Advanced turbine engine simulation technique development and applications to testing

Abstract

the high levels of accuracy provided by previous methods so that meaningful evaluations of engine performance can be made. A joint development efnique (ATEST) is a digital computer simulation sysfort between AEDC and AFWAL has produced the tem for the prediction and correlation of both Advanced Turbine Engine Simulation Technique steady-state and transient engine performance. (ATEST), a program that provides the required capaATEST has been developed using both a modular conbilities and reduces both computer execution time cept which provides the flexibility to simulate and memory requirements (as compared to other profied Newton-Raphson numerical method (matrix solver) to achieve convergence at both design and offDigital computer simulations of turbine endesign operating points. In addition, ATEST has gine systems have steadily progressed over the past the capability to treat the specialized charactertwo decades. In the mid-1960's the Simulation Of istics of a particular engine (turbine coaling air Turbofan Engine (SM0TE)l automated the calculation paths, Reynolds number effects, etc.) to the degree of steady-state off-design performance of singleof detail required by the user. The technique has and dual-spool turbofan engines. Later the Genbeen developed around a collection of the best eralized Engine Programs (GENENG I and GENENG features of current advanced engine simulations extended the capabilities of SMOTE to includethreeused in government and industry. ATEST is intended spool turbofans and both singleand dual-spool for aircraft gas turbine engine cycles (e.g., turbojets. The Dynamic Generalized Engine Program single-spool turbojet, high-bypass turbofan, vari(DYNGEN)4 then provided the capability to calculate able-cycle engine, etc.) but could also be used to both steady-state and transient turbine engine persimulate other physical systems, taking advantage formance for the same engine configurations that of the modular structure. The ATEST has been sucare available in GENENG 11. In the early 1970's cessfully applied to a number of different turbine the Navy Engine Performance Program (NEPCOMP)5'6 engine cycles. These cycles represent a wide specprovided the capability to calculate the steadytrum of the different types of aircraft gas turbine state off-design performance of arbitrary turbine engines and include a single-spool turbojet (579), engine cycles. A few years later the Navy-NASA a dual-spool turbojet (J57), dual-spool mixed-flow turbofans (F100, F110, F109, F404), a dual-spool NEPCOMP to include an optimization technique and a separate-flow turbofan (CFM56), a turboshaft engine technique to represent performance of variable(T64), and an unconventional cycle engine. The geometry components. In the early 1980's a program ATEST has been written in FORTRAN 77 and has been executed on a variety of digital computers, includbility of calculating transient performance for ing the AMDAHL 5860, CRAY-XMP, and CYBER 7600. arbitrary turbine engine cycles. TmBOTRANS also has a capability allowing the user to include a Introduction limited amount of engine control logic in a simulation. ATEST encompasses the capabilities of The Advanced Turbine Engine Simulation Techarbitrary engine cycle configurations and a modigrams). Engine Program (NNEP)7 extended the capabilities of called TURBOTRANS8 provided the additional capaBoth the Arnold Engineering Development NEPCOMP, NNEP, and TURBOTRANS and expands these Center (AEDC) and the Air Force Wright Aeronautical capabilities to simulate the intricacies of adLaborarories ( A N A L ) at Wright-Patterson Air Force vanced turbine engines. These i n t r i c a c i e s inc lude , Base (WAFB) require an advanced turbine engine but are not limited to, engine specific operational simulation that can be effectively configured to effects on component performance (e.g., low simulate bath steady-state and transient performReynolds number, active clearance control), multiance of any turbine engine cycle. This advanced ple bleed paths from individual components and method must be capable of simulating all of the ininter-component interactions (e.g., pressure losses tricacies of modern turbine engines (e.g., multiple and/or heat rejection between components). cooling bleed paths, active clearance control, and effects of low Reynolds number) i n order to Support both engine testing at AEDC and exploratory cycle analysis at AFWAL. This means that previous simuA generalized program capable of simulating lation techniques must be expanded to include these both steady-state and transient performance of arintricacies and improved to reduce both execution bitrary turbine engine cycles is required. The time requirements and computer memory requirements. capability to simulate the complexities of a speThe advanced simulation method must also maintain cific cycle (such as radial variations of fan *The research reported herein was performed by the Arnold Engineering Development Center (AEDC), Air Work and analysis for this research were done by personnel of Svrrdrup Technology, General ATEST Development Force Systems Command. Inc./AEDC Group, operating contractor fox the AEDC propulsion test facilities and of the Air Force Wright Aeronautical Laboratories. Government. Further reproduction is authorized to satisfy needs of the United States