Data processing method and system for industrial Internet of Things

Abstract

The invention provides a data processing method and system for an industrial internet of things, and the method comprises the steps: S1, collecting the data of various industrial internet, carrying out the preprocessing of the data of the industrial internet, and obtaining an industrial internet data set; s2, training a sensitive database based on the industrial internet data set, extracting sensitive data of the industrial internet data set, storing the sensitive data in the sensitive database, and storing residual non-sensitive data in a non-sensitive database; s3, encrypting the sensitive database through an encryption device, and creating a fast channel for each network through a data exchange center of the industrial internet to realize data exchange and sharing of the industrial internet; s4, the monitoring center monitors the data of the industrial internet in real time, and the integrity of the data is kept. The method is relatively high in practicability and relatively good in accuracy and flexibility, and can be used for identification and access control of sensitive data in data leakage prevention.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial Internet data processing, and in particular to a data processing method and system for industrial Internet of Things. Background Art

[0002] The Industrial Internet is a new type of infrastructure, application model and industrial ecology that deeply integrates the new generation of information and communication technology with the industrial economy. Through the comprehensive connection of people, machines, objects, systems, etc., it builds a new manufacturing and service system covering the entire industrial chain and the entire value chain, providing a way to achieve the digitalization, networking and intelligent development of industry and even industry. It is an important cornerstone of the Fourth Industrial Revolution. The essence and core of the Industrial Internet is to closely connect and integrate equipment, production lines, factories, suppliers, products and customers through the Industrial Internet platform. It can help the manufacturing industry to extend the industrial chain and form interconnection across equipment, systems, factories and regions, thereby improving efficiency and promoting the intelligence of the entire manufacturing service system. It is also conducive to promoting the integrated development of the manufacturing industry, realizing the leapfrog development between the manufacturing industry and the service industry, and enabling the efficient sharing of various elements and resources of the industrial economy. The data monitoring and collection of the Industrial Internet plays a decisive role in managers' analysis of the industrial operation process.

[0003] However, when traditional methods are used to analyze industrial big data, the diversity and cross-domain nature of the required data leads to low application efficiency of data sharing and extremely high cost of data sharing. In addition, when traditional monitoring systems encounter abnormal access, they do not have corresponding protection measures for abnormal access, which can easily lead to data leakage and data loss. Summary of the invention

[0004] In view of the above technical problems, the present invention discloses a data processing method and system for industrial Internet of Things, which are used to solve the technical problems arising from the above background technology, and the technical solutions adopted are as follows:

[0005] A data processing method for industrial Internet of Things, the method comprising:

[0006] S1: Collect various industrial Internet data, pre-process the industrial Internet data, and obtain an industrial Internet data set;

[0007] S2: Based on the industrial Internet data set, the sensitive database is trained, the sensitive data of the industrial Internet data set is extracted, and the sensitive data is stored in the sensitive database, and the remaining non-sensitive data is stored in the non-sensitive database;

[0008] S3: Encrypt sensitive databases through encryption devices and create fast channels to various networks through the data exchange center of the industrial Internet to realize data exchange and sharing of the industrial Internet;

[0009] S4: The monitoring center monitors the data of the industrial Internet in real time to maintain the integrity of the data.

[0010] Preferably, the S1 includes:

[0011] S11: Collecting data from the industrial Internet and preprocessing the data, wherein the preprocessing includes cleaning the data, removing duplicate records, and then extracting key features of the data;

[0012] S12: Use the decision tree algorithm to divide the data into different categories based on the features, and recursively generate more child nodes, each of which represents the judgment condition of a feature.

[0013] Preferably, the S2 includes:

[0014] S21: adding sensitive words and sensitive terms to the sensitive database, and determining sensitive features of the data according to multiple sub-nodes of the data. When the data is determined to be sensitive data, it is moved into the sensitive database;

[0015] S22: When the data is determined to be non-sensitive data, it is moved into a non-sensitive database.

[0016] Preferably, S3 includes:

[0017] S31: Delete the nodes related to the sensitive data, and add a key to the sensitive database data for encryption;

[0018] S32: Authentication of the visitor is performed, and the key held by the visitor is compared with the key set for the encrypted data. After the authentication is completed and the keys are consistent, the sensitive database opens access restrictions.

[0019] Preferably, the S4 includes:

[0020] S41: The monitoring center monitors and protects the sensitive database in real time, and sets an access failure threshold for the sensitive database. When the number of access failures N of the visitor is greater than the warning value, the warning device sends an alarm to the relevant personnel, warning the relevant personnel that the database may be in danger of being invaded, and reminds the visitor of the remaining access opportunities;

[0021] S42: When the number of failed access attempts N by the visitor is greater than a critical value, the security device locks the sensitive database, and the visitor loses access rights. At the same time, the visitor's IP is uploaded to the monitoring center, and the monitoring center sends the visitor's IP information to relevant personnel, who then proceed to the next step.

[0022] A data processing system for industrial Internet of Things, the system comprising:

[0023] Data collection system: collects data from various industrial Internets, pre-processes the data from the industrial Internet, and obtains industrial Internet data sets;

[0024] Data storage system: Based on the industrial Internet data set, the sensitive database is trained, the sensitive data of the industrial Internet data set is extracted, and the sensitive data is stored in the sensitive database, and the remaining non-sensitive data is stored in the non-sensitive database;

[0025] Data exchange system: Encrypt sensitive databases through encryption devices, and create fast channels to various networks through the data exchange center of the industrial Internet to realize data exchange and sharing of the industrial Internet;

[0026] Data monitoring system: The monitoring center monitors the data of the industrial Internet in real time to maintain the integrity of the data.

[0027] Preferably, the data acquisition system comprises:

[0028] Data preprocessing device: collects data from the industrial Internet and preprocesses the data, wherein the preprocessing includes cleaning the data, removing duplicate records, and then extracting key features of the data;

[0029] Data decision tree: Use the decision tree algorithm to divide the data into different categories based on features, and recursively generate more child nodes, each of which represents the judgment condition of a feature.

[0030] Preferably, the data storage system comprises:

[0031] Sensitive database: Add sensitive words and sensitive terms to the sensitive database, and determine the sensitive characteristics of the data based on multiple sub-nodes of the data. When the data is determined to be sensitive data, it is moved into the sensitive database;

[0032] Non-sensitive database: When the data is judged to be non-sensitive data, it is moved into the non-sensitive database.

[0033] Preferably, the data exchange system comprises:

[0034] Encryption system: Delete nodes related to sensitive data and add keys to sensitive database data for encryption;

[0035] Verification system: authenticates the visitor and compares the key held by the visitor with the key set for the encrypted data. After the identity authentication is completed and the keys are consistent, the sensitive database opens access restrictions.

[0036] Preferably, the data monitoring system comprises:

[0037] Early warning system: The monitoring center monitors and protects sensitive databases in real time, and sets an access failure threshold for the sensitive databases. When the number of failed accesses N by a visitor is greater than the early warning value, the early warning device sends an alarm to the relevant personnel, warning them that the database may be in danger of being invaded, and reminds the visitor of the remaining access opportunities;

[0038] Locking system: When the number of failed access attempts N by a visitor is greater than a critical value, the security device locks the sensitive database, and the visitor loses access rights. At the same time, the visitor's IP is uploaded to the monitoring center, and the monitoring center sends the visitor's IP information to relevant personnel, who then proceed to the next step.

[0039] Beneficial effects of the invention: The invention uses machine learning or deep learning algorithms to train a sensitive database based on an industrial Internet data set to identify and extract sensitive data containing personal privacy, commercial secrets, or the operating status of critical infrastructure. The extracted sensitive data is stored in an encrypted sensitive database, and encryption algorithms and access rights control can effectively prevent data leakage and unauthorized access. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a data processing method for industrial Internet of Things described in the present invention. DETAILED DESCRIPTION

[0041] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0042] One embodiment of the present invention provides a data processing method for industrial Internet of Things, the method comprising:

[0043] S1: Collect various industrial Internet data, pre-process the industrial Internet data, and obtain an industrial Internet data set;

[0044] S2: Based on the industrial Internet data set, the sensitive database is trained, the sensitive data of the industrial Internet data set is extracted, and the sensitive data is stored in the sensitive database, and the remaining non-sensitive data is stored in the non-sensitive database;

[0045] S3: Encrypt sensitive databases through encryption devices and create fast channels to various networks through the data exchange center of the industrial Internet to realize data exchange and sharing of the industrial Internet;

[0046] S4: The monitoring center monitors the data of the industrial Internet in real time to maintain the integrity of the data.

[0047] The working principle and effect of the above technical solution are: through various sensors, equipment controllers and other IoT devices deployed in industrial production environments, industrial Internet data such as temperature, pressure, flow, equipment status and output are collected in real time. The collected raw data contains noise, duplicates, missing values ​​and other problems, and it is necessary to improve the data quality through pre-processing steps such as cleaning, deduplication, interpolation, and normalization to lay the foundation for subsequent data analysis and processing. Using machine learning or deep learning algorithms, sensitive databases are trained based on industrial Internet data sets to identify and extract sensitive data containing personal privacy, commercial secrets or the operating status of critical infrastructure. The extracted sensitive data is stored in an encrypted sensitive database to ensure that the data is not accessed without authorization. At the same time, the remaining non-sensitive data is stored in a non-sensitive database for subsequent analysis and sharing. The sensitive database is encrypted by an encryption device, and advanced encryption algorithms are used to enhance data security to prevent data from being stolen or tampered with during transmission and storage. Using the data exchange center of the industrial Internet, a fast channel is created for each network node to achieve efficient exchange and sharing of data. These channels are based on secure communication protocols to ensure the security of data during transmission. At the same time, the data exchange center is also responsible for the routing, forwarding and management of data to ensure the real-time and reliability of data. The monitoring center monitors the industrial Internet data in real time, including the transmission status, storage status and access records of the data. Through real-time monitoring, potential security threats and abnormal behaviors can be discovered in time, and corresponding countermeasures can be taken. Through regular data verification, backup and recovery mechanisms, the integrity and consistency of industrial Internet data are ensured. In the event of data damage or loss, data can be quickly restored to reduce losses. At the same time, technologies such as blockchain are used to record the modification history of data to increase the traceability and credibility of data. The industrial Internet of Things collects data from machines and assets through a real-time data acquisition system and processes it immediately to achieve rapid insights and timely responses. This real-time nature greatly improves production efficiency and enables enterprises to respond quickly to changes in the production process. Automated data processing processes reduce manual intervention and improve processing efficiency. At the same time, combined with intelligent algorithms such as machine learning and statistical modeling, patterns, trends and anomalies in data can be more accurately identified, providing strong support for decision-making. The industrial Internet of Things data processing method emphasizes data encryption and access control to ensure the security of sensitive data during transmission and storage. Through encryption algorithms and access control, data leakage and unauthorized access can be effectively prevented. Monitor data in real time to detect anomalies and trigger alerts when problems may occur. This mechanism helps to promptly detect and respond to potential security threats and ensure the stable operation of production systems.

[0048] In one embodiment of the present invention, the S1 includes:

[0049] S11: Collecting data from the industrial Internet and preprocessing the data, wherein the preprocessing includes cleaning the data, removing duplicate records, and then extracting key features of the data;

[0050] S12: Use the decision tree algorithm to divide the data into different categories based on the features, and recursively generate more child nodes, each of which represents the judgment condition of a feature.

[0051] The working principle and effect of the above technical solution are as follows: Data cleaning is an important part of the preprocessing process, the purpose of which is to remove or correct errors, inconsistencies and outliers in the data. This includes: for records with missing key information, it is necessary to delete or fill them. Check whether there are exactly the same records in the data set, and delete redundant copies to avoid deviations in subsequent analysis. Identify and process outliers in the data, which are caused by measurement errors, data entry errors, or extreme cases of the data itself. After data cleaning, it is necessary to extract features (or attributes) that are important for subsequent analysis from the remaining data. These features should be variables that can reflect the essential characteristics of the data, such as temperature, pressure, speed, etc. in the production process. Feature extraction helps to simplify the data set and improve the accuracy and efficiency of subsequent tasks such as classification and prediction. When constructing each node of the decision tree, the algorithm evaluates each feature in the data set and selects a feature that can best divide the data set into different categories as the judgment condition of the current node. According to the selected features and their corresponding thresholds, the data set is divided into two or more subsets. Each subset contains records with the same or similar values ​​on the selected features. Repeat the above process for each subset, that is, select the best features, split the data set, and generate new child nodes. This process will continue until the stopping condition is met. Through the above recursive process, a complete decision tree is finally constructed. This tree structure starts with a root node, each internal node represents a judgment condition for a feature, and each leaf node represents a category or prediction result. After the decision tree is constructed, it can be used to classify new data records. By traversing the new data record from the root node down the tree structure, selecting the corresponding branch according to the judgment condition of each node, and finally reaching a leaf node, the category or prediction result represented by the leaf node is the classification result of the new data record. Through data cleaning, errors, inconsistencies and outliers in the data can be removed or corrected, significantly improving the data quality. This helps to reduce noise and bias in subsequent analysis and improve the accuracy of the results. After removing duplicate records and irrelevant data, the data set is streamlined, thereby reducing the consumption of computing resources in the subsequent analysis process. This not only improves processing speed, but also reduces storage costs. The decision tree model displays the classification process in a tree structure, which is intuitive and easy to understand. This enables people with non-technical backgrounds to understand the decision basis and results of the model, improving the interpretability of the model. The decision tree algorithm automatically selects the best features for segmentation during the construction process, which helps identify the features that have the greatest impact on the classification results. This automatic feature selection capability reduces the need for manual intervention and improves the efficiency and accuracy of classification.

[0052] In one embodiment of the present invention, S2 includes:

[0053] S21: adding sensitive words and sensitive terms to the sensitive database, and determining sensitive features of the data according to multiple sub-nodes of the data. When the data is determined to be sensitive data, it is moved into the sensitive database;

[0054] S22: When the data is determined to be non-sensitive data, it is moved into a non-sensitive database.

[0055] The specific implementation method of the sensitive database identifying sensitive data is as follows:

[0056] A: After the sensitive words and sensitive vocabulary are added to the sensitive database, vector space weight calculation is performed on the sensitive words and sensitive vocabulary, and the text data is converted into a numerical form, that is, a feature vector, and the vector space weight is obtained by the following calculation formula:

[0057] P=TFi,j×IDFi

[0058] Where TFi,j represents the frequency of word i in document j, and IDFi represents the inverse document frequency of word i. The inverse document frequency represents a measure of the universal importance of a word, reflecting the balance between the universality and uniqueness of a word.

[0059] And, the TFi,j is obtained by the following formula:

[0060]

[0061] Where ni,j represents the number of times word i appears in document j, ∑knk,j represents the sum of the number of times all words appear in document j, that is, the total number of words in document j; Lj represents the length of document j;

[0062] The IDFi is obtained by the following formula:

[0063]

[0064] Where N represents the total number of documents, |{j:ti∈dj}| represents the number of documents containing word i;

[0065] B: Importing the vector space weights of the sensitive words and sensitive vocabulary into a sensitive database, and setting a threshold Q for the sensitive database, where the threshold represents the minimum similarity of two data that are sufficiently similar;

[0066] C: Extract features from the data to be judged, obtain the vector space weights through the algorithm in step A, perform cosine calculation on P, and compare the obtained cosine value with Q.

[0067] When the obtained cosine value is greater than Q, it indicates that the data to be judged is sensitive data and is moved into the sensitive database;

[0068] When the obtained cosine value is less than Q, it indicates that the data to be judged is non-sensitive data and is moved into the non-sensitive database.

[0069] The working principle and effect of the above technical solution are as follows: sensitive words and sensitive vocabulary are predefined and added in the sensitive database. The weights of these sensitive vocabulary are calculated using the vector space model (VSM). The sensitive vocabulary is converted into a numerical form, i.e., a feature vector. Each dimension of the feature vector represents a vocabulary, and its weight reflects the importance or discrimination of the vocabulary in the text. The calculated vector space weights of the sensitive vocabulary are imported into the sensitive database. A threshold Q is set for the sensitive database, and the similarity between two data is measured by Q. When the similarity of two data exceeds Q, they are considered to be sufficiently similar. Feature extraction is performed on the data to be judged, and it is also converted into the form of a feature vector. The cosine similarity algorithm is used to calculate the similarity between the feature vector of the data to be judged and the feature vector of the sensitive vocabulary or data already in the sensitive database. The cosine similarity algorithm is used to evaluate the similarity between the feature vectors of the data to be judged and the feature vectors of the sensitive vocabulary or data already in the sensitive database. The cosine similarity evaluates the similarity of two vectors by measuring the degree of proximity in direction, and the larger the value, the higher the similarity. If the calculated cosine value is greater than Q, it means that the data to be judged is sufficiently similar to one or some sensitive vocabulary or data in the sensitive database, so it is judged as sensitive data and moved into the sensitive database. If the cosine value is less than Q, it means that the data to be judged is not similar to the sensitive data in the sensitive database, so it is judged as non-sensitive data and moved into the non-sensitive database. Sensitive data is identified by vector space weight calculation and cosine similarity calculation, which reduces the need for manual intervention and improves processing efficiency and accuracy. Using mathematical methods such as vector space model and cosine similarity, the similarity between data can be more accurately evaluated, so that sensitive data can be more accurately identified. The sensitive data identification requirements under different scenarios and needs can be adapted by adjusting parameters such as sensitive vocabulary list, vector space weight calculation method, threshold Q, etc. Separating sensitive data from non-sensitive data and storing it in a dedicated sensitive database helps to implement more stringent security management and access control strategies and reduce the risk of data leakage and abuse.

[0070] In one embodiment of the present invention, S3 includes:

[0071] S31: Delete the nodes related to the sensitive data, and add a key to the sensitive database data for encryption;

[0072] S32: Authentication of the visitor is performed, and the key held by the visitor is compared with the key set for the encrypted data. After the authentication is completed and the keys are consistent, the sensitive database opens access restrictions.

[0073] The working principle and effect of the above technical solution are as follows: the system will identify and delete nodes directly related to sensitive data. These nodes may include specific fields, records or files that store sensitive information. Deleting these nodes can reduce the exposure of sensitive data and reduce the risk of data leakage. Before a visitor attempts to access a sensitive database, the system will authenticate the visitor. Authentication is the process of confirming the identity of the visitor, including username and password verification, multi-factor authentication, etc. Through authentication, the system can ensure that only legitimate users can access sensitive data. After the authentication is passed, the system will also compare the key held by the visitor with the key set for the encrypted data. This step is to ensure that the visitor has the authority to access sensitive data and the correct key. If the keys are consistent, it means that the visitor has passed the authentication and has the legitimate authority to access sensitive data. When the authentication is completed and the keys are consistent, the system will lift the access restrictions on the sensitive database and allow the visitor to access the sensitive data. This step is the last line of defense in the secure access process, ensuring that only legitimate users who have undergone strict authentication and key comparison can access sensitive data. Visitor authentication and key comparison ensure that only legitimate users who have undergone strict authentication can access sensitive databases. This dual authentication mechanism based on identity and key improves the accuracy of access control and reduces the risk of illegal access and data leakage. By setting different levels of access rights and keys, access control of sensitive data can be further refined to ensure the secure sharing and transmission of data between different users. Encryption and access control mechanisms are usually highly flexible and scalable. Enterprises can flexibly adjust encryption policies and access control rules according to their own business needs and regulatory requirements. At the same time, as technology develops and business expands, these mechanisms can also be easily upgraded and expanded to adapt to new security threats and business needs.

[0074] In one embodiment of the present invention, the S4 includes:

[0075] S41: The monitoring center monitors and protects the sensitive database in real time, and sets an access failure threshold for the sensitive database. When the number of access failures N of the visitor is greater than the warning value, the warning device sends an alarm to the relevant personnel, warning the relevant personnel that the database may be in danger of being invaded, and reminds the visitor of the remaining access opportunities;

[0076] S42: When the number of failed access attempts N by the visitor is greater than a critical value, the security device locks the sensitive database, and the visitor loses access rights. At the same time, the visitor's IP is uploaded to the monitoring center, and the monitoring center sends the visitor's IP information to relevant personnel, who then proceed to the next step.

[0077] The working principle and effect of the above technical solution are as follows: the monitoring center continuously monitors the sensitive database in real time to detect any abnormal activities or potential security threats. This includes but is not limited to monitoring the database access log, monitoring the database performance, and detecting potential malicious behaviors. At the same time, the monitoring center also takes a series of protective measures to protect the sensitive database, including firewall settings, deployment of intrusion detection systems (IDS) and intrusion prevention systems (IPS), etc., to prevent unauthorized access and malicious attacks. In order to prevent attacks such as brute force cracking, the monitoring center sets an access failure threshold (i.e., warning value) for the sensitive database. When the number of consecutive access failures N of the visitor reaches or exceeds this warning value, the system considers that there may be an attempt to illegally invade. And the warning value is 3. Once the warning condition is triggered, the warning device will immediately issue an alarm to the relevant personnel. The alarm content usually includes warning information, relevant information of the visitor, and remaining access opportunities. When the number of consecutive access failures N of the visitor further increases and reaches or exceeds the set critical value, the system considers that the visitor has posed a security threat to the sensitive database. At this time, the critical value is 5. The security device will immediately lock the sensitive database to prevent the visitor from continuing to access. At the same time, the system will upload the visitor's IP address to the monitoring center for subsequent security analysis and processing. The monitoring center sends the visitor's IP information to relevant personnel for further investigation and processing. By monitoring sensitive databases in real time, the system can immediately detect any abnormal activities or potential security threats, thereby realizing rapid perception of security events. Once an abnormality is detected or an early warning condition is triggered, the system can respond quickly, such as issuing an alarm, locking the database, etc., effectively shortening the time interval from detection to response. Setting an access failure threshold can effectively prevent attacks such as brute force cracking, increasing the difficulty of illegal intrusion. When the number of visitor access failures exceeds the critical value, the system automatically locks the database to prevent potential malicious behavior from causing further damage to the database. The early warning device can issue an alarm to relevant personnel when the number of visitor access failures reaches the early warning value, reminding them that the database may be in danger of being invaded, and reminding the visitor of the remaining access opportunities. This helps relevant personnel take timely measures to deal with potential security threats. Uploading the visitor's IP information to the monitoring center and notifying relevant personnel will help with subsequent security analysis and processing. By tracking the attacker's IP address, the enterprise can further understand the source and means of the attack, thereby strengthening security protection measures.

[0078] One embodiment of the present invention provides a data processing system for industrial Internet of Things, the system comprising:

[0079] Data collection system: collects data from various industrial Internets, pre-processes the data from the industrial Internet, and obtains industrial Internet data sets;

[0080] Data storage system: Based on the industrial Internet data set, the sensitive database is trained, the sensitive data of the industrial Internet data set is extracted, and the sensitive data is stored in the sensitive database, and the remaining non-sensitive data is stored in the non-sensitive database;

[0081] Data exchange system: Encrypt sensitive databases through encryption devices, and create fast channels to various networks through the data exchange center of the industrial Internet to realize data exchange and sharing of the industrial Internet;

[0082] Data monitoring system: The monitoring center monitors the data of the industrial Internet in real time to maintain the integrity of the data.

[0083] The working principle and effect of the above technical solution are: through various sensors, equipment controllers and other IoT devices deployed in industrial production environments, industrial Internet data such as temperature, pressure, flow, equipment status and output are collected in real time. The collected raw data contains noise, duplicates, missing values ​​and other problems, and it is necessary to improve the data quality through pre-processing steps such as cleaning, deduplication, interpolation, and normalization to lay the foundation for subsequent data analysis and processing. Using machine learning or deep learning algorithms, sensitive databases are trained based on industrial Internet data sets to identify and extract sensitive data containing personal privacy, commercial secrets or the operating status of critical infrastructure. The extracted sensitive data is stored in an encrypted sensitive database to ensure that the data is not accessed without authorization. At the same time, the remaining non-sensitive data is stored in a non-sensitive database for subsequent analysis and sharing. The sensitive database is encrypted by an encryption device, and advanced encryption algorithms are used to enhance data security to prevent data from being stolen or tampered with during transmission and storage. Using the data exchange center of the industrial Internet, a fast channel is created for each network node to achieve efficient exchange and sharing of data. These channels are based on secure communication protocols to ensure the security of data during transmission. At the same time, the data exchange center is also responsible for the routing, forwarding and management of data to ensure the real-time and reliability of data. The monitoring center monitors the industrial Internet data in real time, including the transmission status, storage status and access records of the data. Through real-time monitoring, potential security threats and abnormal behaviors can be discovered in time, and corresponding countermeasures can be taken. Through regular data verification, backup and recovery mechanisms, the integrity and consistency of industrial Internet data are ensured. In the event of data damage or loss, data can be quickly restored to reduce losses. At the same time, technologies such as blockchain are used to record the modification history of data to increase the traceability and credibility of data. The industrial Internet of Things collects data from machines and assets through a real-time data acquisition system and processes it immediately to achieve rapid insights and timely responses. This real-time nature greatly improves production efficiency and enables enterprises to respond quickly to changes in the production process. Automated data processing processes reduce manual intervention and improve processing efficiency. At the same time, combined with intelligent algorithms such as machine learning and statistical modeling, patterns, trends and anomalies in data can be more accurately identified, providing strong support for decision-making. The industrial Internet of Things data processing method emphasizes data encryption and access control to ensure the security of sensitive data during transmission and storage. Through encryption algorithms and access control, data leakage and unauthorized access can be effectively prevented. Monitor data in real time to detect anomalies and trigger alerts when problems may occur. This mechanism helps to promptly detect and respond to potential security threats and ensure the stable operation of production systems.

[0084] In one embodiment of the present invention, the data acquisition system comprises:

[0085] Data preprocessing device: collects data from the industrial Internet and preprocesses the data, wherein the preprocessing includes cleaning the data, removing duplicate records, and then extracting key features of the data;

[0086] Data decision tree: Use the decision tree algorithm to divide the data into different categories based on features, and recursively generate more child nodes, each of which represents the judgment condition of a feature.

[0087] The working principle and effect of the above technical solution are as follows: Data cleaning is an important part of the preprocessing process, the purpose of which is to remove or correct errors, inconsistencies and outliers in the data. This includes: for records with missing key information, it is necessary to delete or fill them. Check whether there are exactly the same records in the data set, and delete redundant copies to avoid deviations in subsequent analysis. Identify and process outliers in the data, which are caused by measurement errors, data entry errors, or extreme cases of the data itself. After data cleaning, it is necessary to extract features (or attributes) that are important for subsequent analysis from the remaining data. These features should be variables that can reflect the essential characteristics of the data, such as temperature, pressure, speed, etc. in the production process. Feature extraction helps to simplify the data set and improve the accuracy and efficiency of subsequent tasks such as classification and prediction. When constructing each node of the decision tree, the algorithm evaluates each feature in the data set and selects a feature that can best divide the data set into different categories as the judgment condition of the current node. According to the selected features and their corresponding thresholds, the data set is divided into two or more subsets. Each subset contains records with the same or similar values ​​on the selected features. Repeat the above process for each subset, that is, select the best features, split the data set, and generate new child nodes. This process will continue until the stopping condition is met. Through the above recursive process, a complete decision tree is finally constructed. This tree structure starts with a root node, each internal node represents a judgment condition for a feature, and each leaf node represents a category or prediction result. After the decision tree is constructed, it can be used to classify new data records. By traversing the new data record from the root node down the tree structure, selecting the corresponding branch according to the judgment condition of each node, and finally reaching a leaf node, the category or prediction result represented by the leaf node is the classification result of the new data record. Through data cleaning, errors, inconsistencies and outliers in the data can be removed or corrected, significantly improving the data quality. This helps to reduce noise and bias in subsequent analysis and improve the accuracy of the results. After removing duplicate records and irrelevant data, the data set is streamlined, thereby reducing the consumption of computing resources in the subsequent analysis process. This not only improves processing speed, but also reduces storage costs. The decision tree model displays the classification process in a tree structure, which is intuitive and easy to understand. This enables people with non-technical backgrounds to understand the decision basis and results of the model, improving the interpretability of the model. The decision tree algorithm automatically selects the best features for segmentation during the construction process, which helps identify the features that have the greatest impact on the classification results. This automatic feature selection capability reduces the need for manual intervention and improves the efficiency and accuracy of classification.

[0088] In one embodiment of the present invention, the data storage system comprises:

[0089] Sensitive database: Add sensitive words and sensitive terms to the sensitive database, and determine the sensitive characteristics of the data based on multiple sub-nodes of the data. When the data is determined to be sensitive data, it is moved into the sensitive database;

[0090] Non-sensitive database: When the data is judged to be non-sensitive data, it is moved into the non-sensitive database.

[0091] The working principle and effect of the above technical solution are as follows: sensitive words and sensitive vocabulary are predefined and added in the sensitive database. The weights of these sensitive vocabulary are calculated using the vector space model (VSM). The sensitive vocabulary is converted into a numerical form, i.e., a feature vector. Each dimension of the feature vector represents a vocabulary, and its weight reflects the importance or discrimination of the vocabulary in the text. The calculated vector space weights of the sensitive vocabulary are imported into the sensitive database. A threshold Q is set for the sensitive database, and the similarity between two data is measured by Q. When the similarity of two data exceeds Q, they are considered to be sufficiently similar. Feature extraction is performed on the data to be judged, and it is also converted into the form of a feature vector. The cosine similarity algorithm is used to calculate the similarity between the feature vector of the data to be judged and the feature vector of the sensitive vocabulary or data already in the sensitive database. The cosine similarity algorithm is used to evaluate the similarity between the feature vectors of the data to be judged and the feature vectors of the sensitive vocabulary or data already in the sensitive database. The cosine similarity evaluates the similarity of two vectors by measuring the degree of proximity in direction, and the larger the value, the higher the similarity. If the calculated cosine value is greater than Q, it means that the data to be judged is sufficiently similar to one or some sensitive vocabulary or data in the sensitive database, so it is judged as sensitive data and moved into the sensitive database. If the cosine value is less than Q, it means that the data to be judged is not similar to the sensitive data in the sensitive database, so it is judged as non-sensitive data and moved into the non-sensitive database. Sensitive data is identified by vector space weight calculation and cosine similarity calculation, which reduces the need for manual intervention and improves processing efficiency and accuracy. Using mathematical methods such as vector space model and cosine similarity, the similarity between data can be more accurately evaluated, so that sensitive data can be more accurately identified. The sensitive data identification requirements under different scenarios and needs can be adapted by adjusting parameters such as sensitive vocabulary list, vector space weight calculation method, threshold Q, etc. Separating sensitive data from non-sensitive data and storing it in a dedicated sensitive database helps to implement more stringent security management and access control strategies and reduce the risk of data leakage and abuse.

[0092] In one embodiment of the present invention, the data exchange system comprises:

[0093] Encryption system: Delete nodes related to sensitive data and add keys to sensitive database data for encryption;

[0094] Verification system: authenticates the visitor and compares the key held by the visitor with the key set for the encrypted data. After the identity authentication is completed and the keys are consistent, the sensitive database opens access restrictions.

[0095] The working principle and effect of the above technical solution are as follows: the system will identify and delete nodes directly related to sensitive data. These nodes may include specific fields, records or files that store sensitive information. Deleting these nodes can reduce the exposure of sensitive data and reduce the risk of data leakage. Before a visitor attempts to access a sensitive database, the system will authenticate the visitor. Authentication is the process of confirming the identity of the visitor, including username and password verification, multi-factor authentication, etc. Through authentication, the system can ensure that only legitimate users can access sensitive data. After the authentication is passed, the system will also compare the key held by the visitor with the key set for the encrypted data. This step is to ensure that the visitor has the authority to access sensitive data and the correct key. If the keys are consistent, it means that the visitor has passed the authentication and has the legitimate authority to access sensitive data. When the authentication is completed and the keys are consistent, the system will lift the access restrictions on the sensitive database and allow the visitor to access the sensitive data. This step is the last line of defense in the secure access process, ensuring that only legitimate users who have undergone strict authentication and key comparison can access sensitive data. Visitor authentication and key comparison ensure that only legitimate users who have undergone strict authentication can access sensitive databases. This dual authentication mechanism based on identity and key improves the accuracy of access control and reduces the risk of illegal access and data leakage. By setting different levels of access rights and keys, access control of sensitive data can be further refined to ensure the secure sharing and transmission of data between different users. Encryption and access control mechanisms are usually highly flexible and scalable. Enterprises can flexibly adjust encryption policies and access control rules according to their own business needs and regulatory requirements. At the same time, as technology develops and business expands, these mechanisms can also be easily upgraded and expanded to adapt to new security threats and business needs.

[0096] In one embodiment of the present invention, the data monitoring system comprises:

[0097] Early warning system: The monitoring center monitors and protects sensitive databases in real time, and sets an access failure threshold for the sensitive databases. When the number of failed accesses N by a visitor is greater than the early warning value, the early warning device sends an alarm to the relevant personnel, warning them that the database may be in danger of being invaded, and reminds the visitor of the remaining access opportunities;

[0098] Locking system: When the number of failed access attempts N by a visitor is greater than a critical value, the security device locks the sensitive database, and the visitor loses access rights. At the same time, the visitor's IP is uploaded to the monitoring center, and the monitoring center sends the visitor's IP information to relevant personnel, who then proceed to the next step.

[0099] The working principle and effect of the above technical solution are as follows: the monitoring center continuously monitors the sensitive database in real time to detect any abnormal activities or potential security threats. This includes but is not limited to monitoring the database access log, monitoring the database performance, and detecting potential malicious behaviors. At the same time, the monitoring center also takes a series of protective measures to protect the sensitive database, including firewall settings, deployment of intrusion detection systems (IDS) and intrusion prevention systems (IPS), etc., to prevent unauthorized access and malicious attacks. In order to prevent attacks such as brute force cracking, the monitoring center sets an access failure threshold (i.e., warning value) for the sensitive database. When the number of consecutive access failures N of the visitor reaches or exceeds this warning value, the system considers that there may be an attempt to illegally invade. And the warning value is 3. Once the warning condition is triggered, the warning device will immediately issue an alarm to the relevant personnel. The alarm content usually includes warning information, relevant information of the visitor, and remaining access opportunities. When the number of consecutive access failures N of the visitor further increases and reaches or exceeds the set critical value, the system considers that the visitor has posed a security threat to the sensitive database. At this time, the critical value is 5. The security device will immediately lock the sensitive database to prevent the visitor from continuing to access. At the same time, the system will upload the visitor's IP address to the monitoring center for subsequent security analysis and processing. The monitoring center sends the visitor's IP information to relevant personnel for further investigation and processing. By monitoring sensitive databases in real time, the system can immediately detect any abnormal activities or potential security threats, thereby realizing rapid perception of security events. Once an abnormality is detected or an early warning condition is triggered, the system can respond quickly, such as issuing an alarm, locking the database, etc., effectively shortening the time interval from detection to response. Setting an access failure threshold can effectively prevent attacks such as brute force cracking, increasing the difficulty of illegal intrusion. When the number of visitor access failures exceeds the critical value, the system automatically locks the database to prevent potential malicious behavior from causing further damage to the database. The early warning device can issue an alarm to relevant personnel when the number of visitor access failures reaches the early warning value, reminding them that the database may be in danger of being invaded, and reminding the visitor of the remaining access opportunities. This helps relevant personnel take timely measures to deal with potential security threats. Uploading the visitor's IP information to the monitoring center and notifying relevant personnel will help with subsequent security analysis and processing. By tracking the attacker's IP address, the enterprise can further understand the source and means of the attack, thereby strengthening security protection measures.

[0100] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A data processing method for industrial Internet of Things, characterized in that: The method comprises: S1: Collect various industrial Internet data, pre-process the industrial Internet data, and obtain an industrial Internet data set; S2: Based on the industrial Internet data set, the sensitive database is trained, the sensitive data of the industrial Internet data set is extracted, and the sensitive data is stored in the sensitive database, and the remaining non-sensitive data is stored in the non-sensitive database; S3: Encrypt sensitive databases through encryption devices and create fast channels to various networks through the data exchange center of the industrial Internet to realize data exchange and sharing of the industrial Internet; S4: The monitoring center monitors the data of the industrial Internet in real time to maintain the integrity of the data.

2. A data processing method for industrial Internet of Things according to claim 1, characterized in that: The S1 includes: S11: Collecting data from the industrial Internet and preprocessing the data, wherein the preprocessing includes cleaning the data, removing duplicate records, and then extracting key features of the data; S12: Use the decision tree algorithm to divide the data into different categories based on the features, and recursively generate more child nodes, each of which represents the judgment condition of a feature.

3. The data processing method for industrial Internet of Things according to claim 1, characterized in that: The S2 includes: S21: adding sensitive words and sensitive terms to the sensitive database, and determining sensitive features of the data according to multiple sub-nodes of the data. When the data is determined to be sensitive data, it is moved into the sensitive database; S22: When the data is determined to be non-sensitive data, it is moved into a non-sensitive database.

4. The data processing method for industrial Internet of Things according to claim 1, characterized in that: The S3 includes: S31: Delete the nodes related to the sensitive data, and add a key to the sensitive database data for encryption; S32: Authentication of the visitor is performed, and the key held by the visitor is compared with the key set for the encrypted data. After the authentication is completed and the keys are consistent, the sensitive database opens access restrictions.

5. The data processing method for industrial Internet of Things according to claim 1, characterized in that: The S4 includes: S41: The monitoring center monitors and protects the sensitive database in real time, and sets an access failure threshold for the sensitive database. When the number of access failures N of the visitor is greater than the warning value, the warning device sends an alarm to the relevant personnel, warning the relevant personnel that the database may be in danger of being invaded, and reminds the visitor of the remaining access opportunities; S42: When the number of failed access attempts N by the visitor is greater than a critical value, the security device locks the sensitive database, and the visitor loses access rights. At the same time, the visitor's IP is uploaded to the monitoring center, and the monitoring center sends the visitor's IP information to relevant personnel, who then proceed to the next step.

6. A data processing system for industrial Internet of Things, characterized in that: The system comprises: Data collection system: collects data from various industrial Internets, pre-processes the data from the industrial Internet, and obtains industrial Internet data sets; Data storage system: Based on the industrial Internet data set, the sensitive database is trained, the sensitive data of the industrial Internet data set is extracted, and the sensitive data is stored in the sensitive database, and the remaining non-sensitive data is stored in the non-sensitive database; Data exchange system: Encrypt sensitive databases through encryption devices, and create fast channels to various networks through the data exchange center of the industrial Internet to realize data exchange and sharing of the industrial Internet; Data monitoring system: The monitoring center monitors the data of the industrial Internet in real time to maintain the integrity of the data.

7. A data processing system for industrial Internet of Things according to claim 6, characterized in that: The data acquisition system comprises: Data preprocessing device: collects data from the industrial Internet and preprocesses the data, wherein the preprocessing includes cleaning the data, removing duplicate records, and then extracting key features of the data; Data decision tree: Use the decision tree algorithm to divide the data into different categories based on features, and recursively generate more child nodes, each of which represents the judgment condition of a feature.

8. A data processing system for industrial Internet of Things according to claim 6, characterized in that: The data storage system comprises: Sensitive database: Add sensitive words and sensitive terms to the sensitive database, and determine the sensitive characteristics of the data based on multiple sub-nodes of the data. When the data is determined to be sensitive data, it is moved into the sensitive database; Non-sensitive database: When the data is judged to be non-sensitive data, it is moved into the non-sensitive database.

9. A data processing system for industrial Internet of Things according to claim 6, characterized in that: The data exchange system comprises: Encryption system: Delete nodes related to sensitive data and add keys to sensitive database data for encryption; Verification system: authenticates the visitor and compares the key held by the visitor with the key set for the encrypted data. After the identity authentication is completed and the keys are consistent, the sensitive database opens access restrictions.

10. A data processing system for industrial Internet of Things according to claim 6, characterized in that: The data monitoring system comprises: Early warning system: The monitoring center monitors and protects sensitive databases in real time, and sets an access failure threshold for the sensitive databases. When the number of failed accesses N by a visitor is greater than the early warning value, the early warning device sends an alarm to the relevant personnel, warning them that the database may be in danger of being invaded, and reminds the visitor of the remaining access opportunities; Locking system: When the number of failed access attempts N by a visitor is greater than a critical value, the security device locks the sensitive database, and the visitor loses access rights. At the same time, the visitor's IP is uploaded to the monitoring center, and the monitoring center sends the visitor's IP information to relevant personnel, who then proceed to the next step.

Images