Early detection and diagnosis of thermal runaway reactions using model-based approaches in batch reactors

Since the early 1990s, considerable efforts have been made in researching and applying approaches for faults detection and diagnosis in chemical processes. The occurrence of chemical accidental events due to thermal runaways has serious consequences for the human, environmental and economic and encourage these efforts. This is the main motivation for this paper that details a fault detection and diagnosis approach for exothermic reactions. The proposed approach is based on the temperature measurements. It compares the collected measurements with a reference behavior for early thermal runaway detection. Once a fault is detected, characteristic features are extracted from the successive measurements collected within a time window. The fault isolation is based on a classification of these features with respect to several faulty modes. The studied faults are the most responsible for thermal runaway events. A multiset of linear classifiers and binary decision diagrams indexed with respect to the time are used for that purpose. The synthesis of peroxyformic acid in a batch reactor is considered to validate the proposed method by numerical simulations and experiments. The performance of the proposed method is carried out in a systematic way in order to evaluate its robustness and efficiency.