Industrial production intelligent data analysis and monitoring system based on micro-service architecture

Abstract

The system is divided into a data acquisition service module, a production management service module, a quality control service module, a risk assessment service module, a data analysis service module and a system monitoring and management service module by adopting a micro-service architecture; the data acquisition service module is efficiently connected with production equipment through a first protocol and is used for acquiring production data of the production equipment in real time; the quality control service module analyzes data in the production process in real time by using a machine learning algorithm, automatically identifies potential quality problems in production and positions the potential quality problems; the data analysis service module is integrated with an efficient distributed computing platform, and the distributed computing platform is Hadoop or Spark and supports deep analysis of historical data and real-time data; the risk assessment service module can assess and predict potential risks in the production process by combining big data analysis and an artificial intelligence model.

Description

Technical Field

[0001] The present invention relates to an intelligent data analysis and real-time monitoring system for industrial production, and in particular to an intelligent data analysis and monitoring system for industrial production based on a microservice architecture. Background Art

[0002] In the context of Industry 4.0, with the continuous maturity of technologies such as the Internet of Things, big data, cloud computing, and artificial intelligence, the modern manufacturing industry has an increasing demand for automation and intelligence, and the production model is transforming towards high intelligence, digitization, and networking.

[0003] Traditional manufacturing management systems often adopt a single or multi-layer architecture, which is complex and has poor scalability, making it difficult to meet the company's requirements for flexibility, efficiency and intelligence. In addition, problems such as quality control and risk assessment in the industrial production process are becoming increasingly prominent, and traditional systems are unable to intelligently optimize production processes and issue warnings through big data analysis.

[0004] As an emerging software architecture model, microservice architecture improves the flexibility, scalability and maintainability of the system by splitting complex software systems into a set of small and independent services, each of which focuses on a specific business function. In the field of industrial intelligent manufacturing, microservice architecture can effectively cope with complex production environments, support real-time data processing, intelligent analysis and automated control, and provide a solid technical foundation for industrial production, intelligent manufacturing, quality management and risk assessment.

[0005] At present, some enterprises have tried to use microservice architecture to improve the shortcomings of traditional manufacturing systems. Some breakthroughs have been made, and the scalability and maintainability of the system have been improved, but there are still many problems. For example, the focus of these solutions is only on how to solve the modular management of production processes, simplify the complexity and maintainability of the system, and do not really solve the actual needs of enterprises for the flexibility, efficiency, and intelligence of industrial intelligent manufacturing platforms, as well as quality control and risk warning in the production process. The following are the specific shortcomings of the current technical solutions:

[0006] 1. Lack of intelligence: The existing system relies too much on manual experience and static statistical models in terms of quality problem location and risk warning, and lacks the ability of real-time data analysis and intelligent prediction. This makes it impossible to identify and proactively respond to quality problems and potential risks in the production process in a timely manner, affecting production efficiency and product quality.

[0007] 2. Difficulty in coping with complex quality control needs: Industrial production processes are complex and involve many variables, and existing systems are unable to quickly and accurately locate quality issues, especially in terms of equipment fault diagnosis and problem repair recommendations.

[0008] 3. Limited data integration capabilities: Although the existing platform adopts a microservice architecture, there are still limitations in data integration and real-time sharing between different services, resulting in a serious phenomenon of information islands.

[0009] 4. Insufficient integration: The existing platform has bottlenecks in cross-system collaboration and multi-source data integration, which results in the inability to seamlessly connect data and business processes, limiting the overall effectiveness of the platform. Summary of the invention

[0010] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide an industrial production intelligent data analysis and monitoring system based on a microservice architecture. The system aims to utilize the flexibility, scalability and distributed computing capabilities of the microservice architecture, combined with big data analysis and artificial intelligence technology, to provide functions such as industrial production, intelligent manufacturing, quality problem location, and risk warning assessment.

[0011] The technical solution adopted by the present invention is: the system adopts a microservice architecture to split the present invention into a data acquisition service module, a production management service module, a quality control service module, a risk assessment service module, a data analysis service module and a system monitoring and management service module, and the data acquisition service module, the production management service module, the quality control service module, the risk assessment service module, the data analysis service module and the system monitoring and management service module all realize real-time data sharing and analysis among the modules through a data stream transmission mechanism;

[0012] The data acquisition service module is efficiently connected to the production equipment through the first protocol, and is used to collect production data of the production equipment in real time, and distribute the production data to the production management service module, the quality control service module and the data analysis service module through the message queue;

[0013] Compared with traditional data collection methods, the service of the present invention has the advantages of high concurrent processing capability and real-time data transmission, ensuring timely acquisition and efficient processing of production data; and the first protocol supports multiple protocols, such as MQTT, OPC UA and other protocols;

[0014] The production management service module is used to realize the collection of management data for automatic allocation of production plans, work order status tracking and equipment status monitoring through Spring Cloud after receiving the production data from the data acquisition service module and the analysis data from the data analysis service module, and distribute the management data to the data analysis service module and the system monitoring and management service module;

[0015] Compared with traditional production management systems, the present invention achieves higher flexibility and scalability through a microservice architecture, can quickly adapt to changes in the production environment, and provide more refined work order management and equipment maintenance services;

[0016] The quality control service module is used to perform real-time analysis on the data in the production process by using a machine learning algorithm after receiving the production data from the data acquisition service module and the analysis data from the data analysis service module, so as to automatically identify and locate potential quality problems in production. At the same time, the data of the quality control service module can be transmitted to the risk assessment service module and the data analysis service module;

[0017] Compared with the traditional method that relies on manual inspection, the present invention uses a data-driven quality control method, which not only improves the accuracy of detection, but also greatly reduces human errors, and improves product consistency and quality pass rate;

[0018] The data analysis service module integrates an efficient distributed computing platform, which is Hadoop or Spark, and supports in-depth analysis of historical data and real-time data. The data analysis service module shares data with the other five service modules in real time.

[0019] Compared with traditional data analysis methods, the present invention can process larger-scale production data and provide more accurate analysis results through efficient computing power, supporting deeper business insights;

[0020] The risk assessment service module can assess and predict potential risks in the production process by combining big data analysis and artificial intelligence models; the risk assessment service module can receive data from the quality control service module and the data analysis service module, and can also transmit risk assessment data to the data analysis service module and the system monitoring and management service module;

[0021] Different from the traditional experience-based risk management method, the present invention can provide more accurate risk warnings through the combination of real-time data and historical data, and generate actionable optimization suggestions to avoid possible production problems in advance;

[0022] The system monitoring and management service module implements system status monitoring through Prometheus and combines with ElasticSearch for log analysis. This module can monitor the operating status of all microservices in real time; the system monitoring and management service module can receive data from the production management service module, the data analysis service module, and the risk assessment service module, and can also transmit the monitored data to the data analysis service module;

[0023] Compared with the traditional single monitoring system, the present invention improves the response speed of fault detection through a distributed monitoring platform, and realizes in-depth diagnosis of system health status through log analysis.

[0024] Furthermore, the data analysis service module includes a data deep mining module and an intelligent data analysis module. The data deep mining module is used to deeply analyze the massive data generated in the production process, accurately identify potential problems and optimize the production process, and improve production efficiency and product quality; the intelligent data analysis module is used to combine machine learning and deep learning models to improve data processing capabilities and analysis accuracy, improve production efficiency, reduce resource waste, and reduce production failure rates.

[0025] Furthermore, the data deep mining module includes a multidimensional data analysis unit, an association rule mining unit and a time series analysis unit; the multidimensional data analysis unit supports fast multidimensional query and complex data summary analysis through OLAP technology. Compared with traditional methods, this technology can complete data analysis within a few seconds, which is about 3 times faster than traditional data analysis. It can analyze production data from multiple angles at the same time and quickly discover potential production bottlenecks and problems; the association rule mining unit uses the Apriori algorithm to mine potential associations in the production process. For example, by analyzing the association between production equipment failures and environmental changes, the potential connection between ambient temperature changes and equipment failures is discovered. This technology can improve the accuracy of equipment failure prediction and warn of failure risks 30%-40% in advance; the time series analysis unit uses the ARIMA model to perform time series prediction on production data, identify production trends or fluctuations in advance, and help management adjust strategies in time during the production process. The accuracy of the time prediction model is improved by 25%, greatly reducing the risk of emergencies in the production process.

[0026] Furthermore, the intelligent data analysis module cleans, converts and selects features of industrial data through feature engineering, combines supervised learning, unsupervised learning and deep learning models for quality prediction and anomaly detection, and through refined feature engineering and model optimization, achieves more accurate quality prediction and production process anomaly detection, further improving production efficiency and product quality.

[0027] The feature engineering process for cleaning industrial data is to clean dirty data such as missing values ​​and outliers, and use algorithms to select features and extract key information. This process ensures that the data quality is improved by 20%-30%, which greatly improves the accuracy of subsequent analysis models. Through precise feature extraction, the present invention can automatically identify more than 80% of key factors, effectively improving the reliability of prediction. The construction of the learning model is for labeled data, such as product quality prediction or abnormal fault detection. The supervised learning model is used for training through historical data. The model can accurately identify quality problems in different production links. Compared with traditional methods, the accuracy rate is improved by about 15%. For unlabeled data, unsupervised learning (such as cluster analysis and principal component analysis) is used for pattern discovery and anomaly detection, which can identify more than 80% of anomalies in the production process. At the same time, deep learning technology is also used: through the convolutional neural network (CNN) to analyze the product appearance image, defects and flaws are automatically identified. Compared with the traditional manual inspection method, the defect detection rate is increased by 30%. Recursive neural network (RNN) is used for time series data analysis, which can accurately capture the dynamic changes of production equipment and predict equipment failure trends, reducing the equipment failure rate by 20%. In addition, the model is optimized by cross-validation, ROC curve and other technologies to optimize the model parameters to ensure the efficiency of the model in the actual production environment. After the model is optimized, the prediction accuracy of production quality is improved by 15%-20%. After being deployed in the microservice architecture, the system can realize real-time data analysis and response, and modules such as production management and quality control can obtain analysis results in real time to assist decision-making.

[0028] Furthermore, the risk assessment service module includes an identification and classification unit, a modeling and prediction unit, and a real-time monitoring and early warning unit; the identification and classification unit accurately identifies potential risk factors by analyzing equipment status, production progress and quality control data in combination with historical data. The identification and classification unit uses a risk matrix to classify risks according to the probability of occurrence and the degree of impact, helping management to quickly identify high-risk areas and take countermeasures in advance. The risk identification rate is 40%-50% higher than that of traditional methods. The modeling and prediction unit uses regression analysis and machine learning methods to establish a risk prediction model, analyze the impact of various risk factors on production, and use model effect evaluation charts (such as ROC curves and confusion matrices) to continuously optimize model parameters and improve prediction accuracy. Compared with traditional empirical judgments, the prediction accuracy rate is increased by 25%-30%, significantly reducing sudden problems in the production process; the real-time monitoring and early warning unit combines sensor data with historical analysis results to monitor key indicators in the production process in real time. Once the indicators exceed the set threshold, the system automatically triggers an early warning and promptly notifies relevant personnel to take measures. Through this mechanism, the incidence of production accidents has been reduced by 35%-40%. The monitoring panel displays changes in key indicators in real time, helping management make faster and more accurate decisions during the production process.

[0029] Furthermore, the risk assessment service module transmits the risk assessment results to the system monitoring and management service module, which helps the management to intuitively understand the data analysis results and make scientific decisions through the visualization of multi-dimensional data analysis results, such as production efficiency graphs, quality monitoring graphs, etc. By integrating a variety of data visualization methods and tools, the present invention realizes the efficient combination of data display in the production process and management decision support.

[0030] The beneficial effects of the present invention are as follows: Due to the modular design of the present invention, the platform supports rapid deployment and iteration, optimizes the production process and improves equipment utilization. At the same time, combined with data mining and intelligent analysis technology, it is possible to discover and solve quality problems in real time, significantly improving the product qualification rate. The risk assessment method is adopted to achieve real-time monitoring and early warning of the entire production process, effectively reducing the risk of sudden accidents. Through multi-dimensional production data analysis and visual reports, management can obtain more scientific decision-making support and further optimize production decisions. Finally, the system ensures that even if some modules fail, the overall system can still maintain stable operation through high independence and fault isolation mechanism between modules, ensuring the high reliability and sustainability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a schematic diagram of the principle structure of the present invention;

[0032] Figure 2It is a flowchart of deep data mining and intelligent analysis of the data analysis service model;

[0033] Figure 3 It is the risk assessment flow chart of the risk assessment service module. DETAILED DESCRIPTION

[0034] like Figures 1 to 3 As shown, in this embodiment, the system adopts a microservice architecture to split the intelligent data analysis and monitoring system into a data acquisition service module 1, a production management service module 2, a quality control service module 3, a risk assessment service module 4, a data analysis service module 5 and a system monitoring and management service module 6. The data acquisition service module 1, the production management service module 2, the quality control service module 3, the risk assessment service module 4, the data analysis service module 5 and the system monitoring and management service module 6 all realize real-time data sharing and analysis between the modules through a data stream transmission mechanism;

[0035] The data acquisition service module 1 is efficiently connected to the production equipment through the first protocol, and is used to collect the production data of the production equipment in real time, and distribute the production data to the production management service module 2, the quality control service module 3 and the data analysis service module 5 through a message queue (such as Kafka);

[0036] The production management service module 2 realizes the automatic allocation of production plans, work order status tracking and the collection of management data for equipment status monitoring through Spring Cloud, where Spring Cloud is an ordered collection of a series of frameworks. It uses the development convenience of Spring Boot to cleverly simplify the development of distributed system infrastructure, such as service discovery registration, configuration center, message bus, load balancing, circuit breaker, data monitoring, etc., which can all be started and deployed with one click using the SpringBoot development style;

[0037] The quality control service module 3 uses machine learning algorithms to perform real-time analysis on data in the production process, automatically identifying potential quality problems in production and locating them;

[0038] The data analysis service module 5 integrates an efficient distributed computing platform (such as Hadoop and Spark), also known as a data warehouse, to support in-depth analysis of historical data and real-time data;

[0039] The risk assessment service module 4 combines big data analysis and artificial intelligence models to assess and predict potential risks in the production process, and provides more accurate risk warnings through the combination of real-time data and historical data, and generates operational optimization suggestions to avoid possible production problems in advance;

[0040] The system monitoring and management service module 6 implements system status monitoring through Prometheus (Prometheus is an open source service monitoring system and time series database), and combines ElasticSearch (Elasticsearch is a distributed search and analysis engine at the core of the Elastic Stack) for log analysis. This module can monitor the operating status of all microservices in real time.

[0041] The present invention adopts advanced multi-protocol data acquisition technology to obtain data from production equipment and sensors in real time, and transmits and distributes data through efficient message queues. Subsequently, the data is deeply processed using efficient processing technology of real-time data streams to form an efficient data processing chain. This technology combination greatly improves the concurrent processing capability of data acquisition and the efficiency of data transmission, and can achieve fast and stable real-time data processing in a large-scale production environment, ensuring smooth and timely transmission and processing of data in the production process; more specifically, it includes the following technical points: 1. Application method and technical implementation of MQTT protocol in production equipment data acquisition; 2. Application method and technical implementation of OPC UA protocol in data transmission between industrial equipment and sensors; 3. Heterogeneous device access method and multi-protocol data integration solution of data acquisition system; 4. Application method and technical implementation of Kafka in data stream transmission; 5. Real-time data filtering, aggregation and analysis technology solution based on stream processing framework (Apache Flink, Spark Streaming); 6. Implementation and optimization technology of real-time monitoring, early warning and decision support functions of data stream.

[0042] In this embodiment, the data analysis service module 5 includes a data deep mining module and an intelligent data analysis module. The data deep mining module is used to deeply analyze the massive data generated in the production process, accurately identify potential problems and optimize the production process, and improve production efficiency and product quality; the intelligent data analysis module is used to combine machine learning and deep learning models to improve data processing capabilities and analysis accuracy, improve production efficiency, reduce resource waste, and reduce production failure rates.

[0043] The present invention combines convolutional neural network (CNN) and recurrent neural network (RNN) to realize the detection of product image quality and dynamic analysis of time series data. By optimizing the deep learning model, the accuracy and efficiency of quality control are improved. CNN is used for image quality detection, and RNN is used to analyze time series data in the production process, further improving the automation and accuracy of quality control. The specific technical points include the following: 1. Application method and optimization technology of convolutional neural network (CNN) in product image quality detection; 2. Application and optimization technology of recurrent neural network (RNN) in dynamic analysis of time series data; 3. Specific application of deep learning model in quality control and its optimization method.

[0044] In this embodiment, the data deep mining module includes a multidimensional data analysis unit, an association rule mining unit and a time series analysis unit; the multidimensional data analysis unit supports fast multidimensional query and complex data summary analysis through OLAP technology; the association rule mining unit uses the Apriori algorithm to mine potential associations in the production process; the time series analysis unit uses the ARIMA model to perform time series forecasting on production data, identify production trends or fluctuations in advance, and help management adjust strategies in a timely manner during the production process.

[0045] The present invention combines OLAP technology with association rule mining algorithms (such as Apriori, FP-Growth) and time series analysis technology (such as ARIMA) to conduct in-depth mining and trend prediction of production data; OLAP technology supports efficient query and analysis of multi-dimensional production data, helping users to explore data from multiple angles, greatly improving data analysis efficiency and problem discovery capabilities; association rule mining algorithms accurately identify potential associations between different variables, and ARIMA models predict trends in production data to discover potential risks in advance. Through these technologies, the present invention can provide efficient and accurate monitoring and decision support for the production process, optimize production efficiency and quality control. More specifically, the following technical points are included: 1. Innovative application methods of OLAP technology in industrial data analysis; 2. Optimized application of association rule mining algorithms (Apriori, FP-Growth) in production data analysis; 3. Innovative methods and applications of ARIMA models in production data trend prediction.

[0046] In this embodiment, the intelligent data analysis module cleans, converts and selects features of industrial data through feature engineering, combines supervised learning, unsupervised learning and deep learning models for quality prediction and anomaly detection, and through refined feature engineering and model optimization, achieves more accurate quality prediction and production process anomaly detection, further improving production efficiency and product quality.

[0047] The present invention uses feature engineering to clean, transform and select features of industrial data, and combines supervised learning, unsupervised learning and deep learning models for quality prediction and anomaly detection; this method not only improves the accuracy of quality control, but also greatly reduces the errors of manual intervention; through refined feature engineering and model optimization, more accurate quality prediction and production process anomaly detection are achieved, further improving production efficiency and product quality. More specifically, the following technical points are included: 1. Application and optimization methods of feature engineering in industrial data analysis; 2. Application technology of supervised learning, unsupervised learning and deep learning models in quality prediction and anomaly detection; 3. Model optimization methods and their application in production data analysis.

[0048] In this embodiment, the risk assessment service module 4 includes an identification and classification unit, a modeling and prediction unit, and a real-time monitoring and early warning unit; the identification and classification unit accurately identifies potential risk factors by combining historical data to analyze equipment status, production progress and quality control data. The identification and classification unit uses a risk matrix to classify risks according to the probability of occurrence and the degree of impact, helping management to quickly identify high-risk areas and take countermeasures in advance. The risk identification rate is increased by 40%-50% compared with traditional methods. The modeling and prediction unit uses regression analysis and machine learning methods to establish a risk prediction model, analyze the impact of each risk factor on production, and use model effect evaluation diagrams (such as ROC curves and confusion matrices) to continuously optimize model parameters and improve prediction accuracy. Compared with traditional empirical judgments, the prediction accuracy is increased by 25%-30%, significantly reducing sudden problems in the production process; the real-time monitoring and early warning unit combines sensor data with historical analysis results to monitor key indicators in the production process in real time. Once the indicator exceeds the set threshold, the system automatically triggers an early warning and promptly notifies relevant personnel to take measures. Through this mechanism, the incidence of production accidents has been reduced by 35%-40%. The monitoring panel displays changes in key indicators in real time, helping management make faster and more accurate decisions during the production process.

[0049] Among them, the present invention combines logistic regression and machine learning models, uses risk matrix to realize real-time risk assessment, sets early warning thresholds for key indicators, and establishes an automated early warning system, which can timely warn of potential production risks, reduce the incidence of sudden accidents, and improve production safety. This method provides enterprises with a dynamic and intelligent risk management mechanism, which enhances the stability and safety of the production process. More specifically, it includes the following technical points: 1. Risk assessment method based on logistic regression and machine learning models; 2. Setting early warning thresholds for key indicators and realizing and optimizing the automated early warning system; 3. Application technology of real-time risk assessment and early warning mechanism.

[0050] In this embodiment, the risk assessment service module 4 transmits the risk assessment results to the system monitoring and management service module 6. The system monitoring and management service module 6 helps the management intuitively understand the data analysis results and make scientific decisions through the visual display of multi-dimensional data analysis results, such as production efficiency charts, quality monitoring charts, etc. By integrating various data visualization methods and tools, the present invention realizes the efficient combination of data display in the production process and decision-making support for the management.

[0051] More specifically, it includes the following technical points: 1. The integration technology of data visualization methods and tools; 2. The specific application of data visualization in production management and quality control.

[0052] In addition, the present invention uses Prometheus and ElasticSearch to implement real-time monitoring and log analysis of microservices. By quickly locating faults and taking isolation measures, the high availability of the system is ensured. This method improves the reliability and maintainability of the system. The points to be protected include the real-time monitoring mechanism based on the microservice architecture, the fault isolation method, and the log analysis technology. More specifically, it includes the following technical points: 1. The real-time monitoring and log analysis technology based on the microservice architecture; 2. The fault isolation method and the system reliability improvement plan; 3. The application technology of Prometheus and ElasticSearch in microservice monitoring.

[0053] Compared with the prior art, the present invention has significant advantages in multiple key fields, especially in system flexibility, decision-making support, data processing efficiency, and risk control. Through innovative design and technology application, the present invention not only solves the deficiencies in traditional systems but also realizes improvements in multiple dimensions.

[0054] First of all, by adopting the microservice architecture, the present invention splits the system into multiple independent functional service modules, significantly improving the flexibility and scalability of the system. Compared with the traditional monolithic architecture, the modular design enables the introduction of new functions and system upgrades to be faster, and the average iteration cycle is shortened by more than 30%. In addition, the modular architecture can support the expansion of up to hundreds of concurrent modules, avoiding the performance bottlenecks faced by traditional systems when the demand expands.

[0055] Secondly, the present invention introduces machine learning and deep learning technologies. Through multi-dimensional data analysis and time series prediction, it can accurately identify potential quality problems and risks in the production process. This technology improves the automation level of quality control, making the problem identification accuracy in the production process increase by about 40%. Compared with the traditional method that relies on manual inspection, the present invention reduces human errors while increasing the product quality pass rate by 15%. Based on the real-time data and risk assessment system, the management can predict potential problems 30%-50% earlier, greatly reducing the probability of production accidents.

[0056] In terms of data processing, the present invention combines distributed computing (such as Spark, Hadoop) and streaming data processing technology (such as Kafka), which can efficiently process several TB of production data per day and ensure real-time response capabilities in a high-concurrency environment. Compared with traditional technologies, the data processing speed is increased by about 3 times, and the system response time is shortened from several minutes to a few seconds, meeting the high requirements of modern industrial production for real-time and performance.

[0057] Finally, the modular design of the present invention also brings stronger system stability. The fault isolation mechanism ensures that when an individual module fails, other modules can continue to operate normally, and the average availability of the system is improved to 99.9%. The real-time risk monitoring and early warning system can identify more than 90% of potential failures in advance, greatly reducing production interruptions and equipment losses.

[0058] In summary, the present invention solves the bottlenecks in traditional systems through innovative technical architecture and intelligent analysis methods, improves production efficiency, quality control, data processing and risk management capabilities, and provides a more efficient, flexible and reliable solution for the field of industrial intelligent manufacturing.

[0059] The present invention is applied to the technical field of industrial intelligent manufacturing.

[0060] Although the embodiments of the present invention are described with practical solutions, they do not constitute limitations on the meaning of the present invention. For those skilled in the art, it is obvious to modify the implementation scheme and combine it with other solutions based on this description.

Claims

1. An industrial production intelligent data analysis and monitoring system based on microservice architecture, characterized by: The system adopts a microservice architecture to split the intelligent data analysis and monitoring system into a data acquisition service module (1), a production management service module (2), a quality control service module (3), a risk assessment service module (4), a data analysis service module (5) and a system monitoring and management service module (6); the data acquisition service module (1), the production management service module (2), the quality control service module (3), the risk assessment service module (4), the data analysis service module (5) and the system monitoring and management service module (6) all realize real-time data sharing and analysis among the modules through a data stream transmission mechanism; The data acquisition service module (1) is connected to the production equipment through a first protocol, and is used to collect production data of the production equipment in real time, and distribute the production data to the production management service module (2), the quality control service module (3) and the data analysis service module (5) through a message queue; The production management service module (2) realizes the collection of management data for automatic allocation of production plans, work order status tracking and equipment status monitoring through Spring Cloud; The quality control service module (3) uses a machine learning algorithm to perform real-time analysis on data in the production process, automatically identifying potential quality problems in production and locating them; The data analysis service module (5) integrates a distributed computing platform to support in-depth analysis of historical data and real-time data; The risk assessment service module (4) can evaluate and predict potential risks in the production process by combining big data analysis and artificial intelligence models, and provide more accurate risk warnings by combining real-time data and historical data, and generate operational optimization suggestions to avoid possible production problems in advance; The system monitoring and management service module (6) implements system status monitoring through Prometheus and combines with ElasticSearch for log analysis. This module can monitor the operating status of all microservices in real time.

2. According to claim 1, an industrial production intelligent data analysis and monitoring system based on microservice architecture is characterized by: The data analysis service module (5) includes a data deep mining module and an intelligent data analysis module. The data deep mining module is used to deeply analyze the massive data generated during the production process, accurately identify potential problems and optimize the production process, and improve production efficiency and product quality; the intelligent data analysis module is used to combine machine learning and deep learning models to improve data processing capabilities and analysis accuracy.

3. The industrial production intelligent data analysis and monitoring system based on microservice architecture according to claim 2 is characterized by: The data deep mining module includes a multidimensional data analysis unit, an association rule mining unit and a time series analysis unit; the multidimensional data analysis unit supports fast multidimensional query and complex data summary analysis through OLAP technology; the association rule mining unit uses the Apriori algorithm to mine potential associations in the production process; the time series analysis unit uses the ARIMA model to perform time series forecasting on production data, identify production trends or fluctuations in advance, and help management adjust strategies in a timely manner during the production process.

4. The industrial production intelligent data analysis and monitoring system based on microservice architecture according to claim 3 is characterized by: The intelligent data analysis module cleans, converts and selects features of industrial data through feature engineering, combines supervised learning, unsupervised learning and deep learning models for quality prediction and anomaly detection, and achieves more accurate quality prediction and production process anomaly detection through refined feature engineering and model optimization.

5. The industrial production intelligent data analysis and monitoring system based on microservice architecture according to claim 4 is characterized by: The risk assessment service module (4) includes an identification and classification unit, a modeling and prediction unit, and a real-time monitoring and early warning unit; the identification and classification unit accurately identifies potential risk factors by analyzing equipment status, production progress and quality control data in combination with historical data; the identification and classification unit uses a risk matrix to classify risks according to probability of occurrence and degree of impact, thereby helping management to quickly identify high-risk areas and take countermeasures in advance; the modeling and prediction unit uses regression analysis and machine learning methods to establish a risk prediction model, analyze the impact of various risk factors on production, and use model effect evaluation diagrams to continuously optimize model parameters and improve prediction accuracy; the real-time monitoring and early warning unit combines sensor data with historical analysis results to monitor key indicators in the production process in real time. Once an indicator exceeds a set threshold, the system automatically triggers an early warning.

6. The industrial production intelligent data analysis and monitoring system based on microservice architecture according to claim 5 is characterized by: The risk assessment service module (4) transmits the risk assessment results to the system monitoring and management service module (6), and the system monitoring and management service module (6) helps the management to intuitively understand the data analysis results and make scientific decisions through the visual display of multi-dimensional data analysis results.

Images