The main objective is to establish a Wind Energy specific knowledge hub in form of a cloud-infrastructure, architecture, shared repository, and database. This will be based on the results of DC11, in which new semantic artefacts and knowledge graph based approaches will be developed.

This shall include a data-lake, which can handle different formats of structured and unstructured data; automated data and standardisation pipelines that enable an automated data ingestion into the knowledge hub and ensure a uniformity of the different data sources according to established taxonomies, industry data standards and latest research findings.

A method shall be developed to dynamically compress the databases in accordance with the ingested data volumes, to save computing resources and allocated memory and to allow high data volumes to be processed in a fast and effective manner.

An API for automated data ingestion shall be developed, giving third parties the possibility to directly ingest data into the knowledge hub, as well as their access to automated and standardised data pipelines.

The development will be based on an existing database of wind turbine specific data sources. Furthermore, the hub will be developed using infrastructure-as-a-code principles, modern open-source DevOps frameworks ensuring reproducibility and scalability, while implementing Continuous Integration, Continuous Delivery, and Continuous Deployment (CI/CD) pipeline principles.

The validation of the advances will be done with a prototype integrating data and knowledge resources of the other WPs. The hub will be connected with the ecosystem based on the WeDoWind Framework from DC11.

The fellow will develop a design proposal for the infrastructure-as-a-code solution as well as the architecture, data-lake, pipelines, and code repositories based on modern DevOps principles.

Using the deep understanding of all the different data sources collected in wind parks and meteorological stations, the fellow will establish a proposal for uniformity and standardisation of these data, to be used across the wind energy industry, a reference implementation for the knowledge hub that will be publicly available, and a tutorial for the deployment and operation of the knowledge hub based on the prototype that will integrate data and knowledge resources identified in the other WPs.

3 months in UNIZAR to attend PhD courses and coordinate research activities (F. Javier Zarazaga-Soria, M10). Short visits to researchers 1-11 equivalent to around 4 months, to understand the pilots developed to support their implementation in the Hub platform.