Mónica Fossati

Numerical Study of the Effect of a Power Plant Cooling Water Discharge in the Montevideo Bay

The numerical simulation of the water temperature in the Rio de la Plata River and Montevideos Bay was done using the numerical model of finite elements RMA-10 in its 2D vertical integrated mode. Parameters involved in the formulations of thermal exchange with the atmosphere were adjusted using measurements of water temperature in several locations of the water body. After calibrating the model, it was used to represent the operation of a power plant located in Montevideos Bay. This central takes water from the bay in order to cool its generators and also discharges high-temperature water into the bay. The correct representation of temperatures at the water intake and discharge of the plant reflects that the model is able to represent the operation of the central. Several analysis were made to study the thermal plume, the effects of the water discharge on the water intake of the power plant, and the effect on environmental variables of the study area like currents.

Taking advantage of HPC techniques in the operational forecast of the Río de la Plata

In this paper we address the use of high performance computing techniques with the aim of accelerating the runtime of a numerical model for the South Atlantic Ocean. This numerical model is a component of a larger system, that includes higher precision models to compute the hidrodynamics of the Río de la Plata river. Our work includes, in a first stage, a thorough study of the use of traditional HPC techniques on the CPU, and in a second stage we perform a preliminar study about the use of GPUs to accelerate the optimized version of the model. Specifically, for the first stage we made a profiling of the original model to identify the routines with larger computational cost. After that, we design and implement some variants for the three most expensive ones. Finally, we validate the proposals with an experimental evaluation. The results obtained show important acceleration values for the routines (up to 11 x), and these accelerations impact in the whole model with a runtime reduction of more than three times. Finally, we migrated one of the most costly routines of the optimized version to the GPU as a proof of concept.