Main :: Research Gallery :: Technical papers :: About FPI :: Know the Truth


NOTE: Website has moved to

Related paper: An Integrated Cloud Platform for Rapid Interface Generation, Job Scheduling, Monitoring, Plotting, and Case Management of Scientific Applications”, in IEEE Proceedings of the International Conference on Cloud Computing Research Initiatives, Singapore, October 26-27, 2015.

An Integrated Cloud Platform for Scientific Applications

Sponsor: FMS Foundation

The Scientific Platform for the Cloud (SPC) presents a framework to support the rapid design and deployment of scientific applications (apps) in the cloud. It provides common infrastructure for running typical IXP (Input-eXecute-Plot) style apps, including: a web interface, post-processing and plotting capabilities, job scheduling, real-time monitoring of running jobs, and case manager. It has been demonstrated to work with a number of scientific applications in a diverse range of scientific fields including computational genetics, computational geophysics, computational fluid dynamics, and also in big data analytics. It has also been integrated to work with Amazon AWS instances and Docker containers.

Numerical Weather Simulations with a Warm Ocean

Sponsor: ICR

In this research, the Weather Research and Forecasting (WRF) Model was modified to be able to prescribe a constant sea surface temperature. This modified code was then used to simulate a number of historical storms, hurricanes and typhoons occurring in various locations through the world. We report exponential increases in precipitation output with temperature, as well as storm strength and duration. Moreover, in some extreme cases, we observe the formation of hypercanes.

Related paper: "Numerical Simulation of Precipitation in Yosemite National Park with a Warm Ocean" (two-part paper), Answers Research Journal, 3(2010).

Related paper: "Mendel's Accountant: a biologically realistic forward-time population genetics program. Scalable Computing: Practice and Experience. 8(2). June 2007."

Mendel's Accountant: a biologically realistic forward-time population genetics program

Sponsor: FMS Foundation

Mendel's accountant is a numerical simulation program that was developed in collaboration with a Cornell University geneticist in order to study fundamental questions within the field of population dynamics. It has been developed to run on parallel cluster computers in order to simulate multiple tribes. It accurately models the mutation/selection process, and thus can be used to investigate fundamental questions about genetic load, tribal competition, group selection, and to predict extinction of species. It has also become a powerful teaching tool used by professors and students around the world.

Automated vortex detection and characterization

Sponsor: National Science Foundation with Ohio State University and Mississippi State University

In this research, a general approach is developed for characterizing vortices in computed flow fields. The vortex characterization system consists of three stages: The first stage processes the detection results to make them more applicable for computing vortex characteristics. The second stage computes the physical extent of a vortex using the processed detection results. The third stage describes the kinematical strength of a vortex in terms of its sense of rotation and strength of rotation, in addition to other useful properties. The method has been demonstrated for a variety of flowfields.

Related paper: "A Multistage Vortex Visualization Algorithm", AIAA 44th Aerospace Sciences Meeting and Exhibit, Reno, NV, January 9-12, 2006.

Related paper: "Manufacturing tolerance effects on ship rudder force/cavitation performance", presented at the 2005 SNAME Maritime Technology Conference & Expo and Ship Production Symposium, October 19-22, 2005 Houston, TX.

Predicting fully-appended ship rudder cavitation

Sponsor: Northrop Grumman Ship Systems with Mississippi State University

In this study, an unstructured RANS solver was coupled to a boundary element cavitation prediction software called PROPCAV. In this way, fully appended ship rudder cavitation can be predicted. This was then used to predict surface cavitation inception envelopes, or ship speed curves where cavitation would not occur. This method was used to investigate the effects of manufacturing tolerances on cavitation performance. The results showed that leading-edge droop manufacturing variation had the most significant effect, causing a 25% reduction in cavitation inception ship speed.

Advanced numerical techniques for trailing vortex simulations

Sponsor: National Science Foundation with Ohio State University and Mississippi State University

Advanced numerical techniques are developed for conducting vortical flow simulations. The techniques provide for: (1) automated core detection, (2) automated solution-adaptive grid refinement, (3) time-accurate vortex core tracking, and (4) vortex analysis. This system of techniques will allow for reduced simulation time at minimal computational cost with high solution accuracy of the flow solution in the vortex core.

Presented at the Harvey Mudd College Mathematics conference on Scientific Computing, Claremont, CA, November 2005.