Sound characteristic signal identification method and device, electronic equipment and storage medium

Abstract

The invention discloses a sound characteristic signal identification method and device, electronic equipment and a storage medium. The method is characterized by comprising the following steps: acquiring a to-be-processed sound signal through a communication unit; performing wavelet denoising on the to-be-processed sound signal to determine a first sound signal; denoising the first sound signal through empirical mode decomposition, and determining a target sound signal; performing spectrogram calculation on the target sound signal through discrete Fourier transform, and determining a target time-frequency graph; and uploading the target time-frequency diagram to a communication unit, performing signal identification on the target time-frequency diagram through the communication unit, and determining a sound identification result. Through two times of denoising, a plurality of different noises existing in the equipment operation sound can be effectively removed, and the accuracy of power grid equipment identification is improved.

Description

Technical Field

[0001] The present invention relates to the field of sound processing technology, and in particular to a sound feature signal recognition method, device, electronic device and storage medium. Background Art

[0002] Power grid equipment such as substation main transformers, reactors, switchgear, overhead transmission lines, cables, distribution transformers, pole switches, switchgear, and ring main units will produce different sounds during operation. These sounds contain important information such as the operating status and potential faults of the equipment. The operating sounds of the equipment can be collected and sound recognition technology can be used to diagnose and monitor the operating conditions of the power equipment and identify faults in the power equipment. However, when using sound recognition technology for identification, the collected equipment operating sounds will be affected by the factors of the equipment operating environment, resulting in the presence of a variety of different noises in the collected equipment operating sounds. Excessive noise results in a low recognition rate for the equipment operating sounds when identifying them, making it difficult to judge the operating status of the power grid equipment. Summary of the invention

[0003] The present invention provides a method, device, electronic device and storage medium for recognizing sound characteristic signals, so as to solve the technical problem that the accuracy of recognizing the operation sound of power grid equipment in the prior art is too low.

[0004] According to one aspect of the present invention, there is provided a method for identifying a sound characteristic signal, comprising:

[0005] Acquiring a sound signal to be processed through a communication unit;

[0006] Perform wavelet denoising on the sound signal to be processed to determine a first sound signal;

[0007] Denoising the first sound signal by empirical mode decomposition to determine a target sound signal;

[0008] Calculate the spectrogram of the target sound signal through discrete Fourier transform to determine the target time-frequency graph;

[0009] The target time-frequency graph is uploaded to the communication unit, and the communication unit performs signal recognition on the target time-frequency graph to determine the sound recognition result.

[0010] According to another aspect of the present invention, there is provided a device for identifying a sound characteristic signal, comprising:

[0011] A data communication module, used for acquiring a sound signal to be processed through a communication unit;

[0012] A first denoising module, used for performing wavelet denoising on the sound signal to be processed to determine a first sound signal;

[0013] A second denoising module, used to denoise the first sound signal by empirical mode decomposition to determine a target sound signal;

[0014] A spectrogram calculation module, used to perform spectrogram calculation on the target sound signal through discrete Fourier transform to determine the target time-frequency graph;

[0015] The data uploading module is used to upload the target time-frequency diagram to the communication unit, perform signal recognition on the target time-frequency diagram through the communication unit, and determine the sound recognition result.

[0016] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0017] at least one processor; and

[0018] a memory communicatively connected to the at least one processor; wherein,

[0019] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the sound feature signal recognition method described in any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the sound feature signal recognition method described in any embodiment of the present invention when executed.

[0021] The technical solution of the embodiment of the present invention obtains a sound signal to be processed through a communication unit, establishes communication with the communication unit, obtains the sound information to be processed through the communication unit, performs wavelet denoising on the sound signal to be processed, and determines a first sound signal. The wavelet denoising can ensure the key features of the sound signal, and then perform efficient denoising to improve the denoising effect; denoises the first sound signal through empirical mode decomposition to determine the target sound signal, and decomposes and denoises the sound signal again through empirical mode decomposition, which can remove noise in the sound through multi-scale and multi-resolution recognition to achieve a more refined denoising effect; performs spectrogram calculation on the target sound signal through discrete Fourier transform to determine the target time-frequency diagram, and obtains the target time-frequency diagram through spectrogram calculation, which can display the sound in the form of a visual picture, can improve the accuracy of feature extraction, and effectively improve the recognition accuracy of the sound signal; uploads the target time-frequency diagram to the communication unit, and performs signal recognition on the target time-frequency diagram through the communication unit to determine the sound recognition result. The present invention solves the technical problem of low accuracy in the recognition of the operation sound of power grid equipment in the prior art. By performing two denoising operations, various different noises in the operation sound of the equipment can be effectively removed, thereby improving the accuracy of the recognition of power grid equipment.

[0022] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0024] Figure 1 A flow chart of a method for identifying a sound characteristic signal is provided for an embodiment of the present invention;

[0025] Figure 2 A flowchart of another method for identifying a sound characteristic signal provided by an embodiment of the present invention;

[0026] Figure 3 A schematic diagram of the structure of a sound characteristic signal recognition device provided by an embodiment of the present invention;

[0027] Figure 4 A schematic structural diagram of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments 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 without creative work should fall within the scope of protection of the present invention.

[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0030] Figure 1 A flow chart of a method for identifying a sound characteristic signal is provided for an embodiment of the present invention. This embodiment is applicable to the case where the sound signal of a power grid device is identified and judged through a sound-synesthesia integrated terminal. The method can be executed by a sound characteristic signal identification device, which can be implemented in the form of hardware and / or software, and can be configured in a sound-synesthesia integrated terminal. Figure 1 As shown, the method includes:

[0031] S110: Acquire a sound signal to be processed through a communication unit.

[0032] Optionally, the sound synaesthesia integrated terminal includes an intelligent voiceprint perception function, which is composed of an audio control processing unit and a communication interface. The audio control processing unit is connected to the communication unit through the communication interface. The communication unit has a three-layer architecture of terminal layer, edge layer and cloud layer. The terminal layer generates different types of data and transmits the data to the cloud layer for processing through wireless communication functions; the edge layer is composed of edge nodes with limited computing power, and the edge nodes include small base stations connected to edge servers, and the edge servers are connected to the cloud layer through optical links; the cloud layer includes a data center layer and a service layer. The sound synaesthesia integrated terminal is connected to the vector sample library through the communication unit, and the vector sample library is used to store the sound signals to be processed. It should be noted that the vector sample library is constructed by the following steps:

[0033] Collect various types of data from power grid equipment, including sound signals from power grid equipment;

[0034] Preprocess the collected data, including data cleaning to remove duplicate, incomplete or erroneous data; data correction and registration to eliminate deviations, errors or inconsistencies in the data; and data format conversion to a unified data format suitable for vectorization.

[0035] Vectorize the pre-processed data and convert the data into vector data. The vectorization method may include automated vectorization algorithm and point cloud processing; the vector data may be points, lines, or surfaces;

[0036] For vector data, data modeling is performed, an appropriate data model is selected for data modeling, and database design is performed to establish a data model, define data tables, fields, and constraints. For example, the data model for data modeling can select a geospatial data model; the database can select a Milvus vector database.

[0037] For the vector database, the original sample library, model information library, label sample library, original data detection record library, noise anomaly detection record library, abnormal discharge detection record library, mechanical fault detection record library and other databases are established. Among them, the original sample library stores the sound signal to be processed.

[0038] Specifically, the vector database is connected via a communication unit to obtain the sound signal to be processed in the original sample library.

[0039] S120 , performing wavelet denoising on the sound signal to be processed to determine a first sound signal.

[0040] The first sound signal may be a sound signal obtained by wavelet transforming the sound signal to be processed. It should be noted that wavelet denoising is to perform wavelet transform on the sound signal to be processed by wavelet basis function, perform threshold denoising on the wavelet transformed signal by selecting a threshold function, and then perform wavelet reconstruction on the denoised signal to obtain the first sound signal.

[0041] Specifically, wavelet denoising is performed on the sound signal to be processed to determine the first sound signal.

[0042] Optionally, in another optional embodiment of the present invention, the wavelet denoising is performed on the sound signal to be processed to determine the first sound signal, including:

[0043] The sound signal to be processed is convolved and downsampled by a discrete wavelet filter to obtain a discrete wavelet signal; the discrete wavelet signal is threshold denoised by a preset threshold function to determine a denoised wavelet signal; the discrete wavelet signal is upsampled and convolved and reconstructed by the wavelet coefficients corresponding to the discrete wavelet filter to obtain a first sound signal.

[0044] The discrete wavelet filter can be a filter group used for signal decomposition and signal reconstruction of the processed sound signal in wavelet denoising; the discrete wavelet filter is composed of a low-pass filter and a high-pass filter, the low-pass filter can be used to extract the low-frequency part of the processed signal, the high-pass filter is used to extract the high-frequency part of the processed signal, and the discrete wavelet filter can decompose the processed sound signal multiple times to obtain a decomposed signal with finer granularity. The discrete wavelet signal can be a decomposed signal obtained by multiple decompositions of the discrete wavelet filter.

[0045] Among them, the preset threshold function can be a threshold function preset for denoising the decomposed signal. Exemplarily, the preset threshold function can be a soft threshold or a hard threshold. The denoised wavelet signal can be a decomposed signal obtained by processing the discrete wavelet signal through the preset threshold function. It should be noted that the discrete wavelet signal is directly subjected to a function operation with the preset threshold function to obtain the denoised wavelet signal. Exemplarily, when the preset threshold function is set to a hard threshold function, the decomposed signal less than the threshold in the discrete wavelet signal is set to 0, and the decomposed signal greater than the threshold is retained.

[0046] Optionally, the wavelet denoising process is as follows: the sound signal to be processed is filtered and calculated through a low-pass filter and a high-pass filter in a discrete wavelet filter, and down-sampled to obtain a low-frequency part and a high-frequency part of the first decomposition, and the low-frequency part of the first decomposition is filtered and calculated again through a low-pass filter and a high-pass filter, and down-sampled, and a discrete wavelet signal is obtained through multiple decomposition cycles, and the discrete wavelet signal is threshold denoised by a preset threshold function to obtain a denoised wavelet signal, and then the inverse discrete wavelet filter is used as the wavelet coefficient, and the denoised wavelet signal is convolved and reconstructed and up-sampled multiple times to obtain a first sound signal.

[0047] S130 , denoising the first sound signal by empirical mode decomposition to determine a target sound signal.

[0048] Among them, Empirical Mode Decomposition (EMD) is an adaptive time-frequency analysis method that can decompose complex signals into several modal functions and effectively separate the different frequency components of the signal.

[0049] The target sound signal may be a sound signal obtained by performing empirical mode decomposition and denoising on the first sound signal. It should be noted that after the first sound signal is subjected to empirical mode decomposition, a plurality of mode functions and residuals are obtained, and the mode functions are denoised, so that the target sound signal is formed by the denoised mode functions and residuals.

[0050] Specifically, the first sound signal is denoised by empirical mode decomposition to determine the target sound signal.

[0051] S140, performing spectrogram calculation on the target sound signal through discrete Fourier transform to determine the target time-frequency graph.

[0052] Among them, Discrete Fourier Transform (DFT) can perform discrete Fourier transform on the signal, and the target sound signal can be converted into a time-frequency diagram through discrete Fourier transform. The target time-frequency diagram can be obtained by converting the target sound signal into a time-frequency diagram. The target video diagram is a two-dimensional diagram, the horizontal axis represents time, the vertical axis represents frequency, and the grayscale represents the energy of the frequency at a time point; the target time-frequency diagram can represent the timbre characteristics of the target sound signal.

[0053] Specifically, the target sound signal is spectrogram-calculated by discrete Fourier transform to determine the target time-frequency graph.

[0054] Optionally, in another optional embodiment of the present invention, the performing spectrogram calculation on the target sound signal by discrete Fourier transform to determine the target time-frequency graph includes:

[0055] The target sound signal is subjected to discrete Fourier transform to determine the sound spectrum signal; a spectrogram is drawn based on the sound spectrum signal to obtain a target time-frequency diagram.

[0056] The sound spectrum signal may be a frequency domain representation of the target sound signal after Fourier transformation.

[0057] Optionally, the specific process of performing Fourier transform on the target sound signal is as follows: the continuous target sound signal is framed into multiple short frames, and r sampling points are set for each short frame. The framed signal can be expressed as e(s,r), where s represents the sth frame and r represents the sampling point in the frame; Discrete Fourier Transform (DFT) is performed on each frame e(s,r) to obtain a sound spectrum signal e(s,k), where k represents a frequency index; the periodogram U(s,k) is calculated based on the sound spectrum signal e(s,k), the periodogram U(s,k) is converted into a logarithmic scale Z(s,k), and a spectrogram is drawn based on the logarithmic scale Z(s,k) to obtain a target time-frequency diagram.

[0058] S150, uploading the target time-frequency graph to a communication unit, performing signal recognition on the target time-frequency graph through the communication unit, and determining a sound recognition result.

[0059] The sound recognition result may be the state information of the power grid device corresponding to the sound signal to be processed. For example, when the power grid device is a transformer, the sound to be processed may be high-frequency noise, and the sound recognition result may be partial discharge of the transformer.

[0060] Specifically, the target time-frequency graph is sent to the communication unit through the communication interface, the edge server of the middle edge layer of the communication unit performs signal recognition, and the target time-frequency graph is subjected to signal recognition through the neural network model to determine the sound recognition result. The neural network model may be a convolutional neural network with a self-attention layer; the convolutional neural network includes an input layer, a convolutional layer, an attention layer, a multi-scale layer, a residual layer, and an output layer.

[0061] The technical solution of the embodiment of the present invention obtains a sound signal to be processed through a communication unit, establishes communication with the communication unit, obtains the sound information to be processed through the communication unit, performs wavelet denoising on the sound signal to be processed, and determines a first sound signal. The wavelet denoising can ensure the key features of the sound signal, and then perform efficient denoising to improve the denoising effect; denoises the first sound signal through empirical mode decomposition to determine the target sound signal, and decomposes and denoises the sound signal again through empirical mode decomposition, which can remove noise in the sound through multi-scale and multi-resolution recognition to achieve a more refined denoising effect; performs spectrogram calculation on the target sound signal through discrete Fourier transform to determine the target time-frequency diagram, and obtains the target time-frequency diagram through spectrogram calculation, which can display the sound in the form of a visual picture, can improve the accuracy of feature extraction, and effectively improve the recognition accuracy of the sound signal; uploads the target time-frequency diagram to the communication unit, and performs signal recognition on the target time-frequency diagram through the communication unit to determine the sound recognition result. The present invention solves the technical problem of low accuracy in the recognition of the operation sound of power grid equipment in the prior art. By performing two denoising operations, various different noises in the operation sound of the equipment can be effectively removed, thereby improving the accuracy of the recognition of power grid equipment.

[0062] Figure 2 This is a flowchart of another method for identifying a sound characteristic signal provided by an embodiment of the present invention. The relationship between this embodiment and the above embodiment is the specific process of performing denoising again through empirical mode decomposition. Figure 2 As shown, the method includes:

[0063] S210: Acquire a sound signal to be processed through a communication unit.

[0064] S220 , performing wavelet denoising on the sound signal to be processed to determine a first sound signal.

[0065] S230: Perform empirical mode decomposition on the first sound signal to determine a first empirical mode function.

[0066] The first empirical mode function may be a function composed of at least one mode function and a residual. It should be noted that after performing empirical mode decomposition on the first sound signal, at least one mode function and a residual are obtained, and the function composed of the mode function and the residual is determined as the first empirical mode function.

[0067] Specifically, empirical mode decomposition is performed on the first sound signal to determine a first empirical mode function.

[0068] Optionally, in another optional embodiment of the present invention, according to the method of claim 3, it is characterized in that performing empirical mode decomposition on the first sound signal to determine the first empirical mode function includes:

[0069] The first sound signal is noised multiple times by using a preset random Gaussian noise to obtain a noised sound signal set; wherein the noised sound signal set is a noised sound signal obtained by noise-adding the first sound signal each time;

[0070] Performing empirical mode decomposition on the noisy sound signals one by one to determine the multi-order mode function and residual corresponding to each noisy sound signal; wherein the multi-order mode function is composed of each mode function component;

[0071] Averaging the modal function components corresponding to each order of all noise-added sound signals to determine the average modal function component corresponding to the modal function components corresponding to each order;

[0072] Averaging the residuals corresponding to each noise-added sound signal to determine an average residual corresponding to the first sound signal;

[0073] A first empirical mode function is constructed based on a plurality of average mode function components and an average residual.

[0074] The random Gaussian noise may be randomly generated Gaussian noise for denoising; the amplitude distribution of the random Gaussian noise complies with the normal distribution.

[0075] Among them, the noisy sound signal set can be a set of noisy sound signals generated by adding random Gaussian noise to the first sound signal. It should be noted that when the first sound signal is noisyed by random Gaussian noise, the first sound signal will be noisyed multiple times, and the random Gaussian noise will change randomly during each noise addition, and then the sound signal obtained each time the noise is added will be saved as a noisy sound signal, and then multiple noisy sound signals will be obtained, and then the noisy sound signal set will be constructed based on the multiple noisy sound signals.

[0076] Among them, the multi-order modal function can be multiple modal functions obtained by empirical mode decomposition; since the empirical mode decomposition is performed in sequence, the modal function obtained by the first decomposition is called the first-order modal function component, the modal function obtained by the second decomposition is called the second-order modal function component, the modal function obtained by the j-th decomposition is called the j-th modal function component, and all modal function components obtained from the first to the n-th decomposition are called multi-order modal functions.

[0077] Optionally, when performing empirical mode decomposition, since the first sound signal is noisyed multiple times to form multiple noisy sound signals, after performing empirical mode decomposition on the multiple noisy sound signals, multi-order modal functions and residuals corresponding to each noisy sound signal are obtained, and each order modal function component of each noisy sound signal and the modal function components corresponding to other noisy sound signals are averaged to calculate the average modal function component corresponding to the modal function component corresponding to each order. Exemplarily, assuming that there are three noisy sound signals, three multi-order modal functions and residuals corresponding to the three noisy sound signals are obtained through empirical mode decomposition, each noisy sound signal is decomposed into four modal function categories, then the multi-order modal function is a fourth-order modal function, and the way to average the fourth-order modal functions of the three noisy sound signals is: average the first-order modal function components of the three noisy sound signals to obtain the first-order average modal function component corresponding to the first-order modal function component, and repeat the calculation to obtain the second-order average modal function component, the third-order average modal function component and the fourth-order average modal function component.

[0078] The average residual may be an average value of the residuals corresponding to all the noise-added sound signals.

[0079] Specifically, the first sound signal is noised multiple times by means of random Gaussian noise, and the signal obtained by each noise addition is used as the noised sound signal corresponding to the first sound signal, thereby forming a set of noised sound signals; each noised sound signal is subjected to empirical mode decomposition to obtain multi-order modal functions and residuals corresponding to each noisy sound signal, and the modal function components and residuals corresponding to each order are averaged respectively to determine the average modal function components and average residuals corresponding to the modal function components corresponding to each order, and a first empirical mode function is formed based on the average modal function components and average residuals of each order.

[0080] Exemplarily, the first sound signal is passed through x(t), and the noise sound signal is passed through x i (t), i represents the ith noise addition, and w i (t) represents the random Gaussian white noise added for the i-th time. After decomposing the noise signal by EMD, the j-order modal function component and residual are obtained. The j-order modal function component is represented by c ij (t), j is 1, 2, 3, ... m, and the residual is r i (t) is represented; the formula for adding noise to the first sound signal by random Gaussian white noise is as follows:

[0081] x i (t) = x(t) + w i (t), i = 1, 2, 3, ... N;

[0082] The formula for decomposing the noisy sound signal by EMD is as follows:

[0083]

[0084] Furthermore, the j-th order average mode function component is expressed as c j (t) is used to represent the average residual, and r(t) is used to represent the average residual. The specific calculation method is as follows:

[0085]

[0086]

[0087] S240: Perform frequency denoising and energy denoising on the first empirical mode function to determine a target sound signal.

[0088] Specifically, the first empirical mode function is first subjected to frequency denoising, and then subjected to capacity denoising to obtain a target sound signal.

[0089] Optionally, in another optional embodiment of the present invention, the frequency denoising and energy denoising of the first empirical mode function to determine the target sound signal includes: performing noise identification on the average modal function component of the first empirical mode function according to a preset frequency threshold to determine the first noise modal function component; removing the first noise modal function component from the first empirical mode function to obtain a second empirical mode function; and performing energy denoising on the second empirical mode function to determine the target sound signal.

[0090] The preset frequency threshold may be a frequency threshold pre-set for denoising the average modal function component. It should be noted that, since the highest frequency of the mechanical response is a fixed frequency value, the highest frequency of the mechanical response is selected as the preset frequency threshold, and the average modal function component with a frequency greater than the preset frequency threshold in the average modal function component is noise, which is determined as the first noise modal function component. That is, the first noise modal function component may be any number of average modal function components, and the average modal function components with a frequency greater than the preset frequency threshold are all determined as the first noise modal function component.

[0091] The second empirical mode function is the first empirical mode function obtained by removing the first noise mode function component.

[0092] Specifically, the frequency of each average modal function component of the first empirical mode function is compared with a preset frequency threshold. If the frequency of the average modal function component is greater than the preset frequency threshold, it is determined to be a first noise modal function component. The first noise modal function component is subtracted from the first empirical mode function to obtain a second empirical mode function. The second empirical mode function is energy denoised to determine the target sound signal.

[0093] Optionally, in another optional embodiment of the present invention, performing energy denoising on the second empirical mode function to determine the target sound signal includes:

[0094] Calculating energy density according to the amplitude of each modal function component in the second empirical mode function to obtain signal energy density corresponding to each modal function component;

[0095] Determining the energy corresponding to each modal function component based on the signal energy density corresponding to each modal function component;

[0096] Calculate the energy ratio of each modal function component in turn to determine the signal energy ratio corresponding to each modal function component;

[0097] Based on the signal energy ratio of each modal function classification, the energy corresponding to each modal function component is compared to perform noise identification to determine a second noise modal function component;

[0098] The second noise mode function component is removed from the second empirical mode function to obtain a target sound signal.

[0099] The signal energy density can describe the energy distribution of each modal function component within a unit frequency range. It should be noted that the signal energy density of each modal function component can be calculated based on the amplitude of each modal function component.

[0100] Optionally, after obtaining the signal energy density of each modal function component, since the modal function components are continuous-time signals, time domain calculation or frequency domain calculation can be performed based on the signal energy density of each modal function component to obtain the energy of each modal function component.

[0101] Optionally, after obtaining the energy of each modal function component, for each modal function component, the energy average of all modal function components preceding each modal function component is calculated, the energy average comparison value of each modal function component is determined, and the energy of each modal function component and the energy average comparison value are calculated to obtain the signal energy ratio of each modal function component. Exemplarily, the calculation formula of the signal energy ratio of each modal function component is as follows: The energy of the j-th order modal function component is expressed by p j The energy average comparison value corresponding to the j-th order modal function component is the average energy of the first j-1 order modal function components, expressed as q j-1 , signal energy ratio is expressed as R pj The specific calculation formula is as follows:

[0102]

[0103] Optionally, after obtaining the signal energy ratio of each modal function component, noise identification is performed on the modal function component based on the signal energy ratio of each modal function component. The specific identification process is as follows: if the signal energy ratio of the modal function component is not less than 1, the modal function component is considered to be a normal sound signal; if the signal energy ratio of the modal function component is less than 1, the modal function component is determined to be a second noise modal function component. The second noise modal function component is a modal function component whose signal energy is noise.

[0104] Specifically, energy density is calculated according to the amplitude of each modal function component in the second empirical mode function to obtain the signal energy density corresponding to each modal function component; the energy corresponding to each modal function component is determined based on the signal energy density corresponding to each modal function component; energy ratios are calculated for the energies corresponding to each modal function component in turn to determine the signal energy ratio corresponding to each modal function component; noise identification is performed on the energy corresponding to each modal function component based on the signal energy ratio of each modal function classification to determine the second noise modal function component; the second noise modal function component is removed in the second empirical mode function to obtain the target sound signal.

[0105] S250, performing spectrogram calculation on the target sound signal by discrete Fourier transform to determine the target time-frequency graph;

[0106] S260, uploading the target time-frequency graph to a communication unit, performing signal recognition on the target time-frequency graph through the communication unit, and determining a sound recognition result.

[0107] The technical solution of the embodiment of the present invention denoises the first sound signal through empirical mode decomposition, determines the target sound signal, decomposes the first sound signal multiple times through empirical mode decomposition, and denoises the sound signal again through frequency denoising and energy denoising, which can remove noise in the sound through multi-scale and multi-resolution recognition, and achieve a more refined denoising effect. The technical problem of the low accuracy of the equipment operation sound recognition of power grid equipment in the prior art is solved. Through two denoising, the various different noises in the equipment operation sound can be effectively removed, and the accuracy of power grid equipment recognition can be improved.

[0108] Figure 3 A schematic diagram of the structure of a sound feature signal recognition device provided by an embodiment of the present invention. Figure 3 As shown, the device includes: a data communication module 310, a first denoising module 320, a second denoising module 330, a spectrogram calculation module 340 and a data upload module 350; wherein,

[0109] The data communication module 310 is used to obtain the sound signal to be processed through the communication unit;

[0110] A first denoising module 320, configured to perform wavelet denoising on the sound signal to be processed to determine a first sound signal;

[0111] A second denoising module 330, configured to denoise the first sound signal by empirical mode decomposition to determine a target sound signal;

[0112] The spectrogram calculation module 340 is used to perform spectrogram calculation on the target sound signal through discrete Fourier transform to determine the target time-frequency graph;

[0113] The data uploading module 350 is used to upload the target time-frequency diagram to the communication unit, and perform signal recognition on the target time-frequency diagram through the communication unit to determine the sound recognition result.

[0114] The technical solution of the embodiment of the present invention obtains a sound signal to be processed through a communication unit, establishes communication with the communication unit, obtains the sound information to be processed through the communication unit, performs wavelet denoising on the sound signal to be processed, and determines a first sound signal. The wavelet denoising can ensure the key features of the sound signal, and then perform efficient denoising to improve the denoising effect; denoises the first sound signal through empirical mode decomposition to determine the target sound signal, and decomposes and denoises the sound signal again through empirical mode decomposition, which can remove noise in the sound through multi-scale and multi-resolution recognition to achieve a more refined denoising effect; performs spectrogram calculation on the target sound signal through discrete Fourier transform to determine the target time-frequency diagram, and obtains the target time-frequency diagram through spectrogram calculation, which can display the sound in the form of a visual picture, can improve the accuracy of feature extraction, and effectively improve the recognition accuracy of the sound signal; uploads the target time-frequency diagram to the communication unit, and performs signal recognition on the target time-frequency diagram through the communication unit to determine the sound recognition result. The present invention solves the technical problem of low accuracy in the recognition of the operation sound of power grid equipment in the prior art. By performing two denoising operations, various different noises in the operation sound of the equipment can be effectively removed, thereby improving the accuracy of the recognition of power grid equipment.

[0115] Optionally, the first denoising module is specifically used to:

[0116] Convolution and down-sampling of the sound signal to be processed are performed through a discrete wavelet filter to obtain a discrete wavelet signal;

[0117] Perform threshold denoising on the discrete wavelet signal by using a preset threshold function to determine a denoised wavelet signal;

[0118] The discrete wavelet signal is up-sampled and convolved to reconstruct the discrete wavelet signal using the wavelet coefficients corresponding to the discrete wavelet filter to obtain a first sound signal.

[0119] Optionally, the second denoising module is specifically used to:

[0120] Performing empirical mode decomposition on the first sound signal to determine a first empirical mode function;

[0121] Frequency denoising and energy denoising are performed on the first empirical mode function to determine the target sound signal.

[0122] Optionally, the second denoising module is further configured to:

[0123] The first sound signal is noised multiple times by using a preset random Gaussian noise to obtain a noised sound signal set; wherein the noised sound signal set is a noised sound signal obtained by noise-adding the first sound signal each time;

[0124] Performing empirical mode decomposition on the noisy sound signals one by one to determine the multi-order mode function and residual corresponding to each noisy sound signal; wherein the multi-order mode function is composed of each mode function component;

[0125] Performing an average processing on the modal function components corresponding to each order of all the noise-added sound signals, and determining the average modal function component corresponding to the modal function components corresponding to each order;

[0126] Averaging the residuals corresponding to each noise-added sound signal to determine an average residual corresponding to the first sound signal;

[0127] A first empirical mode function is constructed based on a plurality of average mode function components and an average residual.

[0128] Optionally, the second denoising module is further configured to:

[0129] Performing noise identification on an average modal function component of the first empirical mode function according to a preset frequency threshold to determine a first noise modal function component;

[0130] Removing the first noise mode function component from the first empirical mode function to obtain a second empirical mode function;

[0131] Energy denoising is performed on the second empirical mode function to determine the target sound signal.

[0132] Optionally, the second denoising module is further configured to:

[0133] Calculating energy density according to the amplitude of each modal function component in the second empirical mode function to obtain signal energy density corresponding to each modal function component;

[0134] Determining the energy corresponding to each modal function component based on the signal energy density corresponding to each modal function component;

[0135] Calculate the energy ratio of each modal function component in turn to determine the signal energy ratio corresponding to each modal function component;

[0136] Based on the signal energy ratio of each modal function classification, the energy corresponding to each modal function component is compared to perform noise identification to determine a second noise modal function component;

[0137] The second noise mode function component is removed from the second empirical mode function to obtain a target sound signal.

[0138] Optionally, the spectrum calculation module is specifically used for:

[0139] Performing discrete Fourier transform on the target sound signal to determine the sound spectrum signal;

[0140] The spectrogram is drawn based on the sound spectrum signal to obtain the target time-frequency diagram.

[0141] The sound characteristic signal recognition device provided in the embodiment of the present invention can execute the sound characteristic signal recognition method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0142] Figure 4 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their modes are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0143] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0144] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer grid such as the Internet and / or various telecommunication grids.

[0145] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 performs the various methods and processes described above, such as the sound feature signal recognition method.

[0146] In some embodiments, the sound characteristic signal recognition method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the sound characteristic signal recognition method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the sound characteristic signal recognition method in any other appropriate manner (e.g., by means of firmware).

[0147] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0148] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the modes / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0149] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0150] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0151] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a grid browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication grid). Examples of communication grids include: a local area network (LAN), a wide area network (WAN), a blockchain grid, and the Internet.

[0152] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication grid. The client and server relationship is generated by computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0153] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0154] This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the steps of the method for identifying a sound characteristic signal provided in any embodiment of the present invention are implemented. The method includes:

[0155] Acquiring a sound signal to be processed through a communication unit;

[0156] Perform wavelet denoising on the sound signal to be processed to determine a first sound signal;

[0157] Denoising the first sound signal by empirical mode decomposition to determine a target sound signal;

[0158] Calculate the spectrogram of the target sound signal through discrete Fourier transform to determine the target time-frequency graph;

[0159] The target time-frequency graph is uploaded to the communication unit, and the communication unit performs signal recognition on the target time-frequency graph to determine the sound recognition result.

[0160] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0161] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0162] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0163] Computer program code for performing the operation of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of grid, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0164] It should be understood by those skilled in the art that the above modules or steps of the present invention can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a grid composed of multiple computing devices, optionally, they can be implemented by a program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0165] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0166] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for identifying a sound characteristic signal, characterized in that: include: Acquiring a sound signal to be processed through a communication unit; Performing wavelet denoising on the sound signal to be processed to determine a first sound signal; Denoising the first sound signal by empirical mode decomposition to determine a target sound signal; Performing spectrogram calculation on the target sound signal by discrete Fourier transform to determine a target time-frequency graph; The target time-frequency graph is uploaded to a communication unit, and signal recognition is performed on the target time-frequency graph by the communication unit to determine a sound recognition result.

2. The method according to claim 1, characterized in that The step of performing wavelet denoising on the sound signal to be processed to determine the first sound signal comprises: Convolving and downsampling the sound signal to be processed through a discrete wavelet filter to obtain a discrete wavelet signal; Performing threshold denoising on the discrete wavelet signal by using a preset threshold function to determine a denoised wavelet signal; The discrete wavelet signal is up-sampled and convolved to reconstruct the discrete wavelet signal using the wavelet coefficients corresponding to the discrete wavelet filter to obtain the first sound signal.

3. The method according to claim 1, characterized in that: The denoising the first sound signal by empirical mode decomposition to determine the target sound signal includes: Performing empirical mode decomposition on the first sound signal to determine a first empirical mode function; Frequency denoising and energy denoising are performed on the first empirical mode function to determine the target sound signal.

4. The method according to claim 3, characterized in that The performing empirical mode decomposition on the first sound signal to determine a first empirical mode function includes: Noising the first sound signal multiple times with a preset random Gaussian noise to obtain a set of noise-added sound signals; wherein the set of noise-added sound signals is a set of noise-added sound signals obtained by noise-adding the first sound signal each time; Performing empirical mode decomposition on the noisy sound signals one by one to determine the multi-order modal function and residual corresponding to each of the noisy sound signals; wherein the multi-order modal function is composed of each modal function component; Performing an average processing on the modal function components corresponding to each order of all the noise-added sound signals, and determining an average modal function component corresponding to the modal function components corresponding to each order; Performing an average process on the residuals corresponding to each of the noise-added sound signals to determine an average residual corresponding to the first sound signal; The first empirical mode function is constructed based on a plurality of the average mode function components and the average residual.

5. The method according to claim 4, characterized in that The performing frequency denoising and energy denoising on the first empirical mode function to determine the target sound signal includes: Performing noise identification on the average modal function component of the first empirical mode function according to a preset frequency threshold to determine a first noise modal function component; removing the first noise mode function component from the first empirical mode function to obtain a second empirical mode function; Energy denoising is performed on the second empirical mode function to determine the target sound signal.

6. The method according to claim 5, characterized in that The performing energy denoising on the second empirical mode function to determine the target sound signal includes: Performing energy density calculation according to the amplitude of each of the modal function components in the second empirical mode function to obtain a signal energy density corresponding to each of the modal function components; Determining the energy corresponding to each of the modal function components based on the signal energy density corresponding to each of the modal function components; Calculating the energy ratio of each of the energies corresponding to the modal function components in turn to determine the signal energy ratio corresponding to each of the modal function components; Performing noise identification on the energy corresponding to each of the modal function components based on the signal energy ratio of each of the modal function classifications to determine a second noise modal function component; The target sound signal is obtained by removing the second noise mode function component from the second empirical mode function.

7. The method according to claim 1, characterized in that The step of performing spectrogram calculation on the target sound signal by discrete Fourier transform to determine the target time-frequency graph includes: Performing Fourier transform on the target sound signal to determine the sound spectrum signal; A spectrogram is drawn based on the sound spectrum signal to obtain the target time-frequency graph.

8. A sound characteristic signal recognition device, characterized in that: include: A data communication module, used for acquiring a sound signal to be processed through a communication unit; A first denoising module, configured to perform wavelet denoising on the sound signal to be processed to determine a first sound signal; A second denoising module, configured to denoise the first sound signal by empirical mode decomposition to determine a target sound signal; A spectrogram calculation module, used to perform spectrogram calculation on the target sound signal through discrete Fourier transform to determine a target time-frequency graph; The data uploading module is used to upload the target time-frequency diagram to the communication unit, perform signal recognition on the target time-frequency diagram through the communication unit, and determine the sound recognition result.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the sound characteristic signal recognition method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the sound feature signal recognition method according to any one of claims 1 to 7 when executed.

Images