BEGIN:VCALENDAR VERSION:2.0 PRODID:researchseminars.org CALSCALE:GREGORIAN X-WR-CALNAME:researchseminars.org BEGIN:VEVENT SUMMARY:Prof. Dr. Sandor Baran (Head of the Department of Applied Mathemat ics and Probability Theory\, Faculty of Informatics\, University of Debrec en\, Hungary) DTSTART:20261118T150000Z DTEND:20261118T160000Z DTSTAMP:20261115T003348Z UID:AMIS/20 DESCRIPTION:Title: Re newable energy-related aspects of statistical post-processing\nby Prof . Dr. Sandor Baran (Head of the Department of Applied Mathematics and Prob ability Theory\, Faculty of Informatics\, University of Debrecen\, Hungary ) as part of Asymptotic Methods in Statistics\n\nInteractive livestream: h ttps://knu-ua.zoom.us/j/89643295643?pwd=eTBZZSt0d0thZzFyaUhDUFNGTVE3QT09 Passcode (if necessary) 785163\n\nAbstract\nBy the end of 2025\, the renew able energy share of the global electricity capacity reached\n49.4% and th e new installations are dominated by solar energy. Due to the high volatil ity of\nphotovoltaic (PV) energy sources\, their successful integration in to the electrical grid\nrequires accurate short-term PV power forecasts. P V power is predicted from the\npredictions of solar irradiance\, where the most advanced method is the probabilistic\napproach based on ensemble for ecasts. However\, ensemble forecasts are often\nunderdispersive and subjec t to systematic bias\, so they require some form of statistical\npost-proc essing. First\, we introduce a general two-step machine learning-based app roach\nto calibrating ensemble forecasts. In a case study based on global horizontal irradiance\nforecasts of the operational ensemble prediction sy stem of the Hungarian Meteorological\nService\, the predictive performance of this novel method is compared with the forecast\nskill of the raw ense mble and the state-of-the-art parametric approaches. Then\, we\ninvestigat e the problem of direct post-processing of PV power ensemble forecasts. Wi th\nthe help of power output data of seven PV power plants in Hungary and corresponding PV\npower ensemble forecasts obtained from ensemble weather predictions of the European\nCentre for Medium-Range Weather Forecasts\, w e compare the skill of various modern\nparametric and nonparametric approa ches.\nThis is a joint work with Ágnes Baran (University of Debrecen)\, N ina Horat (Karlsruhe Institute of Technology)\, Sebastian Lerch (Marburg U niversity\, HITS)\, and Martin János Mayer (Budapest University of Techno logy and Economics).\n LOCATION:https://researchseminars.org/talk/AMIS/20/ URL:https://knu-ua.zoom.us/j/89643295643?pwd=eTBZZSt0d0thZzFyaUhDUFNGTVE3Q T09 Passcode (if necessary) 785163 END:VEVENT END:VCALENDAR