How do industrial big data be processed? I drew a relatively generic architecture diagram to explain: Industrial big data is similar to Internet big data in the big layer of business logic. It is generally divided into three parts, data acquisition layer, data processing layer and data presentation layer. Of course, it is specific to an actual case, or according to different Application scenarios can be divided into more layers. For example, data processing can be divided into metadata management layer, data interaction layer, data analysis layer, etc. If you are interested, you can divide big data into N layers. At the same time of stratification, there are many equally important things, such as security, operation and maintenance services, test specifications, etc., all of which are clearly stated, and it is basically unclear what is going on with industrial big data. So today we only discuss general, simplified, and general-purpose architectures, and they are only business and logical. We will elaborate on the technical aspects later. Industrial big data systems are very important, and the basic level is data collection. Without data collection, all the powerful, gorgeous, and technically NB-like stuff can only hide in the dark and cry, the hero is useless. At this level, it is also a very different layer between industrial big data and Internet big data. The data collection of Internet big trees mainly depends on various operations of users, such as web browsing, system login, information interaction, mouse click, etc. The data sources of industrial big data are more diverse. The most basic ones are sensors for collecting various industrial signals. Through the collection of sensors, the operating status of the equipment, the environmental indicators, and the operating behavior of the operators can be obtained. Class information. In addition to the information collected by the sensor, it also includes live video information, pictures taken by various image devices (for example, equipment taken by the inspector with the handheld device, pictures of environmental information), voice and voice information (for example, the call of the operator) , the volume of equipment operation, etc., remote sensing telemetry information, etc., all of which are transmitted through various types of equipment. In addition, there are various types of information manually entered by operators, information on the intranet captured by the software, information related to the enterprise on the Internet, and so on. Together, these pieces of information form the source of data collection. The field of data collection is also the area with the most intersection of automation and informationization. In fact, it is also a headache for general IT practitioners. Interdisciplinary, ha! After the data collection is completed, it enters the data processing stage. At this stage, the software personnel can show their talents. Most of the technologies are not too different from the Internet big data technology. Usually, we will first process the collected data to a certain extent. Various industrial protocols need to be parsed, the video stream needs to be decoded, the voice needs to be recognized, etc. After the identification and processing, the data is standardized and cleaned. After the washing brush is completed, Clean data is easy to use. Data that needs to be processed immediately, such as alarms, monitoring, etc., is sent directly to the real-time processing system for processing. If you are not in a hurry, please enter the warehouse and save it according to the type. Then slowly analyze the processing. Storage may use a variety of databases, including industrial real-time databases, relational databases, geodatabases, distributed databases, non-relational databases, in-memory databases, etc. The integration, management, and linking of these databases is a problem for industrial big data. We will go into the details of this later. Let's talk about macros today. Washed, cut into pieces, frozen braised pork, ah, no, the stored industrial data is so massive, how to push it to the user becomes the data visualization layer to do. Data visualization has also been relatively hot recently. The original report was now called Big Data Visualization. In fact, pure big data visualization is meaningless. Without an in-depth understanding of industrial business, it is difficult to make a satisfactory visualization of customers. result. Visualization includes many methods, such as reports, 2D maps, 3D maps, 3D models, SMS, mobile APP, WeChat, big screen, etc. In short, let the user see the data they want to see by all means. Air Powered Stirrer,Vertical Air Mixer,Vertical Pneumatic Stirrer,Industrial Pneumatic Air Mixer RUDONG HONGXIN MACHINERY CO.,LTD , https://www.rdhxmfr.com