Home » Projects » Novel methods for a knowledge graph generation approach within the Virtual Wind Farm Hub

Doctoral Candidate – no. 11

Ostschweizer Fachhochschule – OST

Novel methods for a knowledge graph generation approach within the Virtual Wind Farm Hub

Scope and Objectives

The goal of this project is to investigate, design and implement a knowledge graph generation approach for the Virtual Wind Farm Hub. This will be done by first identifying and characterising the knowledge and data resources needed by the other researchers, based on the results of DC12.

This will allow identification, assessment and evaluation of existing semantic artefacts that could be useful for the needs of the researchers in this project. The available data sets will be prepared according to the FAIR Data Maturity Model41, and the most important improvements to the existing semantic artefacts will be identified.

Based on these results, the relevant semantic artefacts will be further developed in alignment with standardisation communities including IEA Wind Task 43, TIM Wind, TechnoPortal, the Research Data Alliance and the WindEurope Digitalisation Taskforce, by applying the Unified Process for Ontology Building and Ontology Development 101.

These could include, a new wind turbine system sensors schema, a wind turbine system ontology, a wind turbine system components taxonomy, a blade damage ontology and an airfoil data model schema.

A model for autonomous ontology evolution will also be designed based on published methodologies. Parallel to this, a new knowledge graph generation approach will be designed, based on a review of previous approaches, including ontology-based data integration, knowledge graph generation methods44, ontology-based data access (OBDA), data-science knowledge graph approaches and data-to-text generation systems utilising transformers for generating corrective maintenance strategies for faults using wind turbine operational data.

These methods will be integrated into the Hub. Finally, a global ecosystem of wind energy experts and researchers will be created within the WeDoWind Framework to incentivise and motivate data and knowledge sharing within the Hub.

Expected Results

New semantic artefacts created and published on an ontology-hosting website. A new knowledge graph generation approach for the Virtual Wind Farm Hub will be designed. A global ecosystem of wind energy experts and researchers to share data will be built up.

Planned secondments

3 months in RTDT Laboratories AG (Dr. Imad Abdallah) to gain experience in real applications of ontologies and handle real measurement data (M28-30). 3 months in ETH, attend PhD courses and coordinate research acts. (Prof. Eleni Chatzi M10-M12).

Develop your career

Don’t miss this opportunity to pursue your passion and make a meaningful contribution

Doctoral Candidate

Name Candidate

Supervisor

Prof. Sarah Barber

  • sarah.barber@ost.ch

Supervisor

Prof. Eleni Chatzi

  • chatzi@ibk.baug.ethz.ch
Institution
OST