Exploiting process topology for fault diagnosis in a simulated pressure leaching system

In this paper, a fault detection and identification framework based on feature extraction and topology extraction methods is proposed, and evaluated on a simulated base metals refinery high pressure leaching process. Feature extraction methods included linear and kernel principal component analysis, while process topology information was extracted with linear cross-correlation, partial cross-correlation and transfer entropy. Connectivity maps were constructed to identify possible fault propagation paths to aid root cause analysis. The results indicate that process topology can enhance the fault identification capability of data-based process monitoring methods. The study of the considered base metals refinery pressure leaching process is ongoing, with future work including the creation of a fault library which would lead to further critical evaluation of the proposed fault detection and identification framework.