Data is the foundation of intelligent manufacturing and the only way forward for Industry 4.0. With data, product quality management analysis, fault diagnosis and prediction can be carried out . However, in the industrial field, up to 80% of data is still hidden in the company's own production operations and other links . How to collect and mine these data is directly related to product quality, production operation and factory management efficiency, and to a large extent affects the smooth development of industrial Internet construction.
In order to fully unleash the "potential" of industrial data, Datangchanglian has built an integrated data center from data collection to data value mining to accelerate the company's automatic collection, storage, analysis, and calculation of real-time data from the entire plant's equipment. Production procedures help improve production efficiency and resource utilization, and quickly meet the needs of various industrial Internet scenarios.
Introduction to industrial collection
Industrial data collection plays an important role in realizing smart factories, improving production efficiency, optimizing resource management and quality control . With the continuous development and application of information technology, industrial data collection will bring more opportunities and challenges to enterprises and promote innovation and development in the industrial field.
1. Definition of industrial collection
Industrial data collection uses sensing technology to collect and gather in the cloud efficient and real-time information from multi-source equipment, heterogeneous systems, operating environments, people and other elements . Its goal is to obtain various types of data from internal and external data sources of the enterprise . The effective data information obtained is the basis for industrial big data processing, analysis and application. Industrial data collection includes equipment operation data, production process data, logistics management data, human-computer interaction data, environmental monitoring data, energy management data, information system data and Internet data. It is an important part of industrial big data . It is of great significance to the optimization and management of industrial production.
2. The importance of collection
Industrial data collection is the foundation of the industrial Internet platform . The development of industrial data collection is the starting point for China to promote the comprehensive and in-depth application of the industrial Internet platform, and is also a necessary condition for the transformation and upgrading of the manufacturing industry. With the deep integration of informatization and industrialization, information technology has penetrated into every link of the industrial chain of industrial enterprises, promoting the development of emerging intelligent manufacturing models represented by " intelligent production, personalized customization, networked collaboration and service extension " The core of development is comprehensive perception based on massive industrial data.
3. Current status of industrial collection
The development status of industrial data collection is as follows:
1) The intelligence of data collection equipment continues to improve, and the type and number of collection equipment continue to increase;
2) With the continuous development of communication technology, the transmission rate and stability of data collection have been greatly improved;
3) The widespread application of cloud computing, big data, artificial intelligence and other technologies has greatly improved the processing and analysis capabilities of industrial data collection .
Difficulties in industrial collection
In industrial production processes, data collection plays a vital role. By collecting and analyzing various industrial data, companies can better monitor production processes, optimize operational efficiency, improve product quality, and ensure stable operation of equipment . During the data collection process, the large amount of data collection, different protocol standards, and difficulty in ensuring real-time are the difficulties that need to be overcome.
1. Large amount of data collection
The applications of the Industrial Internet of Things are becoming more and more abundant, the amount of data is increasing, and the formats are diverse. The requirements for data storage and computing capabilities are very high. The data also contains a large number of time attributes. Data collection must ensure the accuracy and traceability of spatiotemporal information. nature, and the amount of data collected is large.
The amount of data is different, and the technical difficulty required for data acquisition is also different. However, completing the preliminary data acquisition is only the first step. After the collection, a series of data processing is required, because data standardization and cleaning must be considered. A large number of industrial The data is "dirty" and cannot be used for analysis if it is stored directly. It must be processed before being stored, which increases the technical difficulty.
2. Different protocol standards
Currently, in the field of industrial data collection, there are a variety of industrial protocol standards . Each automation equipment manufacturer and integrator will also develop various private industrial protocols. The various protocol standards are not unified and incompatible with each other .
In the industrial field, various types of industrial protocols such as ModBus, OPC, CAN, ControlNet, DeviceNet, Profibus, and Zigbee will appear. Moreover, various automation equipment manufacturers and integrators will also develop various private industrial protocols themselves, resulting in a lack of industrial protocols. There are great difficulties in interconnection. The biggest problem that many enterprises encounter in the process of digital transformation is the information islands caused by numerous protocols. The high investment in customized data mining products has deterred many enterprises.
3. It is difficult to guarantee real-time
A major feature of industrial data collection is real-time, including real-time data collection and data processing . As the most common and frequent demand in production, data collection directly determines the quality and efficiency of subsequent data processing. With the intelligent upgrading of production equipment and related technologies, the standards for real-time data collection and calculation in the industry have been raised to Second level requirement.
Data center solution
The data center solution is mainly to meet the needs of enterprise data governance and analysis, manage the enterprise's basic data, realize the synchronous distribution of basic data, and at the same time build the enterprise's data warehouse, highly aggregate and integrate basic data and business data, and use graphics to The specialized analysis dashboards and components display the data intuitively to satisfy the senior management of the enterprise in understanding the business and making decisions.
1. Overall introduction of the plan
The data middle platform is mainly to open up the data links of the enterprise and achieve full life cycle data management. It builds a data management system through the data middle platform to achieve effective integration of data in various business systems and ensures the consistency of the underlying basic data through basic data governance. Business indicators are dynamically displayed on the front end, combined with multi-dimensional penetration of data indicators, to achieve analysis and presentation in different forms and dimensions. While analyzing business indicators, it is also a sorting out of the enterprise's business, discovering problems in the business, realizing the upgrading and transformation of the enterprise's business, ensuring that the leadership can control the development of the enterprise's business in real time, understand the business pain points, improve the business management method, and realize the business and Data support each other to provide data support for high-level decision-making of enterprises.
2. Plan scenario description
In the data middle platform solution composed of "MDM basic data platform + ESB enterprise data bus + DAP data analysis platform", it emphasizes the value of data, the management of data assets, the sorting and analysis of business themes and business indicators, and the realization of business indicators - based Data analysis examines the source and impact of data indicators, and realizes data penetration of multiple topics and multiple businesses based on data indicators, as well as the changes and development trends of indicators in different businesses, so as to make predictions and decisions for business development.
3. Involving product functions
The data middle platform is divided into three modules, data analysis platform (DAP), master data management platform (MDM), and data integration platform (ESB). The deployment environment of the data middle platform is deployed on the K8S cloud platform , which mainly implements the middle platform based on data governance analysis. After the basic data within the enterprise are standardized, normalized and integrated through MDM, DAP is used to construct the data warehouse and visualize the final data analysis. ESB completes the master data distribution and data warehouse of the entire data center through data integration. of data aggregation.
Construction key points
The construction of industrial data collection needs to be considered from many aspects, including realizing device access, flexible cloud collaboration, and data analysis and presentation. Only by fully considering these points and implementing them can we ensure the smooth construction and effective operation of the industrial data collection system.
1. Implement device access
In order to solve the problem of real-time integration of production site systems, support real-time collection of DCS/SCADA data, and go deep into the manufacturing industry, the ESB enterprise service bus has expanded device access-related functions, including: OPCUA resource management, OPCUA data interface and OPCUA data subscription . It supports complete industrial protocols parsed by mainstream protocols such as Modbus and PLC, and supports third-party protocol docking. It cleverly solves the problem of complex access protocols, realizes unified management and control of terminal equipment, and can easily complete high-quality collection and standardized use of data.
1) Resource management : Mainly manages information related to OPCUA resources, including connection URL, security policy, message mode, user name and password and other information.
2) Data interface : manages information related to the OPCUA interface, including coding, belonging applications, node information, etc. After successful configuration, the interface can be called by adding a configuration service in the API service module.
3) Data subscription : You can select the node node of the OPCUA resource to subscribe, and you can configure the forwarding processing mechanism. When the subscribed node changes, the configured interceptor will be called to obtain relevant information.
2. Flexible cloud collaboration
Collaboration on the cloud reduces data transmission costs and improves data usage efficiency. The actual industrial production environment covers a large area, has a large number of workshops, and diverse equipment, and the data presents characteristics such as "massive" and "dispersed". How to efficiently complete massive data collection? How to reduce transmission costs? Issues such as how to perform edge computing and response without occupying central cloud resources have become a test that enterprises have to face in data collection and value mining. The cloud collaboration capabilities of the data center bring solutions. It can be flexibly deployed in private clouds or public clouds to conduct in-depth big data collection and analysis, effectively reducing enterprise data transmission costs and usage costs. At the same time, it supports enterprises to deploy lightweight intelligent edge applications on demand, helping to achieve local data collection, calculation and real-time response, and improving data usage efficiency.
3. Data analysis display
Visual data analysis can comprehensively improve the efficiency of production operation management. The data is invisible and traceless. To release the "potential" of data , it is necessary to transform "intangible numbers" into "visual analysis" to provide decision-making support for the refined operation of industrial production. The Industrial Internet platform provides a set of visual big data analysis tools that can help enterprises achieve comprehensive big data calculation and analysis, save data processing time, improve equipment utilization, reduce faults and downtime, and improve problem-solving efficiency. Through continuous data value mining and application, enterprise production and manufacturing will be promoted and accelerated towards digitalization and intelligence.
Significance of program construction
With the continuous advancement and innovation of technology, industrial data collection will play an increasingly important role in the industrial field, promoting the development of industry in the direction of digitalization and intelligence. Industrial data collection is of great significance in strengthening production control, fault diagnosis and prediction, improving production efficiency, and ensuring production safety.
1. Strengthen production control
Relying on industrial big data technology for all-round management, not only the data collection is more accurate, but also by analyzing the data, waste and excessive consumption of resources can be identified, which can improve management capabilities and production efficiency for the production process, making the entire The production process meets information transparency management standards and meets standardized management requirements.
2. Fault diagnosis and prediction
Industrial data collection can be used to implement status monitoring and fault diagnosis of equipment and systems. Through the analysis of equipment operation data, abnormalities and potential faults can be discovered in time, and repair and maintenance measures can be taken in advance to avoid shutdowns and production losses caused by equipment failure. Various data can be analyzed and coordinated, and large amounts of data can be managed to achieve efficient standards, helping managers make accurate decisions.
3. Improve production efficiency
Collecting various data in the industrial production process can monitor key indicators and operating parameters in the production process in real time, such as output, quality, energy consumption, etc. These data can be used to analyze and identify problems and bottlenecks in the production process, and provide optimization plans and improvement measures to improve production efficiency, ensure product quality, and reduce production costs.
4. Ensure production safety
Industrial data collection plays an important role in improving production safety . Production management can be optimized by real-time monitoring and analysis of equipment operation data, process parameter data, product quality data, energy consumption data, personnel information data, production safety data, logistics and transportation data, supplier data, customer relationship data, and environmental protection data. , improve production efficiency, prevent production accidents, and further improve production safety.
The importance of industrial data collection to industrial digitalization is self-evident. It can help enterprises achieve a positive cycle and effectively improve their overall strength and management capabilities. Access different production equipment through different communication methods, collect data from production equipment, and build a data foundation for building an industrial Internet of Things platform through edge computing and other technologies.
Datatong's enterprise data middle-end solution uses advanced industrial data collection methods to build a unified and integrated data management platform to realize the collection, storage, processing and sharing of various data in the industrial field . It can not only solve the problems of data islands and data fragmentation in traditional industries, but also provide comprehensive and efficient data support and decision-making analysis capabilities.