Juraj Kardoš

Juraj Kardoš

Auto-generated profile, based on publicly available information

Juraj Kardoš

Institute of Computing (CI Università della Svizzera italiana Lugano Switzerland

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
23 September 2019 in IEEE Transactions on Power Systems
Verified
Journal Article
Closed access
JKJuraj Kardos
Juraj Kardos
DKDrosos Kourounis
Drosos Kourounis
OSOlaf Schenk
Olaf Schenk
Modern power grids incorporate renewable energy at an increased pace, placing greater stress on the power grid equipment and shifting their operational conditions towards their limits. As a result, failures of any network component, such as a transmission line or power generator, can be critical to the overall grid operation. The security constrained optimal power flow (SCOPF) aims for the long term precontingency operating state, such that in the event of any contingency, the power grid will remain secure. For a realistic power network, however, with numerous contingencies considered, the overall problem size becomes intractable for single-core optimization tools in short time frames established by real-time industrial operations. We propose a parallel distributed memory structure exploiting framework, BELTISTOS-SC, which accelerates the solution of SCOPF problems over state of the art techniques. The acceleration on single-core execution is achieved by a structure- exploiting interior point method, employing successive Schur complement evaluations to further reduce the size of the systems solved while maintaining sparsity, resulting in lower computational resources for the linear systems. Additionally the parallel, distributed memory implementation of the proposed framework is also presented in detail and validated through several large-scale examples, demonstrating its efficiency for large-scale SCOPF problems.

29 December 2022 in Proceedings of the National Academy of Sciences
Verified
Journal Article
Open access
IHIllia Horenko
Illia Horenko
EVEdoardo Vecchi
Edoardo Vecchi
JKJuraj Kardoš
Juraj Kardoš
AWAndreas Wächter
Andreas Wächter
OSOlaf Schenk
Olaf Schenk
SGSusanne Gerber
Susanne Gerber
Regression learning is one of the long-standing problems in statistics, machine learning, and deep learning (DL). We show that writing this problem as a probabilistic expectation over (unknown) feature probabilities – thus increasing the number of unknown parameters and seemingly making the problem more complex—actually leads to its simplification, and allows incorporating the physical principle of entropy maximization. It helps decompose a very general setting of this learning problem (including discretization, feature selection, and learning multiple piece-wise linear regressions) into an iterative sequence of simple substeps, which are either analytically solvable or cheaply computable through an efficient second-order numerical solver with a sublinear cost scaling. This leads to the computationally cheap and robust non-DL second-order Sparse Probabilistic Approximation for Regression Task Analysis (SPARTAn) algorithm, that can be efficiently applied to problems with millions of feature dimensions on a commodity laptop, when the state-of-the-art learning tools would require supercomputers. SPARTAn is compared to a range of commonly used regression learning tools on synthetic problems and on the prediction of the El Niño Southern Oscillation, the dominant interannual mode of tropical climate variability. The obtained SPARTAn learners provide more predictive, sparse, and physically explainable data descriptions, clearly discerning the important role of ocean temperature variability at the thermocline in the equatorial Pacific. SPARTAn provides an easily interpretable description of the timescales by which these thermocline temperature features evolve and eventually express at the surface, thereby enabling enhanced predictability of the key drivers of the interannual climate.

30 May 2022 in Sustainable Energy, Grids and Networks
Verified
Journal Article
Open access
JKJuraj Kardoš
Juraj Kardoš
THTimothy Holt
Timothy Holt
VFVincenzo Fazio
Vincenzo Fazio
LFLuca Fabietti
Luca Fabietti
FSFilippo Spazzini
Filippo Spazzini
OSOlaf Schenk
Olaf Schenk

31 May 2023 in arXiv
Verified
Preprint
Open access
PDF file
JKJuraj Kardos
Juraj Kardos
WEWouter Edeling
Wouter Edeling
DSDiana Suleimenova
Diana Suleimenova
DGDerek Groen
Derek Groen
OSOlaf Schenk
Olaf Schenk
Sensitivity analysis is an important tool used in many domains of computational science to either gain insight into the mathematical model and interaction of its parameters or study the uncertainty propagation through the input-output interactions. In many applications, the inputs are stochastically dependent, which violates one of the essential assumptions in the state-of-the-art sensitivity analysis methods. Consequently, the results obtained ignoring the correlations provide values which do not reflect the true contributions of the input parameters. This study proposes an approach to address the parameter correlations using a polynomial chaos expansion method and Rosenblatt and Cholesky transformations to reflect the parameter dependencies. Treatment of the correlated variables is discussed in context of variance and derivative-based sensitivity analysis. We demonstrate that the sensitivity of the correlated parameters can not only differ in magnitude, but even the sign of the derivative-based index can be inverted, thus significantly altering the model behavior compared to the prediction of the analysis disregarding the correlations. Numerous experiments are conducted using workflow automation tools within the VECMA toolkit.

11 July 2019 in arXiv
Verified
Preprint
Open access
PDF file
JKJuraj Kardoš
Juraj Kardoš
DKDrosos Kourounis
Drosos Kourounis
OSOlaf Schenk
Olaf Schenk
Interior point methods are among the most popular techniques for large scale nonlinear optimization, owing to their intrinsic ability of scaling to arbitrary large problem sizes. Their efficiency has attracted in recent years a lot of attention due to increasing demand for large scale optimization in industry and engineering. A parallel interior point method is discussed that exploits the intrinsic structure of large-scale nonlinear optimization problems so that the solution process can employ massively parallel high-performance computing infastructures. Since the overall performance of interior point methods relies heavily on scalable sparse linear algebra solvers, particular emphasis is given to the underlying algorithms for the distributed solution of the associated sparse linear systems obtained at each iteration from the linearization of the optimality conditions. The interior point algorithm is implemented in a object-oriented parallel IPM solver and applied for the solution of large scale optimal control problems solved in a daily basis for the secure transmission and distribution of electricity in modern power grids.

2 October 2023 in arXiv
Verified
Preprint
Open access
PDF file
NJNaga Venkata Sai Jitin Jami
Naga Venkata Sai Jitin Jami
JKJuraj Kardoš
Juraj Kardoš
OSOlaf Schenk
Olaf Schenk
HKHarald Köstler
Harald Köstler
Accurate price predictions are essential for market participants in order to optimize their operational schedules and bidding strategies, especially in the current context where electricity prices become more volatile and less predictable using classical approaches. The Locational Marginal Pricing (LMP) pricing mechanism is used in many modern power markets, where the traditional approach utilizes optimal power flow (OPF) solvers. However, for large electricity grids this process becomes prohibitively time-consuming and computationally intensive. Machine learning (ML) based predictions could provide an efficient tool for LMP prediction, especially in energy markets with intermittent sources like renewable energy. This study evaluates the performance of popular machine learning and deep learning models in predicting LMP on multiple electricity grids. The accuracy and robustness of these models in predicting LMP is assessed considering multiple scenarios. The results show that ML models can predict LMP 4-5 orders of magnitude faster than traditional OPF solvers with 5-6\% error rate, highlighting the potential of ML models in LMP prediction for large-scale power models with the assistance of hardware infrastructure like multi-core CPUs and GPUs in modern HPC clusters.