Industrial time series data enhancement method and system based on time-frequency domain feature fusion

Abstract

The invention relates to an industrial time series data enhancement method and system based on time-frequency domain feature fusion. The method comprises the following steps: acquiring industrial time series data in an industrial production process; carrying out data preprocessing on the industrial time series data by combining a weighted weight adjustment strategy in a time domain and a data interpolation method; randomly slicing the pre-processed extended data and aligning the pre-processed extended data with the original data; the industrial time sequence data are converted into frequency domain data, time domain features and frequency domain features are extracted from the industrial time sequence data and the frequency domain data respectively, and the time domain features and the frequency domain features are fused. According to the method, the weighted weight adjustment, the data interpolation and the time-frequency domain feature fusion in the time domain are combined, so that the comprehensive utilization of the time sequence data in the time domain and the frequency domain is realized, the enhanced time sequence data not only comprises the dynamic change information of the data on the time axis, but also comprises the static structure information of the data on the frequency distribution, and the accuracy of the enhanced time sequence data is improved. And important data support is provided for subsequent data analysis, anomaly detection and other tasks.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an industrial time series data enhancement method and system based on time-frequency domain feature fusion. Background Art

[0002] In order to mine the potential important feature information in the data, machine learning, deep learning and other methods are currently used to analyze and process the data. In this process, the data enhancement method is a reliable way to improve model performance and generalization ability in the field of machine learning and deep learning, but it is now mainly used to process two-dimensional image data in computer vision tasks, while the data enhancement method for one-dimensional data collected and uploaded from the Internet cloud with the upload time as the index number is not widely used.

[0003] Most of the existing data enhancement schemes are data enhancement methods for two-dimensional image data. Through jittering, scaling, rotation, arrangement, cropping and other methods in image preprocessing, the original data is preprocessed to generate new data to improve the sample richness of training data in machine learning and deep learning processes.

[0004] When existing data enhancement methods for images are applied to time series data in the field of industrial manufacturing, they face the problems of data deformation and feature loss during the execution of data enhancement methods, which affects the accuracy of data feature mining. For example, flipping or rotating an image to enhance image data does not significantly destroy the meaning of the original data. Therefore, how to propose a data enhancement scheme suitable for time series data has become an urgent problem to be solved. Summary of the invention

[0005] In view of this, it is necessary to provide an industrial time series data enhancement method based on time-frequency domain feature fusion to solve the above-mentioned defects of the prior art.

[0006] In order to solve the above problems, in a first aspect, an embodiment of the present invention provides an industrial time series data enhancement method based on time-frequency domain feature fusion, comprising:

[0007] S1, collects industrial time series data in the industrial production process to obtain the original data set;

[0008] S2, combining the weighted weight adjustment strategy in the time domain and the data interpolation method, performs data preprocessing on the industrial time series data in the original data set;

[0009] S3, randomly slice the preprocessed extended data and align the sliced ​​data with the time length of the original data;

[0010] S4, converting the industrial time series data after data preprocessing and slice alignment into frequency domain data, extracting time domain features and frequency domain features from the industrial time series data and frequency domain data respectively, and fusing the time domain features and frequency domain features to obtain data enhanced industrial time series data.

[0011] Preferably, when collecting industrial time series data in an industrial production process, the method further includes:

[0012] Based on Shannon sampling theorem and prior knowledge, determine the optimal sampling frequency of industrial time series data; wherein the industrial time series data includes voltage, current, temperature, humidity and speed;

[0013] Determine the data capacity that needs to be expanded for industrial time series data based on the optimal sampling frequency.

[0014] Preferably, the weighted weight adjustment strategy and the data interpolation method in the time domain are combined to perform data preprocessing on the industrial time series data in the original data set, including:

[0015] According to the characteristics of the industrial time series data in the original data set, a weight value is assigned to each data point in the original data set;

[0016] Considering the weight value of each data point, data interpolation is performed on the industrial time series data in the original data set.

[0017] Preferably, after performing data interpolation on the industrial time series data in the original data set, the method further includes:

[0018] The data after data interpolation is optimized; the optimization includes data rotation, data jitter, data scaling and data rearrangement.

[0019] Preferably, randomly slicing the preprocessed extended data and aligning the sliced ​​data with the time length of the original data includes:

[0020] According to the time length of the industrial time series data in the original data set, the preprocessed extended data is randomly sliced ​​to obtain sliced ​​data with the same time length as the original data;

[0021] Make sure that the label of each slice data corresponds to the label of the corresponding time period in the original dataset.

[0022] Preferably, step S4 specifically includes:

[0023] Through fast Fourier transform, the industrial time series data after data preprocessing and slice alignment is converted into frequency domain data;

[0024] Extracting time domain features from the industrial time series data after data preprocessing and slice alignment, and extracting frequency domain features from the frequency domain data;

[0025] The time domain features and frequency domain features are fused to obtain data-enhanced industrial time series data.

[0026] In a second aspect, an embodiment of the present invention provides an industrial time series data enhancement system based on time-frequency domain feature fusion, including:

[0027] The data acquisition module is used to collect industrial time series data in the industrial production process to obtain the original data set;

[0028] The data preprocessing module is used to preprocess the industrial time series data in the original data set by combining the weighted weight adjustment strategy in the time domain and the data interpolation method;

[0029] A data slice alignment module is used to randomly slice the preprocessed extended data and align the sliced ​​data with the time length of the original data;

[0030] The time-frequency domain feature fusion module is used to convert the industrial time series data after data preprocessing and slice alignment into frequency domain data, extract time domain features and frequency domain features from the industrial time series data and frequency domain data respectively, and fuse the time domain features and frequency domain features to obtain data-enhanced industrial time series data.

[0031] In a third aspect, the present invention further provides an electronic device, including a memory and a processor, wherein:

[0032] The memory is used to store programs;

[0033] The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the industrial time series data enhancement method based on time-frequency domain feature fusion as described in the embodiment of the first aspect of the present invention.

[0034] In a fourth aspect, the present invention further provides a computer-readable storage medium for storing computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the industrial time series data enhancement method based on time-frequency domain feature fusion as described in the embodiment of the first aspect of the present invention.

[0035] Compared with the prior art, the industrial time series data enhancement method and system based on time-frequency domain feature fusion provided by the present invention have the following beneficial effects:

[0036] 1) The present invention improves the density and diversity of data by combining weighted weight adjustment and data interpolation methods in the time domain. Through random slicing and alignment strategies, the present invention can increase the diversity of data while maintaining the data length, avoiding the data deformation and feature loss problems that may be caused when traditional image data enhancement methods are directly applied to time series data.

[0037] 2) The present invention performs data enhancement through the method of time-frequency domain feature fusion, which can make full use of the information of the data in the time domain and frequency domain. The enhanced time series data not only contains the dynamic change information of the data on the time axis, but also contains the static structure information of the data in the frequency distribution, providing important data support for subsequent data analysis, anomaly detection and other tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 A flow chart of the industrial time series data enhancement method based on time-frequency domain feature fusion provided by the present invention;

[0039] Figure 2 A structural block diagram of the industrial time series data enhancement system based on time-frequency domain feature fusion provided by the present invention;

[0040] Figure 3 This is a structural block diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0041] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.

[0042] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0043] Traditional data enhancement schemes are mostly data enhancement methods for two-dimensional image data. Through jittering, scaling, rotation, arrangement, cropping and other methods in image preprocessing, the original data is preprocessed to generate new data to improve the sample richness of training data in machine learning and deep learning processes.

[0044] However, when existing data enhancement methods for images are applied to time series data in the field of industrial manufacturing, they face the problems of data deformation and feature loss during the execution of data enhancement methods, which affects the accuracy of data feature mining. For example, flipping or rotating an image to enhance image data does not significantly destroy the meaning of the original data. Therefore, how to propose a data enhancement scheme suitable for time series data has become an urgent problem to be solved.

[0045] In view of this, the present invention provides an industrial time series data enhancement method based on time-frequency domain feature fusion, which improves the density and diversity of data by combining weighted weight adjustment and data interpolation methods in the time domain. Through random slicing and alignment strategies, while maintaining the data length, the diversity of data is increased, avoiding the data deformation and feature loss problems that may be caused when the traditional image data enhancement method is directly applied to time series data. The following will be explained and introduced through multiple embodiments.

[0046] Figure 1 The flowchart of the industrial time series data enhancement method based on time-frequency domain feature fusion provided by the present invention is as follows. Figure 1 As shown, the industrial time series data enhancement method based on time-frequency domain feature fusion includes the following steps S1 to S4, wherein:

[0047] Step S1, collect industrial time series data in the industrial production process to obtain an original data set.

[0048] Specifically, in the industrial production process, sensors deployed on the equipment collect time series data of detection indicators such as voltage, current, temperature, humidity, speed, etc. These data are usually recorded in discrete form, and each data point contains a timestamp and the corresponding measurement value. T , which can be defined as

[0049] X T ={x 1 , x 2 , …, x i}

[0050] Where, X T Represents the measured value of the detection index X at time T, x i is the measurement value of node i in the industrial time series data, i∈[1, 2, …, T].

[0051] When collecting industrial time series data in an industrial production process, the optimal sampling frequency of the industrial time series data is determined based on Shannon's sampling theorem and prior knowledge; wherein the industrial time series data includes voltage, current, temperature, humidity and speed.

[0052] It is understandable that Shannon's sampling theorem states that in order to ensure the complete recovery of the original continuous-time signal, the sampling frequency must be at least twice the highest frequency of the signal. In industrial production, the highest frequency of the signal is usually determined by equipment characteristics and process requirements.

[0053] In this embodiment, the highest frequency fH of the signal can be obtained based on prior knowledge, that is, the highest frequency of the signal can be estimated based on the historical data of the equipment or process and expert experience. According to Shannon sampling theorem, the maximum frequency fH of the signal can be obtained to ensure the effective reconstruction of industrial time series data X T The ideal sampling frequency f, where f≥2fH, is set as the optimal sampling frequency f=2fH for industrial time series data in this embodiment, which is a constant parameter.

[0054] In a preferred embodiment of the present invention, after determining the optimal sampling frequency of the industrial time series data, the data capacity that needs to be expanded for the industrial time series data can be determined based on the optimal sampling frequency.

[0055] Specifically, for the discrete sparse time series data X that has been collected T ={x 1 , x 2 , …, x i}, the original data length is N, and the discrete sampling time series index is T = {1, 2, ..., t}. In order to meet the requirements of the optimal sampling frequency, it is necessary to expand the data capacity. According to the requirements of the ideal sampling frequency f, the newly synthesized data through the data enhancement method is defined as Its data length is expanded to N * =2N*f-f+1, the expanded time series index becomes in:

[0056]

[0057] In this embodiment, after determining the optimal sampling frequency of industrial time series data, the discrete sparse time series data in the original data set is expanded to a data capacity that meets the requirements of the optimal sampling frequency based on the optimal sampling frequency, providing a richer data foundation for subsequent data enhancement.

[0058] Step S2, combining the weighted weight adjustment strategy and data interpolation method in the time domain, performs data preprocessing on the industrial time series data in the original data set.

[0059] Step S2 may include the following steps S21 to S23, wherein:

[0060] Step S21, assigning a weight value to each data point in the original data set according to the characteristics of the industrial time series data in the original data set.

[0061] In this embodiment, different weights may be assigned according to the time distance or importance of the data points. In this embodiment, the weights of the data points are assigned according to the time distance of the data points. Recent data points are assigned higher weights because they are more able to reflect the current trend or status.

[0062] Step S22, considering the weight value of each data point, and performing data interpolation on the industrial time series data in the original data set.

[0063] In this embodiment, a variety of data interpolation methods such as linear interpolation, polynomial interpolation, and Lagrange interpolation can be used. When performing data interpolation, the weight of the data point needs to be considered. For example, when using linear interpolation or polynomial interpolation, the coefficients or parameters of the interpolation formula can be adjusted according to the weight of the data point to obtain a more accurate interpolation result.

[0064] This embodiment improves the density and diversity of data by combining weighted weight adjustment and data interpolation methods in the time domain.

[0065] Step S23, optimizing the data after data interpolation; the optimization processing includes data rotation, data dithering, data scaling and data rearrangement.

[0066] Specifically, in order to further improve the diversity of data, image preprocessing methods in computer vision (such as rotation, jitter, scaling, reordering, etc.) are introduced into the optimization processing of industrial time series data.

[0067] In this embodiment, the intermediate data generated by interpolation For optimization processing, assume that the data before industrial time series data optimization processing is The optimized data is X' T ;

[0068] Intermediate data generated by data interpolation Four data operations are performed: data rotation, data dithering, data scaling, and data rearrangement, among which:

[0069] Data rotation operation:

[0070]

[0071] Where R is the rotation parameter set.

[0072] Data jitter operation:

[0073]

[0074] Where, ∈ is the set of jitter parameters, ∈ t represents the sum of the superimposed noise at each time t.

[0075] Data scaling operations:

[0076]

[0077] Where W is the amplitude scaling operation parameter, is a sequence, when they are a uniform value α, the data are scaled synchronously; or it can be obtained by cubic spline interpolation method, first through a normal distribution with a standard deviation of σ To select the node u = {u 1 ,u 2 ,…,u t}, and then use cubic spline interpolation CS(u) to find α t :

[0078]

[0079] Data rearrangement operation:

[0080]

[0081] Where i, j, k represent the first index slice of each window, so each slice is selected only once, using slices of uniform window size, w represents the slice window size, Where n is the total number of slices.

[0082] In the intermediate data In the process of performing the four data optimization processing operations of data rotation, data dithering, data scaling and data rearrangement, each data optimization processing operation has two possibilities of being used / not used, with a total of 16 possibilities.

[0083] In this embodiment, the industrial time series data after data interpolation is optimized, thereby improving the diversity of the data while retaining the data characteristics.

[0084] Step S3, randomly slice the preprocessed extended data, and align the sliced ​​data with the time length of the original data.

[0085] In this embodiment, the preprocessed extended data is randomly sliced ​​according to the time length of the industrial time series data in the original data set to obtain sliced ​​data with the same time length as the original data; ensuring that the label of each sliced ​​data corresponds one-to-one to the label of the corresponding time period in the original data set.

[0086] Specifically, the preprocessed extended data The length is N * When creating a data set, it is necessary to align the length of the original data set. In this embodiment, the random slicing method is used for alignment. For the time series index T of the new data *, slice it into time series data segments with the same length N as the original data. Suppose a slice on the extended data is defined as: New data can be obtained The slices are:

[0087]

[0088] After the time length is aligned, each slice data S (.) The labels correspond one-to-one to the original data labels.

[0089] In this embodiment, through random slicing and alignment strategies, the present invention can increase data diversity while maintaining data length.

[0090] Step S4, converting the industrial time series data after data preprocessing and slice alignment into frequency domain data, extracting time domain features and frequency domain features from the industrial time series data and frequency domain data respectively, and fusing the time domain features and frequency domain features to obtain data enhanced industrial time series data.

[0091] Specifically, unlike image data, time series data can be transformed in the frequency domain and the time domain. In this embodiment, first, the industrial time series data after data preprocessing and slice alignment is converted into frequency domain data through fast Fourier transform;

[0092] Then, time domain features are extracted from the industrial time series data after data preprocessing and slice alignment, and frequency domain features are extracted from the frequency domain data. As an optional implementation, a neural network model can be used to extract time domain features and frequency domain features.

[0093] Finally, the time domain features and frequency domain features are fused to obtain data-enhanced industrial time series data. In this embodiment, the time domain features and frequency domain features can be fused by methods such as direct splicing, weighted summation, or the use of a neural network model. Input, use a convolutional neural network to extract frequency domain features, use a recurrent neural network or a long short-term memory network to extract time domain features, and then fuse the two through a fully connected layer. Introduce a cross-attention mechanism in the neural network model, which enables the model to pay more attention to important feature information when fusing time domain and frequency domain features.

[0094] In this embodiment, data enhancement is performed through the method of time-frequency domain feature fusion, which can make full use of the information of the data in the time domain and frequency domain. The enhanced time series data not only includes the dynamic change information of the data on the time axis, but also includes the static structure information of the data in the frequency distribution, providing important data support for subsequent data analysis, anomaly detection and other tasks.

[0095] Figure 2The structural block diagram of the industrial time series data enhancement system based on time-frequency domain feature fusion provided by the present invention is shown in FIG. Figure 2 , the industrial time series data enhancement system based on time-frequency domain feature fusion includes:

[0096] The data collection module 201 is used to collect industrial time series data in the industrial production process to obtain an original data set;

[0097] The data preprocessing module 202 is used to perform data preprocessing on the industrial time series data in the original data set by combining the weighted weight adjustment strategy and the data interpolation method in the time domain;

[0098] The data slice alignment module 203 is used to randomly slice the preprocessed extended data and align the slice data with the time length of the original data;

[0099] The time-frequency domain feature fusion module 204 is used to convert the industrial time series data after data preprocessing and slice alignment into frequency domain data, extract time domain features and frequency domain features from the industrial time series data and frequency domain data respectively, and fuse the time domain features and frequency domain features to obtain data-enhanced industrial time series data.

[0100] The industrial time series data enhancement system based on time-frequency domain feature fusion provided by the present invention executes the industrial time series data enhancement method based on time-frequency domain feature fusion provided by the above-mentioned embodiments through the above-mentioned modules. The industrial time series data enhancement method based on time-frequency domain feature fusion has been described in detail in the above-mentioned embodiments, and this embodiment will not be repeated here.

[0101] Figure 3 The structural block diagram of the electronic device provided by the present invention is as follows: Figure 3 As shown, the present invention also provides an electronic device, and the electronic device 300 can be a computing device such as a mobile terminal, a desktop computer, a notebook, a palm computer, and a server. The electronic device 300 includes a processor 301 and a memory 302, wherein the memory 302 stores an industrial time series data enhancement program 303 based on time-frequency domain feature fusion.

[0102] In some embodiments, the memory 302 may be an internal storage unit of a computer device, such as a hard disk or memory of the computer device. In other embodiments, the memory 302 may also be an external storage device of the computer device, such as a plug-in hard disk equipped on the computer device, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. Furthermore, the memory 302 may also include both an internal storage unit of the computer device and an external storage device. The memory 302 is used to store application software and various types of data installed on the computer device, such as program codes for installing the computer device. The memory 302 can also be used to temporarily store data that has been output or is to be output. In one embodiment, when the industrial time series data enhancement program 303 based on time-frequency domain feature fusion is executed by the processor 301, the following steps are implemented:

[0103] S1, collects industrial time series data in the industrial production process to obtain the original data set;

[0104] S2, combining the weighted weight adjustment strategy in the time domain and the data interpolation method, performs data preprocessing on the industrial time series data in the original data set;

[0105] S3, randomly slice the preprocessed extended data and align the sliced ​​data with the time length of the original data;

[0106] S4, converting the industrial time series data after data preprocessing and slice alignment into frequency domain data, extracting time domain features and frequency domain features from the industrial time series data and frequency domain data respectively, and fusing the time domain features and frequency domain features to obtain data enhanced industrial time series data.

[0107] In some embodiments, the processor 301 may be a central processing unit (CPU), a microprocessor or other data processing chip, used to run program codes or process data stored in the memory 302, such as executing an industrial time series data enhancement program based on time-frequency domain feature fusion.

[0108] This embodiment further provides a computer-readable storage medium, on which an industrial time series data enhancement program based on time-frequency domain feature fusion is stored. When the industrial time series data enhancement program based on time-frequency domain feature fusion is executed by a processor, the following steps are implemented:

[0109] S1, collects industrial time series data in the industrial production process to obtain the original data set;

[0110] S2, combining the weighted weight adjustment strategy in the time domain and the data interpolation method, performs data preprocessing on the industrial time series data in the original data set;

[0111] S3, randomly slice the preprocessed extended data and align the sliced ​​data with the time length of the original data;

[0112] S4, converting the industrial time series data after data preprocessing and slice alignment into frequency domain data, extracting time domain features and frequency domain features from the industrial time series data and frequency domain data respectively, and fusing the time domain features and frequency domain features to obtain data enhanced industrial time series data.

[0113] In summary, the industrial time series data enhancement method and system based on time-frequency domain feature fusion provided by the present invention have the following beneficial effects compared with the prior art:

[0114] 1) The present invention improves the density and diversity of data by combining weighted weight adjustment and data interpolation methods in the time domain. Through random slicing and alignment strategies, the present invention can increase the diversity of data while maintaining the data length, avoiding the data deformation and feature loss problems that may be caused when traditional image data enhancement methods are directly applied to time series data.

[0115] 2) The present invention performs data enhancement through the method of time-frequency domain feature fusion, which can make full use of the information of the data in the time domain and frequency domain. The enhanced time series data not only contains the dynamic change information of the data on the time axis, but also contains the static structure information of the data in the frequency distribution, providing important data support for subsequent data analysis, anomaly detection and other tasks.

[0116] The above-described embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the attached claims.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention 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 invention.

Claims

1. An industrial time series data enhancement method based on time-frequency domain feature fusion, characterized in that: include: S1, collects industrial time series data in the industrial production process to obtain the original data set; S2, combining the weighted weight adjustment strategy in the time domain and the data interpolation method, performs data preprocessing on the industrial time series data in the original data set; S3, randomly slice the preprocessed extended data and align the sliced ​​data with the time length of the original data; S4, converting the industrial time series data after data preprocessing and slice alignment into frequency domain data, extracting time domain features and frequency domain features from the industrial time series data and frequency domain data respectively, and fusing the time domain features and frequency domain features to obtain data enhanced industrial time series data.

2. The industrial time series data enhancement method based on time-frequency domain feature fusion according to claim 1 is characterized in that: When collecting industrial time series data in an industrial production process, the method further includes: Based on Shannon sampling theorem and prior knowledge, determine the optimal sampling frequency of industrial time series data; wherein the industrial time series data includes voltage, current, temperature, humidity and speed; Determine the data capacity that needs to be expanded for industrial time series data based on the optimal sampling frequency.

3. The industrial time series data enhancement method based on time-frequency domain feature fusion according to claim 1 is characterized in that: The weighted weight adjustment strategy and data interpolation method in the time domain are combined to preprocess the industrial time series data in the original data set, including: According to the characteristics of the industrial time series data in the original data set, a weight value is assigned to each data point in the original data set; Considering the weight value of each data point, data interpolation is performed on the industrial time series data in the original data set.

4. The industrial time series data enhancement method based on time-frequency domain feature fusion according to claim 3 is characterized in that: After data interpolation of the industrial time series data in the original data set, it also includes: The data after data interpolation is optimized; the optimization includes data rotation, data jitter, data scaling and data rearrangement.

5. The industrial time series data enhancement method based on time-frequency domain feature fusion according to claim 1 is characterized in that: The method of randomly slicing the preprocessed extended data and aligning the sliced ​​data with the time length of the original data includes: According to the time length of the industrial time series data in the original data set, the preprocessed extended data is randomly sliced ​​to obtain sliced ​​data with the same time length as the original data; Make sure that the label of each slice data corresponds to the label of the corresponding time period in the original dataset.

6. The industrial time series data enhancement method based on time-frequency domain feature fusion according to claim 1 is characterized in that: Step S4 specifically includes: Through fast Fourier transform, the industrial time series data after data preprocessing and slice alignment is converted into frequency domain data; Extracting time domain features from the industrial time series data after data preprocessing and slice alignment, and extracting frequency domain features from the frequency domain data; The time domain features and frequency domain features are fused to obtain data-enhanced industrial time series data.

7. An industrial time series data enhancement system based on time-frequency domain feature fusion, characterized in that: include: The data acquisition module is used to collect industrial time series data in the industrial production process to obtain the original data set; The data preprocessing module is used to preprocess the industrial time series data in the original data set by combining the weighted weight adjustment strategy in the time domain and the data interpolation method; A data slice alignment module is used to randomly slice the preprocessed extended data and align the sliced ​​data with the time length of the original data; The time-frequency domain feature fusion module is used to convert the industrial time series data after data preprocessing and slice alignment into frequency domain data, extract time domain features and frequency domain features from the industrial time series data and frequency domain data respectively, and fuse the time domain features and frequency domain features to obtain data-enhanced industrial time series data.

8. An electronic device, It is characterized in that comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the industrial time series data enhancement method based on time-frequency domain feature fusion as described in any one of claims 1 to 6 above.

9. A computer-readable storage medium, characterized in that: Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the industrial time series data enhancement method based on time-frequency domain feature fusion as described in any one of claims 1 to 6 above.

Images