Production log analysis methods, systems, equipment, and media based on the Industrial Internet of Things

Abstract

This application discloses a method, system, device, and medium for production log analysis based on the Industrial Internet of Things (IIoT), relating to the technical field of IIoT. The method includes: extracting an environmental parameter set, a standard setting parameter set, and an actual processing parameter set corresponding to the target production line from the production logs; performing regression modeling based on the environmental parameter set, the standard setting parameter set, and the actual processing parameter set to obtain an initial SVM model; using the standard setting parameter set, the environmental parameter set, and the actual processing parameter set as model training data, and training and testing the initial SVM model based on the model training data to obtain an actual processing parameter prediction model; acquiring the environmental parameters of the area where the target production line is located and the processing parameters of the products, setting standard setting parameter constraints, calling the actual processing parameter prediction model, and selecting a set of standard setting parameters based on an annealing algorithm. This application has the effect of improving production efficiency.

Description

Technical Field

[0001] This application relates to the technical field of industrial Internet of Things (IIoT), and in particular to methods, systems, devices, and media for production log analysis based on IIoT. Background Technology

[0002] With the continuous application of Industrial Internet of Things (IIoT) technology, manufacturing enterprises are able to collect large amounts of production log data in real time, including environmental parameters, equipment status, and processing information. In-depth analysis of this data not only enables comprehensive monitoring of the production process but also helps identify potential problems in key areas, thereby optimizing processes, reducing production costs, and improving overall efficiency. Simultaneously, data-driven production log analysis can also uncover factors affecting product quality, providing strong support for intelligent quality control and ensuring that products meet design standards and customer requirements.

[0003] However, current production log-based analysis still has certain limitations. Many enterprises struggle to accurately predict and dynamically adjust production parameters in practice, resulting in a lack of flexibility and adaptability in production processes, and consequently, reduced production efficiency and product quality. Furthermore, the lack of intelligent analysis and optimization methods that integrate multi-source data makes it difficult to detect and correct anomalies and deviations in the production process in a timely manner, leading to a generally low level of production. Therefore, there is an urgent need to introduce more advanced algorithms and models to achieve continuous optimization of production processes and quality. Summary of the Invention

[0004] To improve production levels, this application provides a method, system, device, and medium for production log analysis based on the Industrial Internet of Things.

[0005] Firstly, this application provides a production log analysis method based on the Industrial Internet of Things, employing the following technical solution:

[0006] A production log analysis method based on the Industrial Internet of Things (IIoT) is applied to an IIoT system, which includes a management platform, a sensor network platform, and an object platform connected in sequence. The method is executed by the management platform and includes:

[0007] Obtain the production logs of the target production line, and extract the environmental parameter set, standard setting parameter set, and actual processing parameter set corresponding to the target production line based on the production logs;

[0008] Based on the SVM algorithm, regression modeling is performed according to the environmental parameter set, the standard setting parameter set and the actual processing parameter set to obtain the initial SVM model;

[0009] The standard parameter set, environmental parameter set, and actual processing parameter set are used as model training data. The initial SVM model is trained and tested based on the model training data to obtain the actual processing parameter prediction model.

[0010] The system obtains environmental parameters and product processing parameters for the target production line area, sets standard parameter limits, calls the actual processing parameter prediction model, and filters a set of standard parameters based on the annealing algorithm so that the predicted values ​​corresponding to the standard parameters and environmental parameters meet the preset conditions.

[0011] By adopting the above technical solution, the production logs of the target production line are first obtained, and the corresponding environmental parameter set, standard setting parameter set, and actual processing parameter set are extracted from the production logs. Then, based on the SVM algorithm, regression modeling is performed on the environmental parameter set, standard setting parameter set, and actual processing parameter set to obtain an initial SVM model. The standard setting parameter set, environmental parameter set, and actual processing parameter set are then used as model training data, and the initial SVM model is trained and tested based on the model training data to obtain an actual processing parameter prediction model. Finally, the environmental parameters of the area where the target production line is located and the processing parameters of the product are obtained, standard setting parameter constraints are set, the actual processing parameter prediction model is called, and a set of standard setting parameters is selected based on the annealing algorithm so that the predicted values ​​corresponding to the set of standard setting parameters and environmental parameters meet the preset conditions of the processing parameters. Through the above method, data-driven production parameter prediction and optimization are realized. By establishing an accurate regression model and using the annealing algorithm to select processing parameters that meet production requirements, the stability of the production process, the rationality of parameter configuration, and the overall production efficiency are improved, thereby improving the production level and the intelligence level of the production process.

[0012] Optionally, the steps of extracting the environmental parameter set, standard setting parameter set, and actual processing parameter set corresponding to the target production line from the production logs include:

[0013] Preprocessing operations are performed on the production logs to obtain corresponding preprocessed data. The preprocessing operations include structured operations.

[0014] Based on predefined parameter categories, and by extracting and classifying parameters from preprocessed data according to regular expressions, corresponding environmental parameter data, standard setting parameter data, and actual processing parameter data are obtained.

[0015] Data filtering and cleaning are performed on environmental parameter data, standard setting parameter data, and actual processing parameter data to obtain the environmental parameter set, standard setting parameter set, and actual processing parameter set corresponding to the target production line.

[0016] By adopting the above technical solution, in order to extract various parameters from the production log, the production log is first preprocessed to obtain corresponding preprocessed data. The preprocessing operation includes structured operations. Then, based on predefined parameter categories, the preprocessed data is extracted and classified according to regular expressions to obtain corresponding environmental parameter data, standard setting parameter data, and actual processing parameter data. Finally, the environmental parameter data, standard setting parameter data, and actual processing parameter data are filtered and cleaned to obtain the environmental parameter set, standard setting parameter set, and actual processing parameter set corresponding to the target production line.

[0017] Optionally, the steps for obtaining an initial SVM model based on the SVM algorithm, using environmental parameter sets, standard parameter sets, and actual processing parameter sets, include:

[0018] The exclusivity of each parameter in the standard parameter set and the actual processing parameter set is evaluated by chi-square test to determine the first and second exclusive parameter groups.

[0019] A linear transformation is performed on the first set of exclusive parameters to obtain the first fusion parameters, and a linear transformation is performed on the second set of exclusive parameters to obtain the second fusion parameters;

[0020] Based on the SVM algorithm, and using regression modeling based on the first fusion parameter, the second fusion parameter, the environmental parameter set, the standard setting parameter set, and the actual processing parameter set, an initial SVM model is obtained.

[0021] By adopting the above technical solution, in order to obtain the initial SVM model, the exclusivity of each parameter in the standard parameter set and the actual processing parameter set is evaluated by chi-square test to determine the first exclusive parameter set and the second exclusive parameter set. Then, the first exclusive parameter set is linearly transformed to obtain the first fusion parameter, and the second exclusive parameter set is linearly transformed to obtain the second fusion parameter. Finally, based on the SVM algorithm, regression modeling is performed according to the first fusion parameter, the second fusion parameter, the environmental parameter set, the standard parameter set, and the actual processing parameter set to obtain the initial SVM model.

[0022] Optionally, the steps of training and testing the initial SVM model based on the model training data to obtain the actual processing parameter prediction model include:

[0023] The model training data is divided into a training set and a test set according to a preset ratio;

[0024] The hyperparameters of the initial SVM model are iterated using a random network search algorithm, and the performance of the initial SVM model under different hyperparameters is evaluated using k-fold cross-validation to obtain the optimal hyperparameters of the initial SVM model. The root mean square error (RMSE) and coefficient of determination (R²) are then compared. 2 As an evaluation indicator;

[0025] The initial SVM model is trained using the training set to obtain a trained SVM model.

[0026] The trained SVM model is tested on the test set, and the error is judged according to the evaluation index to determine whether it is within the preset range. If so, the trained SVM model is used as the actual processing parameter prediction model.

[0027] By adopting the above technical solution, in order to obtain the actual processing parameter prediction model, the model training data is divided into training set and test set according to a preset ratio. Then, the hyperparameters of the initial SVM model are traversed according to the random network search algorithm, and the performance of the initial SVM model under different hyperparameters is evaluated by k-fold cross-validation to obtain the optimal hyperparameters of the initial SVM model. The root mean square error (RMSE) and coefficient of determination (R²) are used as evaluation indicators. Then, the initial SVM model is trained on the training set to obtain the trained SVM model. Finally, the trained SVM model is tested on the test set, and the error is judged according to the evaluation indicators to determine whether it is within the preset range. If the error is within the preset range, the trained SVM model is used as the actual processing parameter prediction model.

[0028] Optionally, the step of selecting a set of standard setting parameters based on the annealing algorithm, so that the predicted values ​​corresponding to the set of standard setting parameters and environmental parameters meet the preset conditions of the processing parameters, includes:

[0029] In the initial state, the initial temperature T, cooling rate α, and iteration number i are set, and each standard setting parameter is initialized according to the standard setting parameter constraints.

[0030] After annealing, the predicted values ​​of the standard setting parameters and environmental parameters under the current state and the processing parameters are vectorized to obtain the corresponding predicted vector and target vector, and the vector product of the predicted vector and the target vector is calculated.

[0031] Determine whether the difference between the magnitude of the vector product and the magnitude of the target vector is less than a first preset value, and determine whether the magnitude of the vector product is less than a second preset value. If both are true, then take the corresponding standard setting parameters in the current state as a set of standard setting parameters.

[0032] Otherwise, within the standard parameter setting limits, n parameter points are randomly selected, and for each parameter point, the distance between the predicted value corresponding to the parameter point and the environmental parameter and the processing parameter is calculated.

[0033] The shortest distance is determined based on the distance, and the parameter points corresponding to the shortest distance are used as a set of standard setting parameters.

[0034] By adopting the above technical solution, in order to select a set of standard setting parameters, in the initial state, the initial temperature T, cooling rate α, and iteration number i are set, and each standard setting parameter is initialized according to the constraints of the standard setting parameters. Then, after annealing, the predicted values ​​and processing parameters corresponding to each standard setting parameter and environmental parameter in the current state are vectorized to obtain the corresponding predicted vector and target vector. The vector product of the predicted vector and the target vector is calculated. Then, it is determined whether the difference between the magnitude of the vector product and the magnitude of the target vector is less than a first preset value, and whether the magnitude of the vector product is less than a second preset value. If both are true, the standard setting parameters corresponding to the current state are taken as a set of standard setting parameters. Otherwise, n parameter points are randomly selected within the constraints of the standard setting parameters. For each parameter point, the distance between the predicted value of the parameter point and the environmental parameter and the processing parameter is calculated. Then, the shortest distance is determined based on the distance, and the parameter point corresponding to the shortest distance is taken as a set of standard setting parameters.

[0035] Optionally, after the step of selecting a set of standard setting parameters based on the annealing algorithm so that the predicted values ​​corresponding to the set of standard setting parameters and environmental parameters meet the preset conditions of the processing parameters, the method further includes:

[0036] The system generates production line control instructions based on a set of standard parameters to control the target production line to process products according to the set of standard parameters.

[0037] By adopting the above technical solution, production line control instructions are generated based on a set of standard setting parameters, thereby controlling the target production line to process products according to a set of standard setting parameters.

[0038] Optionally, the method also includes:

[0039] Obtain target environmental parameters and target processing parameters, wherein the target environmental parameters include at least one of target temperature, target humidity, target ventilation rate, target light intensity, and target pressure;

[0040] The target environmental parameters and target processing parameters are input into the actual processing parameter prediction model to obtain the corresponding predicted values ​​of the actual processing parameters.

[0041] By adopting the above technical solution, in order to predict the actual processing parameters, target environmental parameters and target processing parameters are obtained. The target environmental parameters include at least one of target temperature, target humidity, target ventilation rate, target light intensity, and target pressure. Then, the target environmental parameters and target processing parameters are input into the actual processing parameter prediction model to obtain the corresponding predicted values ​​of the actual processing parameters.

[0042] Secondly, this application also provides a production log analysis system based on the Industrial Internet of Things, which adopts the following technical solution:

[0043] The industrial IoT-based production log analysis system includes a management platform, a sensor network platform, and an object platform that are connected in sequence. The management platform is configured with:

[0044] The parameter extraction module is used to obtain the production logs of the target production line and extract the environmental parameter set, standard setting parameter set and actual processing parameter set corresponding to the target production line based on the production logs.

[0045] The model building module is used to perform regression modeling based on the SVM algorithm, according to the environmental parameter set, the standard parameter set, and the actual processing parameter set, to obtain the initial SVM model.

[0046] The model generation module is used to take the standard parameter set, environmental parameter set and actual processing parameter set as model training data, and train and test the initial SVM model based on the model training data to obtain the actual processing parameter prediction model.

[0047] The standard setting parameter filtering module is used to obtain the environmental parameters and product processing parameters of the target production line area, set standard setting parameter limits, call the actual processing parameter prediction model, and filter a set of standard setting parameters based on the annealing algorithm so that the predicted values ​​corresponding to the set of standard setting parameters and environmental parameters meet the preset conditions of the processing parameters.

[0048] Thirdly, this application also provides a computer device, which adopts the following technical solution:

[0049] A computer device includes a memory and a processor, the memory storing a computer program executable on the processor, the processor executing the computer program to implement the method described in the first aspect.

[0050] Fourthly, this application also provides a computer-readable storage medium, which adopts the following technical solution:

[0051] A computer-readable storage medium storing a computer program capable of being loaded by a processor and executing the method described in the first aspect.

[0052] In summary, this application includes at least the following beneficial technical effects: First, the production logs of the target production line are obtained, and the environmental parameter set, standard setting parameter set, and actual processing parameter set corresponding to the target production line are extracted from the production logs. Then, based on the SVM algorithm, regression modeling is performed on the environmental parameter set, standard setting parameter set, and actual processing parameter set to obtain an initial SVM model. Then, the standard setting parameter set, environmental parameter set, and actual processing parameter set are used as model training data, and the initial SVM model is trained and tested based on the model training data to obtain an actual processing parameter prediction model. Finally, the environmental parameters of the area where the target production line is located and the processing parameters of the product are obtained, standard setting parameter constraints are set, the actual processing parameter prediction model is called, and a set of standard setting parameters is selected based on the annealing algorithm so that the predicted values ​​corresponding to a set of standard setting parameters and environmental parameters meet the preset conditions with the processing parameters. Through the above method, data-driven production parameter prediction and optimization are realized. By establishing an accurate regression model and using the annealing algorithm to select processing parameters that meet production requirements, the stability of the production process, the rationality of parameter configuration, and the overall production efficiency are improved, thereby improving the production level and the intelligence level of the production process. Attached Figure Description

[0053] Figure 1 This is a schematic diagram of the overall process of an embodiment of this application.

[0054] Figure 2 This is a structural diagram of one application scenario of the system according to an embodiment of this application.

[0055] Figure 3 This is a structural diagram of another application scenario of the system according to an embodiment of this application.

[0056] Figure 4 This is a structural block diagram of the computer device described in this application. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0058] This application discloses a production log analysis method based on the Industrial Internet of Things (IIoT).

[0059] Reference Figure 1 A production log analysis method based on the Industrial Internet of Things (IIoT) is characterized by its application to an IIoT system, which includes a management platform, a sensor network platform, and an object platform connected in sequence. The method is executed by the management platform and includes:

[0060] Step S11: Obtain the production log of the target production line, and extract the environmental parameter set, standard setting parameter set, and actual processing parameter set corresponding to the target production line based on the production log.

[0061] It should be noted that in step S11, the production logs of the target production line need to be obtained first. Then, based on these logs, the corresponding environmental parameter set (such as production environment data such as temperature and humidity), standard setting parameter set (i.e. the preset process parameters of the production line) and actual processing parameter set (reflecting the parameters during actual product processing) are extracted to provide detailed basic data for subsequent production optimization, quality control and process adjustment.

[0062] Step S12: Based on the SVM algorithm, regression modeling is performed according to the environmental parameter set, the standard setting parameter set, and the actual processing parameter set to obtain the initial SVM model.

[0063] It should be noted that in step S12, the SVM (Support Vector Machine) algorithm is used, with the environmental parameter set, standard setting parameter set and actual processing parameter set extracted in step S11 as input features, to perform regression modeling on the target variable (such as product quality, processing efficiency, etc.) and train an initial SVM model. This model can be used to predict or analyze key indicators in the production process, thereby helping to optimize production parameters and improve quality.

[0064] Step S13: Use the standard setting parameter set, environmental parameter set, and actual processing parameter set as model training data, and train and test the initial SVM model based on the model training data to obtain the actual processing parameter prediction model.

[0065] It should be noted that in step S13, the standard setting parameter set, environmental parameter set and actual processing parameter set extracted in step S11 are used as training data. The initial SVM regression model is trained using these data, and then the model performance is verified through the test set. Finally, a reliable model for predicting actual processing parameters is obtained, thereby realizing the optimization and control of the production process.

[0066] Step S14: Obtain the environmental parameters of the target production line area and the processing parameters of the product, set standard setting parameter limits, call the actual processing parameter prediction model, and filter a set of standard setting parameters based on the annealing algorithm so that the predicted values ​​corresponding to a set of standard setting parameters and environmental parameters meet the preset conditions of the processing parameters.

[0067] It should be noted that in step S14, the environmental parameters of the target production line area and the processing parameters of the product are first obtained, and the corresponding standard setting parameter limits are set according to the standard setting parameter set or the standard limits of each equipment in the target production line. Then, the previously trained actual processing parameter prediction model is called, and the annealing algorithm is used to optimize the search within the standard limit range to select a set of optimal standard setting parameters so that the predicted values ​​of this set of parameters and the environmental parameters meet the preset production conditions, thereby achieving the optimal configuration of production parameters.

[0068] In the above implementation, the production logs of the target production line are first obtained, and the environmental parameter set, standard setting parameter set, and actual processing parameter set corresponding to the target production line are extracted from the production logs. Then, based on the SVM algorithm, regression modeling is performed on the environmental parameter set, standard setting parameter set, and actual processing parameter set to obtain an initial SVM model. The standard setting parameter set, environmental parameter set, and actual processing parameter set are then used as model training data, and the initial SVM model is trained and tested based on the model training data to obtain an actual processing parameter prediction model. Finally, the environmental parameters of the area where the target production line is located and the processing parameters of the product are obtained, standard setting parameter constraints are set, the actual processing parameter prediction model is called, and a set of standard setting parameters is selected based on the annealing algorithm so that the predicted values ​​corresponding to the set of standard setting parameters and environmental parameters meet the preset conditions of the processing parameters. Through the above method, data-driven production parameter prediction and optimization are realized. By establishing an accurate regression model and using the annealing algorithm to select processing parameters that meet production requirements, the stability of the production process, the rationality of parameter configuration, and the overall production efficiency are improved, thereby improving the production level and the intelligence level of the production process.

[0069] As a further implementation of the method, the step of extracting the environmental parameter set, standard setting parameter set, and actual processing parameter set corresponding to the target production line based on the production log includes:

[0070] Step S21: Perform preprocessing operations on the production logs to obtain corresponding preprocessed data, wherein the preprocessing operations include structured operations.

[0071] It should be noted that in step S21, the production logs are preprocessed, mainly including structuring, which transforms unstructured or semi-structured log data into a unified structured data format, thereby facilitating subsequent analysis and modeling.

[0072] Step S22: Based on predefined parameter categories, parameters are extracted and classified from the preprocessed data according to regular expressions to obtain corresponding environmental parameter data, standard setting parameter data, and actual processing parameter data.

[0073] Step S23 involves filtering and cleaning the environmental parameter data, standard setting parameter data, and actual processing parameter data to obtain the environmental parameter set, standard setting parameter set, and actual processing parameter set corresponding to the target production line.

[0074] In the above implementation, in order to extract various parameters from the production log, the production log is first preprocessed to obtain corresponding preprocessed data. The preprocessing operation includes structured operations. Then, based on predefined parameter categories, the preprocessed data is extracted and classified according to regular expressions to obtain corresponding environmental parameter data, standard setting parameter data, and actual processing parameter data. Finally, the environmental parameter data, standard setting parameter data, and actual processing parameter data are filtered and cleaned to obtain the environmental parameter set, standard setting parameter set, and actual processing parameter set corresponding to the target production line.

[0075] As a further implementation of the method, the steps of obtaining an initial SVM model by performing regression modeling based on the SVM algorithm, according to the environmental parameter set, the standard setting parameter set, and the actual processing parameter set, include:

[0076] Step S31: The exclusivity of each parameter in the standard parameter set and the actual processing parameter set is evaluated by chi-square test to determine the first exclusive parameter group and the second exclusive parameter group.

[0077] It should be noted that the goal of step S31 is to assess the exclusive relationship between parameters in the two parameter sets using a chi-square test, thereby determining the first exclusive parameter group in the standard specification parameter set and the second exclusive parameter group in the actual processing parameter set. Specifically, for each pair of parameters in set A (standard specification parameter set A and actual processing parameter set B), the joint frequency of the two parameters in the data is calculated (cross-tabulation), and their independence is tested using a chi-square test (null hypothesis: the two parameters are independent and have no exclusive relationship; alternative hypothesis: the two parameters have an exclusive relationship (i.e., not independent)). If the p-value is less than the significance level, the null hypothesis is rejected, indicating that the two parameters have a statistically significant dependency, reflecting potential exclusivity or association, and the two parameters form an exclusive parameter group. If the p-value is greater than or equal to α, the parameters are considered independent and have no exclusivity. Furthermore, each pair of parameters in set B is also processed accordingly.

[0078] Step S32: Perform a linear transformation on the first exclusive parameter group to obtain the first fusion parameter, and perform a linear transformation on the second exclusive parameter group to obtain the second fusion parameter.

[0079] Step S33: Based on the SVM algorithm, regression modeling is performed according to the first fusion parameter, the second fusion parameter, the environmental parameter set, the standard setting parameter set, and the actual processing parameter set to obtain the initial SVM model.

[0080] In the above implementation, in order to obtain the initial SVM model, the exclusivity of each parameter in the standard setting parameter set and the actual processing parameter set is evaluated by chi-square test to determine the first exclusive parameter set and the second exclusive parameter set. Then, the first exclusive parameter set is linearly transformed to obtain the first fusion parameter, and the second exclusive parameter set is linearly transformed to obtain the second fusion parameter. Finally, based on the SVM algorithm, regression modeling is performed according to the first fusion parameter, the second fusion parameter, the environmental parameter set, the standard setting parameter set and the actual processing parameter set to obtain the initial SVM model.

[0081] As a further implementation of the method, the step of training and testing the initial SVM model based on the model training data to obtain the actual processing parameter prediction model includes:

[0082] Step S41: Divide the model training data into a training set and a test set according to a preset ratio.

[0083] It should be noted that the preset ratio can be 7:3 or 8:2.

[0084] Step S42: The hyperparameters of the initial SVM model are traversed according to the random network search algorithm, and the performance of the initial SVM model under different hyperparameters is evaluated by k-fold cross-validation to obtain the optimal hyperparameters of the initial SVM model. The root mean square error (RMSE) and the coefficient of determination (R²) are used as evaluation indicators.

[0085] Step S43: Train the initial SVM model based on the training set to obtain the trained SVM model.

[0086] Step S44: Test the trained SVM model according to the test set, and determine whether the error is within the preset range according to the evaluation index. If so, use the trained SVM model as the actual processing parameter prediction model.

[0087] In the above implementation, in order to obtain the actual processing parameter prediction model, the model training data is divided into a training set and a test set according to a preset ratio. Then, the hyperparameters of the initial SVM model are traversed according to the random network search algorithm, and the performance of the initial SVM model under different hyperparameters is evaluated according to the k-fold cross-validation method to obtain the optimal hyperparameters of the initial SVM model. The root mean square error (RMSE) and the coefficient of determination (R²) are used as evaluation indicators. Then, the initial SVM model is trained according to the training set to obtain the trained SVM model. Finally, the trained SVM model is tested according to the test set, and the error is judged according to the evaluation indicators to determine whether it is within the preset range. If the error is within the preset range, the trained SVM model is used as the actual processing parameter prediction model.

[0088] As a further implementation of the method, the step of selecting a set of standard setting parameters based on the annealing algorithm, so that the predicted values ​​corresponding to the set of standard setting parameters and environmental parameters meet the preset conditions of the processing parameters, includes:

[0089] Step S51: In the initial state, set the initial temperature T, cooling rate α, and iteration number i, and initialize each standard setting parameter according to the standard setting parameter constraints.

[0090] Step S52: After annealing, the predicted values ​​and processing parameters corresponding to the standard setting parameters and environmental parameters under the current state are vectorized to obtain the corresponding predicted vector and target vector, and the vector product of the predicted vector and target vector is calculated.

[0091] Step S53: Determine whether the difference between the magnitude of the vector product and the magnitude of the target vector is less than a first preset value, and determine whether the magnitude of the vector product is less than a second preset value. If both are true, then take the corresponding standard setting parameters in the current state as a set of standard setting parameters.

[0092] Step S54: Otherwise, randomly select n parameter points within the standard parameter setting limits, and for each parameter point, calculate the distance between the predicted value corresponding to the parameter point and the environmental parameter and the processing parameter.

[0093] Step S55: Determine the shortest distance based on the distance, and use the parameter points corresponding to the shortest distance as a set of standard setting parameters.

[0094] In the above implementation, to select a set of standard setting parameters, in the initial state, the initial temperature T, cooling rate α, and iteration number i are set, and each standard setting parameter is initialized according to the standard setting parameter constraints. Then, after annealing, the predicted values ​​and processing parameters corresponding to each standard setting parameter and environmental parameter in the current state are vectorized to obtain the corresponding predicted vector and target vector. The vector product of the predicted vector and target vector is calculated. Then, it is determined whether the difference between the magnitude of the vector product and the magnitude of the target vector is less than a first preset value, and whether the magnitude of the vector product is less than a second preset value. If both are true, the standard setting parameters corresponding to the current state are taken as a set of standard setting parameters. Otherwise, n parameter points are randomly selected within the standard setting parameter constraints. For each parameter point, the distance between the predicted value of the parameter point and environmental parameter and the processing parameter is calculated. Then, the shortest distance is determined based on the distance, and the parameter point corresponding to the shortest distance is taken as a set of standard setting parameters.

[0095] As a further implementation of the method, after the step of selecting a set of standard setting parameters based on the annealing algorithm so that the predicted values ​​corresponding to the set of standard setting parameters and environmental parameters meet the preset conditions of the processing parameters, the method further includes:

[0096] The system generates production line control instructions based on a set of standard parameters to control the target production line to process products according to the set of standard parameters.

[0097] In the above implementation, production line control instructions are generated based on a set of standard setting parameters to control the target production line to process the product according to a set of standard setting parameters.

[0098] As a further implementation of the method, the method also includes:

[0099] Step S61: Obtain target environmental parameters and target processing parameters, wherein the target environmental parameters include at least one of target temperature, target humidity, target ventilation rate, target light intensity, and target pressure.

[0100] Step S62: Input the target environmental parameters and target processing parameters into the actual processing parameter prediction model to obtain the corresponding predicted values ​​of the actual processing parameters.

[0101] In the above embodiments, in order to predict the actual processing parameters, target environmental parameters and target processing parameters are obtained. The target environmental parameters include at least one of target temperature, target humidity, target ventilation rate, target light intensity, and target pressure. Then, the target environmental parameters and target processing parameters are input into the actual processing parameter prediction model to obtain the corresponding predicted values ​​of the actual processing parameters.

[0102] This application also discloses a production log analysis system based on the Industrial Internet of Things.

[0103] refer to Figure 2 The industrial IoT-based production log analysis system includes a management platform, a sensor network platform, and an object platform that are connected in sequence. The management platform is configured with:

[0104] The parameter extraction module is used to obtain the production logs of the target production line and extract the environmental parameter set, standard setting parameter set and actual processing parameter set corresponding to the target production line based on the production logs.

[0105] The model building module is used to perform regression modeling based on the SVM algorithm, according to the environmental parameter set, the standard parameter set, and the actual processing parameter set, to obtain the initial SVM model.

[0106] The model generation module is used to take the standard parameter set, environmental parameter set and actual processing parameter set as model training data, and train and test the initial SVM model based on the model training data to obtain the actual processing parameter prediction model.

[0107] The standard setting parameter filtering module is used to obtain the environmental parameters and product processing parameters of the target production line area, set standard setting parameter limits, call the actual processing parameter prediction model, and filter a set of standard setting parameters based on the annealing algorithm so that the predicted values ​​corresponding to the set of standard setting parameters and environmental parameters meet the preset conditions of the processing parameters.

[0108] The overall framework of another application scenario of the production log analysis system based on the Industrial Internet of Things in this application is as follows: Figure 3 As shown, it can include a user platform, service platform, management platform, sensor network platform, and object platform that interact sequentially, forming a five-platform architecture based on the Industrial Internet of Things. The sensor network platform includes n sensor network sub-platforms, each with its own sensor sub-database.

[0109] Specifically, in the aforementioned application scenario, the production log analysis method based on the Industrial Internet of Things (IIoT) includes a management platform configured to: acquire production logs of the target production line and extract the corresponding environmental parameter set, standard setting parameter set, and actual processing parameter set based on the production logs; perform regression modeling based on the SVM algorithm, using the environmental parameter set, standard setting parameter set, and actual processing parameter set to obtain an initial SVM model; use the standard setting parameter set, environmental parameter set, and actual processing parameter set as model training data, and train and test the initial SVM model based on the model training data to obtain an actual processing parameter prediction model; acquire the environmental parameters of the area where the target production line is located and the processing parameters of the product, set standard setting parameter constraints, call the actual processing parameter prediction model, and filter a set of standard setting parameters based on the annealing algorithm so that the predicted values ​​corresponding to the set of standard setting parameters and environmental parameters meet preset conditions with the processing parameters.

[0110] By leveraging the interaction between the various functional platforms of the industrial IoT-based production log analysis system, which is based on the aforementioned three or five platforms, a complete closed-loop information operation logic is established, ensuring the orderly operation of perceived and control information and realizing intelligent equipment management.

[0111] The production log analysis system based on the Industrial Internet of Things (IIoT) of the present invention can implement any method in the production log analysis system based on the Industrial Internet of Things (IIoT), and the specific working process of the production log analysis method based on the Industrial Internet of Things (IIoT) of the present invention can refer to the corresponding process in the above-mentioned production log analysis method based on the Industrial Internet of Things (IIoT).

[0112] This application also discloses a computer device.

[0113] refer to Figure 4 A computer device includes a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement any of the above-described methods for production log analysis based on the Industrial Internet of Things.

[0114] This application also discloses a computer-readable storage medium.

[0115] A computer-readable storage medium storing a computer program capable of being loaded by a processor and executing any of the above-described methods for production log analysis based on the Industrial Internet of Things.

[0116] The computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device; the program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0117] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.

Claims

1. A production log analysis method based on an industrial internet of things, characterized by, The method is applied to an industrial Internet of Things system, the industrial Internet of Things system comprising a management platform, a sensor network platform and an object platform which are sequentially communicatively connected, the method being executed by the management platform and comprising: obtaining a production log of a target production line, and extracting an environment parameter set, a standard setting parameter set and an actual processing parameter set corresponding to the target production line according to the production log; based on an SVM algorithm, performing regression modeling according to the environment parameter set, the standard setting parameter set and the actual processing parameter set to obtain an initial SVM model; taking the standard setting parameter set, the environment parameter set and the actual processing parameter set as model training data, and training and testing the initial SVM model according to the model training data to obtain an actual processing parameter prediction model; obtaining environment parameters of a region where the target production line is located and processing parameters of a product, setting a standard setting parameter limit, calling the actual processing parameter prediction model, and screening a group of standard setting parameters based on an annealing algorithm, so that a predicted value corresponding to the group of standard setting parameters and the environment parameters and the processing parameters satisfy a preset condition; the step of obtaining an initial SVM model based on an SVM algorithm according to the environment parameter set, the standard setting parameter set and the actual processing parameter set, comprising: evaluating the exclusivity of each parameter in the standard setting parameter set and the actual processing parameter set respectively through a chi-square test to determine a first exclusive parameter group and a second exclusive parameter group; performing linear transformation on the first exclusive parameter group to obtain a first fusion parameter, and performing linear transformation on the second exclusive parameter group to obtain a second fusion parameter; based on an SVM algorithm, and according to the first fusion parameter, the second fusion parameter, the environment parameter set, the standard setting parameter set and the actual processing parameter set, performing regression modeling to obtain an initial SVM model; the step of screening a group of standard setting parameters based on an annealing algorithm, so that a predicted value corresponding to the group of standard setting parameters and the environment parameters and the processing parameters satisfy a preset condition, comprising: in an initial state, setting an initial temperature T, a cooling rate a and an iteration number i, and initializing each standard setting parameter according to the standard setting parameter limit; after annealing is completed, vectorizing each standard setting parameter and a predicted value corresponding to the environment parameter and the processing parameter in a current state respectively to obtain a corresponding prediction vector and a target vector, and calculating a vector product corresponding to the prediction vector and the target vector; judging whether a difference corresponding to a module length of the vector product and a module length of the target vector is less than a first preset value, and whether the module length of the vector product is less than a second preset value, if both are, corresponding each standard setting parameter in the current state is taken as a group of standard setting parameters; otherwise, randomly taking n parameter points within the standard setting parameter limit, for each parameter point, calculating a distance between the parameter point and a predicted value corresponding to the environment parameter and the processing parameter; A shortest distance is determined according to the distances, and the parameter point corresponding to the shortest distance is taken as a set of standard setting parameters. 2.The industrial Internet of Things based production log analysis method according to claim 1, characterized in that, The step of extracting the environmental parameter set, the standard setting parameter set and the actual processing parameter set corresponding to the target production line according to the production log comprises: A preprocessing operation is performed on the production log to obtain corresponding preprocessed data, wherein the preprocessing operation comprises a structuring operation; Parameter extraction and classification are performed on the preprocessed data based on a pre-defined parameter category and according to a regular expression to obtain corresponding environmental parameter data, standard setting parameter data and actual processing parameter data; Data filtering and data cleaning are performed on the environmental parameter data, the standard setting parameter data and the actual processing parameter data to obtain the environmental parameter set, the standard setting parameter set and the actual processing parameter set corresponding to the target production line. 3.The industrial Internet of Things based production log analysis method according to claim 1, characterized in that, The step of training and testing the initial SVM model according to the model training data to obtain an actual processing parameter prediction model comprises: The model training data is divided into a training set and a test set according to a pre-set proportion; According to the random network search algorithm, the hyperparameters of the initial SVM model are traversed, and the performance of the initial SVM model under different hyperparameters is evaluated according to the k-fold cross-validation method, to obtain the optimal hyperparameters of the initial SVM model, and the root mean square error RMSE and the determination coefficient R 2 as an evaluation index; The initial SVM model is trained according to the training set to obtain a trained SVM model; The trained SVM model is tested according to the test set, and it is determined whether the error is within a pre-set range according to the evaluation index; if yes, the trained SVM model is taken as an actual processing parameter prediction model. 4.The industrial Internet of Things based production log analysis method according to claim 1, characterized in that, After the step of screening a set of standard setting parameters based on an annealing algorithm so that the predicted values corresponding to the set of standard setting parameters and the environmental parameters and the processing parameters satisfy a pre-set condition, the method further comprises: A production line control instruction is generated according to the set of standard setting parameters to control the target production line to process the product according to the set of standard setting parameters. 5.The industrial Internet of Things based production log analysis method according to claim 1, characterized in that, The method further comprises: Target environmental parameters and target processing parameters are obtained, wherein the target environmental parameters comprise at least one of a target temperature, a target humidity, a target ventilation rate, a target illumination intensity and a target pressure; The target environmental parameters and the target processing parameters are input into the actual processing parameter prediction model to obtain corresponding actual processing parameter prediction values.

6. A production log analysis system based on an industrial internet of things, characterized by, The system comprises a management platform, a sensing network platform and an object platform which are sequentially communicatively connected, and the management platform is configured with: A parameter extraction module configured to obtain a production log of a target production line and extract an environmental parameter set, a standard setting parameter set and an actual processing parameter set corresponding to the target production line according to the production log; A model construction module configured to perform regression modeling according to the environmental parameter set, the standard setting parameter set and the actual processing parameter set based on an SVM algorithm to obtain an initial SVM model; A model generation module configured to take the standard setting parameter set, the environmental parameter set and the actual processing parameter set as model training data and train and test the initial SVM model according to the model training data to obtain an actual processing parameter prediction model; and The system further comprises a parameter extraction module configured to obtain target environmental parameters and target processing parameters, wherein the target environmental parameters comprise at least one of a target temperature, a target humidity, a target ventilation rate, a target illumination intensity and a target pressure; and a model generation module configured to input the target environmental parameters and the target processing parameters into the actual processing parameter prediction model to obtain corresponding actual processing parameter prediction values. The standard setting parameter screening module is configured to acquire environmental parameters of a region where the target production line is located and processing parameters of a product, set standard setting parameter limits, call an actual processing parameter prediction model, and screen a set of standard setting parameters based on an annealing algorithm, so that predicted values corresponding to the set of standard setting parameters and the environmental parameters and the processing parameters satisfy a preset condition. The step of obtaining an initial SVM model based on the SVM algorithm according to the set of environmental parameters, the set of standard setting parameters, and the set of actual processing parameters includes: The exclusivity of each parameter in the set of standard setting parameters and the set of actual processing parameters is respectively evaluated by means of chi-square test to determine a first exclusive parameter group and a second exclusive parameter group. The first exclusive parameter group is subjected to linear transformation to obtain a first fusion parameter, and the second exclusive parameter group is subjected to linear transformation to obtain a second fusion parameter. An initial SVM model is obtained based on the SVM algorithm and according to the first fusion parameter, the second fusion parameter, the set of environmental parameters, the set of standard setting parameters, and the set of actual processing parameters. The step of screening a set of standard setting parameters based on the annealing algorithm so that predicted values corresponding to the set of standard setting parameters and the environmental parameters and the processing parameters satisfy a preset condition includes: In an initial state, an initial temperature T, a cooling rate α, and an iteration number i are set, and each standard setting parameter is initialized according to the standard setting parameter limits; After annealing is completed, each standard setting parameter and the predicted value corresponding to the environmental parameter in the current state and the processing parameter are respectively vectorized to obtain a corresponding prediction vector and a target vector, and a vector product corresponding to the prediction vector and the target vector is calculated; It is determined whether the difference between the length of the vector product and the length of the target vector corresponds to a first preset value, and whether the length of the vector product corresponds to a second preset value, and if both are true, the corresponding each standard setting parameter in the current state is taken as a set of standard setting parameters; Otherwise, n parameter points are randomly taken within the standard setting parameter limits, and for each parameter point, the distance between the parameter point and the predicted value corresponding to the environmental parameter and the processing parameter is calculated; The shortest distance is determined according to the distance, and the parameter point corresponding to the shortest distance is taken as a set of standard setting parameters.

7. A computer device, characterized by The memory and the processor, the memory has computer programs that can be run on the processor, the processor executes the computer programs to realize the method of any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The memory has computer programs that can be loaded and executed by the processor to realize the method of any one of claims 1-5.

Images