Problem description:
Digital System Models (DSMs) are specific types of software artefacts that aims to reduce data heterogeneity and improve data understanding, accessibility, processing, analytics and sharing.to optimize the management and monitoring of complex systems. DSMs are instrumental in the realization of (Spatio-Temporal) Digital Twins (DTs), which can be defined as digital replicas of a real-world entity or system through its life-cycle. A DSM commonly provides a graph-oriented representation of the relevant physical components of an asset. Each component can have associated: signals (sensor data), alarms and events, technical documentation, relevant design and operational requirements (defined by standards and recommended practices), purchase and delivery information, operational and business processes, etc.
One of the main problems of DSMs is that they are very costly and difficult to build, reuse and share, which is very discouraging for many companies and practitioners. Moreover, it is not totally clear how to better apply DSMs to maximise the benefits on these software artefacts on the daily operations of companies and large organizations.
Goal:
The aim of this thesis project is two-fold: (a) reduce implementation costs of building digital system models using deep learning and semantic technologies, and (b) ensure digital systems models enable domain experts (e.g., engineers and researchers) to better monitor and optimize the performances of real complex systems such as energy production facilities, energy grid infrastructures or container vessels, and complex ecosystems such as the Norwegian freshwaters and continental shelf.
Qualifications:
Candidates should have a good understanding on deep learning techniques, data engineering, and semantic technologies. Moreover, it will be recommended some experience programming in Python with libraries for data processing (e.g., Pandas, SQLAlchemy, etc.), data analytics (NumPy, Scikit-learn, TensorFlow, PyTorch, etc.) and data visualisation (e.g., Matplotlib, Seaborne, etc.).
Some relevant courses at UiO: TEK5040, IN3060, IN2090, IN5800 and IN3110.
References:
Boje C. et al., 2020. Towards a semantic Construction Digital Twin: Directions for future research.
D’Amico R.D. et al., 2021. Cognitive digital twin: An approach to improve the maintenance management.
Kharlamov E. et al., 2018. Towards semantically enhanced digital twins.