29th Symposium on Fusion Technology (SOFT 2016)

P4.010 Speeding up predictive electromagnetic simulations for tokamak application

8 Sep 2016, 14:20

1h 40m

Foyer 2A (2nd floor), 3A (3rd floor) (Prague Congress Centre)

Board: 10

Poster A. Experimental Fusion Devices and Supporting Facilities P4 Poster session

Speaker

Sergei Sytchevsky (JSC «NIIEFA»)

Description

Large state-of-the-art fusion devices involve extensive computations throughout the engineering design process from the concept to the commissioning. A variety of well-established software tools, such as ANSYS, OPERA, CARIDDY, TYPHOON, TORNADO has produced a range of simulation techniques and approaches for electro-magnetic (EM) simulations of principal components of tokamaks. The installation activities make it possible to adjust computational models using inspection and measurement data. A computational technique is presented that is targeted to speed up parametric and predictive EM simulations with intensive data communication. A large number of parameters on various scenarios should be efficiently correlated before, during and after experiments on fusion devices in order to generate a consistent operational database. The paper presents an attempt to proceed to a general concept of software environment for fast and consistent multi-task simulation of EM transients (engineering simulator for tokamak applications). The strategy exploits parallel processing with optimized simulation algorithms, based on using of influence functions and superposition principle, and improved intertask communication to take full advantage of parallelism. The software has been tested on a multi-core supercomputer. The results were compared with data obtained in standard TYPHOON computations. A discrepancy was found to be below 0.4%. The computation cost for the simulator is proportional to the number of observation points. An average computation time with the simulator is found to be by hundreds times less than the time required to solve numerically a relevant system of differential equations for known software tools.