel around a to-be-measured frequency point for a set time based on the additional excitation signal;

[0159] Step S2: denoising and optimizing the online identification result by using a filtering algorithm;

[0160] Step S3: determining a fusion identification result based on the optimized new and old identification results by fusion operation, as a final identification result;

[0161] Step S4: processing the active control signal of the corresponding frequency point based on the determined online identification result of the secondary path, to realize active vibration noise control.

[0162] Considering that if the identification result is poor, the accuracy of the filtered reference signal is still not high, it is necessary to evaluate the identification result, and reapply the better result to the active vibration control system to filter the reference signal and improve the active vibration control effect. Therefore, in this embodiment, after step S1, it further includes step S1-2: quality evaluation of the online identification result of the secondary path, and selection of the identification result meeting the set requirement to realize vibration noise control.

[0163] Figure 4 A flowchart for realizing quality evaluation of the identification result in the vibration active control secondary path online identification method based on fusion operation provided by the embodiment of the application is shown, as shown in Figure 4 The process of quality evaluation of the online identification result of the secondary path in an embodiment includes:

[0164] Obtaining the online identification result and performing filtering processing, and selecting an identification result sample set based on the filtered online identification result;

[0165] According to the identification result sample set, the identification quality is evaluated by using the weight band method.

[0166] Based on this, the application of the online identification result of the secondary path in the active vibration noise control system is further decided in combination with the evaluation result.

[0167] The online identification result of the secondary path is filtered to remove the influence of random interference, and a weight band is obtained to evaluate the identification effect, and the steps are as follows:

[0168] (I) filtering the weight result in the online identification process to make the result smoother;

[0169] In the embodiment of the application, the filtering result of the sine / cosine component weight in the set time before the current time is taken as the mean value of the sine / cosine component weight at the current time, but the processing mode can be other feasible modes, and the application is not limited thereto. The online secondary path identification result filtering mode can be any applicable mode.

[0170] (ii) taking the points in the filtered weight result where all gradients are 0, i.e. local extrema, as the sample set of the weight result for evaluation;

[0171] The points in the weight result where the gradient is 0 are the extrema of the weight in the entire recognition process, i.e. the maximum (extreme) of the weight in a certain time period in the recognition process, and also represent the range of the weight in this time period, and have certain representativeness. However, in fact, a series of weights obtained by using any standard can be taken as samples to obtain the weight band, and the application is not limited thereto, and all statistics related to the sine and cosine component weight in the recognition process can be taken as the evaluation parameter of the weight band.

[0172] (iii) calculating the mean μ0 and variance σ0 of the sample set of the weight result for evaluation and calculating the weight band ratio bandwidth, and determining the evaluation result according to the weight band ratio bandwidth and the set threshold value.

[0173] In the embodiment of the application, the weight band refers to the range of the weight in a certain time period obtained by a certain method, i.e. it is considered that the weight obtained in a certain time period is always within this range. The weight band can be obtained in the following ways:

[0174] (1) directly subtracting the minimum weight from the maximum weight in a certain time period to obtain the weight band bandwidth.

[0175] (2) obtaining the weight samples in a certain time period by some method, calculating the mean and variance, and obtaining the equivalent maximum and equivalent minimum from the mean and variance, and obtaining the weight band (equivalent) bandwidth from the difference between the equivalent minimum values, and considering that the weight in this time period is within this range.

[0176] In addition to the above two ways of obtaining the weight band, in fact, all statistics related to the sine and cosine component weight in the recognition process and the operation results thereof can be taken as the value range of the weight obtained in a certain time period, i.e. the weight band.

[0177] In the preferred embodiment, the process of using the weight band to evaluate the recognition quality according to the sample set of the recognition result includes:

[0178] calculating the mean and variance of the sine weight data and the cosine weight data based on the sample set of the recognition result at different times, and calculating the equivalent ratio bandwidth of the weight band according to the mean and variance;

[0179] obtaining the equivalent ratio bandwidth of the weight band from the difference between the equivalent minimum values, comparing the equivalent ratio bandwidth of the weight band with the set bandwidth threshold, and determining the evaluation result of the sine weight and the cosine weight.

[0180] In the optional embodiment, the equivalent ratio bandwidth of the weight band is calculated according to the following formula:

[0181] Θ = 6σ0 / μ0

[0182] Θ = 6σ0 / μ0

[0183] In the formula, Θ is the equivalent weight band width, μ0 is the mean of the weight result sample set, and σ0 is the standard deviation of the weight result sample set.

[0184] The greater the weight band ratio bandwidth at the end of identification, the lower the quality of the identification process; the smaller the weight band ratio bandwidth at the end of identification, the higher the quality of the identification process, and when the weight band ratio bandwidth is greater than a certain critical value, the corresponding identification process should be considered to be of poor quality, and the current online identification result should be discarded.

[0185] Regarding the critical value, the variance divided by the mean is the coefficient of variation in statistics, and the weight band ratio bandwidth in the present application is 6 times the coefficient of variation. Generally, when the coefficient of variation is less than 0.2, it can be considered that the variation degree of the data is small, and therefore, if the specific distribution of the data sample is unknown, the critical value can be set to 1.2. However, this critical value can also be set by empirical values or specific conditions of the data sample.

[0186] The embodiment of the present application evaluates the quality of the identification result by calculating the weight band ratio bandwidth corresponding to the secondary channel identification result sample, discards the secondary channel identification result that does not meet the set quality requirement, filters the control signal based on the secondary channel identification result that meets the set quality requirement, selects the online identification result, and uses the statistical parameters to select the identification result with good quality for application, thereby participating in the operation of the active control system, avoiding the online identification result with poor quality causing the instability of the active control system, and effectively improving the stability and accuracy of the vibration noise active control.

[0187] Embodiment five:

[0188] The online identification method for vibration noise active control based on fusion operation provided by the embodiment of the present application includes the following operations.

[0189] Step S1: obtaining an active control signal based on a secondary channel using an active vibration noise control system algorithm, combining an additional excitation signal as an actuator driving signal, and performing online identification of the secondary channel around a frequency point to be measured for a set time based on the additional excitation signal;

[0190] Step S2: denoising and optimizing the online identification result using a filtering algorithm;

[0191] Step S3: determining a fusion identification result based on the optimized new and old identification results through fusion operation, as the final identification result;

[0192] Step S4: processing the active control signal of the corresponding frequency point based on the determined secondary path online identification result, to realize active vibration noise control.

[0193] After step S1, step S1-2 is further included: quality evaluation is performed on the secondary path online identification result, and the identification result meeting the set requirement in quality is selected to realize vibration noise control.

[0194] In one embodiment, the process of quality evaluation on the secondary path online identification result includes:

[0195] The online identification result is obtained and filtered, and the identification result sample set is selected based on the filtered online identification result.

[0196] The identification quality is evaluated by using the weight band method according to the identification result sample set. Embodiment five is a variant of the foregoing embodiments, and the same or corresponding technical features in the above embodiments will not be described again. Hereinafter, only the different technical features of the present embodiment will be described.

[0197] Figure 5 is a flowchart of the vibration noise active control online identification method based on fusion operation provided by another embodiment of the present application; in this embodiment, the process of quality evaluation on the secondary path online identification result in step S1-2 further includes:

[0198] After the identification quality is evaluated by using the weight band method according to the identification result sample set, if the result is passed, the identification result is further converted into the amplitude-phase form, and then the converted identification result is evaluated in two levels, and the final evaluation result is determined by combining the obtained two-level evaluation parameters.

[0199] The amplitude attenuation coefficient and the phase lag angle range of the secondary path online identification result are evaluated by the two-level evaluation process in the embodiment of the present application. In an optional embodiment, the process of two-level evaluation on the converted identification result includes:

[0200] The sine and cosine weight extreme values of the secondary path online identification result are selected, the corresponding amplitude attenuation coefficient extreme value and phase lag angle extreme value are calculated based on the sine weight extreme value and the cosine weight extreme value, and then the amplitude attenuation coefficient evaluation parameter and the phase lag angle evaluation parameter are calculated.

[0201] In the process of evaluation by using the weight band method in the above embodiment, the mean μ ws and the variance of the sine component weight are calculated respectively. wc and the variance of the cosine component weight are calculated respectively. Based on this, the equivalent maximum value w s_max of the sine component weight, the equivalent minimum value w s_minand the equivalent maximum value of the cosine component weight w c_max , the equivalent minimum value of the cosine component weight w c_min There are four values, and the weight value range is obtained, and the equivalent maximum and minimum value calculation formula of the weight is:

[0202] w s_max = μ ws + 3σ ws

[0203] w s_min = μ ws - 3σ ws

[0204] w c_max = μ wc + 3σ wc

[0205] w c_min = μ wc - 3σ wc

[0206] In actual application, the maximum value and the minimum value of the amplitude attenuation coefficient or the phase lag angle can be obtained according to the weight value range obtained by calculation, and specifically, the equivalent maximum value of the sine component weight w s_max , the equivalent minimum value of the sine component weight w s_min can be taken as w s,j (k), the equivalent maximum value of the cosine component weight w c_max , the equivalent minimum value of the cosine component weight w c_min can be taken as w c,j (k), and the corresponding equivalent maximum value and the equivalent minimum value of the amplitude attenuation coefficient and the equivalent maximum value and the equivalent minimum value of the phase lag angle are calculated by substituting the following formula:

[0207]

[0208] In the formula, Amp j (k) is the amplitude attenuation coefficient in the secondary channel identification result of the jth frequency point at the kth moment, is the phase lag angle in the secondary channel identification result of the jth frequency point at the kth moment, w s,j (k) represents the sine component weight of the secondary channel identification result of the jth frequency point at the kth moment, and w c,j (k) represents the cosine component weight of the secondary channel identification result of the jth frequency point at the kth moment. Figure 6 The equivalent extreme value range distribution diagram of the sine and cosine component weights of the online identification method for active control of vibration noise based on fusion operation provided by the embodiment of the application is shown.

[0209] Preferably, in one embodiment, the amplitude attenuation coefficient evaluation parameter and the phase lag angle evaluation parameter are calculated as follows:

[0210] A = |Amp| min / |Amp| max

[0211]

[0212] wherein A is the amplitude attenuation coefficient evaluation parameter, B is the phase lag angle evaluation parameter, |Amp| min represents the amplitude attenuation coefficient minimum value, represents the phase lag angle maximum value.

[0213] The amplitude attenuation coefficient evaluation parameter and the phase lag angle evaluation parameter are respectively analyzed and determined according to the set evaluation rules to obtain corresponding evaluation results; wherein the amplitude attenuation coefficient evaluation parameter A is compared and analyzed with the set amplitude attenuation coefficient evaluation threshold, the phase lag angle evaluation parameter B is compared and analyzed with the set phase lag angle evaluation threshold, the evaluation parameter A is greater than the threshold to determine that the evaluation pass condition is met, and the evaluation parameter B is less than the threshold to determine that the evaluation pass condition is met.

[0214] The threshold is set according to the selected evaluation parameter, and in the embodiment of the present application, the evaluation parameters adopted are the amplitude attenuation coefficient extreme value ratio (minimum value ratio maximum value) and the phase lag angle angle difference. In actual engineering, the acceptable amplitude attenuation coefficient extreme value ratio threshold and the phase lag angle angle difference threshold can be obtained through a large number of tests, and this range is related to the acceptable degree of vibration control effect; for example, in the embodiment of the present application, the amplitude attenuation coefficient extreme value ratio threshold can be set to 0.9 and the phase lag angle angle difference threshold can be set to 10° by the technical personnel according to experience and knowledge.

[0215] In actual application, the value of the phase lag angle evaluation parameter B should be not greater than π, and if the calculated parameter B is greater than π, the transformation formula can be used for transformation to meet the parameter B range requirement. Alternatively, the value of B can be reduced by an integer multiple of 2π until the parameter B is not greater than π.

[0216] In this embodiment, if the evaluation result of the weight band method and the secondary evaluation result of the amplitude attenuation coefficient or the phase lag angle are both passed, the mean value of the current weight result sample set is taken as the online identification result of the current frequency point, and the subsequent step S3 is performed for active vibration noise control, otherwise, the online identification result is discarded.

[0217] The secondary channel online identification method with identification quality evaluation provided in the scheme of the present application is applied to an automobile active suspension system (an active vibration control system) as an embodiment of the scheme of the present application: in the embodiment, a trap wave FxLMS algorithm is used to obtain an actuator driving voltage signal to reduce the vibration caused by the excitation of a vehicle power source; meanwhile, a sweep frequency sinusoidal excitation signal far away from the current engine excitation frequency is added to perform online identification on the secondary channels at some frequency points.

[0218] The present application is further described below in combination with an embodiment. The scope of the present application is not limited by the embodiment, and the scope of the present application is proposed in the claims.

[0219] The secondary channel online identification method with identification quality evaluation in the scheme of the present application is applied to an automobile active suspension system (an active vibration control system) as an embodiment of the scheme of the present application: in the embodiment, a trap wave FxLMS algorithm is used to obtain an actuator driving voltage signal to reduce the vibration caused by the excitation of a vehicle power source; meanwhile, a sweep frequency sinusoidal excitation signal far away from the current engine excitation frequency is added to perform online identification on the secondary channels at some frequency points.

[0220] In the embodiment, the secondary channel changes greatly in the range of 100-120 Hz, leading to the deterioration of the vibration reduction effect, so it is determined that the secondary channel in the range of 100-120 Hz needs to be identified online. When the engine speed is at idle speed (about 750 rpm), the secondary channel in the range of 100-120 Hz is identified using the online secondary channel identification method with identification quality evaluation provided in the scheme of the present application while the engine is actively controlled, and the change of the engine speed during the whole process is as shown in Figure 7 , and the time domain change of the error sensor signal is as shown in Figure 8 . In the online identification process, the sweep step is 2.5 Hz in the range of 100-120 Hz, that is, the change of the secondary channel identification results (phase lag angle and amplitude attenuation coefficient) of a total of 9 frequency points is as shown in Figure 9 (a) and Figure 9 (b).

[0221] After the online identification of the secondary channel in the range of 100-120 Hz is completed, the engine speed is increased to the corresponding speed range, and the active vibration control effect of the frequency range is improved; the engine speed and the vibration frequency have a certain relationship, and for the second-order vibration suppression in the embodiment of the present application, the speed range corresponding to "100-120 Hz" is "3000-3600 rpm" optionally.

[0222] The frequency domain comparison diagram of the vibration in the frequency range before and after the online identification is as shown in Figure 10 , which proves the effectiveness of the secondary channel identification method in the scheme.

[0223] For the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited by the order of the described actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily necessary for the present application.

[0224] It should be noted that in other embodiments of the present application, the method can also be obtained by combining one or several of the above embodiments to obtain a new secondary channel online identification method, so as to realize high-quality optimization of the active vibration control effect.

[0225] It should be noted that based on the method in any one or more of the above embodiments of the present application, the present application also provides a storage medium having a program code for realizing the method in any one or more of the above embodiments stored thereon, and the code can realize the vibration noise active control online identification method based on fusion operation when executed by an operating system.

[0226] Embodiment six:

[0227] The above embodiments of the present application disclose the method in detail, and the method of the present application can be realized in various forms of devices or systems, so based on other aspects of the method in any one or more of the above embodiments, the present application also provides a vibration noise active control online identification system based on fusion operation, which is used to execute the vibration noise active control online identification method based on fusion operation in any one or more of the above embodiments. The following specific embodiments are given for detailed description.

[0228] Specifically, Figure 11 The structure diagram of the vibration noise active control online identification system based on fusion operation provided in the embodiments of the present application is shown in FIG. 1, as shown in FIG. 1, the system comprises: Figure 11

[0229] The memory is used to store computer programs and secondary channel models;

[0230] The processor is used to realize the steps of the vibration noise active control online identification method based on fusion operation as disclosed above when executing the computer program.

[0231] The processor comprises:

[0232] ​a vibration control module configured to obtain an active control signal based on the secondary channel using an active vibration and noise control system algorithm, combine the additional excitation signal as an actuator drive signal, and perform online identification of the secondary channel at a set time around a frequency point to be measured based on the additional excitation signal;

[0233] A correction optimization module configured to perform denoising optimization on the online identification results using a filtering algorithm;

[0234] a fusion operation module configured to determine a fused identification result based on the optimized new and old identification results through a fusion operation as a final identification result;

[0235] The update execution module is configured to process the active control signal of the corresponding frequency point based on the determined secondary channel online identification result to achieve active vibration noise control.

[0236] Furthermore, in one embodiment, the correction optimization module performs denoising optimization on the online identification results using a filtering algorithm according to the following operations:

[0237] Based on the secondary channel online identification result at the previous moment, the state matrix of the secondary channel online identification result is used for prior prediction to obtain the prior estimation value of the secondary channel identification result at the current moment of the corresponding frequency point;

[0238] The prior estimate of the system state error covariance matrix corresponding to the current frequency point is predicted, and then the Kalman gain is calculated by combining the observation matrix of the corresponding frequency point and the measurement noise covariance matrix of the secondary channel online identification;

[0239] The identification result obtained by this identification is filtered using the Kalman gain in combination with the prior estimation value of the secondary channel identification result to obtain a correction value of the identification result as the optimized secondary channel online identification result.

[0240] Optionally, in one embodiment, the fusion operation module determines the fusion identification result based on the optimized new and old identification results by using the fusion algorithm shown in the following formula:

[0241]

[0242] Among them, w ja (k) is the target secondary channel identification result of the jth frequency point at time k actually used in the active control process, M f is the fusion factor of the new and old secondary channels, is the correction value of the secondary channel online identification result at the jth frequency point at time k.

[0243] Specifically, in an optional embodiment, the correction optimization module is configured to calculate the Kalman gain according to the following formula:

[0244]

[0245] Among them, R j (k) is the measurement noise covariance matrix of the secondary channel online identification at the jth frequency point at time k, K j (k) is the Kalman gain of the secondary channel online identification process at the jth frequency point at time k, H j (k) is the observation matrix of the jth frequency point at time k, is the prior estimate of the system state quantity error covariance matrix at the jth frequency point at time k, and T represents the transposed matrix.

[0246] Furthermore, in one embodiment, the correction optimization module is configured to calculate the identification result correction value according to the following formula:

[0247]

[0248] in, is the correction value of the secondary channel online identification result at the jth frequency point at time k, w j (k) is the online identification result of the secondary channel at the jth frequency point at time k, K j (k) is the Kalman gain of the secondary channel online identification process at the jth frequency point at time k, H j (k) is the observation matrix of the jth frequency point at time k, is the prior estimate of the secondary channel identification result at the jth frequency point at time k.

[0249] Preferably, in one embodiment, the system further includes a result amplification module, which is configured to analyze the secondary channel online identification results corresponding to the frequency points to be predicted using an interpolation algorithm based on the fusion identification results, wherein the frequency points to be predicted include frequency points for which secondary channel online identification has not been performed but which meet identification requirements.

[0250] Optionally, in one embodiment, the system further includes a variable step size configuration module, which is configured to set the identification time of the secondary channel online identification in stages for the frequency point to be measured before performing online identification on the secondary channel, set identification time coefficients for the identification time of different stages, and calculate the identification step sizes matched in different stages based on the identification time coefficients for performing online identification on the secondary channel.

[0251] Further, in one embodiment, the system further comprises a quality evaluation module configured to: after online identification of the secondary channel, acquire the online identification result and perform filtering processing, select an identification result sample set based on the filtered online identification result; perform identification quality evaluation using a weight band method according to the identification result sample set, select an identification result meeting a set requirement, and further start a correction and optimization module to perform subsequent operations.

[0252] Specifically, in one preferred embodiment, the quality evaluation module is further configured to:

[0253] After performing identification quality evaluation using a weight band method according to the identification result sample set, if the result is passed, the method further comprises: transforming the identification result into a magnitude-phase form, performing secondary evaluation on the transformed identification result; selecting sine and cosine weight extreme values of the online identification result of the secondary channel, calculating corresponding magnitude attenuation coefficient extreme values and phase lag angle extreme values based on the sine weight extreme values and the cosine weight extreme values, respectively, and then calculating secondary evaluation parameters determined by combining the magnitude attenuation coefficient evaluation parameters and the phase lag angle evaluation parameters to obtain a final evaluation result, and selecting an identification result meeting the evaluation requirement.

[0254] Based on the online identification system of vibration noise active control based on fusion operation according to the above embodiments, the present application further provides a vibration noise active control system, which comprises the online identification system of vibration noise active control based on fusion operation according to the above embodiments.

[0255] In the online identification system of vibration noise active control based on fusion operation according to the embodiments of the present application, each module or unit structure can be independently operated or combined to operate according to actual data processing and operation decision requirements, so as to realize corresponding technical effects.

[0256] It should be understood that the disclosed embodiments of the present application are not limited to the specific structures, processing steps or materials disclosed herein, but should extend to equivalent alternatives of these features understood by those skilled in the relevant art. It should also be understood that the terms used herein are only for the purpose of describing specific embodiments and do not mean limitation.

[0257] The phrase "one embodiment" mentioned in the specification means that the specific features, structures or characteristics described in connection with the embodiment are included in at least one embodiment of the present application. Therefore, the phrase "one embodiment" appearing throughout the specification does not necessarily mean the same embodiment.

[0258] Although the embodiments of the present application have been disclosed with reference to the above embodiments, the above description is merely used to understand the present application and is not used to limit the present application. Any person skilled in the art, without departing from the spirit and scope of the present application, can make any modification and change in the form and details of the embodiments, but the patent protection scope of the present application should be subject to the scope defined by the appended claims.

Claims

1. A vibration noise active control online identification method based on fusion operation, characterized in that: The method comprises: Step S1: using an active vibration noise control system algorithm to obtain an active control signal based on the secondary channel, combining it with an additional excitation signal as an actuator drive signal, and performing online identification of the secondary channel for a set time around a frequency point to be measured based on the additional excitation signal; Step S2: denoising and optimizing the online recognition results using a filtering algorithm; Step S3: Determine a fused identification result based on the optimized new and old identification results through a fusion operation as the final identification result; Step S4: processing the active control signal of the corresponding frequency point based on the determined secondary channel online identification result to achieve active vibration noise control; In step S3, the fusion identification result is determined based on the optimized new and old identification results using the fusion algorithm shown in the following formula: in, It is actually used in the active control process k Moment j The target secondary channel identification results of frequency points, is the fusion factor of the new and old secondary channels, for k Moment j Correction value of the secondary channel online identification result at each frequency point.

2. The method according to claim 1, characterized in that In step S2, the process of performing denoising optimization on the online recognition result using a filtering algorithm includes: Based on the secondary channel online identification result at the previous moment, the state matrix of the secondary channel online identification result is used for prior prediction to obtain the prior estimation value of the secondary channel identification result at the current moment of the corresponding frequency point; The prior estimate of the system state error covariance matrix corresponding to the current frequency point is predicted, and then the Kalman gain is calculated by combining the observation matrix of the corresponding frequency point and the measurement noise covariance matrix of the secondary channel online identification; The identification result obtained by this identification is filtered using the Kalman gain in combination with the prior estimation value of the secondary channel identification result to obtain a correction value of the identification result as the optimized secondary channel online identification result.

3. The method according to claim 2, characterized in that In step S2, the Kalman gain is calculated according to the following formula: ； in, for k Moment j The measurement noise covariance matrix of the secondary channel online identification at frequency points is for k Moment j The Kalman gain of the secondary channel online identification process at each frequency point, for k Moment j The observation matrix of frequency points, for k Moment j The prior estimate of the system state quantity error covariance matrix at each frequency point, T represents the transposed matrix.

4. The method according to claim 2, characterized in that In step S2, the identification result correction value is calculated according to the following formula: in, is the correction value of the secondary channel online identification result at the jth frequency point at time k, is the online identification result of the secondary channel at the jth frequency point at time k, is the Kalman gain of the secondary channel online identification process at the jth frequency point at time k, is the observation matrix of the jth frequency point at time k, is the prior estimate of the secondary channel identification result at the jth frequency point at time k.

5. The method according to claim 1, wherein The method further includes step S3-1: after step S3, using an interpolation algorithm to analyze the secondary channel online identification results corresponding to the frequency points to be predicted based on the fusion identification results, the frequency points to be predicted include frequency points for which secondary channel online identification has not been performed but which meet identification requirements.

6. The method according to claim 1, characterized in that The method further includes step S1-1: before performing online identification on the secondary channel, setting the identification time of the secondary channel online identification in stages for the frequency point to be measured, setting identification time coefficients for the identification time in different stages, and calculating the identification step lengths for matching in different stages based on the identification time coefficients for performing online identification on the secondary channel.

7. The method according to claim 1, characterized in that The method further includes steps 1-2: after step S1, obtaining online recognition results and performing filtering processing, and selecting a recognition result sample set based on the filtered online recognition results; performing recognition quality assessment using a weight band method based on the recognition result sample set, selecting recognition results whose quality meets the set requirements, and further executing step S2.

8. The method according to claim 7, characterized in that Step S1-2 further includes: After the identification quality is evaluated using the weight band method based on the identification result sample set, if the result is passed, the identification result is also converted into amplitude-phase form, and the transformed identification result is evaluated at a secondary level; the sine and cosine weight extremes of the secondary channel online identification result are selected to determine the corresponding amplitude attenuation coefficient extremes and phase lag angle extremes, and then the amplitude attenuation coefficient evaluation parameters and phase lag angle evaluation parameters are calculated. Based on this comprehensive evaluation, the identification result that meets the evaluation requirements is selected.

9. A storage medium, characterized in that: The storage medium stores program code that can implement the method according to any one of claims 1 to 8.

10. A vibration noise active control online identification system based on fusion operation, characterized in that: The system comprises: a memory for storing a computer program and a secondary channel model; A processor, configured to implement the steps of the online identification method for active vibration noise control based on fusion operation as described in any one of claims 1 to 8 when executing a computer program.

11. A vibration noise active control system, characterized in that: The system includes the vibration noise active control online identification system based on fusion operation as described in claim 10.

Images