© RTG 2323

Doctoral researcher

Name

Nazanin Hamedi M.Sc.

Send email

Send encrypted email via the SecureMail portal (for TUD external users only).

Contact Information

Organization Name

Research Training Group 2323

Research Training Group 2323

Address work

Visiting address:

Institutsgebäude S7A
Raum / Room 205 Georg-Schumann-Str. 7a

01187 Dresden

Show map of this location.

work Tel.

Research Topic

Modular Plants (MPs), composing of predesigned modules offering specific process goals, is a promising approach to fulfil the requirements of the volatile global market of today’s chemical industry. Nevertheless, the flexible feature of the MPs brings about a huge complexity. To help operators to deal with this complex system and exploit its outstanding features, the right information, with the best format should be provided to them. This research project aims at investigating various hierarchical structures in chemical industry and their potential to be used as a means to increase the operators’ comprehension of the MPs.

This research aims at using different hierarchies that has been applied for various purposes in the field of process engineering, and develop a knowledge graph of the modular processes. The developed knowledge graph consists of an A Box (instances) and a T Box (ontologies), and it considers the behaviour (phenomena) of the processes. Afterwards, some agents are required to communicate with the knowledge graph and infer the requested information. This will be helpful in many adaptation scenarios of modular plants. For example, for fault diagnosis purposes, it can identify the root cause of the problem and therefore help the operator make a better decision, or it can be used to provide information about the active automation system and by presenting this information to the operator, he will be more aware of the situation.  The following  questions will be addressed:

• How should we define the process phenomena in a hierarchical, computer understandable way, in the knowledge graph of a process?
• What are the ontologies that we need to consider as the basis of the knowledge graph? Are the ontologies in the literature sufficient or we need to modify them?
• How should the related knowledge be extracted (inferred) from the knowledge graph?