Data-based monitoring of model-based PLT safety functions Anomaly detection in reproducible, chemical batch processes

In complex and model-based PCT safety functions, uncertain process measurement values are used to calculate shutdown values that are not directly observable, leading to high safety margins when using manufacturers' safety measurement accuracy specifications and common error propagation methods. This paper outlines an approach to data-based monitoring and anomaly detection for reproducible chemical batch processes. which would justify the use of lower safety margins based on measurement repeatability, and thus have a highly positive impact on the operability of complex and model-based PLT safety functions. The potential of this method to increase the degree of diagnosis and detection of unknown anomalies in the production process is demonstrated and discussed using historical process data.