Method for predicting the lifespan of a disposable electronic tag based on big data analysis and artificial intelligence

Abstract

The present invention provides a method for predicting the lifespan of disposable electronic tags based on big data analysis and artificial intelligence, which relates to the technical field of electronic tag lifespan prediction. By using graph neural networks and generative adversarial networks to accurately preprocess historical data and working condition data, the present invention significantly improves the data quality. A multiple linear regression model and a deep learning model are respectively used to construct a lifespan prediction model and a fault prediction model, enhancing the accuracy and reliability of the models. Through optimization processing and combination of multiple models, this method can more accurately predict the lifespan and fault risks of electronic tags, thereby improving the practicality and stability of the overall solution and providing strong support for tag maintenance and decision-making in industrial and logistics management.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic tag life prediction, and specifically to a method for predicting the life of disposable electronic tags based on big data analysis and artificial intelligence. Background Art

[0002] In modern industrial and logistics management, disposable electronic tags are widely used in fields such as product tracking, inventory management, and asset monitoring. Due to their complex and variable working environments, the life and reliability of electronic tags have become an important issue. However, in the research on the life of electronic tags, there are still some significant drawbacks in traditional methods, such as: the limitations of single environmental stress and the lack of data support. Traditional methods usually conduct life prediction only under single or a few environmental stresses, and fail to fully consider the synergistic coupling effects of various stresses in complex and variable environments, resulting in the prediction results being difficult to accurately reflect the life changes of electronic tags in practical applications. At the same time, traditional methods often rely on limited experimental data or assumed premises, and these data are prone to introducing biases or losing useful information during the process of collection and processing, resulting in insufficient accuracy in model training and prediction, and limited generalization ability, which further affects the accuracy and robustness of the prediction model.

[0003] Therefore, in order to solve the problems existing in traditional methods, it is necessary to provide a method that can efficiently process complex data and accurately predict the life and failure status of disposable electronic tags.

[0004] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure, and therefore it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for predicting the life of disposable electronic tags based on big data analysis and artificial intelligence to solve the problems raised in the above background art.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] A method for predicting the life of disposable electronic tags based on big data analysis and artificial intelligence, the specific steps include:

[0008] S1: Take several groups of disposable electronic tags as samples, divide the samples into a first test group and a second test group on average for life tests, collect the historical data of all samples during the life tests, and preprocess the historical data;

[0009] S2: Construct a life prediction model and a failure prediction model, and use the historical data of the first test group and the second test group respectively to train and optimize the life prediction model and the failure prediction model;

[0010] S3: Collect the working condition data of the disposable electronic tag in its actual working environment and preprocess the working condition data;

[0011] S4: Substitute the preprocessed working condition data into the optimized life prediction model and fault prediction model in sequence to preliminarily predict the life and fault status of the disposable electronic tag, generate a risk index according to the preliminary prediction results, and finally evaluate the life of the disposable electronic tag according to the risk index.

[0012] Preferably, the historical data includes the test temperature, test humidity, test pressure, and test amplitude during the test; the working condition data includes the working temperature, working humidity, working pressure, and working amplitude in the actual working environment; the preprocessing of the historical data and the working condition data includes data cleaning, data denoising, missing value filling, and normalization processing.

[0013] Preferably, use a preset graph neural network with graph convolutional layers to perform data cleaning and data denoising on the historical data and the working condition data. The output data of the graph neural network is expressed as:

[0014] ;

[0015] In the formula represents the output data of node in the th graph convolutional layer, represents the activation function of the graph neural network, represents the set of neighbor nodes of node , represents the index of the neighbor node, represents the normalization coefficient, represents the th weight matrix of the graph convolutional layer;

[0016] Substitute the output data of the graph neural network into a preset adversarial network model for missing value filling. The adversarial network model is expressed as:

[0017] ;

[0018] In the formula represents the loss function of the adversarial network model, where , respectively represent the discriminator and the generator of the adversarial network model, , respectively represent the distribution of the output data of the graph neural network and the distribution of the noise, represents the noise, , respectively represent the expected value of the distribution of the output data of the graph neural network and the expected value of the distribution of the noise;

[0019] After the missing value filling is completed, the output data of the adversarial network model is normalized, and the calculation method is:

[0020] ;

[0021] In the formula represents the normalized data, , , respectively represent the output data of the adversarial network model, and the maximum and minimum values of the output data of the adversarial network model;

[0022] Then, use the preset label generation algorithm to label the historical data after the normalization process, classify the historical data of each sample according to the fault type generated in the life test, and number and sort the fault labels based on the severity of the fault type.

[0023] Preferably, the logic for constructing the life prediction model and the fault prediction model is:

[0024] Construct a life prediction model for the disposable electronic tag based on the multiple linear regression model, and construct a fault prediction model for the disposable electronic tag based on the deep learning model and the softmax function;

[0025] Divide the historical data of the first test group and the second test group into the first training set, the first validation set, the second training set, and the second validation set according to the ratio of 8:2 respectively;

[0026] Use the first training set and the first validation set to train and optimize the life prediction model, and use the second training set and the second validation set to train and optimize the fault prediction model.

[0027] Preferably, the life prediction model is expressed as:

[0028] ;

[0029] In the formula represents the predicted life of the disposable electronic tag, represents the model parameters of the life prediction model, , , , respectively represent temperature, humidity, pressure, and amplitude, represents the preset first margin.

[0030] Preferably, the construction logic of the fault prediction model is:

[0031] Input the normalized data into the deep learning model to calculate its unnormalized score, and the unnormalized score is expressed as:

[0032] ;

[0033] Then substitute the unnormalized score into the softmax function to convert it into a predicted probability distribution, and the predicted probability distribution is expressed as:

[0034] ;

[0035] ;

[0036] In the formula represents the predicted probability of the th disposable electronic tag within the th fault tag, represents the unnormalized score of the th disposable electronic tag within the th fault tag, represents the number of fault tags, represents the predicted probability distribution of the th disposable electronic tag.

[0037] Preferably, when training the life prediction model and the fault prediction model, calculate the loss according to the loss functions of the life prediction model and the fault prediction model, where:

[0038] The loss function of the life prediction model is expressed as:

[0039] ;

[0040] The loss function of the fault prediction model is expressed as:

[0041] ;

[0042] In the formula , respectively represent the losses of the life prediction model and the fault prediction model, , are respectively the number of samples in the first training set and the second training set, , respectively represent the predicted life and the actual life of the th disposable electronic tag, represents the actual probability of the th disposable electronic tag within the th fault tag;

[0043] Use the backpropagation algorithm to calculate the gradients of the loss functions of the life prediction model and the fault prediction model respectively. Then, use the gradient descent method to update the life prediction model and the fault prediction model, and substitute the first validation set and the second validation set into the life prediction model and the fault prediction model for optimization. When the loss of the life prediction model is less than the preset first threshold , and the loss of the fault prediction model is less than the preset second threshold , it is considered that the optimization of the life prediction model and the fault prediction model is completed.

[0044] Preferably, the logic for risk assessment of disposable electronic tags is as follows:

[0045] Substitute the working condition data of the disposable electronic tags into the life prediction model and the fault prediction model respectively to obtain the corresponding predicted life and predicted probability distribution;

[0046] Generate a risk index based on the predicted life and the predicted probability distribution. The calculation method is:

[0047] ;

[0048] wherein represents the risk index of the th disposable electronic tag;

[0049] Compare the risk index with the preset risk threshold. When the following conditions are met:

[0050] , it is considered that the th disposable electronic tag is of low risk and the predicted life is relatively long;

[0051] , it is considered that the th disposable electronic tag is of medium risk and the predicted life is medium;

[0052] , it is considered that the th disposable electronic tag is of high risk and the predicted life is relatively short;

[0053] wherein , respectively represent the upper and lower limits of the risk threshold interval, and .

[0054] Compared with the prior art, the beneficial effects of the present invention are:

[0055] The present invention significantly improves the data quality by precisely preprocessing historical data and working condition data using graph neural networks and generative adversarial networks. A multiple linear regression model and a deep learning model are respectively used to construct a life prediction model and a fault prediction model, enhancing the accuracy and reliability of the models. Through optimization processing and combination of multiple models, this method can more accurately predict the life and fault risks of electronic tags, thereby improving the practicability and stability of the overall solution and providing strong support for label maintenance and decision-making in industrial and logistics management. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a schematic diagram of the overall method flow of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] To make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with specific embodiments.

[0058] It should be noted that unless otherwise defined, the technical terms or scientific terms used in the present invention should have the ordinary meanings understood by those with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar terms used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. The terms such as "including" or "comprising" mean that the elements or objects appearing before this term cover the elements or objects listed after this term and their equivalents, without excluding other elements or objects. The terms such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left", "right" are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0059] Embodiment:

[0060] Please refer to Figure 1 , the present invention provides a technical solution:

[0061] A method for predicting the life of disposable electronic tags based on big data analysis and artificial intelligence, the specific steps include:

[0062] S1: Using a number of disposable electronic tags as samples, dividing the samples into a first test group and a second test group on average for life tests, collecting the historical data of all samples in the life tests, and preprocessing the historical data.

[0063] The historical data includes the test temperature, test humidity, test pressure, and test amplitude during the test. The preprocessing of the historical data includes data cleaning, data denoising, missing value filling, and normalization processing.

[0064] During preprocessing, a graph neural network with a preset number of graph convolutional layers is used to clean and denoise historical data and working condition data. The output data of the graph neural network is expressed as:

[0065] ;

[0066] where represents the output data of node in the -th graph convolutional layer, represents the activation function of the graph neural network, represents the set of neighbor nodes of node , represents the index of the neighbor node, represents the normalization coefficient, represents the weight matrix of the -th graph convolutional layer. Here, the node represents an observation state or data point of a disposable electronic tag in a life test or an actual working environment. For example, the node can represent a set of environmental parameters such as temperature, humidity, pressure, vibration, etc. at a certain moment. These nodes form a graph structure to capture and represent the complex relationships between environmental factors. After being processed by the graph neural network, the accuracy and consistency of the data are improved, providing a more reliable input for subsequent life and fault prediction. Specifically, assuming that the environmental parameters collected at time point are temperature , humidity , pressure and amplitude , then the corresponding node feature can be represented as the vector . Substituting this vector into the GNN through the graph structure, after layers of convolution, the denoised feature can be obtained.

[0067] The graph neural network, that is, GNN, can utilize the graph structure information of the data to effectively capture the complex associations and patterns in the data, so it performs excellently in processing multi-dimensional and unstructured data. Compared with traditional denoising methods, GNN can more accurately retain useful information and remove noise.

[0068] Substitute the output data of the graph neural network into a preset adversarial network model for missing value filling. The adversarial network model is expressed as:

[0069] ;

[0070] where represents the loss function of the adversarial network model, where , respectively represent the discriminator and the generator of the adversarial network model, , respectively represent the distribution of the output data of the graph neural network and the distribution of the noise, represents the noise, , respectively represent the expected value of the distribution of the output data of the graph neural network and the expected value of the distribution of the noise.

[0071] In the adversarial network model, that is, GAN, this term calculates the expected value of the logarithm of the output probability of the discriminator for the input data of the graph neural network, that is, the real data, one term calculates the expected value of the logarithm of the output probability of the discriminator for the noise, that is, the generated data. The goal of the discriminator is to maximize the sum of these two parts so that the discriminator can more accurately identify real data and generated data, while the goal of the generator is to minimize this part so that the probability that the generated data is considered real data by the discriminator is as large as possible, thereby generating more realistic data to deceive the discriminator. GAN realizes missing value filling through the way of generating confrontation, and can generate more complete information closer to the real data. Compared with simple mean filling or interpolation method, the generator and discriminator of GAN are optimized simultaneously, improving the accuracy of filling.

[0072] After the missing value filling is completed, the output data of the adversarial network model is normalized, and the calculation method is:

[0073] ;

[0074] In the formula represents the normalized data, , , respectively represent the output data of the adversarial network model, and the maximum and minimum values of the output data of the adversarial network model. Combining the normalization process after the adversarial network output ensures the stability and consistency when the data is input into the model, and improves the effect of subsequent model training.

[0075] Here is an example to illustrate. For the vector when using the graph neural network, after layers of convolution, the denoised feature is obtained, and then the denoised feature is input into the adversarial network model for missing value filling. Assuming the data after missing value filling is , then at this time is equivalent to the output data of the adversarial network model, which is also equivalent to , the data after normalization can be expressed as .

[0076] Then, use the preset label generation algorithm to label the historical data after normalization. According to the fault types generated by the samples in the life test, generate different fault labels to classify the historical data of each sample, and sort and number the fault labels based on the severity of the fault types. Automatically generate fault labels and number them, and sort them based on the severity of the fault types, which can effectively support the training of subsequent classification models and improve the model's recognition ability for different fault types.

[0077] In this step, the preset label generation algorithm can be implemented by a combined model of the K-Means clustering analysis method plus a custom sorting algorithm. The K-Means clustering analysis method can group similar fault features in the historical data, and in the dataset after normalization, apply the clustering algorithm to identify the naturally distributed groups in the data. Each group represents a fault type, and finally assign each data to a specific clustering group to initially identify the possible fault types and obtain the corresponding fault labels. In the above embodiment, assume that the dataset after normalization is , , where represents the number of samples in the dataset. Execute the K-Means algorithm on the dataset, set the number of clusters to 3, and randomly select 3 data as the initial cluster centers, denoted as respectively, and calculate the Euclidean distance between each data and the cluster center , and the subscript represents the index of the cluster center, and the calculation method is expressed as:

[0078] ;

[0079] In the formula represents the index of the dimension. In this embodiment, since there are four features: temperature, humidity, pressure, and amplitude, the number of dimensions is 4. Assign each data to the nearest cluster center, that is, complete one cluster classification, and then update each cluster. Update the cluster center to the mean value of all data in the cluster, and the calculation method is expressed as:

[0080] ;

[0081] In the formula represents the data set in the cluster. Repeat the cluster classification and cluster center update for iteration. When the cluster center no longer changes significantly, it is considered convergent, and the clustering label corresponding to each sample can be output, that is, the fault label.

[0082] The custom sorting algorithm can use numerical scores or weighting rules to evaluate the severity of each fault. Each fault label is assigned a severity score and numbered in ascending order. The specific numerical scores or weighting rules can be obtained by combining expert scoring and domain knowledge. Suppose the fault labels include C1 overheating temperature, C2 unstable humidity, and C3 sensor drift, and the corresponding severity scores are S1 = 3, S2 = 2, and S3 = 1, respectively, which represent the severity of the three faults. Then, the fault labels can be sorted according to the severity scores and renumbered accordingly.

[0083] Here is a simple example to illustrate. Suppose there is a batch of disposable electronic tags during the life test. Due to sensor failures, some data is missing, and the collected data is affected by environmental noise. Traditional methods may use the mean to fill in the missing values and perform simple filtering to remove noise. This may introduce biases during the filling process and lose valid information during noise removal. In this invention, through precise data preprocessing and missing value filling, the high quality of the input data is ensured, which can improve the performance of the life prediction model and the fault prediction model, thereby enhancing the accuracy and robustness of the prediction model. Moreover, noise is effectively removed and missing values are filled during the preprocessing stage, enabling the model to maintain a stable performance when dealing with data under different environments and working conditions, thus enhancing the generalization ability of the model.

[0084] S2: Construct a life prediction model and a fault prediction model, and use the historical data of the first test group and the second test group to train and optimize the life prediction model and the fault prediction model respectively.

[0085] The logic for constructing the life prediction model and the fault prediction model is as follows:

[0086] Construct a life prediction model for disposable electronic tags based on a multiple linear regression model, and construct a fault prediction model for disposable electronic tags based on a deep learning model and the softmax function;

[0087] Divide the historical data of the first test group and the second test group into a first training set, a first validation set, a second training set, and a second validation set at a ratio of 8:2 respectively;

[0088] Use the first training set and the first validation set to train and optimize the life prediction model, and use the second training set and the second validation set to train and optimize the fault prediction model.

[0089] The samples are divided into two groups for different model trainings respectively, making full use of data resources, improving the pertinence of the model. The separately optimized models can be adjusted and extended according to different application scenarios, enabling the solution to maintain good performance in various environments. Compared with the traditional method of using a single model for unified prediction, the present invention can take into account the requirements of both lifespan and failure, thus predicting the lifespan and failure types of electronic tags more accurately and supporting more effective decision-making and maintenance plans.

[0090] The lifespan prediction model is expressed as:

[0091] ;

[0092] In the formula represents the predicted lifespan of a disposable electronic tag, represents the model parameters of the lifespan prediction model, , , , respectively represent temperature, humidity, pressure and amplitude, represents a preset first margin, and the specific size of the first margin can be set according to the user's experience in model training.

[0093] The construction logic of the failure prediction model is:

[0094] The normalized data is input into the deep learning model to calculate its unnormalized score, and the unnormalized score is expressed as:

[0095] ;

[0096] The unnormalized score represents the score of the model for a specific sample in a specific category during forward propagation. It is the internal evaluation value of the model for the sample belonging to this category. By performing a softmax transformation on these scores, the prediction probability of each sample in each category can be obtained.

[0097] Then, the unnormalized score is substituted into the softmax function to be converted into a prediction probability distribution, and the prediction probability distribution is expressed as:

[0098] ;

[0099] ;

[0100] In the formula represents the th disposable electronic tag in the th failure tag, represents the th disposable electronic tag in the th failure tag, Indicates the number of fault labels, Indicates the predicted probability distribution of the

[0101] Here is a simple example for illustration. Suppose there are 3 types of fault labels, labeled as fault label No. 1, No. 2, and No. 3 according to severity, and the corresponding fault types are poor contact, partial damage, and overall damage respectively. 、 、 It means that the probabilities of the sample No. 1 having poor contact, partial damage, and overall damage are 10%, 70%, and 20% respectively. The predicted probability distribution of the first sample can be expressed as 。

[0102] The loss function of the life prediction model is expressed as:

[0103] ;

[0104] The loss function of the fault prediction model is expressed as:

[0105] ;

[0106] In the formula 、 respectively represent the losses of the life prediction model and the fault prediction model, 、 are the numbers of samples in the first training set and the second training set respectively, 、 respectively represent the predicted life and actual life of the th disposable electronic label, represents the th actual probability of the

[0107] When training the life prediction model and the fault prediction model, calculate the losses according to the loss functions of the life prediction model and the fault prediction model, where:

[0108] Use the backpropagation algorithm to calculate the gradients of the loss functions of the life prediction model and the fault prediction model respectively, then use the gradient descent method to update the life prediction model and the fault prediction model, and substitute the first validation set and the second validation set into the life prediction model and the fault prediction model for optimization. When the loss of the life prediction model is less than the preset first threshold and the loss When it is considered that the optimization of the life prediction model and the fault prediction model is completed, both the first threshold and the second threshold here can be set according to the user's experience in model training.

[0109] Life prediction is a multiple regression task, and its goal is to explore the influence of environmental factors such as temperature, humidity, pressure, and amplitude on the life of disposable electronic tags. Therefore, a multiple linear regression model is used to construct the life prediction model, and the root mean square error is used as the loss function of the life prediction model to minimize the gap between the predicted life and the actual life. It can be understood that temperature, humidity, pressure, and amplitude in the life prediction model are a generalized concept, that is, when the input data is historical data, temperature, humidity, pressure, and amplitude correspond to the test temperature, test humidity, test pressure, and test amplitude in the historical data; when the data is working condition data, temperature, humidity, pressure, and amplitude correspond to the working temperature, working humidity, working pressure, and working amplitude in the working condition data.

[0110] Fault prediction is a classification task. Therefore, a method of coupling a deep learning model and a softmax function is used to construct the fault prediction model, and cross-entropy loss is used as the loss function of the fault prediction model to maximize the consistency between the predicted probability distribution and the actual probability distribution.

[0111] In this step, by applying multiple linear regression to life prediction and combining a deep learning model and a softmax function for fault prediction, the accuracy and reliability of the prediction can be significantly improved. Training and optimizing the models for different tasks separately ensures the comprehensiveness and adaptability of the solution, thereby enhancing the practicality and effectiveness of the overall solution.

[0112] S3: Collect the working condition data of the disposable electronic tag in its actual working environment and preprocess the working condition data.

[0113] The working condition data includes the working temperature, working humidity, working pressure, and working amplitude in the actual working environment. Similar to the preprocessing method of historical data, the preprocessing of the working condition data also includes data cleaning, data denoising, missing value filling, and normalization processing.

[0114] S4: Substitute the preprocessed working condition data into the optimized life prediction model and fault prediction model in turn to make a preliminary prediction of the life and fault status of the disposable electronic tag, generate a risk index according to the preliminary prediction result, and make a final evaluation of the life of the disposable electronic tag according to the risk index.

[0115] The logic for risk assessment of the disposable electronic tag is as follows:

[0116] Substitute the operating condition data of the disposable electronic tag into the life prediction model and the fault prediction model respectively to obtain the corresponding predicted life and predicted probability distribution;

[0117] Generate a risk index based on the predicted life and predicted probability distribution, and the calculation method is:

[0118] ;

[0119] In the formula represents the risk index of the th disposable electronic tag;

[0120] Compare the risk index with the preset risk threshold. When the following condition is met:

[0121] , it is considered that the th disposable electronic tag is of low risk and the predicted life is relatively long;

[0122] , it is considered that the th disposable electronic tag is of medium risk and the predicted life is medium;

[0123] , it is considered that the th disposable electronic tag is of high risk and the predicted life is relatively short;

[0124] In the formula , respectively represent the upper and lower limits of the risk threshold interval, and .

[0125] It can be seen from the calculation formula of the risk index that the risk index is directly proportional to the severity of the failure of the disposable electronic tag and inversely proportional to the predicted life of the disposable electronic tag, reflecting the ability of the disposable electronic tag to work stably under actual operating conditions. The larger the risk index, the greater the probability of failure, the shorter the normal use time, and the weaker the ability to work stably under actual operating conditions; conversely, the smaller the risk index, the smaller the probability of failure, the longer the normal use time, and the stronger the ability to work stably under actual operating conditions.

[0126] To sum up, the present invention significantly improves the data quality by using graph neural networks and generative adversarial networks to accurately preprocess historical data and operating condition data. The life prediction and fault prediction models are respectively constructed by using the multiple linear regression model and the deep learning model, enhancing the accuracy and reliability of the models. Through optimization processing and combination of multiple models, this method can more accurately predict the life and fault risk of electronic tags, thereby improving the practicability and stability of the overall solution, and providing strong support for label maintenance and decision-making in industrial and logistics management.

[0127] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0128] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0129] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units. They can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0130] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application.

Claims

1. A disposable electronic tag life prediction method based on big data analysis and artificial intelligence, characterized in that: The specific steps include: S1: Take several groups of disposable electronic tags as samples, divide the samples into the first test group and the second test group for life test, collect historical data of all samples in the life test, and pre-process the historical data; S2: constructing a life prediction model and a fault prediction model, and using the historical data of the first test group and the second test group to train and optimize the life prediction model and the fault prediction model respectively; The logic for building the life prediction model and the failure prediction model is: A life prediction model for disposable electronic tags is built based on a multivariate linear regression model, and a fault prediction model for disposable electronic tags is built based on a deep learning model and a softmax function. The historical data of the first test group and the second test group are divided into the first training set, the first validation set and the second training set, and the second validation set in a ratio of 8:2 respectively; The life prediction model is trained and optimized using the first training set and the first validation set, and the fault prediction model is trained and optimized using the second training set and the second validation set; The life prediction model is expressed as: In the formula represents the predicted life of a disposable electronic tag, represents the model parameters of the life prediction model, , , , represent temperature, humidity, pressure and amplitude respectively, Indicates the preset first margin; The construction logic of the fault prediction model is: The normalized data is input into the deep learning model to calculate its unnormalized score, which is expressed as: Then substitute the unnormalized score into the softmax function to convert it into a predicted probability distribution, which is expressed as: In the formula Indicates A disposable electronic tag is The predicted probability within the fault label, Indicates A disposable electronic tag is The unnormalized scores within the fault labels, Indicates the number of fault labels, Indicates The predicted probability distribution of disposable electronic tags; S3: Collect the working condition data of the disposable electronic tag in the actual working environment and pre-process the working condition data; S4: Substitute the preprocessed working condition data into the optimized life prediction model and fault prediction model in turn, make a preliminary prediction of the life and fault status of the disposable electronic tag, generate a risk index based on the preliminary prediction results, and make a final evaluation of the life of the disposable electronic tag based on the risk index.

2. The method for predicting the life of a disposable electronic tag based on big data analysis and artificial intelligence according to claim 1 is characterized in that: The historical data include the test temperature, test humidity, test pressure, and test amplitude during the test; the operating condition data include the working temperature, working humidity, working pressure, and working amplitude under the actual working environment; the preprocessing of the historical data and operating condition data includes data cleaning, data denoising, missing value filling, and normalization processing.

3. The method for predicting the life of a disposable electronic tag based on big data analysis and artificial intelligence according to claim 2 is characterized in that: Use the preset A graph neural network with a graph convolutional layer is used to clean and denoise the historical data and working condition data. The output data of the graph neural network is expressed as: In the formula Representation Node In the The output data of the graph convolution layer, represents the activation function of the graph neural network, Representation Node The set of neighbor nodes of Represents the index of the neighbor node, represents the normalization coefficient, Indicates The weight matrix of the graph convolutional layer; Substitute the output data of the graph neural network into the preset adversarial network model to fill in the missing values. The adversarial network model is expressed as: In the formula Represents the loss function of the adversarial network model, where , Respectively represent the discriminator and generator of the adversarial network model, , They represent the distribution of the output data and the distribution of noise of the graph neural network respectively. represents noise, , They represent the expected value of the distribution of the graph neural network output data and the expected value of the distribution of noise respectively; After the missing values ​​are filled, the output data of the adversarial network model is normalized, and the calculation method is: In the formula represents normalized data, , , Respectively represent the output data of the adversarial network model, and the maximum and minimum values ​​of the output data of the adversarial network model; The preset label generation algorithm is then used to label the normalized historical data. According to the type of fault generated by the sample in the life test, different fault labels are generated to classify the historical data of each sample, and the fault labels are numbered and sorted based on the severity of the fault type.

4. The method for predicting the life of a disposable electronic tag based on big data analysis and artificial intelligence according to claim 1 is characterized in that: When training the life prediction model and the fault prediction model, the loss is calculated according to the loss function of the life prediction model and the fault prediction model, where: The loss function of the life prediction model is expressed as: The loss function of the fault prediction model is expressed as: In the formula , represent the losses of the life prediction model and the failure prediction model respectively, , are the number of samples in the first training set and the second training set respectively, , Respectively represent The predicted and actual lifespan of a disposable electronic tag. Indicates A disposable electronic tag is The actual probability of a fault label; The back propagation algorithm is used to calculate the gradient of the loss function of the life prediction model and the fault prediction model respectively, and then the gradient descent method is used to update the life prediction model and the fault prediction model. The first validation set and the second validation set are respectively substituted into the life prediction model and the fault prediction model for optimization. When the loss of the life prediction model Less than the preset first threshold , and the loss of the fault prediction model Less than the preset second threshold When , the optimization of life prediction model and failure prediction model is considered complete.

5. The method for predicting the life of a disposable electronic tag based on big data analysis and artificial intelligence according to claim 4 is characterized in that: The logic of risk assessment for disposable electronic tags is: Substitute the working condition data of the disposable electronic tag into the life prediction model and the fault prediction model respectively to obtain the corresponding predicted life and predicted probability distribution; The risk index is generated based on the predicted life span and the predicted probability distribution, and is calculated as follows: In the formula Indicates The risk index of a disposable electronic tag; Compare the risk index with the preset risk threshold, and when it meets: , think that The risk of a disposable electronic tag is low, and the predicted life span is long; , think that The disposable electronic tag is medium risk and the predicted life expectancy is medium; , think that A disposable electronic tag is high risk and has a short predicted lifespan; In the formula , Respectively represent the upper and lower limits of the risk threshold, and .

Images