Information data processing method based on industrial internet

Abstract

The invention relates to the technical field of multi-screen display, in particular to an information data processing method based on the industrial internet. According to the technical scheme, efficient information data processing is achieved through the industrial internet, the data acquisition module automatically identifies and converts heterogeneous protocols such as Modibus, OPC UA and MQTT into a unified JSON or XML format, and data compatibility and consistency are improved. Data processing is carried out in a layered mode, the edge nodes rapidly process real-time data, anomaly detection and filtering are included, and low-delay response is ensured; the cloud computing platform carries out deep analysis and trend prediction, and a machine learning model is used to enhance insight. The communication module adopts a dynamic path optimization algorithm, adaptively selects an optimal transmission path based on real-time network parameters, improves the reliability and efficiency of data transmission, and supports a standby mode to cope with network faults. The modular interface design allows hot plug and standardized interaction, and the expandability and maintainability of the system are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial internet data processing, and in particular to an information data processing method based on industrial internet. BACKGROUND

[0002] In the field of industrial internet, information data processing faces many challenges, mainly due to the diversity of industrial equipment and the complexity of network environment. In the prior art, the data collection link often relies on various industrial communication protocols, such as Modbus, OPCUA and MQTT, etc. These protocols are highly heterogeneous and have non-uniform data formats, making it difficult to directly integrate and process the collected operation data. Traditional methods usually require manual configuration or fixed conversion rules, which are inefficient and prone to errors, and cannot adapt to dynamically changing equipment environments.

[0003] In addition, in terms of data processing, existing systems often use centralized cloud computing mode to upload all data to the cloud for analysis. Although this can perform deep calculation, it ignores scenarios with high real-time requirements, such as equipment anomaly detection, resulting in increased response delay and affecting the safety and efficiency of industrial production. The application of edge computing has not been popularized or optimized, and there is a lack of effective division mechanism, so real-time data cannot be processed quickly, and non-real-time data may occupy too many resources.

[0004] In the data transmission link, the industrial network environment is unstable, with large fluctuations in bandwidth, delay and packet loss rate, while existing communication modules mostly use static or simple heuristic path selection algorithms, which cannot dynamically adapt to network changes. For example, path optimization is usually based on fixed weights or thresholds, and cannot be adaptively adjusted according to real-time network parameters (such as bandwidth value, delay value and packet loss rate), resulting in low transmission efficiency, data loss or increased delay. In particular, under poor network conditions, such as wireless or 5G private networks, existing methods lack backup mechanisms and cannot guarantee the reliability of data transmission.

[0005] In addition, the integration of machine learning and big data analysis in existing technologies is insufficient, and the application of prediction models such as time series analysis (such as ARIMA or LSTM) is less common, and the modular design is imperfect, with low interface standardization, limiting the system scalability and interoperability, making it difficult to achieve hot plug and dynamic loading. These limitations highlight the urgent need for improvement in the automation, intelligence and reliability of industrial internet data processing. SUMMARY

[0006] The present application provides an information data processing method based on industrial internet, which improves the management efficiency of industrial equipment operation data and system performance.

[0007] The present application provides an information data processing method based on industrial internet, which improves the management efficiency of industrial equipment operation data and system performance. S1: The data acquisition module collects the operating data of the industrial equipment. The data acquisition module is configured to convert the heterogeneous operating data into a unified format through an automatic protocol identification and conversion mechanism. The automatic protocol identification and conversion mechanism includes parsing the protocol identifier in the header of the input data packet and performing format conversion according to a predefined protocol mapping table. S2: The collected running data is analyzed by the data processing module. The data processing module performs fast processing on real-time data at the edge computing node and uploads non-real-time data to the cloud computing platform for in-depth analysis. The fast processing includes anomaly detection based on real-time threshold comparison and the in-depth analysis includes data modeling using machine learning algorithms. S3: Data information is transmitted through the communication module, wherein the communication module algorithm selects the optimal transmission path based on real-time monitored network bandwidth, latency, and packet loss rate parameters through a dynamic path optimization algorithm. The calculation method of the dynamic path optimization algorithm includes: for each available transmission path, a weighted sum is performed based on the bandwidth value, the reciprocal of the latency value, and the complement of the packet loss rate to determine a comprehensive score, wherein the weighting coefficient is adaptively adjusted according to the network conditions, and the path with the highest comprehensive score is selected as the optimal path.

[0008] Furthermore, the data acquisition module supports at least one of the industrial communication protocols including Modbus, OPC UA, and MQTT, wherein automatic protocol identification is achieved by matching data packet signatures with the protocol database, and format conversion is accomplished by converting the data into JSON or XML standard format.

[0009] Furthermore, in S2, the fast processing performed at the edge computing node also includes data filtering or aggregation using lightweight algorithms, such as moving average filtering or median filtering; the deep analysis performed on the cloud computing platform also includes trend prediction using time series analysis models, such as autoregressive integral moving average models or long short-term memory networks.

[0010] Furthermore, the weighted summation formula for the dynamic path optimization algorithm is: ; Where S is the path comprehensive score, B is the current bandwidth measurement, D is the current latency measurement, and L is the current packet loss rate measurement; B_max, D_max, and L_max are the preset maximum bandwidth threshold, maximum latency threshold, and maximum packet loss rate threshold, respectively; the weighting coefficients W1, W2, and W3 are dynamically calculated in the following ways: Where B_th, D_th, and L_th are the dynamic adjustment thresholds for bandwidth, latency, and packet loss rate, respectively, and α, β, and γ are configurable sensitivity coefficients.

[0011] Furthermore, the dynamic path optimization algorithm also includes: when the comprehensive score S of a certain path is lower than the preset minimum threshold S_min, the path is marked as an unusable path; when the scores of all available paths are lower than S_min, a backup transmission mode is started, wherein the backup transmission mode adopts a data fragmentation and multi-path parallel transmission strategy.

[0012] Furthermore, for wired networks, α=0.8, β=1.2, and γ=1.0 are set; for wireless networks, α=0.5, β=0.8, and γ=1.5 are set; and for 5G private networks, α=1.2, β=1.5, and γ=0.8 are set.

[0013] Furthermore, the dynamic path optimization algorithm also includes: recording historical transmission performance data for each path and establishing a transmission performance prediction model. ; Where λ is the forgetting factor, S_current is the score calculated in real time, and S_history is the weighted average score based on historical data.

[0014] Furthermore, the method also includes a modular interface design, in which the data acquisition module, data processing module, communication module and user terminal interact with each other through a standardized RESTful API interface. The RESTful API interface supports OAuth2.0 authentication and JSON data format, and allows modules to be dynamically loaded and unloaded through a plugin mechanism to achieve hot-swappable replacement.

[0015] Furthermore, in S2, the in-depth analysis performed on the cloud computing platform also includes trend prediction using a time series analysis model, which is based on an autoregressive integral moving average model or a long short-term memory network.

[0016] The beneficial effects of the technical solution provided in this application include: adopting an automatic protocol identification and conversion mechanism, which can parse the protocol identifiers of heterogeneous data and convert the data into a unified format such as JSON or XML according to a predefined mapping table. This effectively solves the multi-protocol compatibility problem in industrial environments, simplifies the data integration process, improves the accuracy and consistency of data acquisition, and supports multiple industrial communication protocols including Modbus, OPC UA and MQTT.

[0017] Real-time data is processed rapidly at edge computing nodes, including anomaly detection based on real-time thresholds and data filtering or aggregation using lightweight algorithms such as moving average filtering or median filtering. This ensures low-latency response and real-time monitoring capabilities, suitable for critical industrial scenarios. Non-real-time data is uploaded to a cloud computing platform for in-depth analysis, utilizing machine learning algorithms for data modeling and time series analysis models such as autoregressive integral moving average models or long short-term memory networks for trend prediction. This enhances data insights and predictive accuracy, supporting long-term optimization and decision support.

[0018] The algorithm employs a dynamic path optimization mechanism to monitor network bandwidth, latency, and packet loss rate in real time. It calculates a comprehensive path score based on a weighted summation formula, with weighting coefficients adaptively adjusted according to network conditions to select the optimal transmission path. This improves data transmission reliability and efficiency while reducing network congestion and packet loss risks. Furthermore, the algorithm includes a backup transmission mode, employing data fragmentation and multi-path parallel transmission strategies when the path score falls below a threshold, further enhancing system robustness. The algorithm also integrates historical performance data prediction and adjusts for a forgetting factor to achieve more intelligent path selection. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating an information data processing method based on the Industrial Internet, provided as an embodiment of this application. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] Please refer to Figure 1 As shown in the figure, this application provides an information data processing method based on the Industrial Internet, including the following steps: S1: Collect operating data of industrial equipment through the data acquisition module. The data acquisition module is configured to convert heterogeneous operating data into a unified format through an automatic protocol identification and conversion mechanism. The automatic protocol identification and conversion mechanism includes parsing the protocol identifier in the header of the input data packet and performing format conversion according to a predefined protocol mapping table.

[0023] The data acquisition module supports at least one of the industrial communication protocols including Modbus, OPC UA, and MQTT. Automatic protocol identification is achieved by matching data packet signatures with the protocol database, and format conversion is accomplished by converting the data into JSON or XML standard format.

[0024] The data acquisition module is configured to collect operational data from industrial equipment through an automatic protocol identification and conversion mechanism. This mechanism aims to address the compatibility issues of heterogeneous data caused by various communication protocols in industrial environments. Industrial equipment typically uses different communication protocols, such as Modbus, OPC UA, or MQTT. These protocols have different data formats, and without standardized processing, they will hinder subsequent data analysis. Through automatic identification and conversion, this mechanism ensures that operational data enters the system in a standardized format, thereby improving the reliability and consistency of data acquisition and laying the foundation for the entire information data processing workflow.

[0025] The automatic protocol identification and conversion mechanism automatically detects the protocol type by parsing the protocol identifier in the header of the input data packet, without manual intervention. After identification, the mechanism converts heterogeneous data into a unified format, such as JSON or XML standard format, based on a predefined protocol mapping table.

[0026] A standardized data format facilitates real-time, rapid processing at edge computing nodes, such as threshold-based anomaly detection, and deep analysis on cloud computing platforms, such as modeling using machine learning algorithms. This optimizes data processing efficiency, reduces system complexity, and supports real-time monitoring and intelligent decision-making in industrial applications, ultimately improving the performance and practicality of the entire information data processing approach.

[0027] S2: The data processing module analyzes the collected operational data. The data processing module performs rapid processing of real-time data at edge computing nodes, and uploads non-real-time data to the cloud computing platform for in-depth analysis. Rapid processing includes anomaly detection based on real-time threshold comparison, and in-depth analysis includes data modeling using machine learning algorithms.

[0028] The data processing module is configured to analyze the collected operational data, and its purpose is to meet the diverse and real-time requirements of data processing in the Industrial Internet environment. The operational data generated by industrial equipment includes real-time data and non-real-time data. Real-time data requires rapid response to support immediate decision-making, while non-real-time data allows for more complex analysis to extract deeper insights.

[0029] By differentiating processing methods, this module optimizes resource utilization, ensures timely processing of key data, and provides a foundation for long-term trend analysis, thereby improving the adaptability and efficiency of the entire system.

[0030] In its implementation, the data processing module rapidly processes real-time data at edge computing nodes, including anomaly detection based on real-time threshold comparisons. This rapid processing mechanism uses lightweight algorithms such as moving average filtering or median filtering for data preprocessing, thereby quickly identifying abnormal device states, reducing latency, and avoiding the overhead of data transmission to the cloud. For non-real-time data, the module uploads it to a cloud computing platform for in-depth analysis, including data modeling using machine learning algorithms, such as autoregressive integral moving average models or long short-term memory networks for trend prediction. This supports more accurate model building and intelligent decision-making.

[0031] By combining edge computing and cloud computing, it reduces network load, improves response speed, and supports a wide range of applications, from real-time monitoring to predictive maintenance. This enhances the reliability and scalability of industrial internet information data processing methods, provides high-quality data input for subsequent communication module data transmission optimization, and ultimately enables efficient system operation and intelligent upgrades.

[0032] In S2, rapid processing at edge computing nodes also includes data filtering or aggregation using lightweight algorithms, such as moving average filtering or median filtering. Deep analytics on the cloud platform also include trend prediction using time series analysis models based on autoregressive integral moving average models or long short-term memory networks.

[0033] The method provided in this application embodiment also includes a modular interface design, wherein the data acquisition module, data processing module, communication module and user terminal interact with each other through a standardized RESTful API interface. The RESTful API interface supports OAuth2.0 authentication and JSON data format, and allows modules to be dynamically loaded and unloaded through a plugin mechanism to achieve hot-swappable replacement.

[0034] In S2, in-depth analysis performed on the cloud computing platform also includes trend prediction using time series analysis models based on autoregressive integral moving average models or long short-term memory networks.

[0035] Specifically, time series analysis models are based on autoregressive integral moving average (ARIMA) models or long short-term memory (LSTM) networks, which are configured to capture time-series characteristics in the data, such as trends, seasonality, and non-linear relationships.

[0036] The ARIMA model handles stationary and non-stationary time series using differencing, autoregressive, and moving average components, while LSTM, as a type of recurrent neural network, excels at learning long-term dependencies and is suitable for high-dimensional and complex data. In its implementation, the model utilizes the computing resources of a cloud computing platform for training and inference. By analyzing historical operational data, it generates predictive results, such as equipment failure probabilities or energy consumption trends, thereby providing data-driven support for industrial decision-making.

[0037] S3: Data information is transmitted through the communication module. The communication module algorithm selects the optimal transmission path based on real-time monitored network bandwidth, latency, and packet loss rate parameters using a dynamic path optimization algorithm. The calculation method of the dynamic path optimization algorithm includes: for each available transmission path, a weighted sum is performed based on the bandwidth value, the reciprocal of the latency value, and the complement of the packet loss rate to determine the comprehensive score. The weighting coefficient is adaptively adjusted according to the network conditions, and the path with the highest comprehensive score is selected as the optimal path.

[0038] The weighted summation formula for the dynamic path optimization algorithm is: .

[0039] Where S is the path comprehensive score, B is the current bandwidth measurement, D is the current latency measurement, and L is the current packet loss rate measurement. B_max, D_max, and L_max are the preset maximum bandwidth threshold, maximum latency threshold, and maximum packet loss rate threshold, respectively. The weighting coefficients W1, W2, and W3 are dynamically calculated in the following way: ; Where B_th, D_th, and L_th are the dynamic adjustment thresholds for bandwidth, latency, and packet loss rate, respectively, and α, β, and γ are configurable sensitivity coefficients.

[0040] The weighted summation formula of the dynamic path optimization algorithm is configured to calculate the comprehensive path score based on real-time network parameters. Its purpose is to solve the problem of unstable data transmission caused by dynamic changes in network conditions in the industrial Internet environment.

[0041] Industrial applications typically involve the transmission of large amounts of real-time and non-real-time data, and factors such as network bandwidth, latency, and packet loss rate can significantly impact transmission efficiency and quality. This algorithm enables the system to adaptively select the optimal path, ensuring reliable and efficient data transmission, thereby supporting the continuity and performance optimization of the entire information data processing flow.

[0042] In its implementation, the algorithm evaluates each available transmission path using a weighted summation formula. The overall score S is calculated by weighting the bandwidth value B, the reciprocal of the latency value D, and the complement of the packet loss rate L. The weighting coefficients W1, W2, and W3 are not fixed values ​​but are dynamically adjusted according to real-time network conditions, using an exponential function-based formula for adaptive calculation.

[0043] For example, the adjustment of W1 depends on the difference between the bandwidth measurement and the dynamic threshold B_th, and the response speed is controlled by the sensitivity coefficient α, similarly applicable to W2 and W3.

[0044] The dynamic weighting mechanism allows the algorithm to prioritize the most critical parameters under the current network conditions, such as giving higher weight to bandwidth when there is a high demand for bandwidth, or optimizing latency performance when there is a low demand for latency.

[0045] This technical feature plays a crucial role in the overall solution. As a core component of the communication module, it ensures that the output of the data processing module can be efficiently transmitted to the cloud computing platform or user terminal.

[0046] Through real-time monitoring and path optimization, the algorithm reduces data transmission latency and packet loss risk, supporting rapid processing and in-depth analysis of real-time data. This enhances the robustness and scalability of the entire approach, ultimately improving the reliability and intelligent decision-making capabilities of industrial internet applications.

[0047] Furthermore, the dynamic path optimization algorithm also includes: when the overall score S of a path is lower than a preset minimum threshold S_min, the path is marked as an unusable path. When the scores of all available paths are lower than S_min, a backup transmission mode is activated, which employs a data fragmentation and multi-path parallel transmission strategy.

[0048] The dynamic path optimization algorithm also includes marking a path as unavailable when the overall score S of a path is lower than the preset minimum threshold S_min, and activating a backup transmission mode when the scores of all available paths are lower than S_min. The purpose of this setting is to deal with the temporary failures or performance degradation that may occur in network paths in the industrial Internet environment, and to ensure the continuity and reliability of data transmission.

[0049] In its implementation, the algorithm monitors the overall score S of each path in real time and dynamically compares it with a preset threshold S_min. If a path's score is lower than S_min, it is immediately marked as unusable and excluded from the current transmission options, thus avoiding the use of low-quality paths.

[0050] When all paths are unavailable, the system automatically activates the backup transmission mode. This mode employs a data fragmentation and multi-path parallel transmission strategy, dividing the data into multiple segments and sending them simultaneously through different available paths. This strategy increases transmission redundancy and fault tolerance, leveraging network diversity to improve the success rate of data arrival; even if some paths fail, transmission can still be completed through other paths.

[0051] Furthermore, for wired networks, set α=0.8, β=1.2, and γ=1.0. For wireless networks, set α=0.5, β=0.8, and γ=1.5. For 5G private networks, set α=1.2, β=1.5, and γ=0.8.

[0052] The dynamic path optimization algorithm is configured to dynamically adjust the sensitivity coefficients α, β, and γ based on the network type. Its purpose is to optimize the path selection process to adapt to the characteristics of different network environments, thereby improving the accuracy and efficiency of data transmission in the Industrial Internet. Industrial applications involve diverse network infrastructures, such as wired networks, wireless networks, and 5G private networks, each exhibiting different behavioral patterns in terms of bandwidth, latency, and packet loss rate.

[0053] For example, wired networks are typically stable and have high bandwidth, wireless networks are susceptible to interference and have high packet loss rates, while 5G private networks offer high speed and low latency. By selectively setting these coefficients, the algorithm can more accurately weight real-time network parameters, ensuring that the path comprehensive score calculation reflects actual network conditions and avoiding performance degradation caused by fixed weights, thereby enhancing adaptability and reliability.

[0054] Specifically, for wired networks, α=0.8, β=1.2, and γ=1.0 are set. This is based on the high bandwidth stability and low packet loss characteristics of wired networks, which makes the algorithm focus more on bandwidth and latency optimization and reduce unnecessary path switching.

[0055] For wireless networks, α=0.5, β=0.8, and γ=1.5 are set because wireless networks are unstable and have a high risk of packet loss. Therefore, the packet loss rate is given a higher weight to prioritize ensuring data integrity.

[0056] For 5G private networks, α=1.2, β=1.5, and γ=0.8 are set. This leverages the high throughput and low latency advantages of 5G networks, emphasizing bandwidth and latency optimization to support real-time applications. This dynamic adjustment mechanism enables rapid response to network changes through configurable sensitivity coefficients, improving the algorithm's flexibility and accuracy.

[0057] As a crucial component of the communication module, it enables the dynamic path optimization algorithm to adapt to different network types, ensuring that the operational data output from the data processing module is efficiently and reliably transmitted to the cloud computing platform or edge nodes. This supports rapid processing and in-depth analysis of real-time data, reduces transmission latency and packet loss, and improves the overall performance and scalability of the entire information data processing method, making it suitable for diverse application scenarios in complex industrial internet environments.

[0058] Furthermore, the dynamic path optimization algorithm also includes: recording historical transmission performance data for each path and establishing a transmission performance prediction model. .

[0059] Where λ is the forgetting factor, S_current is the score calculated in real time, and S_history is the weighted average score based on historical data.

[0060] The dynamic path optimization algorithm also includes recording historical transmission performance data for each path and establishing a transmission performance prediction model. Its purpose is to solve the problem of time-series fluctuations in network path performance in the industrial internet environment, so as to improve the accuracy and stability of path selection.

[0061] Relying solely on real-time data can lead to path selection being overly sensitive to instantaneous changes, resulting in frequent switching and transmission instability. By incorporating historical data records and predictive models, the algorithm can comprehensively consider current and past performance, smooth out short-term fluctuations, and predict future trends, thereby making more reliable path decisions, reducing unnecessary path switching, and improving the continuity and efficiency of data transmission.

[0062] In its implementation, the prediction model uses the formula S_predict = λ × S_current + (1-λ) × S_history, where λ is the forgetting factor, S_current is the current real-time calculated path composite score, and S_history is the weighted average score based on historical data. The forgetting factor λ allows the algorithm to dynamically adjust the weight of historical data; a value closer to 1 indicates a greater focus on current data, adapting to rapidly changing network environments; a value closer to 0 indicates a greater reliance on historical data, providing more stable predictions.

[0063] This mechanism enables the algorithm to learn long-term behavior patterns of paths by continuously monitoring and updating historical records. For example, it can identify certain paths that perform well during specific periods, thereby optimizing the selection process and avoiding misjudging path quality due to temporary performance degradation.

[0064] As a crucial component of the dynamic path optimization algorithm, it ensures that the operational data output from the data processing module can be efficiently and reliably transmitted to the cloud computing platform or edge computing nodes via the optimal path. This supports the rapid processing and in-depth analysis of real-time data, reducing transmission latency, packet loss, and interruption risks, thereby improving the overall performance, reliability, and scalability of the entire information data processing method. Ultimately, it is suitable for complex industrial internet applications, such as real-time monitoring and predictive maintenance, contributing to the efficient operation and intelligent upgrading of the system.

[0065] This algorithm enables the system to dynamically evaluate available transmission paths and calculate a comprehensive score based on key indicators such as bandwidth, latency, and packet loss rate. This optimizes path selection, reduces transmission interruptions, and supports the continuity and performance of the entire information data processing method.

[0066] The specific logic of the weighted summation formula is based on the normalization of real-time measurements for each path. The ratio of bandwidth value B to the maximum threshold B_max reflects bandwidth utilization; the ratio of the reciprocal of latency value D to the maximum latency D_max emphasizes the importance of low latency; and the ratio of the complement of packet loss rate L to the maximum packet loss rate L_max ensures data integrity. The weighting coefficients W1, W2, and W3 are not fixed but dynamically calculated using an exponential function to adapt to changes in network conditions. For example, the adjustment of W1 depends on the difference between the bandwidth measurement and the dynamic threshold B_th, and the response speed is controlled by a sensitivity coefficient α. A similar mechanism applies to W2 and W3. This design allows the algorithm to prioritize bandwidth when bandwidth is sufficient, optimize latency when latency is sensitive, or focus on reliability when packet loss is high, thus flexibly responding to dynamic conditions in industrial networks.

[0067] By integrating with the communication module, it works in conjunction with the unified format output of the data acquisition module and the real-time analysis of the data processing module, enhancing the robustness and adaptability of data transmission. This ultimately improves the overall performance of the method, supporting intelligent decision-making and scalability in industrial applications, such as real-time monitoring and predictive maintenance, contributing to the efficient operation of the system.

[0068] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0069] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0070] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0071] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0072] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. An information data processing method based on the Industrial Internet, characterized in that, Includes the following steps: S1: The data acquisition module collects the operating data of the industrial equipment. The data acquisition module is configured to convert the heterogeneous operating data into a unified format through an automatic protocol identification and conversion mechanism. The automatic protocol identification and conversion mechanism includes parsing the protocol identifier in the header of the input data packet and performing format conversion according to a predefined protocol mapping table. S2: The collected running data is analyzed by the data processing module. The data processing module performs fast processing on real-time data at the edge computing node and uploads non-real-time data to the cloud computing platform for in-depth analysis. The fast processing includes anomaly detection based on real-time threshold comparison and the in-depth analysis includes data modeling using machine learning algorithms. S3: Data information is transmitted through the communication module, wherein the communication module algorithm selects the optimal transmission path based on real-time monitored network bandwidth, latency, and packet loss rate parameters through a dynamic path optimization algorithm. The calculation method of the dynamic path optimization algorithm includes: for each available transmission path, a weighted sum is performed based on the bandwidth value, the reciprocal of the latency value, and the complement of the packet loss rate to determine a comprehensive score, wherein the weighting coefficient is adaptively adjusted according to the network conditions, and the path with the highest comprehensive score is selected as the optimal path.

2. The method according to claim 1, characterized in that, The data acquisition module supports at least one of the industrial communication protocols including Modbus, OPC UA, and MQTT. Automatic protocol identification is achieved by matching data packet signatures with the protocol database, and format conversion is accomplished by converting the data into JSON or XML standard format.

3. The method according to claim 1, characterized in that, In S2, the fast processing performed at the edge computing node also includes data filtering or aggregation using lightweight algorithms, such as moving average filtering or median filtering; the deep analysis performed on the cloud computing platform also includes trend prediction using time series analysis models, such as autoregressive integral moving average models or long short-term memory networks.

4. The method according to claim 1, characterized in that, The weighted summation formula for the dynamic path optimization algorithm is: ； Where S is the path comprehensive score, B is the current bandwidth measurement, D is the current latency measurement, and L is the current packet loss rate measurement; B_max, D_max, and L_max are the preset maximum bandwidth threshold, maximum latency threshold, and maximum packet loss rate threshold, respectively; the weighting coefficients W1, W2, and W3 are dynamically calculated in the following ways: Where B_th, D_th, and L_th are the dynamic adjustment thresholds for bandwidth, latency, and packet loss rate, respectively, and α, β, and γ are configurable sensitivity coefficients.

5. The method according to claim 4, characterized in that, The dynamic path optimization algorithm further includes: when the comprehensive score S of a certain path is lower than the preset minimum threshold S_min, the path is marked as an unusable path; when the scores of all available paths are lower than S_min, a backup transmission mode is started, wherein the backup transmission mode adopts a data fragmentation and multi-path parallel transmission strategy.

6. The method according to claim 5, characterized in that, For wired networks, set α=0.8, β=1.2, and γ=1.0; for wireless networks, set α=0.5, β=0.8, and γ=1.5; for 5G private networks, set α=1.2, β=1.5, and γ=0.

8.

7. The method according to claim 5, characterized in that, The dynamic path optimization algorithm also includes: recording historical transmission performance data for each path and establishing a transmission performance prediction model. ； Where λ is the forgetting factor, S_current is the score calculated in real time, and S_history is the weighted average score based on historical data.

8. The method according to claim 1, characterized in that, The method also includes a modular interface design, in which the data acquisition module, data processing module, communication module and user terminal interact with each other through a standardized RESTful API interface. The RESTful API interface supports OAuth2.0 authentication and JSON data format, and allows modules to be dynamically loaded and unloaded through a plugin mechanism to achieve hot-swappable replacement.

9. The method according to claim 1, characterized in that, In S2, the in-depth analysis performed on the cloud computing platform also includes trend prediction using a time series analysis model, which is based on an autoregressive integral moving average model or a long short-term memory network.

Images