BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:ENAC Seminar Series by Dr J. P. Matos
DTSTART:20190718T130000
DTEND:20190718T140000
DTSTAMP:20260928T200330Z
UID:3176a378efe70ba702766f812a0a1794a52bdadc0d16c6e06d463a4e
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr José Pedro Matos\n13:00 – 14:00 – Dr José Pedro Matos
 \nConsultant for the hydropower sector\, Stucky SA\, Switzerland\n\nUsing 
 machine learning and efficient computing to model uncertainty in hydraulic
  infrastructures\n\nUncertainty should be at the core of decision-making\,
  particularly so when addressing critical assets such as hydraulic infrast
 ructures. Notwithstanding\, uncertainty is often hard to quantify\, which 
 has led generations of engineers to shy away from explicitly evaluating it
 . Recent developments in machine learning techniques and platforms for eff
 icient computing constitute extremely valuable tools in the quantification
  of uncertainty. Exploiting these developments\, the presentation will hig
 hlight two applications where uncertainty plays a central role.\n \nThe f
 irst application lies at an intersection between machine learning and mult
 i-objective optimization. It explores a breakthrough algorithm capable of 
 making reliable probabilistic predictions based on observed data. Notions 
 of probabilistic forecasting will be introduced and several examples\, ill
 ustrating the broad applicability of the model\, will be discussed. They i
 nclude the operational flood forecasting model employed at the Rogun dam (
 to be the highest in the world)\, an inflow forecasting system for the Kar
 iba reservoir (the largest in the world by volume)\, the estimation of sus
 pended sediment concentrations on the Upper Yangtze River\, and the predic
 tion of the euro-dollar exchange rate (with bittersweet results).\n \nThe
  second application addresses the risk associated with large dams. It prov
 ides a framework capable of estimating risk while accounting for the most 
 important sources of uncertainty\, be it aleatoric or epistemic. It can re
 produce the complex chains of events that may lead to failure and evaluate
  losses when a failure does occur. Through efficient computing\, the most 
 likely paths to failure are found in a dynamic simulation including intera
 ctions between hazards\, dam components and the reservoir. For each simula
 ted failure\, multiple dam-break waves are generated and propagated downst
 ream\, acting on built infrastructure in non-deterministic ways and culmin
 ating on individualized loss of life estimates.\n 
LOCATION:CM 1 4 https://plan.epfl.ch/?room==CM%201%204
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR