Static performance prediction of data-dependent programs

Contention-sensitive static performance prediction for parallel distributed applications

This paper presents a performance estimator, a prototype of which is implemented within the parallel programming environment PCASE. The estimation is based on static performance prediction, using not only the design information of target machines but also benchmarking results. Additionally, communication costs are estimated based on a hierarchical memory machine model, which enables users to understand all the underlying communication costs on distributed memory machines for each parallel loop. With this performance suggestion, it is possible to interactively optimize data distribution and appropriately select vectorized or parallelized loops. Moreover, the skeleton profiling method is presented. It makes high speed trace (execution count of each statement) generation possible by deleting statements which do not affect the execution path of a program.KeywordsExecution PathExecution CostTarget MachineMemory MachineVector InstructionThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

<div class="htmlview paragraph">The model described here is a dynamic mean value model which is small enough to be realized on a microcomputer. Nevertheless it contains significant and very accurate information on the gross internal variables of an engine (the indicated efficiency, scavenge ratio, scavenge efficiency, etc.). This makes the model useful for control and expert system application on line and/or in parallel with an operating diesel engine. Moreover, because of its simplicity, it gives an overall picture of engine operation which is not possible with more complex single engine cycle models. Comparisons with time dependent experimental data show that the model is very accurate both for static and dynamic predictions of engine performance over a large operating range.</div>

Process fluid lubricated thrust bearings (TBs) in a turbomachine control rotor placement due to axial loads arising from pressure fields on the front shroud and back surface of impellers. To date, prediction of aerodynamic induced thrust loads is still largely empirical. Thus needs persist to design and operate proven thrust bearings and to validate predictions of performance derived from often too restrictive computational tools. This paper describes a test rig for measurement of the load performance of water lubricated hydrostatic/hydrodynamic thrust bearings operating under conditions typical of cryogenic turbo pumps. The test rig comprises of a rigid rotor composed of a thick shaft and two end collars. A pair of flexure-pivot hydrostatic journal bearings (38 mm in diameter) support the rotor and quill shaft connected to a drive motor. The test rig hosts two thrust bearings (8 pockets with inner diameter equal to 41 mm and outer diameter equal to 76 mm); one is a test bearing and the other is a slave bearing, both facing the outer side of the thrust collars on the rotor. The slave TB is affixed rigidly to a bearing support. A load system delivers an axial load to the test TB through a non-rotating shaft floating on two aerostatic radial bearings. The test TB displaces to impose a load on the rotor thrust collar and the slave TB reacts to the applied axial load. The paper presents measurements of the TB operating axial clearance, flow rate and pocket pressure for conditions of increasing static load (max. 3,600 N) and shaft speed to 17.5 krpm (tip speed 69.8 m/s) and for an increasing water supply pressure into the thrust bearings, max. 17.2 bar (250 psig). Predictions from a bulk flow model that accounts for both fluid inertia and turbulence flow effects agree well with recorded bearing flow rates (supply and exiting thru the inner diameter), pocket pressure and ensuing film clearance due to the imposed external load. The measurements and predictions show a film clearance decreasing exponentially as the applied load increases. The bearing flow rate also decreases, and at the highest rotor speed and lowest supply pressure, the bearing is starved of lubricant on its inner diameter side, as predicted. The measured bearing flow rate and pocket pressure aid to the empirical estimation of the orifice discharge coefficient for use in the predictive tool. The test data and validation of a predictive tool give confidence to the integration of fluid film thrust bearings in cryogenic turbo pumps as well as in other more conventional (commercial) machinery. The USAF Upper Stage Engine Technology (USET) program funded the work during the first decade of the 21st century.

In this paper, a technology computer-aided design (TCAD) model of a silicon carbide (SiC) insulated-gate bipolar transistor (IGBT) has been calibrated against previously reported experimental data. The calibrated TCAD model has been used to predict the static performance of theoretical SiC IGBTs with ultra-high blocking voltage capabilities in the range of 20-50 kV. The simulation results of transfer characteristics, IC-VGE, forward characteristics, IC-VCE, and blocking voltage characteristics are studied. The threshold voltage is approximately 5 V, and the forward voltage drop is ranging from VF = 4.2-10.0 V at IC = 20 A, using a charge carrier lifetime of τA = 20 μs. Furthermore, the forward voltage drop impact for different process dependent parameters (i.e., carrier lifetimes, mobility/scattering and trap related defects) and junction temperature are investigated in a parametric sensitivity analysis. The wide-range simulation results may be used as an input to facilitate high power converter design and evaluation. In this case, the TCAD simulated static characteristics of SiC IGBTs is compared to silicon (Si) IGBTs in a modular multilevel converter in a general high-power application. The results indicate several benefits and lower conduction energy losses using ultra-high voltage SiC IGBTs compared to Si IGBTs.

The performance of the real-scale LP turbine diffuser is predicted using Numerical Turbine developed by our group. Firstly, The validity of the inlet boundary condition was discussed as the performance prediction of the diffuser. Secondly, the influence of the inlet wetness on the static pressure recovery performance was investigated. Compressible Navier-Stokes equations are solved by the high-order high-resolution finite-difference method based on the fourth-order compact MUSCL TVD scheme, Roe's approximate Riemann solver, and the LU-SGS scheme. The SST-model is also solved for evaluating the eddy-viscosity.

The present work deals with a numerical simulation of a flow in finned tube banks arranged behind a gas turbine. Three models of dual-pressure tube systems are developed and analyzed in order to predict the static system performances by optimizing the utilization of the exhaust gas from an existing gas turbine without heat recovery system. For more precise modeling, the theoretical analysis of finned tube banks systems is based on the non-linear conservation equations of mass, momentum and energy. Simulations are accomplished to prove the effectiveness of the present work in performance prediction of the dual-pressure tube systems. The obtained results clearly show the necessity to take into account all relevant physical phenomena in the simulation of flows in and across finned tube banks installed behind a gas turbine. The results also reveal the different operating behavior of the developed models considering combined effects of the exhaust gas parameters and the tube geometries.

A Bayesian–Gaussian neural network and its applications in process engineering