Modern industrial processes are highly coupled due to extensive heat-mass integration and presence of control loops. Hence, when an anomaly (fault) occurs in any process, it leads to abnormal values of several variables in the process. A challenge is to then identify the culprit variable (root cause of anomaly) so that appropriate corrective action could be taken to rectify the cause. This project aims to look at various approaches and tools in causal AI literature for the same. There will be an opportunity to work with a live industry sponsored problem as part of this project.
UG Project Type
BTP
SLP
Name of Faculty