A knowledge-based reasoning model using causal table for identifying corrosion failure mechanisms in refining and petrochemical plants

Corrosion is one of the main causes of equipment failure in refining and petrochemical plants. For the safety and normal operation of equipment, it is necessary to identify the corrosion mechanisms resulting in equipment failure and take appropriate mitigation measures. However, there are many corrosion mechanisms in refining and petrochemical plants and identifying them becomes a key issue. In this paper, a knowledge-based reasoning model, which uses a causal table rather than traditional IF–THEN rules to organize and represent domain knowledge, is proposed for the identification of corrosion failure mechanisms. Uncertainty in the reasoning process is addressed by using a hybrid method involving a combination of the Bayesian method and certainty factor theory. An application example for the identification of environmentally assisted cracking mechanisms in petroleum refineries is presented. The results demonstrate that the model is effective and feasible.

► We report a knowledge-based reasoning model to identify corrosion failure mechanisms.
► A type of causal table is designed to represent corrosion failure knowledge.
► We propose a hybrid method combining the Bayesian method and certainty factor theory.