Method and system for accessing electronic tag to industrial measurement device, electronic tag and medium
By time-frequency decomposition and neural network model identification of communication signals of industrial measurement devices, the problem that electronic tags cannot be adapted to industrial measurement devices is solved, automatic identification and analysis are realized, and data processing efficiency is improved.
Patent Information
- Application Number
- CN202411189107.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-12-17
AI Technical Summary
Existing electronic tags cannot automatically adapt to the communication protocols of most industrial measuring devices, resulting in inability to access efficiently.
By time-frequency decomposing the communication signal data sent by the industrial measurement device, time-domain and frequency-domain feature vectors are obtained, and communication protocol types are identified using pre-trained neural network models to analyze and display data.
It solves the problem that electronic tags cannot be automatically adapted to industrial measurement devices, realizes automatic identification and data analysis of communication signals of different types of industrial measurement devices, and improves data processing efficiency.
Smart Images
Figure CN119150134A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a method, a system, an electronic tag and a medium for accessing an industrial measuring device with an electronic tag, and belongs to the technical field of Internet of Things. Background Art
[0002] Electronic tags, as an efficient information display and identification tool, can conveniently display detailed information corresponding to specific items in various occasions. Such tags usually have a microelectronic chip embedded in them, which can store information such as the name, specifications, production date, price, and other important attributes of the item. When it is necessary to obtain information about these items, the tag can be quickly read and displayed by scanning the tag with an electronic device such as a reader. The application of this technology has greatly improved the efficiency of information processing, especially in the fields of logistics, retail, warehouse management, and production line tracking.
[0003] At present, some industrial measurement devices have begun to use electronic tags for management, such as transmitters, density meters, flow meters, oxygen meters, etc. Electronic tags are used in industrial measurement devices, which can not only display product information, but also assist in displaying measurement information, with significant application effects. However, due to security and adaptability issues, industrial measurement devices of different categories, models, and manufacturers may use different communication protocols, which results in most electronic tags being unable to be well connected to industrial measurement devices for use. Summary of the invention
[0004] In order to solve the above problems existing in the prior art, the present invention proposes a method, system, electronic tag and medium for connecting an electronic tag to an industrial measurement device.
[0005] The technical solution of the present invention is as follows:
[0006] In one aspect, the present invention provides a method for connecting an electronic tag to an industrial measurement device, comprising the following steps:
[0007] Connect the electronic tag with the industrial measuring device and receive the communication signal data sent by the industrial measuring device;
[0008] Preprocessing the communication signal data to obtain a feature vector;
[0009] Input the feature vector into the pre-trained neural network model and output the communication protocol identification result corresponding to the communication signal data index;
[0010] The electronic tag analyzes the communication data transmitted by the industrial measurement device according to the communication protocol identification result, and displays the corresponding content in response to the analysis result;
[0011] The specific preprocessing method is as follows:
[0012] The communication signal data is processed in segments to obtain a time domain signal sequence after the segments, and a frequency domain signal sequence is obtained based on the time domain signal sequence;
[0013] The time domain signal sequence and the frequency domain signal sequence are convolved respectively, and the convolution results are input into the spatial attention module to output the feature vector considering the spatial attention.
[0014] As a preferred implementation, the steps of performing convolution processing on the time domain signal sequence and the frequency domain signal sequence respectively, inputting the convolution result into the spatial attention module, and outputting the feature vector considering the spatial attention are specifically as follows:
[0015] Perform one-dimensional convolution processing on each sub-signal in the time domain signal sequence, obtain the convolution eigenvalue of each time domain sub-signal, and combine the convolution eigenvalues of each time domain sub-signal into a one-dimensional matrix of time domain features;
[0016] Perform one-dimensional convolution processing on each sub-signal in the frequency domain signal sequence, obtain the convolution eigenvalue of each frequency domain sub-signal, and combine the convolution eigenvalues of each frequency domain sub-signal into a frequency domain feature one-dimensional matrix;
[0017] The one-dimensional matrix of time domain features and the one-dimensional matrix of frequency domain features are input into the spatial attention module. The spatial attention module uses the softmax function to calculate the spatial attention features, as follows:
[0018]
[0019] Among them, q ij X is the spatial attention of the i-th convolution eigenvalue in the one-dimensional matrix of time domain features to the j-th convolution eigenvalue in the one-dimensional matrix of frequency domain features; i is the i-th convolution eigenvalue in the one-dimensional matrix of time domain features; is the jth convolution eigenvalue in the transpose of the one-dimensional matrix of frequency domain features; n is the number of segments;
[0020] Output feature vector based on spatial attention features:
[0021]
[0022] Among them, σ is the set weight.
[0023] As a preferred implementation, the neural network model adopts a convolutional neural network, and the pre-training method of the neural network model is:
[0024] Acquire communication signal sample data sent using different types of communication protocols;
[0025] Performing sample multiplication processing on the communication signal sample data so that the ratio of the number of communication signal sample data of each type of communication protocol is maintained within a preset value;
[0026] Preprocessing the communication signal sample data after sample proliferation to obtain sample feature vectors, and adding labels corresponding to the communication protocol type to each sample feature vector to form a training sample set;
[0027] The training sample set is input into the neural network model, and the parameters of the neural network model are adjusted according to the error between the predicted value output by the neural network model and the corresponding label value to complete the pre-training.
[0028] As a preferred implementation, the method for obtaining a frequency domain signal sequence based on a time domain signal sequence specifically adopts short-time Fourier transform, specifically:
[0029] Convert the time domain signal sequence to the frequency domain and calculate the spectrum amplitude:
[0030]
[0031] Where S(f, t) is the spectrum amplitude, f is the frequency axis, t is the time axis, T is the total length of the time domain signal sequence; τ is the time window, corresponding to the sequence number of the sub-signal in the time domain signal sequence; x t (τ) is the time domain signal of the time window τ at time t; δ(τ) is the window function, e is a natural constant, f s is the sampling frequency.
[0032] In another aspect, the present invention provides a system for accessing an electronic tag to an industrial measurement device, characterized in that it includes:
[0033] A data receiving module is used to connect the electronic tag with the industrial measuring device and receive the communication signal data sent by the industrial measuring device;
[0034] A preprocessing module, used for preprocessing communication signal data to obtain a feature vector;
[0035] The protocol type identification module is used to input the feature vector into the pre-trained neural network model and output the communication protocol identification result corresponding to the communication signal data index;
[0036] A parsing module, which is disposed in the electronic tag and is used to parse the communication data transmitted by the industrial measurement device according to the communication protocol identification result, and to display corresponding content in response to the parsing result;
[0037] The method in which the preprocessing module preprocesses the communication signal data is specifically as follows:
[0038] The communication signal data is processed in segments to obtain a time domain signal sequence after the segments, and a frequency domain signal sequence is obtained based on the time domain signal sequence;
[0039] The time domain signal sequence and the frequency domain signal sequence are convolved respectively, and the convolution results are input into the spatial attention module to output the feature vector considering the spatial attention.
[0040] As a preferred implementation, the preprocessing module performs convolution processing on the time domain signal sequence and the frequency domain signal sequence respectively, inputs the convolution result into the spatial attention module, and outputs the feature vector considering the spatial attention in the following steps:
[0041] Perform one-dimensional convolution processing on each sub-signal in the time domain signal sequence, obtain the convolution eigenvalue of each time domain sub-signal, and combine the convolution eigenvalues of each time domain sub-signal into a one-dimensional matrix of time domain features;
[0042] Perform one-dimensional convolution processing on each sub-signal in the frequency domain signal sequence, obtain the convolution eigenvalue of each frequency domain sub-signal, and combine the convolution eigenvalues of each frequency domain sub-signal into a frequency domain feature one-dimensional matrix;
[0043] The one-dimensional matrix of time domain features and the one-dimensional matrix of frequency domain features are input into the spatial attention module. The spatial attention module uses the softmax function to calculate the spatial attention features, as follows:
[0044]
[0045] Among them, q ij X is the spatial attention of the i-th convolution eigenvalue in the one-dimensional matrix of time domain features to the j-th convolution eigenvalue in the one-dimensional matrix of frequency domain features; i is the i-th convolution eigenvalue in the one-dimensional matrix of time domain features; is the jth convolution eigenvalue in the transpose of the one-dimensional matrix of frequency domain features; n is the number of segments;
[0046] Output feature vector based on spatial attention features:
[0047]
[0048] Among them, σ is the set weight.
[0049] As a preferred implementation, the neural network model adopts a convolutional neural network, and the pre-training method of the neural network model is:
[0050] Acquire communication signal sample data sent using different types of communication protocols;
[0051] Performing sample multiplication processing on the communication signal sample data so that the ratio of the number of communication signal sample data of each type of communication protocol is maintained within a preset value;
[0052] Preprocessing the communication signal sample data after sample proliferation to obtain sample feature vectors, and adding labels corresponding to the communication protocol type to each sample feature vector to form a training sample set;
[0053] The training sample set is input into the neural network model, and the parameters of the neural network model are adjusted according to the error between the predicted value output by the neural network model and the corresponding label value to complete the pre-training.
[0054] As a preferred implementation, the method for the preprocessing module to obtain the frequency domain signal sequence based on the time domain signal sequence specifically adopts short-time Fourier transform, specifically:
[0055] Convert the time domain signal sequence to the frequency domain and calculate the spectrum amplitude:
[0056]
[0057] Where S(f, t) is the spectrum amplitude, f is the frequency axis, t is the time axis, T is the total length of the time domain signal sequence; τ is the time window, corresponding to the sequence number of the sub-signal in the time domain signal sequence; x t (τ) is the time domain signal of the time window τ at time t; δ(τ) is the window function, e is a natural constant, f s is the sampling frequency.
[0058] On the other hand, the present invention also provides an electronic tag, including a main body and an adapter, wherein the adapter is provided with a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the program, the method for connecting the electronic tag to an industrial measurement device as described in any embodiment of the present invention is implemented.
[0059] In another aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for connecting an electronic tag to an industrial measurement device as described in any embodiment of the present invention.
[0060] The present invention has the following beneficial effects:
[0061] The present invention performs time-frequency decomposition on the communication signal of the industrial measurement device to obtain a feature vector containing time domain features and frequency domain features, then identifies the type of communication protocol based on the feature vector and a pre-trained neural network model, and performs data analysis based on the identification result, thereby solving the problem that existing electronic tags cannot automatically adapt to most industrial measurement devices. BRIEF DESCRIPTION OF THE DRAWINGS
[0062] Figure 1 The figure is a schematic diagram of the method flow of the first embodiment of the present invention. DETAILED DESCRIPTION
[0063] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0064] It should be understood that the step numbers used in this document are only for convenience of description and are not intended to limit the order in which the steps are executed.
[0065] It should be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.
[0066] The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.
[0067] The term "and / or" means and includes any and all possible combinations of one or more of the associated listed items.
[0068] Embodiment 1:
[0069] See also Figure 1 This embodiment proposes a method for connecting an electronic tag to an industrial measurement device, which is applicable to various scenarios where electronic tags are connected to industrial measurement devices. The method specifically includes the following steps:
[0070] S100, connecting the electronic tag and the industrial measurement device to realize information exchange and data transmission between the two. In this process, the electronic tag processes and stores the received data for subsequent reading and analysis. At the same time, the electronic tag also sends the data to be transmitted to the industrial measurement device to realize two-way communication. The communication signal data sent by the industrial measurement device is received by the electronic tag or the relay device.
[0071] S200, preprocessing the communication signal data, first filtering, denoising, normalizing and other operations are performed on the signal to ensure the accuracy and stability of the feature vector. Then feature extraction is performed to extract its core characteristics and important information, and a set of feature vectors that can characterize the characteristics of the original signal are obtained.
[0072] S300, input the feature vector into the pre-trained neural network model, and output the communication protocol identification result corresponding to the communication signal data index.
[0073] S400, the electronic tag is pre-installed with a data analysis module, and the corresponding data analysis program can be activated to perform data analysis according to the input communication protocol type. The electronic tag inputs the communication protocol identification result into the data analysis module, and then analyzes the communication data transmitted by the industrial measurement device, and displays the corresponding content in response to the analysis result. This display method can be visual, such as displaying text information on the screen, or auditory, such as reading the information content through a speech synthesizer.
[0074] The method of step S200 is specifically as follows:
[0075] S201. For communication signal data, we adopt a segmentation processing method to divide the entire signal into multiple continuous segments. By processing these segments, we can obtain a segmented time domain signal sequence. This time domain signal sequence reflects the change law of the signal in time.
[0076] On this basis, the principle of Fourier transform is further used to convert the time domain signal sequence into a frequency domain signal sequence. Fourier transform is a method of decomposing a time domain signal into multiple frequency components, which can better understand and analyze the composition of the signal.
[0077] S202: For the original time domain signal sequence and frequency domain signal sequence, they are transformed by convolution processing. By performing convolution operation on these two sequences, their key features and information can be effectively extracted. The result of the convolution processing, i.e., the convolution feature, will be passed as input to the spatial attention module.
[0078] The spatial attention module is a mechanism that can perform fine analysis and screening on the input signal sequence. It can assign different weights to each element in the sequence according to its importance and relevance. In this process, the spatial attention module will comprehensively consider the various features in the time domain signal sequence and the frequency domain signal sequence, and calculate and compare the similarity and correlation between them to obtain a weighted feature vector that can reflect the overall characteristics of the sequence.
[0079] Through such processing, we can obtain a feature vector that contains the key information of the original signal sequence and reflects the spatial attention mechanism. This feature vector can more accurately represent the characteristics of the input signal.
[0080] As a preferred implementation of this embodiment, the steps of performing convolution processing on the time domain signal sequence and the frequency domain signal sequence respectively, inputting the convolution result into the spatial attention module, and outputting the feature vector considering the spatial attention are specifically as follows:
[0081] Perform one-dimensional convolution processing on each sub-signal in the time domain signal sequence, obtain the convolution eigenvalue of each time domain sub-signal, and combine the convolution eigenvalues of each time domain sub-signal into a one-dimensional matrix of time domain features;
[0082] Perform one-dimensional convolution processing on each sub-signal in the frequency domain signal sequence, obtain the convolution eigenvalue of each frequency domain sub-signal, and combine the convolution eigenvalues of each frequency domain sub-signal into a frequency domain feature one-dimensional matrix;
[0083] The one-dimensional matrix of time domain features and the one-dimensional matrix of frequency domain features are input into the spatial attention module. The spatial attention module uses the softmax function to calculate the spatial attention features, as follows:
[0084]
[0085] Among them, q ij X is the spatial attention of the i-th convolution eigenvalue in the one-dimensional matrix of time domain features to the j-th convolution eigenvalue in the one-dimensional matrix of frequency domain features; i is the i-th convolution eigenvalue in the one-dimensional matrix of time domain features; is the jth convolution eigenvalue in the transpose of the one-dimensional matrix of frequency domain features; n is the number of segments;
[0086] Output feature vector based on spatial attention features:
[0087]
[0088] Among them, σ is the set weight.
[0089] As a preferred implementation of this embodiment, the neural network model adopts a convolutional neural network, and the pre-training method of the neural network model is:
[0090] Acquire communication signal sample data sent using different types of communication protocols;
[0091] In the classification task, if the number of samples of some categories is far less than that of other categories, it may cause the model to have poor recognition ability for these minority categories. Therefore, for the communication signal sample data of various types of communication protocols, this embodiment adopts a method of sample proliferation processing to ensure that its quantity ratio in the overall data can be maintained within a preset reasonable range. The method of sample proliferation can be oversampling. The method of sample proliferation can be oversampling or generative adversarial networks (GANs) and autoencoders. Among them, GANs generates fake samples similar to real samples by training a generator and a discriminator. The generator is responsible for generating new samples, while the discriminator is responsible for judging whether the sample is real or generated. Through continuous adversarial training, GANs can generate more and more realistic fake samples, thereby effectively increasing the size of the data set. Autoencoders are an unsupervised learning method that reconstructs original data by learning low-dimensional representations of data. By modifying the structure or training process of the autoencoder, it can generate new samples similar to the original data.
[0092] Preprocessing the communication signal sample data after sample proliferation to obtain sample feature vectors, and adding labels corresponding to the communication protocol type to each sample feature vector to form a training sample set;
[0093] The training sample set is input into the neural network model, and the parameters of the neural network model are adjusted according to the error between the predicted value output by the neural network model and the corresponding label value to complete the pre-training.
[0094] As a preferred implementation of this embodiment, the method for obtaining a frequency domain signal sequence based on a time domain signal sequence specifically adopts short-time Fourier transform, specifically:
[0095] Convert the time domain signal sequence to the frequency domain and calculate the spectrum amplitude:
[0096]
[0097] Where S(f, t) is the spectrum amplitude, f is the frequency axis, t is the time axis, T is the total length of the time domain signal sequence; τ is the time window, corresponding to the sequence number of the sub-signal in the time domain signal sequence; x t (τ) is the time domain signal of the time window τ at time t; δ(τ) is the window function, e is a natural constant, f s is the sampling frequency.
[0098] Embodiment 2:
[0099] This embodiment provides a system for connecting an electronic tag to an industrial measurement device, which includes:
[0100] A data receiving module, used to connect the electronic tag and the industrial measuring device, and receive the communication signal data sent by the industrial measuring device; this module is used to implement the function of step S100 in the first embodiment, and will not be described in detail here;
[0101] A preprocessing module, used for preprocessing the communication signal data to obtain a feature vector; this module is used to implement the function of step S200 in the first embodiment, and will not be described in detail here;
[0102] A protocol type identification module, which is used to input the feature vector into the pre-trained neural network model and output the communication protocol identification result corresponding to the communication signal data index; this module is used to implement the function of step S300 in the first embodiment, which will not be repeated here;
[0103] The parsing module is provided in the electronic tag, and is used to parse the communication data transmitted by the industrial measurement device according to the communication protocol identification result, and display the corresponding content in response to the parsing result; this module is used to implement the function of step S400 in the first embodiment, and will not be described in detail here;
[0104] The method in which the preprocessing module preprocesses the communication signal data is specifically as follows:
[0105] The communication signal data is processed in segments to obtain a time domain signal sequence after the segments, and a frequency domain signal sequence is obtained based on the time domain signal sequence;
[0106] The time domain signal sequence and the frequency domain signal sequence are convolved respectively, and the convolution results are input into the spatial attention module to output the feature vector considering the spatial attention.
[0107] As a preferred implementation of this embodiment, the preprocessing module performs convolution processing on the time domain signal sequence and the frequency domain signal sequence respectively, inputs the convolution result into the spatial attention module, and outputs the feature vector considering the spatial attention in the following steps:
[0108] Perform one-dimensional convolution processing on each sub-signal in the time domain signal sequence, obtain the convolution eigenvalue of each time domain sub-signal, and combine the convolution eigenvalues of each time domain sub-signal into a one-dimensional matrix of time domain features;
[0109] Perform one-dimensional convolution processing on each sub-signal in the frequency domain signal sequence, obtain the convolution eigenvalue of each frequency domain sub-signal, and combine the convolution eigenvalues of each frequency domain sub-signal into a frequency domain feature one-dimensional matrix;
[0110] The one-dimensional matrix of time domain features and the one-dimensional matrix of frequency domain features are input into the spatial attention module. The spatial attention module uses the softmax function to calculate the spatial attention features, as follows:
[0111]
[0112] Among them, q ij X is the spatial attention of the i-th convolution eigenvalue in the one-dimensional matrix of time domain features to the j-th convolution eigenvalue in the one-dimensional matrix of frequency domain features; i is the i-th convolution eigenvalue in the one-dimensional matrix of time domain features; is the jth convolution eigenvalue in the transpose of the one-dimensional matrix of frequency domain features; n is the number of segments;
[0113] Output feature vector based on spatial attention features:
[0114]
[0115] Among them, σ is the set weight.
[0116] As a preferred implementation of this embodiment, the neural network model adopts a convolutional neural network, and the pre-training method of the neural network model is:
[0117] Acquire communication signal sample data sent using different types of communication protocols;
[0118] Performing sample multiplication processing on the communication signal sample data so that the ratio of the number of communication signal sample data of each type of communication protocol is maintained within a preset value;
[0119] Preprocessing the communication signal sample data after sample proliferation to obtain sample feature vectors, and adding labels corresponding to the communication protocol type to each sample feature vector to form a training sample set;
[0120] The training sample set is input into the neural network model, and the parameters of the neural network model are adjusted according to the error between the predicted value output by the neural network model and the corresponding label value to complete the pre-training.
[0121] As a preferred implementation of this embodiment, the method for the preprocessing module to obtain the frequency domain signal sequence based on the time domain signal sequence specifically adopts short-time Fourier transform, specifically:
[0122] Convert the time domain signal sequence to the frequency domain and calculate the spectrum amplitude:
[0123]
[0124] Where S(f, t) is the spectrum amplitude, f is the frequency axis, t is the time axis, T is the total length of the time domain signal sequence; τ is the time window, corresponding to the sequence number of the sub-signal in the time domain signal sequence; x t (τ) is the time domain signal of the time window τ at time t; δ(τ) is the window function, e is a natural constant, f s is the sampling frequency.
[0125] Embodiment three:
[0126] This embodiment provides an electronic tag, including a main body and an adapter. The adapter is provided with a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for connecting an electronic tag to an industrial measurement device as described in any embodiment of the present invention is implemented.
[0127] Embodiment 4:
[0128] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method described in any embodiment of the present invention is implemented.
[0129] In the embodiments of the present application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can be represented by: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, c can be single or multiple.
[0130] Those of ordinary skill in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented in a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0131] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0132] In several embodiments provided in the present application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory; hereinafter referred to as: ROM), random access memory (Random Access Memory; hereinafter referred to as: RAM), disk or optical disk, and other media that can store program codes.
[0133] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.
Claims
1. A method for connecting an electronic tag to an industrial measurement device, characterized in that: The following steps are involved: Connect the electronic tag with the industrial measuring device and receive the communication signal data sent by the industrial measuring device; Preprocessing the communication signal data to obtain a feature vector; Input the feature vector into the pre-trained neural network model and output the communication protocol identification result corresponding to the communication signal data index; The electronic tag analyzes the communication data transmitted by the industrial measurement device according to the communication protocol identification result, and displays the corresponding content in response to the analysis result; The specific preprocessing method is as follows: The communication signal data is processed in segments to obtain a time domain signal sequence after the segments, and a frequency domain signal sequence is obtained based on the time domain signal sequence; The time domain signal sequence and the frequency domain signal sequence are convolved respectively, and the convolution results are input into the spatial attention module to output the feature vector considering the spatial attention.
2. The method for connecting an electronic tag to an industrial measurement device according to claim 1, characterized in that: The steps of performing convolution processing on the time domain signal sequence and the frequency domain signal sequence respectively, inputting the convolution results into the spatial attention module, and outputting the feature vector considering the spatial attention are specifically as follows: Perform one-dimensional convolution processing on each sub-signal in the time domain signal sequence, obtain the convolution eigenvalue of each time domain sub-signal, and combine the convolution eigenvalues of each time domain sub-signal into a one-dimensional matrix of time domain features; Perform one-dimensional convolution processing on each sub-signal in the frequency domain signal sequence, obtain the convolution eigenvalue of each frequency domain sub-signal, and combine the convolution eigenvalues of each frequency domain sub-signal into a frequency domain feature one-dimensional matrix; The one-dimensional matrix of time domain features and the one-dimensional matrix of frequency domain features are input into the spatial attention module. The spatial attention module uses the softmax function to calculate the spatial attention features, as follows: Among them, q ij X is the spatial attention of the i-th convolution eigenvalue in the one-dimensional matrix of time domain features to the j-th convolution eigenvalue in the one-dimensional matrix of frequency domain features; i is the i-th convolution eigenvalue in the one-dimensional matrix of time domain features; Y j T is the jth convolution eigenvalue in the transpose of the one-dimensional matrix of frequency domain features; n is the number of segments; Output feature vector based on spatial attention features: Among them, σ is the set weight.
3. The method for connecting an electronic tag to an industrial measurement device according to claim 1, characterized in that: The neural network model adopts a convolutional neural network, and the pre-training method of the neural network model is: Acquire communication signal sample data sent using different types of communication protocols; Performing sample multiplication processing on the communication signal sample data so that the ratio of the number of communication signal sample data of each type of communication protocol is maintained within a preset value; Preprocessing the communication signal sample data after sample proliferation to obtain sample feature vectors, and adding labels corresponding to the communication protocol type to each sample feature vector to form a training sample set; The training sample set is input into the neural network model, and the parameters of the neural network model are adjusted according to the error between the predicted value output by the neural network model and the corresponding label value to complete the pre-training.
4. The method for connecting an electronic tag to an industrial measurement device according to claim 1, characterized in that: The method for obtaining a frequency domain signal sequence based on a time domain signal sequence specifically adopts short-time Fourier transform, specifically: Convert the time domain signal sequence to the frequency domain and calculate the spectrum amplitude: Where S(f, t) is the spectrum amplitude, f is the frequency axis, t is the time axis, T is the total length of the time domain signal sequence; τ is the time window, corresponding to the sequence number of the sub-signal in the time domain signal sequence; x t (τ) is the time domain signal of the time window τ at time t; δ(τ) is the window function, e is a natural constant, f s is the sampling frequency.
5. A system for connecting electronic tags to industrial measuring devices, characterized in that: include: A data receiving module is used to connect the electronic tag with the industrial measuring device and receive the communication signal data sent by the industrial measuring device; A preprocessing module, used for preprocessing communication signal data to obtain a feature vector; The protocol type identification module is used to input the feature vector into the pre-trained neural network model and output the communication protocol identification result corresponding to the communication signal data index; A parsing module, which is disposed in the electronic tag and is used to parse the communication data transmitted by the industrial measurement device according to the communication protocol identification result, and to display corresponding content in response to the parsing result; The method in which the preprocessing module preprocesses the communication signal data is specifically as follows: The communication signal data is processed in segments to obtain a time domain signal sequence after the segments, and a frequency domain signal sequence is obtained based on the time domain signal sequence; The time domain signal sequence and the frequency domain signal sequence are convolved respectively, and the convolution results are input into the spatial attention module to output the feature vector considering the spatial attention.
6. A system for connecting electronic tags to industrial measurement devices according to claim 5, characterized in that: The preprocessing module performs convolution processing on the time domain signal sequence and the frequency domain signal sequence respectively, inputs the convolution result into the spatial attention module, and outputs the feature vector considering the spatial attention in the following steps: Perform one-dimensional convolution processing on each sub-signal in the time domain signal sequence, obtain the convolution eigenvalue of each time domain sub-signal, and combine the convolution eigenvalues of each time domain sub-signal into a one-dimensional matrix of time domain features; Perform one-dimensional convolution processing on each sub-signal in the frequency domain signal sequence, obtain the convolution eigenvalue of each frequency domain sub-signal, and combine the convolution eigenvalues of each frequency domain sub-signal into a frequency domain feature one-dimensional matrix; The one-dimensional matrix of time domain features and the one-dimensional matrix of frequency domain features are input into the spatial attention module. The spatial attention module uses the softmax function to calculate the spatial attention features, as follows: Among them, q ij X is the spatial attention of the i-th convolution eigenvalue in the one-dimensional matrix of time domain features to the j-th convolution eigenvalue in the one-dimensional matrix of frequency domain features; i is the i-th convolution eigenvalue in the one-dimensional matrix of time domain features; Y j T is the jth convolution eigenvalue in the transpose of the one-dimensional matrix of frequency domain features; n is the number of segments; Output feature vector based on spatial attention features: Among them, σ is the set weight.
7. The system for connecting electronic tags to industrial measurement devices according to claim 5, characterized in that: The neural network model adopts a convolutional neural network, and the pre-training method of the neural network model is: Acquire communication signal sample data sent using different types of communication protocols; Performing sample multiplication processing on the communication signal sample data so that the ratio of the number of communication signal sample data of each type of communication protocol is maintained within a preset value; Preprocessing the communication signal sample data after sample proliferation to obtain sample feature vectors, and adding labels corresponding to the communication protocol type to each sample feature vector to form a training sample set; The training sample set is input into the neural network model, and the parameters of the neural network model are adjusted according to the error between the predicted value output by the neural network model and the corresponding label value to complete the pre-training.
8. The system for connecting electronic tags to industrial measurement devices according to claim 5, characterized in that: The method for the preprocessing module to obtain the frequency domain signal sequence based on the time domain signal sequence specifically adopts short-time Fourier transform, specifically: Convert the time domain signal sequence to the frequency domain and calculate the spectrum amplitude: Where S(f, t) is the spectrum amplitude, f is the frequency axis, t is the time axis, T is the total length of the time domain signal sequence; τ is the time window, corresponding to the sequence number of the sub-signal in the time domain signal sequence; x t (τ) is the time domain signal of the time window τ at time t; δ(τ) is the window function, e is a natural constant, f s is the sampling frequency.
9. An electronic tag, comprising a body and an adapter, wherein the adapter is provided with a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method for connecting an electronic tag to an industrial measurement device according to any one of claims 1 to 4 is implemented.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for connecting an electronic tag to an industrial measurement device as claimed in any one of claims 1 to 4 is implemented.