A method, device, equipment and storage medium for adaptive filtering of environmental noise

Abstract

The present invention discloses a method, apparatus, device, and storage medium for adaptive filtering of environmental noise, which can realize the rapid deployment and application of equipment vibration monitoring systems. Since the entire filtering process is completely free of human intervention, the system using this method does not require manual acquisition of prior information before deployment. After deployment, it can basically iterate to a relatively ideal effect in a very short time, which can effectively improve the deployment speed of the system. In addition, the use of this method is conducive to the large-scale deployment and application of equipment vibration monitoring systems. Due to the adaptive generation of environmental noise and the adaptive iteration mechanism of the equipment working condition identification threshold, the method can quickly adapt to the simultaneous deployment of multiple types of equipment, and thus can support large-scale deployment and application.

Description

Technical Field

[0001] The present invention relates to the field of mechanical processing technology, and in particular to an environmental noise adaptive filtering method, device, equipment and storage medium. Background Art

[0002] In the mechanical processing industry, vibration sensors are usually used to collect vibration data of key equipment components, and then a series of signal processing and data analysis methods are used to detect the status of key equipment components. In this process, the purification and processing of vibration signals is particularly important.

[0003] Existing technologies typically use spectrum synthesis and signal restoration techniques or adaptive filter design to purify vibration signals. Spectrum synthesis and signal restoration techniques (see Patent Document 1: Method, Terminal, and Storage Medium for Training a Spectrum Synthesis Model and Synthesizing Audio; Patent Document 2: Speech Synthesis System, Method, Computer Device, Storage Medium, and Program Product) are currently widely used in the speech or communications field, with typical examples including speech synthesis and signal modulation and demodulation. Taking Patent Documents 1 and 2 as examples, speech synthesis techniques typically identify the effective frequency characteristics of a signal based on text features or existing samples. Then, they extract a portion of these effective frequency characteristics from multiple signal groups and combine them to form a target spectrum. Signal restoration is then used to restore the target signal.

[0004] Signal modulation and demodulation technology is more direct. Usually, the frequency of the noise signal in the modulated signal is a known natural frequency. The target spectrum can be formed by directly screening out the natural frequency points of the signal spectrum, and then the target signal can be restored through inverse Fourier transform.

[0005] Whether in speech synthesis or communications, spectrum superposition and signal restoration share a common feature: the ability to obtain information about the frequency characteristics of background noise in advance. However, this is difficult to achieve in vibration signal recognition in complex scenarios, as shown below:

[0006] (1) The noise that appears in each time period in a complex environment is not fixed, and new noise may appear at any time. It is impossible to know the noise frequency characteristics in advance through prior information;

[0007] (2) Due to the complexity of the production environment, even the data that is initially judged to be noise often contains irregular and occasional noise data. If this part of the misjudged data is directly applied to the adaptive filter, it will directly lead to the loss of effective signals.

[0008] Adaptive filter design (see Patent Document 3: Song melody extraction method, song processing method, computer device, and product; Patent Document 4: An adaptive pulse signal detection and extraction method). Currently, the adaptive filters used in various fields are essentially filters that target the signal itself. During execution, the filters utilize an adaptive principle to process the target signal. Innovation often revolves around this adaptive process.

[0009] Taking Patent Document 3 as an example, Patent Document 3 uses adaptive filters twice in the process of human voice signal extraction and dry sound signal extraction. Both times, the target signal is extracted by inputting the original signal and the noise signal into the adaptive filter. Its essence is to use the difference between the two types of original signals and the background signal to extract the target signal.

[0010] Because equipment in the machining industry is often densely packed and close together, vibrations often interfere with each other. This effect is called composite environmental noise. Composite environmental noise has diverse noise types and a high degree of uncertainty regarding its timing. In the filtering process of vibration composite environmental noise, filtering the raw data solely using noise data is not entirely feasible. The main problem lies in the inability to fully identify noise data. Whether using equipment assistance or manual judgment, it is impossible to fully identify all noise data, leading to errors in noise data identification or too small a sample size. Filtering the raw data may produce erroneous results. Therefore, if this traditional adaptive filter principle is used, no matter how advanced the adaptive algorithm within the filter is, errors in the data source will inevitably lead to erroneous filtering results due to incomplete or erroneous noise data.

[0011] It can be seen that in the traditional manual analysis process, it is difficult to effectively distinguish such noise interference and fault data. Summary of the Invention

[0012] In view of the above problems, the present invention provides an environmental noise adaptive filtering method, apparatus, device, and storage medium for overcoming or at least partially resolving the above problems. Because each device is located in a different environment, using a uniform environmental noise filter to filter all devices can easily lead to the negative consequence of losing the real signal. Therefore, the embodiments of the present application use a composite environmental noise generation and iteration method with environmental adaptability. This method can effectively distinguish environmental noise interference from fault data without the need for traditional manual analysis.

[0013] The present invention provides the following solutions:

[0014] An environmental noise adaptive filtering method, comprising:

[0015] After determining that the device is in an unprocessed state, a noise signal in the unprocessed state collected by a vibration sensor is received;

[0016] Acquiring characteristic values ​​of the noise signal, performing cluster analysis on the noise signal according to the characteristic values, and classifying the noise signal to obtain a multi-class noise signal set;

[0017] performing spectrum calculation on noise data of multiple types of noise signal sets respectively to obtain multiple spectrums, and superimposing all the spectrums to generate a composite noise spectrum;

[0018] Performing an inverse Fourier transform on the composite noise spectrum to generate composite environmental noise time domain data;

[0019] After determining that the device is in a processing state, receiving a basic signal in the processing state collected by a vibration sensor;

[0020] The composite ambient noise time domain data is used as the input of an adaptive filter based on target spectrum elimination, the basic signal is used as the expected signal of the adaptive filter; and the difference between the expected signal and the output signal of the adaptive filter is used as an estimate of the signal source.

[0021] Preferably, whether the device is in the unprocessed state or the processed state is determined based on the relationship between the current and voltage signals of the device collected by the current and voltage monitoring sensor and the preset target current and voltage thresholds.

[0022] Preferably, the target current and voltage thresholds are adjusted by receiving an adjustment method input by a user to obtain an adjusted current and voltage threshold; the adjustment method is determined by the user performing a secondary identification of the noise data based on the variance change rate of the data before and after filtering;

[0023] It is determined whether the device is in an unprocessed state or a processed state according to the relationship between the current and voltage signals of the device collected by the current and voltage monitoring sensor and the adjusted current and voltage thresholds.

[0024] Preferably, the characteristic values ​​include mean, kurtosis, and main frequency characteristics.

[0025] Preferably: performing random sampling on each type of noise signal set, extracting 20% ​​of the data of each type of noise signal set to obtain multiple noise sample data sets;

[0026] Performing spectrum calculation on the noise data of the plurality of noise sample data sets to obtain a plurality of spectrums, and superimposing all the spectrums to generate a composite noise spectrum;

[0027] All frequency points included in the multiple spectra are compared, and a maximum value is assigned according to the noise frequency amplitude, so as to achieve superposition of all the spectra to generate a composite noise spectrum.

[0028] Preferably: the composite environmental noise time domain data is used to perform targeted filtering operations on all the noise sample data sets, the noise sample data sets are classified according to the filtering results, the noise sample data are divided into recognizable noise data and unrecognizable noise data, and the validity of the composite environmental noise time domain data is determined according to the recognition rate of the noise sample data.

[0029] Preferably, when the recognition rate is lower than 95% or a new type of noise signal is determined to exist, the composite environmental noise time domain data is determined to be invalid.

[0030] An environmental noise adaptive filtering device, comprising:

[0031] an unprocessed state determining unit, configured to determine that the device is in an unprocessed state and then receive a noise signal in the unprocessed state collected by a vibration sensor;

[0032] a characteristic value acquisition unit, configured to acquire characteristic values ​​of the noise signal, and perform cluster analysis on the noise signal according to the characteristic values ​​to classify the noise signal to obtain a multi-class noise signal set;

[0033] a composite noise spectrum generating unit, configured to perform spectrum calculation on noise data of multiple types of noise signal sets respectively to obtain multiple spectra, and to superimpose all the spectra to generate a composite noise spectrum;

[0034] A time domain data conversion unit, configured to perform an inverse Fourier transform on the composite noise spectrum to generate composite environmental noise time domain data;

[0035] a processing state determining unit, configured to determine that the device is in a processing state and receive a basic signal in the processing state collected by the vibration sensor;

[0036] A signal source estimation and acquisition unit is used to use the composite ambient noise time domain data as the input of an adaptive filter based on target spectrum elimination, use the basic signal as the expected signal of the adaptive filter; and use the difference between the expected signal and the output signal of the adaptive filter as an estimate of the signal source.

[0037] An environmental noise adaptive filtering device, the device comprising a processor and a memory:

[0038] The memory is used to store program code and transmit the program code to the processor;

[0039] The processor is configured to execute the above-mentioned environmental noise adaptive filtering method according to instructions in the program code.

[0040] A computer-readable storage medium is used to store program codes, and the program codes are used to execute the above-mentioned environmental noise adaptive filtering method.

[0041] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0042] The embodiments of the present application provide an environmental noise adaptive filtering method, apparatus, equipment, and storage medium that can realize the rapid deployment and application of equipment vibration monitoring systems. Since the entire process requires no human intervention at all, there is no need to manually obtain prior information before deploying a system using this method. After deployment, it can basically iterate to a relatively ideal effect in a very short time, which can effectively improve the deployment speed of the system.

[0043] In addition, the use of this method is conducive to the large-scale deployment and application of equipment vibration monitoring systems. Due to the adaptive generation of environmental noise and the adaptive iteration mechanism of equipment operating condition identification thresholds, this method can quickly adapt to the simultaneous deployment of various types of equipment, and therefore can support large-scale deployment and application.

[0044] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be derived from these drawings without inventive effort.

[0046] Figure 1 This is a flow chart of an environmental noise adaptive filtering method provided by an embodiment of the present invention;

[0047] Figure 2 This is a flowchart of a method according to an implementation mode provided by an embodiment of the present invention;

[0048] Figure 3 is a schematic diagram of an adaptive filter provided by an embodiment of the present invention;

[0049] Figure 4 is a schematic diagram of an environmental noise adaptive filtering device provided by an embodiment of the present invention;

[0050] Figure 5 It is a structural diagram of an environmental noise adaptive filtering device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of the present invention.

[0052] See also Figure 1 , is an environmental noise adaptive filtering method provided by an embodiment of the present invention, such as Figure 1 As shown, the method may include:

[0053] S101: After determining that the device is in an unprocessed state, receiving a noise signal in the unprocessed state collected by a vibration sensor;

[0054] S102: Acquire characteristic values ​​of the noise signal, perform cluster analysis on the noise signal according to the characteristic values, classify the noise signal to obtain multiple types of noise signal sets; specifically, the characteristic values ​​include mean, kurtosis, and main frequency characteristics.

[0055] S103: Spectral calculation is performed on the noise data of the multiple types of noise signal sets to obtain multiple spectra, and all the spectra are superimposed to generate a composite noise spectrum. Since the amount of noise signal data directly obtained is large, in order to reduce the amount of data used for calculation, the embodiment of the present application can also provide random sampling for each type of noise signal set, and 20% of the data of each type of noise signal set is extracted to obtain multiple noise sample data sets.

[0056] Performing spectrum calculation on the noise data of the plurality of noise sample data sets to obtain a plurality of spectrums, and superimposing all the spectrums to generate a composite noise spectrum;

[0057] All frequency points included in the multiple spectra are compared, and a maximum value is assigned according to the noise frequency amplitude, so as to achieve superposition of all the spectra to generate a composite noise spectrum.

[0058] In order to verify whether the composite environmental noise time domain data determined by sampling is accurate, the embodiment of the present application can also provide a method of using the composite environmental noise time domain data to perform targeted filtering operations on all the noise sample data sets, classify the noise sample data sets according to the filtering results, divide the noise sample data into recognizable noise data and unrecognizable noise data, and determine the validity of the composite environmental noise time domain data based on the recognition rate of the noise sample data. Specifically, when the recognition rate is lower than 95% or it is determined that there is a new category of the noise signal, the composite environmental noise time domain data is determined to be invalid. After determining that it is invalid, repeat the above steps to re-determine the composite environmental noise time domain data. The specific operation method will be introduced in detail later.

[0059] S104: performing an inverse Fourier transform on the composite noise spectrum to generate composite environmental noise time domain data;

[0060] S105: After determining that the device is in a processing state, receiving a basic signal in the processing state collected by a vibration sensor;

[0061] S106: Using the composite ambient noise time domain data as input to an adaptive filter based on target spectrum elimination, using the basic signal as a desired signal of the adaptive filter; and using the difference between the desired signal and the output signal of the adaptive filter as an estimate of the signal source.

[0062] Since the environment in which each device is located is different, if a unified environmental noise is used to filter all devices, it is easy to have the negative consequence of losing the real signal. Therefore, the environmental noise adaptive filtering method provided in the embodiment of the present application has the ability to generate and iterate composite environmental noise with environmental adaptability.

[0063] This method can achieve the classification of valid data and the removal of environmental noise through the adaptive iteration of the equipment working condition recognition threshold and the adaptive superposition of the composite environmental noise, and then use the iterated environmental noise to perform adaptive filtering and classification on the normal processing data, ultimately realizing the classification of valid data and the removal of environmental noise. The whole process contains a triple adaptive design principle.

[0064] In order to ensure that the method provided in the embodiment of the present application can make an accurate judgment on whether the device is in a processing state, the embodiment of the present application can provide a method for determining whether the device is in an unprocessed state or a processing state based on the relationship between the current and voltage signals of the device collected by the current and voltage monitoring sensor and the preset target current and voltage thresholds.

[0065] Because the noise signal previously obtained in its raw state may be inaccurate after the device is first deployed or the surrounding environment changes, it is necessary to adjust the current and voltage thresholds based on environmental changes to ensure the accuracy of the noise signal obtained. Specifically, the adjustment method for the target current and voltage thresholds is received from the user and adjusted to obtain the adjusted current and voltage thresholds; the adjustment method is determined by the user performing a secondary identification of the noise data based on the variance change rate of the data before and after filtering;

[0066] The relationship between the current and voltage signals of the device collected by the current and voltage monitoring sensor and the adjusted current and voltage thresholds determines whether the device is in an unprocessed state or a processed state. The user's secondary identification of noise data and the current and voltage threshold adjustment method will be described in detail later.

[0067] This method uses external electrical signal sensors to collect signal data such as current and voltage during device operation. The threshold's automatic iteration capability is achieved through the relationship between the threshold and the signal-to-noise ratio. This method does not require excessive attention to the initial threshold value setting; in the initial stage, a conservative setting with a smaller threshold can be used. This setting will sacrifice a large amount of valid data in the early stages, but it will ensure that only unprocessed data is identified, forming a more accurate basic data set for subsequent iterations. Compared to other methods that use similar methods to determine the power-on and power-off status of devices, this method minimizes the influence of traditional empirical cognition on the judgment results, while also ensuring the quantity and quality of valid data after a certain period of iteration.

[0068] This method also generates composite noise by superimposing noise spectra. This method effectively integrates various unknown environmental noise characteristics, ensuring the characteristic integrity of the resulting composite noise. Conventional methods often rely on manually identifying noise data and then superimposing it. While manual methods can relatively effectively identify noise data and ensure the accuracy of the data source, they suffer from high workloads and incomplete identification of noise types. This method uses clustering, sampling, and comparative verification to form a set of automatic methods for superimposing composite noise, supplemented by electrical signal judgment. This method effectively ensures the accuracy of sample data. The clustering process also ensures the completeness of noise types in the sample data. Through repeated comparative verification and iteration, the automatic composite superposition of environmental noise is achieved.

[0069] The following describes in detail the method provided in the embodiment of the present application by taking one implementation as an example.

[0070] Since the environment in which each device is located is different, if a unified environmental noise is used to filter all devices, it is easy to have the negative consequence of losing the real signal. Therefore, the embodiment of the present application adopts a composite environmental noise generation and iteration method with environmental adaptability.

[0071] See also Figure 2 The specific implementation includes the following steps:

[0072] Step 1: This method first uses current and voltage monitoring sensors to collect the device's current and voltage signals. The device's processing state is identified based on the current and voltage signal thresholds, categorizing the device as either processing or unprocessed. Initially, a conservative threshold setting is used, sacrificing accuracy for processing data in exchange for ensuring that all unprocessed data identified is generally accurate. Obviously, when the device is in the unprocessed state, the data collected by the vibration sensor can be considered ambient noise, essentially interference signals generated by the processing of surrounding equipment. Since the current and voltage signals often have some error in identifying the device's processing state, this means that the processing data category also contains some non-processing noise data, which is processed in Step 9.

[0073] Step 2: This method calculates the mean, kurtosis, main frequency characteristics and other characteristic values ​​of the collected noise signal, and performs cluster analysis based on the characteristic values ​​to classify the noise signal.

[0074] Step 3: Randomly sample each type of noise signal set and extract 20% of the data of each type of noise signal set to obtain multiple noise sample data sets.

[0075] Step 4: Spectral calculation is performed on the noise data of the multiple noise sample data sets to obtain multiple spectra, and all spectra are superimposed to generate a composite noise spectrum. The superposition method is to compare all frequency points and assign the maximum value according to the noise frequency amplitude.

[0076] Step 5: Generate the superimposed composite noise spectrum and perform inverse Fourier transform to regenerate composite environmental noise time domain data.

[0077] Step 6: Use the generated environmental noise to perform targeted filtering operations on the entire noise sample data set. Classify the noise sample data set according to the filtering results, and divide the noise sample data into two categories: recognizable noise data and unrecognizable noise data. When the recognition rate is lower than 95%, repeat Step 3-Step 6.

[0078] Step 7: Reclassify the noise data regularly. When new categories appear, repeat Step 3 to Step 6.

[0079] Step 8: Establish an adaptive filter based on target spectrum elimination. The principle of adaptive filter is relatively simple and is a mature technology. The principle of adaptive filter can be found in Figure 3 Optimization can be used to eliminate unknown interference contained in the base signal (the vibration signal acquired while the equipment is operating). The base signal serves as the desired response of the adaptive filter, and the noise signal serves as the filter input. After multiple iterations, the difference between the desired signal and the adaptive filter's output is an estimate of the signal source.

[0080] Step 9: Perform secondary identification on the noise data based on the variance change rate of the data before and after filtering. Since the noise data are highly correlated, the variance of the noise data will be significantly reduced after filtering, while the variance of the normal processed data will not change much or may even increase after filtering. Based on this principle, secondary identification and classification of the noise data in the processed data is achieved, and the noise data in the processed data is further eliminated, thereby outputting accurate and effective result data.

[0081] Step 10: Adjust the current and voltage thresholds in Step 1 based on the noise ratio in the processed data from the secondary recognition results in Step 9. If the user determines that the noise ratio in the processed data is high, the current and voltage thresholds can be lowered to increase the ratio used for the shutdown judgment. If the noise ratio is low, it may indicate overclassification. In this case, the current and voltage thresholds can be raised to reduce the ratio used for the shutdown judgment. The general principle is to keep the noise ratio in the processed data around 5% to ensure both the quantity and quality of valid data.

[0082] The method provided in the embodiment of the present application can realize the rapid deployment and application of the equipment vibration monitoring system. Since the entire process has no human intervention at all, there is no need to manually obtain prior information before the system deployment using this method. After deployment, it can basically iterate to a relatively ideal effect in a very short time, which can effectively improve the deployment speed of the system.

[0083] In addition, the use of this method is conducive to the large-scale deployment and application of equipment vibration monitoring systems. Due to the adaptive generation of environmental noise and the adaptive iteration mechanism of equipment operating condition identification thresholds, this method can quickly adapt to the simultaneous deployment of various types of equipment, and therefore can support large-scale deployment and application.

[0084] See also Figure 4 , the embodiment of the present application can also provide an environmental noise adaptive filtering device, such as Figure 4 As shown, the device may include:

[0085] The unprocessed state determining unit 401 is configured to determine that the device is in the unprocessed state and then receive a noise signal in the unprocessed state collected by the vibration sensor;

[0086] The eigenvalue acquisition unit 402 is configured to acquire eigenvalues ​​of the noise signal, perform cluster analysis on the noise signal according to the eigenvalues, classify the noise signal, and obtain a plurality of noise signal sets;

[0087] The composite noise spectrum generating unit 403 is configured to perform spectrum calculation on the noise data of the multiple types of noise signal sets to obtain multiple spectra, and to superimpose all the spectra to generate a composite noise spectrum;

[0088] The time domain data conversion unit 404 is used to perform an inverse Fourier transform on the composite noise spectrum to generate composite environmental noise time domain data;

[0089] A processing state determining unit 405 is configured to determine that the device is in a processing state and then receive a basic signal of the processing state collected by the vibration sensor;

[0090] The signal source estimation acquisition unit 406 is used to use the composite ambient noise time domain data as the input of an adaptive filter based on target spectrum elimination, use the basic signal as the expected signal of the adaptive filter; and use the difference between the expected signal and the output signal of the adaptive filter as the estimate of the signal source.

[0091] The embodiment of the present application may further provide an environmental noise adaptive filtering device, characterized in that the device includes a processor and a memory:

[0092] The memory is used to store program code and transmit the program code to the processor;

[0093] The processor is configured to execute the steps of the above-mentioned environmental noise adaptive filtering method according to the instructions in the program code.

[0094] like Figure 5 As shown, an environmental noise adaptive filtering device provided by an embodiment of the present application may include: a processor 10, a memory 11, a communication interface 12, and a communication bus 13. The processor 10, the memory 11, and the communication interface 12 all communicate with each other through the communication bus 13.

[0095] In the embodiment of the present application, the processor 10 may be a central processing unit (CPU), an application-specific integrated circuit, a digital signal processor, a field programmable gate array, or other programmable logic devices.

[0096] The processor 10 may call a program stored in the memory 11 . Specifically, the processor 10 may execute operations in an embodiment of the environmental noise adaptive filtering method.

[0097] The memory 11 is used to store one or more programs. The program may include program code, and the program code includes computer operating instructions. In the embodiment of the present application, the memory 11 stores at least a program for implementing the following functions:

[0098] After determining that the device is in an unprocessed state, a noise signal in the unprocessed state collected by a vibration sensor is received;

[0099] Acquiring characteristic values ​​of the noise signal, performing cluster analysis on the noise signal according to the characteristic values, and classifying the noise signal to obtain a multi-class noise signal set;

[0100] performing spectrum calculation on noise data of multiple types of noise signal sets respectively to obtain multiple spectrums, and superimposing all the spectrums to generate a composite noise spectrum;

[0101] Performing an inverse Fourier transform on the composite noise spectrum to generate composite environmental noise time domain data;

[0102] After determining that the device is in a processing state, receiving a basic signal in the processing state collected by a vibration sensor;

[0103] The composite ambient noise time domain data is used as the input of an adaptive filter based on target spectrum elimination, the basic signal is used as the expected signal of the adaptive filter; and the difference between the expected signal and the output signal of the adaptive filter is used as an estimate of the signal source.

[0104] In one possible implementation, the memory 11 may include a program storage area and a data storage area, wherein the program storage area can store an operating system and application programs required for at least one function (such as a file creation function, a data reading and writing function), etc.; the data storage area can store data created during use, such as initialization data, etc.

[0105] In addition, the memory 11 may include a high-speed random access memory and may also include a non-volatile memory, such as at least one disk storage device or other volatile solid-state storage device.

[0106] The communication interface 12 may be an interface of a communication module, and is used to connect to other devices or systems.

[0107] Of course, it needs to be explained that Figure 5 The structure shown does not constitute a limitation on the environmental noise adaptive filtering device in the embodiment of the present application. In actual applications, the environmental noise adaptive filtering device may include Figure 5 More or fewer components than shown, or combinations of certain components.

[0108] An embodiment of the present application may further provide a computer-readable storage medium, characterized in that the computer-readable storage medium is used to store program code, and the program code is used to execute the steps of the above-mentioned environmental noise adaptive filtering method.

[0109] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0110] Through the description of the above embodiments, it can be seen that those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present application.

[0111] Each embodiment in this specification is described in a progressive manner. The same or similar parts between the embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. In particular, for system or system embodiments, since they are basically similar to method embodiments, the description is relatively simple. For relevant parts, refer to the partial description of the method embodiment. The system and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without expending creative work.

[0112] The above description is only a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the scope of protection of the present invention.

Claims

1. A method for adaptive filtering of environmental noise, characterized in that: include: After determining that the device is in an unprocessed state, a noise signal in the unprocessed state collected by a vibration sensor is received; Acquiring characteristic values ​​of the noise signal, performing cluster analysis on the noise signal according to the characteristic values, and classifying the noise signal to obtain a multi-class noise signal set; performing spectrum calculation on noise data of multiple types of noise signal sets respectively to obtain multiple spectrums, and superimposing all the spectrums to generate a composite noise spectrum; Performing an inverse Fourier transform on the composite noise spectrum to generate composite environmental noise time domain data; After determining that the device is in a processing state, receiving a basic signal in the processing state collected by a vibration sensor; Using the composite ambient noise time domain data as input to an adaptive filter based on target spectrum elimination, using the basic signal as a desired signal of the adaptive filter; and using the difference between the desired signal and the output signal of the adaptive filter as an estimate of the signal source; Determining whether the device is in an unprocessed state or a processed state based on a relationship between a current and voltage signal of the device collected by a current and voltage monitoring sensor and a preset target current and voltage threshold; Receiving the target current and voltage threshold adjustment method input by the user, adjusting the target current and voltage threshold to obtain an adjusted current and voltage threshold; the adjustment method is determined by the user performing secondary identification on the noise data based on the variance change rate of the data before and after filtering; It is determined whether the device is in an unprocessed state or a processed state according to the relationship between the current and voltage signals of the device collected by the current and voltage monitoring sensor and the adjusted current and voltage thresholds.

2. The environmental noise adaptive filtering method according to claim 1, characterized in that: The characteristic values ​​include mean, kurtosis, and main frequency characteristics.

3. The environmental noise adaptive filtering method according to claim 1, characterized in that: Random sampling is performed on each type of noise signal set, and 20% of the data of each type of noise signal set is extracted to obtain multiple noise sample data sets; Performing spectrum calculation on the noise data of the plurality of noise sample data sets to obtain a plurality of spectrums, and superimposing all the spectrums to generate a composite noise spectrum; All frequency points included in the multiple spectra are compared, and a maximum value is assigned according to the noise frequency amplitude, so as to achieve superposition of all the spectra to generate a composite noise spectrum.

4. The environmental noise adaptive filtering method according to claim 3, characterized in that: The composite environmental noise time-domain data is used to perform targeted filtering operations on all the noise sample data sets, and the noise sample data sets are classified according to the filtering results to divide the noise sample data into recognizable noise data and unrecognizable noise data. The validity of the composite environmental noise time-domain data is determined based on the recognition rate of the noise sample data.

5. The environmental noise adaptive filtering method according to claim 4, characterized in that: When the recognition rate is lower than 95% or a new type of noise signal is determined to exist, the composite environmental noise time domain data is determined to be invalid.

6. An environmental noise adaptive filtering device, characterized in that: The apparatus is configured to execute the environmental noise adaptive filtering method according to any one of claims 1 to 5, comprising: an unprocessed state determining unit, configured to determine that the device is in an unprocessed state and then receive a noise signal in the unprocessed state collected by a vibration sensor; a characteristic value acquisition unit, configured to acquire characteristic values ​​of the noise signal, and perform cluster analysis on the noise signal according to the characteristic values ​​to classify the noise signal to obtain a multi-class noise signal set; a composite noise spectrum generating unit, configured to perform spectrum calculation on noise data of multiple types of noise signal sets respectively to obtain multiple spectra, and to superimpose all the spectra to generate a composite noise spectrum; A time domain data conversion unit, configured to perform an inverse Fourier transform on the composite noise spectrum to generate composite environmental noise time domain data; a processing state determining unit, configured to determine that the device is in a processing state and receive a basic signal in the processing state collected by the vibration sensor; A signal source estimation and acquisition unit is used to use the composite ambient noise time domain data as the input of an adaptive filter based on target spectrum elimination, use the basic signal as the expected signal of the adaptive filter; and use the difference between the expected signal and the output signal of the adaptive filter as an estimate of the signal source.

7. An environmental noise adaptive filtering device, characterized in that: The device includes a processor and a memory: The memory is used to store program code and transmit the program code to the processor; The processor is configured to execute the environmental noise adaptive filtering method according to any one of claims 1 to 5 according to instructions in the program code.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium is used to store program code, and the program code is used to execute the environmental noise adaptive filtering method according to any one of claims 1 to 5.

Images