Increasing product quality by means of visual signals
Online process monitoring with ibaVision
process state monitoring
The task is to detect rhomboidity during a continuous casting process. This needs to be done to allow timely process adjustments and consequently avoid problems in the downstream rolling mill.
Continuous casting is a semi-continuous procedure for manufacturing steel blocks. The molten liquid metal is poured in the mold and pre-cooled, so that the strand has a solid shell of a few centimeters and a large part of the cross section still is in a liquid state. The strand is guided by the machines over strand guide rolls and is afterwards cooled. After it has solidified, the strand is cut on the out conveyor in the desired length by means of a cutting torch.
Without real-time monitoring, the geometry of the cast billets can only be measured on a random basis after cooling. Non-squaered billets and a deviation in the process state can only be detected much later in the process.
This is prevented by continuously monitoring the production plant with an image processing system. The process efficiency and also the product quality are increased with a lasting effect.
A camera that is installed behind the cutting torch records the front sides on the billets and provides ibaCapture with the video data. ibaVision processes the video sequences and first of all determines the corner points of the front side of the billet and calculates - using these data - the length and the difference of the diagonal. These values are fed in the ibaPDA process data acquisition system as so called visual signals. ibaPDA then creates a long-term trend of these characteristic values. Thus, ibaVision allows for determining numerical values that do not have sensorics. In case of a significant trend, thus a diagonal-difference above a pre-defined threshold, the operator is being alarmed by means of a virtual traffic light on the user interface. The defective billet can then be eliminated. Simultaneously, the settings of the plant are adapted.
"Image processing delivers information that no other sensor can get – and does so 24/7."
Product Manager, iba AG
ibaCapture, in conjunction with ibaPDA, records video and HMI images synchronously with measurement data. Recording can be done continuously, or triggered by events. Important events can be automatically stored as still images. For historical analytics, ibaAnalyzer can be used to automatically retrieve, align, and display recorded video and HMI images with archived measurement data.
ibaVision integrates professional, industrial image processing into the iba system and enables visual monitoring and analysis of processes. Quality checks can be automated during the production and allow for early intervention in the process before major errors occur.
As the central core of the iba system, ibaPDA has proven itself as one of the most versatile data acquisition systems for maintenance and production for many years. Among the many powerful features of ibaPDA are asset-agnostic acquisition, high-speed synchronous archiving, and real-time streaming analytics.
ibaQPanel is a software addon that allows for the live display of process and quality data in an HMI-picture. ibaQPanel combines the functionality of technological representations of measurement variables with elements typical for HMI screens. It is seamlessly integrated in ibaPDA.
ibaAnalyzer is characterized by its broad capability for analyzing and evaluating data. The application is intuitive and offers an advanced set of functions. The license for analyzing measurement files, which have been generated with the iba system, is free of charge.
The process data can be visualized online by means of the data recorded with ibaCapture and ibaPDA. The data determined in real time with ibaVision are available like all the other process data. The data are displayed on the control panel via ibaQPanel, which combines the functionality of an interactive measurement value display and elements from the field of HMI. In addition to the live video image of the camera, the snapshot with the diagonal calculation, the current trend of the visual signals, and the process quality is displayed by means of a traffic light. The offline analysis of the process data, the visual data, and the time-synchronously recorded video images can be done with ibaAnalyzer for the purpose of root cause analysis.
Explore how iba helps you to understand and master the growing technological complexity of automated processes and mechatronic systems