The 8th International Conference on System Reliability and Safety
Sicily, Italy - November 20-22, 2024

Bahri Uzunoglu, Uppsala University, Sweden

 

Creating knowledge and tools for the field of computational and data-enabled research and engineering in energy systems is Bahri Uzunoğlu's main focus. Bahri Uzunoğlu is a research faculty member in the mathematics department at Florida State University in the United States and an associate professor in the department of electrical engineering at Uppsala University in Sweden, where he specializes in computational science of electricity. With funding from agencies like NSF (USA), DAAD (Germany), EPSRC (UK), STEM (Sweden), he has worked on computational projects for institutions and industries in the UK, such as University of Liverpool and University of Southampton; in the USA, such as Florida State University and Colorado State University; in Germany, such as German Aerospace Centre (DLR); and in Türkiye, with TÜBİTAK in collaboration with industry and institutions.

Speech Title: Data assimilation in resilient energy systems within statistical, machine learning computing

Data assimilation is a discipline that aims to optimally fuse numerical/theoretical models of processes with sparse and inaccurate data, irregularly distributed in space and time to infer the evolving state of the system being modelled. Some of the physical and artificial processes in the energy/power systems can be the atmosphere, ocean, power system, electricity market, energy critical power supply systems etc. while the data of these processes can be wind speed and water speed, radiation, voltage, electricity price, navigation data, neuronal data etc. Data assimilation serves to achieve the balance between the complexity of the model and available data to reduce both the complexity of the model and the data to achieve better accuracy. This serves different goals such as state estimation, parameter estimation, improving initial conditions, prediction, filtering, smoothing, global sensitivity, control and resilience. An introduction to data assimilation methods (Machine Learning equivalents) with its application examples in critical power systems will be presented. The implications within the context of energy/power system resilience, energy essential systems, and how they are utilized in computational instances within emerging computational approaches, will be discussed in relation to the applications.