Options
06 September 2022
Conference Paper
Titel
Integrating an XPath-Enhanced OPC UA Data Collection Into Industrial Communication
Abstract
Recent trends lead to more and more available data from automated manufacturing systems. Many data scientists collect data from production machines via the fast and widespread OPC Unified Architecture (OPC UA) communication protocol. However, heterogeneous vendor-specific configurations require high manual effort for establishing new data collections, and therefore valuable metadata is either extracted hard-coded or not at all. Our previous publication tackled this challenge purely from a computer science perspective and proposed a transformation of the query language XPath to OPC UA to enable more convenient and expressive queries. Remaining open research questions include a more sophisticated data collection from multiple automated manufacturing systems and proper integration into enterprise-scaled industrial communication systems.This paper focuses the manufacturing perspective and covers these research questions with a real-world implementation for a 500 tons High Pressure Die Casting (HPDC) production cell with six embedded OPC UA servers. We formulate XPath queries and apply these to augment the current workflow, which only extracts raw sensor values, to a more comprehensive one that additionally captures metadata such as units, value ranges, and measurement precision. Our main contributions include an aggregation of multiple individual data collection setups into a general one and an embedding of these into a fully-fledged data lake integration. The results demonstrate an integration of an XPath-enhanced OPC UA data collection into industrial communication for automated manufacturing systems, which dramatically reduces complexity and manual effort for experts.
Author(s)