Data processing method and device, electronic equipment and storage medium

Abstract

The invention provides a data processing method and device, electronic equipment and a storage medium, and belongs to the technical field of noise processing, and the method comprises the steps: obtaining original monitoring data in a preset scene; performing modal decomposition operation on the original monitoring data to obtain modal decomposition data representing local characteristics of the original monitoring data at different times and different frequency scales; performing data energy calculation on the modal decomposition data to obtain energy characteristic data containing an entropy value and an entropy value increment; and performing data reconstruction operation on the energy characteristic data to obtain noise reduction reconstruction data. According to the method, the data extraction operation can be effectively executed, the noise reduction function is realized, and the accuracy of the data can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of noise processing, and in particular to a data processing method, device, electronic device and storage medium. Background Art

[0002] The automotive cockpit monitoring technology is an important part of the modern automobile intelligence, which aims to monitor the in-car environment and the comfort and safety of passengers in real time. With the rapid development of automotive electronics and sensor technology, the amount of data generated in the cockpit is increasing, which provides a rich basis for real-time monitoring and analysis. However, these monitoring data are often affected by a variety of noise sources, including engine noise, road noise, and interference from electronic equipment. Summary of the invention

[0003] The present application provides a data processing method, device, electronic device and storage medium, aiming to solve the problem that vehicle cabin monitoring data is affected by multiple noise sources.

[0004] In a first aspect, the present application provides a data processing method, the method comprising:

[0005] Obtain the original monitoring data under the preset scenario;

[0006] Performing a modal decomposition operation on the original monitoring data to obtain modal decomposition data representing local characteristics of the original monitoring data at different times and different frequency scales;

[0007] Performing data energy calculation on the modal decomposition data to obtain energy characteristic data including entropy value and entropy value increment;

[0008] A data reconstruction operation is performed on the energy characteristic data to obtain noise-reduced reconstructed data.

[0009] In a second aspect, the present application further provides a data processing device, the device comprising:

[0010] A data acquisition module is used to obtain raw monitoring data under a preset scenario;

[0011] A modal decomposition module, used to perform a modal decomposition operation on the original monitoring data to obtain modal decomposition data representing local characteristics of the original monitoring data at different times and different frequency scales;

[0012] An energy calculation module, used to perform data energy calculation on the modal decomposition data to obtain energy characteristic data including an entropy value and an entropy value increment, wherein the entropy value represents the characteristics of the modal decomposition data itself, and the entropy value increment represents the relationship between the modal decomposition data;

[0013] The data reconstruction module is used to perform a data reconstruction operation on the energy characteristic data to obtain noise-reduced reconstructed data.

[0014] In a third aspect, the present application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the steps of the data processing method as described in any one of the first aspects are implemented.

[0015] In a fourth aspect, the present application provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the data processing method as described in any one of the first aspects.

[0016] The present application provides a data processing method, device, electronic device and storage medium, which aims to solve the problem that the monitoring data of the car cockpit is affected by multiple noise sources. The method first obtains the original monitoring data under the preset scene of the car cockpit, and then performs a modal decomposition operation on it. This operation can decompose the original monitoring data into modal decomposition data representing local characteristics at different times and different frequency scales, so as to preliminarily distinguish the part that may contain noise and the useful part. Then, the data energy is calculated for the modal decomposition data to obtain energy characteristic data including entropy value and entropy value increment, wherein the entropy value reflects the characteristics of the modal decomposition data itself, and the entropy value increment reflects the relationship between the data. Signals from different sources (including noise and useful signals) have differences in energy characteristics, and noise signals can be further identified through this calculation. Finally, a data reconstruction operation is performed based on the energy characteristic data. The effective data can be reconstructed by setting a threshold to filter the effective data, discard the data part that may be noise, retain the useful part and reconstruct the noise reduction reconstructed data.

[0017] Therefore, in terms of implementation, the data processing method of the present application can achieve multi-scale analysis through modal decomposition, thereby mining the multi-dimensional information of the data; energy calculation can achieve the quantification of data characteristics and relationships; and data reconstruction can effectively extract data and achieve noise reduction. In terms of application value, this method helps to improve the accuracy of the data, and it has noise reduction effect evaluation indicators and is evaluable. In addition, this method is highly versatile and widely applicable to the processing of original monitoring data in a variety of preset scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present application or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 is a flow chart of the data processing method provided by the present application;

[0020] Figure 2 It is a structural schematic diagram of a data processing device provided by the present application;

[0021] Figure 3 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0023] The terms "first", "second", etc. in the specification and claims of this application 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 terms used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein.

[0024] In order to solve the problem that the monitoring data of the car cabin is affected by multiple noise sources, the present application proposes a data processing method, device, electronic device and storage medium, which can improve the accuracy of the monitoring data in the car cabin. For the monitoring of the cabin environment (such as temperature, humidity, air quality) and the monitoring of the physiological state of passengers (such as heart rate, breathing), the noise-reduced monitoring data can more accurately reflect the actual situation, which helps the intelligent control system to make reasonable decisions, such as adjusting the air-conditioning temperature and ventilation volume. In terms of automobile safety-related monitoring, such as driver fatigue status monitoring, abnormal situation monitoring in the cabin, etc., accurate data can reduce misjudgment, timely warning and take measures. At the same time, the method helps to optimize the overall performance of the car cabin. In improving the functional experience of the smart cockpit, accurate monitoring data can make personalized services such as adjusting the seat position and music playback according to the passenger comfort state more accurate. And when the car cabin monitoring data interacts with external devices, accurate data can ensure the effectiveness of the interaction, such as sending the car cabin monitoring data to the cloud for optimizing the cabin design or remote maintenance services.

[0025] Combine the following Figure 1-Figure 3 The present application describes a data processing method, device, electronic device and storage medium.

[0026] Please refer to Figure 1 , Figure 1It is a flow chart of the data processing method provided by the present application. A data processing method includes:

[0027] S110, obtaining original monitoring data under a preset scenario.

[0028] Specifically, taking the car cockpit as an example, it is necessary to obtain the original monitoring data of the car cockpit. The original monitoring data is the original signal data collected in the car cockpit environment, which contains various useful information and noise components. Due to the complex and changeable characteristics of the car cockpit environment, the collected data presents nonlinear and non-stationary signals.

[0029] S120, performing a modal decomposition operation on the original monitoring data to obtain modal decomposition data representing local characteristics of the original monitoring data at different times and different frequency scales.

[0030] Specifically, the modal decomposition operation is a processing method for the original monitoring data. The original monitoring data contains complex information, and the modal decomposition operation aims to deeply analyze this data. It decomposes the original monitoring data to obtain modal decomposition data, which can show the local characteristics of the original monitoring data at different times and frequency scales. On the time scale, it can not only separate short-term characteristics such as short noise pulses in the car cabin or seat adjustment friction sounds, but also decompose long-term characteristics such as vibration noise changes during the entire operating cycle of the engine; on the frequency scale, low-frequency characteristics such as low-frequency engine hum, wind noise, and high-frequency characteristics such as high-frequency whistling sounds generated by electronic equipment can be separated from the original monitoring data, and the obtained modal decomposition data provides a more targeted data basis for subsequent work such as car cabin noise control.

[0031] S130, performing data energy calculation on the modal decomposition data to obtain energy characteristic data including entropy value and entropy value increment.

[0032] Specifically, the data energy is calculated for the modal decomposition data. The purpose of this calculation is to obtain energy characteristic data, which includes entropy value and entropy increment. The entropy value can reflect the characteristics of the modal decomposition data itself, such as the complexity and disorder of the data; while the entropy increment reflects the relationship between the modal decomposition data, which helps to reveal the degree of correlation between different modal decomposition data. Through such calculations, we can deeply understand the intrinsic characteristics of the modal decomposition data and the relationship between them from the perspective of energy characteristics.

[0033] S140, performing a data reconstruction operation on the energy characteristic data to obtain noise-reduced reconstructed data.

[0034] Specifically, data reconstruction refers to the use of energy characteristic data to reconstruct the original monitoring data through a specific algorithm to obtain noise-reduced reconstructed data. The purpose of this step is to remove the noise components in the original data and retain useful signal information, making the reconstructed data clearer and more accurate, which is convenient for subsequent analysis and application.

[0035] In some embodiments, the data processing method further includes:

[0036] S150, performing a data analysis operation on the noise reduction and reconstruction data to obtain a quantitative evaluation result of the noise reduction effect and an analysis result of the noise component.

[0037] Specifically, the data analysis operation is to conduct an in-depth analysis of the denoised and reconstructed data, with the goal of obtaining quantitative evaluation results of the denoising effect and detailed analysis results of the noise components through a specific algorithm. This step can quantitatively reflect the quality improvement of the data after denoising, and further reveal the characteristics and sources of the noise components, thus providing a scientific basis for optimizing denoising strategies and improving data quality.

[0038] The above steps S110 to S150 are described in detail below.

[0039] In some embodiments, in the above S120, the step of performing a modal decomposition operation on the original monitoring data to obtain modal decomposition data representing local characteristics of the original monitoring data at different time and frequency scales includes:

[0040] S121, preprocessing the original monitoring data, and obtaining modal decomposition data by detecting local extreme points, calculating envelopes, and performing iterative processing operations.

[0041] Specifically, the preprocessing of the original monitoring data refers to preliminary smoothing, the purpose of which is to remove the influence of high-frequency noise, which can be done by moving average filtering. Then, the modal decomposition data is obtained by detecting local extreme points, calculating envelopes, and performing iterative processing operations. This preprocessing moving average filtering operation can initially optimize the original monitoring data, reduce fluctuations and noise interference in the data, and lay the foundation for subsequent detection, calculation, and iterative operations, so that the final modal decomposition data can better reflect the local characteristics of the original monitoring data at different times and frequency scales.

[0042] Exemplarily, the modal decomposition operation can adopt time-varying empirical mode decomposition (TVFEMD), and the modal decomposition data is represented in the form of intrinsic mode function (IMF). TVFEMD is a signal processing method specifically used for the analysis of nonlinear and non-stationary signals. Compared with traditional signal processing technology, TVFEMD has stronger adaptability and flexibility, and can better extract the characteristic information of the signal. By decomposing the original monitoring data into a series of IMFs, these IMFs can reflect the local characteristics of the signal on different time scales. And by screening and iterating the original monitoring data multiple times, its inherent frequency components are extracted. It can be seen that the modal decomposition of the present application introduces time-varying characteristics, so that the decomposition process can be dynamically adjusted according to the signal characteristics, and is suitable for signals such as automobile cockpit monitoring data in a dynamic environment.

[0043] Exemplarily, step S121 specifically includes:

[0044] S1211, detecting local extreme value points in the original monitoring data, where the local extreme value points are the maximum and minimum values ​​of the data in a local range.

[0045] Specifically, the local range can be a small interval in the data sequence. In this interval, there are maximum values ​​and minimum values. The maximum value is a value larger than the surrounding data in this local interval, while the minimum value is a value smaller than the surrounding data in this local interval.

[0046] S1212, constructing an envelope based on the detected local extreme points, and for each local extreme point, constructing an upper envelope and a lower envelope respectively: the upper envelope is a curve formed by connecting all local maximum points; the lower envelope is a curve formed by connecting all local minimum points.

[0047] Specifically, in step S1211, local extreme value points have been determined, including local maximum value points and local minimum value points. For each such local extreme value point, an upper envelope and a lower envelope are constructed respectively. Among them, the upper envelope is a curve formed by connecting all the local maximum value points, which reflects the upper limit trend of the data in the local range. The lower envelope, which is opposite to it, is a curve formed by connecting all the local minimum value points, which reflects the lower limit trend of the data in the local range. Constructing these envelopes helps to further analyze the fluctuation range and characteristics of the data.

[0048] S1213, calculating the average values ​​of the corresponding points of the upper envelope and the lower envelope, determining the value corresponding to the new curve formed by the average value as temporary data, and subtracting the temporary data from the original monitoring data to obtain new data.

[0049] Specifically, the average values ​​of the upper and lower envelopes at corresponding points are calculated. For example, the upper envelope has a value at a certain position, and the lower envelope also has a value at the same position. The two values ​​are averaged. Then, these average values ​​are combined into a new curve, and the value corresponding to this new curve is determined as temporary data. Finally, the temporary data is subtracted from the original monitoring data to obtain new data.

[0050] S1214, repeatedly performing the operations of detecting local extreme points and constructing envelopes to continuously iterate until a preset stop condition is reached, and all the new data finally obtained are modal decomposition data.

[0051] It can be understood that in the above steps S1211 to S1214, all modal decomposition data are extracted by step-by-step iteration. In each iteration process, the signal is detected for local extreme points and the envelope is calculated to obtain modal decomposition data. This process will continue until the preset stop condition is met, for example, the preset stop condition is to reach a preset number of decompositions or the remaining data reaches a stable state. The characteristic of modal decomposition data is that it has a relatively good time-frequency localization capability and can effectively capture the instantaneous changes and non-stationary characteristics of the signal. In addition, in the modal decomposition process, by dynamically adjusting the parameters in each iteration step, it is possible to adaptively respond to changes in signal characteristics. Combined with the actual application scenarios of automobile cockpit monitoring data, modal decomposition can not only extract important frequency components, but also identify noise from different sources by analyzing the energy distribution of modal decomposition data. This lays the foundation for the subsequent entropy value increment calculation, making the noise reduction process more accurate and efficient.

[0052] In some embodiments, in the above step S130, the step of performing data energy calculation on the modal decomposition data to obtain energy characteristic data including entropy value and entropy value increment includes:

[0053] S1301, performing an integration operation of square values ​​on modal decomposition data to perform energy calculation, and obtaining an energy value of the modal decomposition data.

[0054] Specifically, the energy calculation for the modal decomposition data can be performed by performing an integration operation on the square value of the modal decomposition data. Integrating the square value of the modal decomposition data can be understood as accumulating and summing the squared data within a certain interval. Through such an operation, the energy value of the modal decomposition data can be obtained. The energy value is a quantitative representation of the energy contained in the modal decomposition data, which reflects the size of the energy contained in the modal decomposition data under its own characteristics.

[0055] S1302, performing standardization processing on the energy value corresponding to each modal decomposition data so as to satisfy the preset probability distribution requirement.

[0056] Specifically, the energy value corresponding to each modal decomposition data obtained in step S1301 is processed. The standardization process is to make these energy values ​​meet the preset probability distribution requirements. Probability distribution is a way to describe the probability law of random variable values. Each modal decomposition data has its corresponding energy value, but these energy values ​​may be in different ranges or have different distribution characteristics. Through standardization, these energy values ​​are adjusted so that they can meet the preset probability distribution requirements.

[0057] S1303, calculating the entropy value of each modal decomposition data according to the standardized energy value, and calculating the entropy value difference between adjacent modal decomposition data to obtain the entropy value increment.

[0058] Specifically, first, the entropy value of each modal decomposition data is calculated based on the energy value standardized in step S1302. The entropy value of each modal decomposition data is calculated using the standardized energy value, and the entropy value can reflect a certain degree of complexity or uncertainty within the modal decomposition data. Next, the entropy value difference between adjacent modal decomposition data is calculated to obtain the entropy value increment. Adjacent modal decomposition data may be associated or have a changing trend to some extent, and calculating the difference in their entropy values ​​can reflect the amplitude or characteristics of the change. The entropy value increment can better understand the dynamic changes of the modal decomposition data in the sequence.

[0059] It can be understood that in the above steps S1301 to S1303, the entropy increment is used for the analysis of non-stationary signals. In the analysis of the car cockpit monitoring data, the present application can more accurately evaluate the noise components therein by introducing the entropy increment, thereby achieving a more efficient noise reduction effect. The entropy increment is the entropy change caused by the change in signal energy distribution at a certain moment, which can reflect the dynamic characteristics of the signal in the time domain. In terms of car cockpit monitoring, the entropy increment can effectively identify and separate noise signals from valid signals, and enhance the adaptability and flexibility of the noise reduction algorithm. In the entire noise reduction process, the introduction of the entropy increment not only improves the accuracy of signal processing, but also provides important feature information for subsequent data reconstruction. By analyzing the entropy increment, the dynamic characteristics of each component in the monitoring data can be better understood, and then a targeted noise reduction strategy can be implemented to improve the quality and reliability of the car cockpit monitoring data.

[0060] In some embodiments, in the above step S140, the step of performing a data reconstruction operation on the energy characteristic data to obtain noise-reduced reconstructed data includes:

[0061] S1401, setting a threshold value according to the entropy value increment in the energy characteristic data.

[0062] Specifically, this step is to set a threshold value based on the entropy increment in the energy characteristic data. The entropy increment reflects the change of entropy value in the energy characteristic data, and the change of entropy value is related to the complexity, uncertainty and other characteristics of the data. By analyzing the entropy increment, a specific value can be determined as a threshold value, which can be a rule for dividing the energy characteristic data.

[0063] S1402, determining the data whose entropy increment is lower than the threshold as data containing noise components and discarding it, while retaining the data whose entropy increment is higher than the threshold.

[0064] Specifically, the data is screened using the previously set threshold. The entropy increment reflects a certain characteristic change of the data. When the entropy increment is lower than the set threshold, it is determined that the data contains noise components. This is because a lower entropy increment may mean that the complexity or uncertainty of the data is lower, and it exhibits noise-like disorder or weak regularity in the entire data system, so it is determined as data containing noise components and discarded. Data with an entropy increment higher than the threshold is retained. A higher entropy increment indicates that the data has a relatively higher complexity or uncertainty, contains more valid information, and is more in line with expectations for valid data, so the data should be retained for use in subsequent operations (such as the data reconstruction operation in step S1403).

[0065] S1403, performing an addition operation on the retained entropy value increments to complete the data reconstruction operation and obtain the monitoring data after noise reduction.

[0066] Specifically, the entropy increments retained after screening in step S1402 are added to complete the reconstruction of the data. The entropy increments that are considered to contain more valid information are screened out through the threshold, and the parts that are judged to be noise are discarded. These retained entropy increments are added and spliced ​​together into a series of useful fragments to reconstruct the noise-reduced monitoring data. The purpose of this step is to remove the noise components in the original monitoring data so that the reconstructed data is purer and closer to the actual valid signal.

[0067] S1404, obtaining a noise reduction effect evaluation index according to the original monitoring data and the noise-reduced monitoring data, wherein the noise reduction and reconstruction data includes the noise-reduced monitoring data and the noise reduction effect evaluation index, and the noise reduction effect evaluation index includes a signal-to-noise ratio and a mean square error.

[0068] Specifically, this step is mainly to quantitatively evaluate the noise reduction effect. First, the original monitoring data is unprocessed data, which contains various information and possible noise. Then, the noise-reduced monitoring data is obtained through step S1403. By comparing the original monitoring data and the noise-reduced monitoring data, the noise reduction effect evaluation index can be obtained. The noise reduction effect evaluation index includes signal-to-noise ratio (SNR) and mean square error (MSE), etc. The signal-to-noise ratio is the power ratio of the signal to the noise, which reflects the proportional relationship between the useful information and the noise in the signal. If the signal-to-noise ratio is higher, it means that the proportion of the useful signal relative to the noise in the noise-reduced monitoring data is larger, which means that the noise reduction effect is better. The mean square error is the average value of the square of the deviation between the observed value and the true value. By calculating the mean square error between the original monitoring data and the noise-reduced monitoring data, the error size generated in the noise reduction process can be intuitively understood. The smaller the mean square error, the better the noise reduction effect. The noise reduction reconstruction data includes the noise-reduced monitoring data and the noise reduction effect evaluation index, so that the result after the noise reduction operation and the quality of the result can be fully reflected.

[0069] In some embodiments, in step 150, a data analysis operation is performed on the denoised reconstructed data to obtain a quantitative evaluation result of the denoising effect and an analysis result of the noise component, including:

[0070] S1501, comparing the changes in the signal-to-noise ratio and the mean square error in the noise reduction effect evaluation index to quantify the noise reduction effect and obtain a quantitative evaluation result of the noise reduction effect.

[0071] Specifically, the noise reduction effect is quantitatively evaluated by analyzing the changes in the signal-to-noise ratio and mean square error in the noise reduction effect evaluation index. The two indicators, signal-to-noise ratio and mean square error, have been calculated in step S1404, and they respectively reflect the proportional relationship between the signal and the noise after noise reduction and the size of the error generated during the noise reduction process. By comparing the changes in the signal-to-noise ratio and mean square error before and after noise reduction, the noise reduction effect can be quantitatively evaluated. For example, if the signal-to-noise ratio after noise reduction is significantly improved and the mean square error is significantly reduced, then it can be considered that the noise reduction effect is good. On the contrary, if the signal-to-noise ratio is not significantly improved or the mean square error is not reduced much, it may indicate that the noise reduction effect is not good.

[0072] S1502, respectively draw signal analysis representation diagrams of the original monitoring data and the monitoring data after noise reduction to obtain analysis results of the noise component, the signal analysis representation diagram including a waveform diagram and a spectrum diagram.

[0073] Specifically, this step mainly analyzes the noise component by drawing a specific signal analysis characterization diagram. Signal analysis characterization diagrams are drawn for the original monitoring data and the monitoring data after noise reduction, respectively. The signal analysis characterization diagrams may include two types, for example, a waveform diagram and a spectrum diagram. The waveform diagram can intuitively show how the signal changes over time. For the waveform diagram of the original monitoring data, various fluctuations contained in the signal can be seen, including irregular fluctuations caused by noise. The waveform diagram of the monitoring data after noise reduction shows the fluctuation of the signal after noise reduction. By comparing the waveform diagrams of the two, the degree of influence of noise on the waveform of the original signal can be intuitively seen. For example, if the waveform becomes smoother, it means that the noise has been removed to a certain extent. The spectrum diagram analyzes the signal from the perspective of frequency. The spectrum diagram of the original monitoring data will show the various frequency components contained in the signal, and the noise will usually appear as some messy frequency distribution in the spectrum diagram. The spectrum diagram of the monitoring data after noise reduction reflects the changes in the frequency components of the signal after noise reduction. By comparing the spectrum diagrams of the two, it can be determined which frequency components are considered to be noise and removed, thereby obtaining the analysis results of the noise components. This analysis result helps to further understand the characteristics of noise and the processing effect of the noise reduction method on noise of different frequencies.

[0074] In summary, the data processing method provided by the present application is based on modal decomposition to obtain modal decomposition data, and performs energy calculation on it to obtain entropy increment. First, in terms of quantitative analysis, the signal-to-noise ratio (SNR) and mean square error (MSE) are calculated to evaluate the noise reduction effect. For example, after adopting this method, the SNR can be increased by about 30%-50%, and the MSE is significantly reduced, which indicates that the ratio of effective components to noise components is enhanced, the interference noise is effectively removed, and the main characteristics of the signal are retained. From the visualization results, the waveform and spectrum before and after noise reduction are drawn. The high-frequency noise of the waveform before noise reduction is obvious and the signal recognizability is low. After noise reduction, the waveform is smoother, the noise peak is reduced, and the signal contour is clear. In the spectrum diagram, the frequency components of the signal after noise reduction are more concentrated and the low-frequency energy is enhanced, which intuitively shows the algorithm's suppression effect on high-frequency noise.

[0075] Compared with traditional methods, this method is superior to wavelet transform and Fourier transform in terms of signal-to-noise ratio and mean square error. Traditional methods have limitations in processing non-stationary signals and cannot effectively capture instantaneous changes, while methods based on modal decomposition can dynamically adjust and accurately extract signal features, with obvious advantages in noise reduction. Finally, in terms of noise component identification, analyzing the entropy value increment can clarify the relationship between noise components and signal characteristics. By setting a threshold and eliminating energy characteristic data below the threshold, the signal can be accurately reconstructed, which not only improves the noise reduction effect and data quality, but also provides a reliable basis for subsequent data processing.

[0076] The data processing device provided by the present application is described below. The data processing device described below and the data processing method described above can be referenced to each other.

[0077] Please refer to Figure 2 , Figure 2 2 is a schematic diagram of the structure of a data processing device provided by the present application. A data processing device 200 includes a data acquisition module 201 , a modal decomposition module 202 , an energy calculation module 203 and a data reconstruction module 204 .

[0078] Exemplarily, the data acquisition module 201 is used to acquire original monitoring data under a preset scenario.

[0079] Exemplarily, the modal decomposition module 202 is used to perform a modal decomposition operation on the original monitoring data to obtain modal decomposition data representing local characteristics of the original monitoring data at different time and frequency scales.

[0080] Exemplarily, the energy calculation module 203 is used to perform data energy calculation on the modal decomposition data to obtain energy characteristic data including an entropy value and an entropy value increment.

[0081] Exemplarily, the data reconstruction module 204 is used to perform a data reconstruction operation on the energy characteristic data to obtain noise-reduced reconstructed data.

[0082] Exemplarily, the data processing device 200 further includes a noise assessment module 205 (not shown in the figure), and the noise assessment module 205 is used to:

[0083] A data analysis operation is performed on the noise reduction and reconstruction data to obtain a quantitative evaluation result of the noise reduction effect and an analysis result of the noise component.

[0084] Exemplarily, the modal decomposition module 202 is further configured to:

[0085] The original monitoring data is preprocessed, and the modal decomposition data is obtained by detecting local extreme points, calculating envelopes, and performing iterative processing operations.

[0086] Exemplarily, the modal decomposition module 202 is further configured to:

[0087] Detecting local extreme value points in the original monitoring data, wherein the local extreme value points are the maximum and minimum values ​​of the data in a local range;

[0088] Constructing envelopes based on the detected local extreme value points, and constructing upper envelopes and lower envelopes for each local extreme value point: the upper envelope is a curve formed by connecting all local maximum value points; the lower envelope is a curve formed by connecting all local minimum value points;

[0089] Calculate the average value of the corresponding points of the upper envelope and the lower envelope, determine the value corresponding to the new curve formed by the average value as temporary data, and subtract the temporary data from the original monitoring data to obtain new data;

[0090] The operations of detecting local extreme points and constructing envelopes are repeatedly performed to continuously iterate until a preset stop condition is reached, and all the new data finally obtained are the modal decomposition data.

[0091] Exemplarily, the energy calculation module 203 is further used for:

[0092] Performing an integration operation of square values ​​on the modal decomposition data to perform energy calculation, thereby obtaining an energy value of the modal decomposition data;

[0093] Perform standardization processing on the energy value corresponding to each modal decomposition data so that it meets the preset probability distribution requirements;

[0094] The entropy value of each modal decomposition data is calculated according to the standardized energy value, and the entropy value difference between adjacent modal decomposition data is calculated to obtain the entropy value increment;

[0095] The entropy value represents the characteristics of the modal decomposition data itself, and the entropy value increment represents the relationship between the modal decomposition data.

[0096] Exemplarily, the data reconstruction module 204 is further used for:

[0097] Setting a threshold value according to the entropy value increment in the energy characteristic data;

[0098] The data with entropy increment lower than the threshold value is judged as data containing noise components and is discarded, while the data with entropy increment higher than the threshold value is retained;

[0099] The retained entropy value increments are added to complete the data reconstruction operation and obtain the noise-reduced monitoring data.

[0100] Exemplarily, the data reconstruction module 204 is further used for:

[0101] A noise reduction effect evaluation index is obtained according to the original monitoring data and the noise-reduced monitoring data, wherein the noise reduction and reconstruction data includes the noise-reduced monitoring data and the noise reduction effect evaluation index, and the noise reduction effect evaluation index includes a signal-to-noise ratio and a mean square error.

[0102] Exemplarily, the noise assessment module 205 is further configured to:

[0103] Comparing the changes in the signal-to-noise ratio and the mean square error in the noise reduction effect evaluation index to quantify the noise reduction effect and obtain a quantitative evaluation result of the noise reduction effect;

[0104] The signal analysis characterization diagrams of the original monitoring data and the monitoring data after noise reduction are respectively drawn to obtain the analysis result of the noise component, wherein the signal analysis characterization diagram includes a waveform diagram and a spectrum diagram.

[0105] Exemplarily, the modal decomposition data is represented in the form of intrinsic modal functions.

[0106] It should be noted here that the above-mentioned data processing device provided in the embodiment of the present application can implement all the method steps implemented in the above-mentioned method embodiment, and can achieve the same technical effect. The parts and beneficial effects that are the same as those in the method embodiment will not be described in detail here.

[0107] Figure 3 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application, such as Figure 3 As shown, the electronic device may include: a processor (Processor) 310, a communication interface (Communications Interface) 320, a memory (Memory) 330 and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other through the communication bus 340. The processor 310 may call the logic instructions in the memory 330 to execute the data processing method.

[0108] In addition, the logic instructions in the above-mentioned memory 330 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art, and the computer software product is stored in a storage medium, including several instructions to ensure that a computer device (which can be a personal computer, a server, or a network device, etc.) executes all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk.

[0109] On the other hand, the present application also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the data processing methods provided by the above methods.

[0110] On the other hand, the present application also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the computer program is executed by a processor to perform the above-mentioned data processing methods.

[0111] The embodiments of the present application provide an electronic device, a computer program product, and a processor-readable storage medium, on which the computer program stored enables the processor to implement all the method steps implemented in the above-mentioned method embodiments and to achieve the same technical effects. The parts and beneficial effects of this embodiment that are the same as those in the method embodiments will not be described in detail herein.

[0112] The device embodiments described above are merely illustrative, 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 may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0113] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions to ensure that a computer device (which can be a personal computer, a server, or a network device, etc.) executes the methods described in each embodiment or some parts of the embodiments.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A data processing method, characterized in that: The data processing method comprises: Obtain the original monitoring data under the preset scenario; Performing a modal decomposition operation on the original monitoring data to obtain modal decomposition data representing local characteristics of the original monitoring data at different times and different frequency scales; Performing data energy calculation on the modal decomposition data to obtain energy characteristic data including entropy value and entropy value increment; A data reconstruction operation is performed on the energy characteristic data to obtain noise-reduced reconstructed data.

2. The data processing method according to claim 1, characterized in that: The data processing method further includes: A data analysis operation is performed on the noise reduction and reconstruction data to obtain a quantitative evaluation result of the noise reduction effect and an analysis result of the noise component.

3. The data processing method according to claim 1, characterized in that: The steps of performing a modal decomposition operation on the original monitoring data to obtain modal decomposition data representing local characteristics of the original monitoring data at different times and different frequency scales include: The original monitoring data is preprocessed, and the modal decomposition data is obtained by detecting local extreme points, calculating envelopes, and performing iterative processing operations.

4. The data processing method according to claim 3, characterized in that: The steps of preprocessing the original monitoring data and obtaining the modal decomposition data by detecting local extreme points, calculating envelopes, and performing iterative processing operations include: Detecting local extreme value points in the original monitoring data, wherein the local extreme value points are the maximum and minimum values ​​of the data in a local range; Constructing envelopes based on the detected local extreme value points, and constructing upper envelopes and lower envelopes for each local extreme value point: the upper envelope is a curve formed by connecting all local maximum value points; the lower envelope is a curve formed by connecting all local minimum value points; Calculate the average value of the corresponding points of the upper envelope and the lower envelope, determine the value corresponding to the new curve formed by the average value as temporary data, and subtract the temporary data from the original monitoring data to obtain new data; The operations of detecting local extreme points and constructing envelopes are repeatedly performed to continuously iterate until a preset stop condition is reached, and all the new data finally obtained are the modal decomposition data.

5. The data processing method according to claim 1, characterized in that: The step of performing data energy calculation on the modal decomposition data to obtain energy characteristic data including entropy value and entropy value increment comprises: Performing an integration operation of square values ​​on the modal decomposition data to perform energy calculation, thereby obtaining an energy value of the modal decomposition data; Perform standardization processing on the energy value corresponding to each modal decomposition data so that it meets the preset probability distribution requirements; The entropy value of each modal decomposition data is calculated according to the standardized energy value, and the entropy value difference between adjacent modal decomposition data is calculated to obtain the entropy value increment; The entropy value represents the characteristics of the modal decomposition data itself, and the entropy value increment represents the relationship between the modal decomposition data.

6. The data processing method according to claim 1, characterized in that: The step of performing a data reconstruction operation on the energy characteristic data to obtain noise-reduced reconstructed data comprises: Setting a threshold value according to the entropy value increment in the energy characteristic data; The data with entropy increment lower than the threshold value is judged as data containing noise components and is discarded, while the data with entropy increment higher than the threshold value is retained; The retained entropy value increments are added to complete the data reconstruction operation and obtain the noise-reduced monitoring data.

7. The data processing method according to claim 6, characterized in that: The step of performing a data reconstruction operation on the energy characteristic data to obtain noise-reduced reconstructed data comprises: A noise reduction effect evaluation index is obtained according to the original monitoring data and the noise-reduced monitoring data, wherein the noise reduction and reconstruction data includes the noise-reduced monitoring data and the noise reduction effect evaluation index, and the noise reduction effect evaluation index includes a signal-to-noise ratio and a mean square error.

8. The data processing method according to claim 2, characterized in that: The step of performing a data analysis operation on the noise reduction and reconstruction data to obtain a quantitative evaluation result of the noise reduction effect and an analysis result of the noise component includes: Comparing the changes in the signal-to-noise ratio and the mean square error in the noise reduction effect evaluation index to quantify the noise reduction effect and obtain a quantitative evaluation result of the noise reduction effect; The signal analysis characterization diagrams of the original monitoring data and the monitoring data after noise reduction are respectively drawn to obtain the analysis result of the noise component, wherein the signal analysis characterization diagram includes a waveform diagram and a spectrum diagram.

9. The data processing method according to claim 1, characterized in that: The modal decomposition data is represented in the form of intrinsic modal functions.

10. A data processing device, characterized in that: The data processing device comprises: A data acquisition module is used to obtain raw monitoring data under a preset scenario; A modal decomposition module, used to perform a modal decomposition operation on the original monitoring data to obtain modal decomposition data representing local characteristics of the original monitoring data at different times and different frequency scales; An energy calculation module, used to perform data energy calculation on the modal decomposition data to obtain energy characteristic data including entropy value and entropy value increment; The data reconstruction module is used to perform a data reconstruction operation on the energy characteristic data to obtain noise-reduced reconstructed data.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the data processing method according to any one of claims 1 to 9 are implemented.

12. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the data processing method according to any one of claims 1 to 9 are implemented.

Images