Nachweis der Betriebsbewährtheit von PLT-Feldgeräten anhand alternativer Datenquellen

In the chemical-pharmaceutical industry, also known as the process industry, process control technology safety devices (PCT safety devices) use automation technology to ensure that production plants (process plants) are brought to a safe state when re-quired. In accordance with IEC 61511, all PCT field devices used in a PCT safety loop must be provided with a proof of operational reliability, irrespective of the existence of a safety integrity level classification in accordance with IEC 61508. The standards IEC 61511 and VDI/VDE 2180 for functional safety in the process industry refer to the NAMUR Recommendation 130 (NE 130), which was amended in 2023, as a possible way of proving operational reliability. In contrast to the previous version, this amended NE 130 enables a case-by-case assessment and thus an applicative proof of operational reliability. In the current report by the Commission for Plant Safety (KAS) of the Federal Ministry for the Environment, Nature Conservation, Nuclear Safety and Consumer Protection (BMUV) on the results of expert inspections in accordance with Section 29b of the Federal Immission Control Act from 2022, the chemical plants category had a defect rate of of 43.8%. The criteria for a deficiency are defined based on the KAS 36 report (see, e.g., the version published 2024). Deviations in the design and documentation of PCT safety devices were listed as one of the main causes of this high defect rate of errors. This dissertation uses a qualitative study to prove that this defect rate also relates to the proof of operational reliability. Based on the V-model in relation to IEC 61508 and on the “Internet of Production” model (IoP) of RWTH Aachen University, a concept for data-based proof of operational reliability is developed. The concept is divided into a prospective information model as a guideline for structuring device data and a retrospective approach for evaluating heterogeneous brownfield data. The case-by-case approach introduced with the amendment of NE 130 published in 2023 supports the proof of operational reliability when only small data sets are available and is therefore also suitable for operators based of small plants with a small number of devices. The result of the present work is a proof of the effectiveness of this concept. In the retrospective approach, input data are structured in a script-based manner and the results can be easily evaluated by experts. In addition to the advantage of being able to process large amounts of field device data in short time, the concept has a higher accuracy and reliability due to the use of statistical data compared to a manual approach.