This paper describes the distributed system, network and software architecture, the application development environment, the performance, and the early lessons learned on the ATM LAN testbed Mercuri established at the Honeywell Technology Center, to develop distributed multimedia technologies for realtime control applications. We have developed a client-serverbased software architecture on Sun Sparcstation-2s connected by a Fore Systems' ASX-100 ATM switch, with video processing handled by Parallax's XVideo cards. The architecture enables network-transparent applications and provides simple primitives for multimedia capture, display, transmission, storage, and retrieval. A real-time multimedia-in-the-loop control application was developed as the vehicle for testing the capabilities and performance of the network. Our test measurements focus on the end-user-level performance metrics such as message throughput and round-trip delay as well as video-frame jitter under no-load and load conditions. Our results show that the maximum burst throughput that can be supported at the user level is 48 Mbls using AAL 5, while round-trip delays for 4-kbyte messages are about 3 ms. Our experience reveals a number of performance bottlenecks and open issues in using commercial ATM switches for practical applications. Our conclusions are: 1) For end-toend performance, the primary bottlenecks are in the protocol processing at layers above ATM (as currently implemented) and the host operating-system's performance for burst data transfers; 2) the current video-processing hardware and its integration with the host operating system are also severe limiting factors; and 3) besides performance issues, other issues that limit ATM for practical application and experimentation are the lack of analysis tools and the support for deadline-driven real-time traffic.
The aerodynamics of a fully cooled axial single stage high-pressure turbine operating at design corrected conditions of corrected speed, flow function, and stage pressure ratio has been investigated. This paper focuses on flow field predictions obtained from the viewpoint of a turbine designer using the computational fluid dynamics (CFD) codes Numeca’s FINE/TURBO and the code TURBO. The predictions were all performed with only knowledge of the stage operating conditions, but without knowledge of the surface pressure measurements. Predictions were obtained with and without distributed cooling flow simulation. The FINE/TURBO model was run in 3-D viscous steady and time-accurate modes; the TURBO model was used to provide only 3-D viscous time-accurate results. Both FINE/TURBO and TURBO utilized phase-lagged boundary conditions to simplify the time-accurate model and to significantly reduce the computing time and resources. The time-accurate surface pressure loadings and steady state predictions are compared to measurements for the blade, vane, and shroud as time-averaged, time series, and power spectrum data. The measurements were obtained using The Ohio State University Gas Turbine Laboratory Turbine Test Facility. The time-average and steady comparisons of measurements and predictions are presented for 50% span on the vane and blade. Comparisons are also presented for several locations along the blade to illustrate local differences in the CFD behavior. The comparisons for the shroud are made across the blade passage at axial blade chord locations corresponding to the pressure transducer locations. The power spectrum decompositions of individual transducers (based on the fast Fourier transform (FFT)) are also included to lend insight into the unsteady nature of the flow. The comparisons show that both computational tools are capable of providing reasonable aerodynamic predictions for the vane, blade, and stationary shroud. The CFD model predictions show the encouraging trend of improved matching to the experimental data with increasing model fidelity from mass averaged to distributed cooling flow inclusion and as the codes change from steady to time-accurate modes.
The aerodynamics of a fully cooled, axial, single stage high-pressure turbine operating at design corrected conditions of corrected speed, flow function, and stage pressure ratio has been investigated experimentally and computationally and presented in Part I of this paper. In that portion of the paper, flow-field predictions obtained using the computational fluid dynamics codes Numeca’s FINE/TURBO and the code TURBO were obtained using different design methodologies that approximated the fully-cooled turbine stage in different ways. These predictions were compared to measurements obtained using the Ohio State University Gas Turbine Laboratory Turbine Test Facility, in a process that was essentially a design methodology validation study, instead of a computational methodology optimization study. The difference between the two is that the designers were given one chance to use their codes (as a designer would normally do) instead of using the existing data to fine-tune their grids/methodologies by doing grid studies and changes in the turbulence models employed. Part I of this paper showed differing results from the two solvers, which appeared to be mainly dependent on the differences in grid resolution and/or modeling features selected by the code users. Examining these occurrences points to places where the design methodology could be improved, but it became clear that metrics were needed to compare overall performance of each approach. In this part of the paper, three criteria are proposed for measuring overall prediction quality of the unsteady predictions, which include the unsteady envelope size, envelope shape, and power spectrum. These measures capture the main characteristics of the unsteady data and allow designers to use the criteria of most interest to them. In addition, these can be used to track how well predictions improve over time as grid resolutions and modeling techniques change.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.