Customers: Magnitogorsk Iron and Steel Works (MMK) Magnitogorsk; Metallurgical industry Contractors: Data Center Automation Product: Data Center Automation: Data-Track Digital PlatformProject date: 2013/03 - 2013/08
|
2013: MES (DCA, PTG). Intelligent system for tracking the movement of rolls using video analytics methods "in the warehouse of heated rolls of sheet rolling shop No. 11 (pilot)
The project includes the following MES technologies (according to MES-11): the DCA (Data Collection/Acquisition) function - data collection and storage, PTG (English Product Tracking and Genealogy) - product tracking and genealogy.
For the first time, the capabilities of intelligent tracking systems (ISS) in the stack of the DATA-TRACK platform were demonstrated during a full-scale experiment at the warehouse of heated rolls of the LPC 11 workshop of the Magnitogorsk Metallurgical Plant.
The system shown was designed to monitor the movement of the rolls and track and condition of the warehouse by fixing the following operations using machine vision:
- moving the roll to the storage point;
- removing the roll from the storage point;
- the current state of the storage point.
For the pilot, an area of the warehouse and a crane were allocated, with the help of which the rolls were moved. Video cameras and a crane positioning system were mounted on the crane. When passing the crane above the warehouse, video cameras recorded the position of the rolls in the warehouse, and when performing operations, the crane recorded the fact of moving the rolls.
The structure of the system is shown in the figure below.
The video monitoring Block (VCD) was directly responsible for recognizing video streams from cameras. The results were transmitted to the server. The warehouse (BVSS unit) was visualized based on 3D models. Warehouse status and product movement reports were performed based on WEB technologies. The figure below shows the video stream from the cameras and the corresponding 3D model built by the system using pattern recognition programs developed by DATA-CENTER Automation.
In the demonstration, the system easily detected both unauthorized roll removal and no roll in stock.
The demonstration and monthly operation of the system in workshop No. 11 proved in a convincing way that the image recognition software developed by DATA-CENTER Automatics is suitable for tracking products in metallurgical production conditions.
Starting from this project, technical vision is used in the DATA-CENTER Automation projects and is used at PAO Severstal for tracking on roller tables and refrigerators, as well as at the Karaganda Metallurgical Plant.