Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “modified separator”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 505 records · Page 28
All sourcesNASA NTRSDOE OSTI
NASA NTRS2013

Analysis of Interactive Conflict Resolution Tool Usage in a Mixed Equipage Environment

A human-in-the-loop simulation was conducted that examined separation assurance concepts in varying levels of traffic density with mixtures of aircraft equipage and automation. This paper's analysis focuses on one of the experimental conditions in which traffic levels were approximately fifty percent higher than today, and approximately fifty percent of the traffic within the test area were equipped with data communications (data comm) capabilities. The other fifty percent of the aircraft required control by voice much like today. Within this environment, the air traffic controller participants were provided access to tools and automation designed to support the primary task of separation assurance that are currently unavailable. Two tools were selected for analysis in this paper: 1) a pre-probed altitude fly-out menu that provided instant feedback of conflict probe results for a range of altitudes, and 2) an interactive auto resolver that provided on-demand access to an automation-generated conflict resolution trajectory. Although encouraged, use of the support tools was not required; the participants were free to use the tools as they saw fit, and they were also free to accept, reject, or modify the resolutions offered by the automation. This mode of interaction provided a unique opportunity to examine exactly when and how these tools were used, as well as how acceptable the resolutions were. Results showed that the participants used the pre-probed altitude fly-out menu in 14% of conflict cases and preferred to use it in a strategic timeframe on data comm equipped and level flight aircraft. The interactive auto resolver was also used in a primarily strategic timeframe on 22% of conflicts and that their preference was to use it on conflicts involving data comm equipped aircraft as well. Of the 258 resolutions displayed, 46% were implemented and 54% were not. The auto resolver was rated highly by participants in terms of confidence and preference. Factors such as aircraft equipage, ownership, and location of predicted separation loss appeared to play a role in the decision of controllers to accept or reject the auto resolver's resolutions.

auto resolver↗
NASA NTRS2004

Computational and Experimental Study of Energetic Materials in a Counterflow Microgravity Environment

Counterflow diffusion flames are studied for various fuels flowing against decomposition products from solid ammonium perchlorate (AP) pellets in order to obtain fundamental understanding of composite propellant flame structure and chemistry. We illustrate this approach through a combined experimental and numerical study of a fuel mixture consisting of C2H4 CO + H2, and C2H2 + C2H4 flowing against solid AP. For these particular AP-fuel systems, the resulting flame zone simulates the various flame structures that are ex+ to exist between reaction products from Ap crystals and a hydrocarbon binder. As in all our experimental studies, quantitative species and temperature profiles have been measured between the fuel exit and AP surface. Species measured included CN, NH, NO, OH, N2, CO2, CO, H2, CO, HCl, and H2O. Temperature was measured using a thermocouple at the exit, spontaneous Raman scattering measurements throughout the flame, OH rotational population distributions, and NO vibrational population distributions. The burning rate of AP was also measured as a function of strain rate, given by the separation distance between the AP surface and the gaseous hydrocarbon fuel tube exit plane. This distance was nominally set at 5 mm, although studies have been performed for variations in separation distance. The measured 12 scalars are compared with predictions from a detailed gas-phase kinetics model consisting of 86 species and 531 reactions. Model predictions are found to be in good agreement with experiment and illustrate the type of kinetic features that may be expected to occur in propellants when AP particle size distributions are varied. Furthermore, the results constitute the continued development of a necessary database and validation of a comprehensive model for studying more complex AP-solid fuel systems in microgravity. Exploratory studies have also been performed with liquid and solid fuels at normal gravity. Because of melting (and hence dripping) and deep thermal wave penetration into the liquid, these experiments were found feasible, but not used for obtaining quantitative data. Microgravity experiments are needed to eliminate the dripping and boiling phenomena of these systems at normal gravity. Microgravity tests in the NASA Glenn 2.2 second drop tower were performed (1) to demonstrate the feasibility of performing propellant experiments using the NASA Glenn microgravity facilities, (2) to develop the operational procedures for safe handing of the energetic materials and disposal of their toxic combustion by-products and (3) to obtain initial measurements of the AP burning rate and flame structure under microgravity conditions. Experiments were conducted on the CH4/AP system previously studied at normal gravity using a modified design of the counterflow burner and a NASA Glenn Pig Rig, i.e., one of the existing drop rigs for general-purpose usage. In these experiments, the AP burning rate was measured directly with a linear variable differential transducer (LVDT) and video imaging of the flame structure was recorded ignition was achieved by hot wires stretched across the AP surfaces. Initial drop tower combustion data show that with the same burner separation distance and flow conditions of the normal gravity experiments, the AP burning rate is approximately a factor of two lower. This difference is likely a result of radiation effects, but further tests with longer test times need to be conducted to verify that steady state conditions were achieved under microgravity conditions.

Takahashi, Fumiaki↗
NASA NTRS1985

Piloted simulation of one-on-one helicopter air combat at NOE flight levels

A piloted simulation designed to examine the effects of terrain proximity and control system design on helicopter performance during one-on-one air combat maneuvering (ACM) is discussed. The NASA Ames vertical motion simulator (VMS) and the computer generated imagery (CGI) systems were modified to allow two aircraft to be independently piloted on a single CGI data base. Engagements were begun with the blue aircraft already in a tail-chase position behind the red, and also with the two aircraft originating from positions unknown to each other. Maneuvering was very aggressive and safety requirements for minimum altitude, separation, and maximum bank angles typical of flight test were not used. Results indicate that the presence of terrain features adds an order of complexiaty to the task performed over clear air ACM and that mix of attitude and rate command-type stability and control augmentation system (SCAS) design may be desirable. The simulation system design, the flight paths flown, and the tactics used were compared favorably by the evaluation pilots to actual flight test experiments.

Lewis, M. S.↗
NASA NTRS2011

Trajectory-Based Complexity (TBX): A Modified Aircraft Count to Predict Sector Complexity During Trajectory-Based Operations

In this paper we introduce a new complexity metric to predict -in real-time- sector complexity for trajectory-based operations (TBO). TBO will be implemented in the Next Generation Air Transportation System (NextGen). Trajectory-Based Complexity (TBX) is a modified aircraft count that can easily be computed and communicated in a TBO environment based upon predictions of aircraft and weather trajectories. TBX is scaled to aircraft count and represents an alternate and additional means to manage air traffic demand and capacity with more consideration of dynamic factors such as weather, aircraft equipage or predicted separation violations, as well as static factors such as sector size. We have developed and evaluated TBX in the Airspace Operations Laboratory (AOL) at the NASA Ames Research Center during human-in-the-loop studies of trajectory-based concepts since 2009. In this paper we will describe the TBX computation in detail and present the underlying algorithm. Next, we will describe the specific TBX used in an experiment at NASA's AOL. We will evaluate the performance of this metric using data collected during a controller-inthe- loop study on trajectory-based operations at different equipage levels. In this study controllers were prompted at regular intervals to rate their current workload on a numeric scale. When comparing this real-time workload rating to the TBX values predicted for these time periods we demonstrate that TBX is a better predictor of workload than aircraft count. Furthermore we demonstrate that TBX is well suited to be used for complexity management in TBO and can easily be adjusted to future operational concepts.

trajectories↗
NASA NTRS1989

Computational fluid dynamics analysis of space shuttle main propulsion feed line 17-inch disconnect valves

A steady incompressible three-dimensional (3-D) viscous flow analysis was conducted for the Space Shuttle Main Propulsion External Tank (ET)/Orbiter (ORB) propellant feed line quick separable 17-inch disconnect flapper valves for liquid oxygen (LO2) and liquid hydrogen (LH2). The main objectives of the analysis were to predict and correlate the hydrodynamic stability of the flappers and pressure drop with available water test data. Computational Fluid Dynamics (CFD) computer codes were procured at no cost from the public domain, and were modified and extended to carry out the disconnect flow analysis. The grid generator codes SVTGD3D and INGRID were obtained. NASA Ames Research Center supplied the flow solution code INS3D, and the color graphics code PLOT3D. A driver routine was developed to automate the grid generation process. Components such as pipes, elbows, and flappers can be generated with simple commands, and flapper angles can be varied easily. The flow solver INS3D code was modified to treat interior flappers, and other interfacing routines were developed, which include a turbulence model, a force/moment routine, a time-step routine, and initial and boundary conditions. In particular, an under-relaxation scheme was implemented to enhance the solution stability. Major physical assumptions and simplifications made in the analysis include the neglect of linkages, slightly reduced flapper diameter, and smooth solid surfaces. A grid size of 54 x 21 x 25 was employed for both the LO2 and LH2 units. Mixing length theory applied to turbulent shear flow in pipes formed the basis for the simple turbulence model. Results of the analysis are presented for LO2 and LH2 disconnects.

Kandula, Max↗
NASA NTRS2010

Regolith Volatile Characterization (RVC) in RESOLVE

Resource investigation in the lunar poles is of importance to the potential impact of in-situ resource utilization (ISRU). The RESOLVE project developed a payload to investigate the permanently shadowed areas of the lunar poles and demonstrate ISRU technology. As a part of the RESOLVE project, the regolith volatile characterization (RVC) subsystem was designed to examine the release of volatiles from sample cores. The test sample was heated in the reactor to release the volatiles where they were analyzed with gas chromatography. Subsequently, the volatile sample was introduced into the lunar water resource demonstration (LWRD) subsystem where the released hydrogen and water were selectively captured. The objective of the Regolith Volatile Characterization (RVC) subsystem was to heat the crushed core sample and determine the desorption of volatile species of interest. The RVC subsystem encompasses the reactor and the system for volatile analysis. The system was designed to analyze H2, He, CO, CO2, N2, 02, CH4, H2S and H2O. The GC chosen for this work is a Siemens MicroSAM process GC with 3 columns and 8 TCD detectors. Neon was chosen as the carrier gas to enhance the analysis of hydrogen and helium.The limit of detection for the gases is approx.1000ppm for H2, CO. CO2 , N2, O2 and H2 S. The limit of detection for CH4 is approx.4000ppm and the water limit of detection is -10000 ppm with a sample analysis time of 2-3 minutes. These values (with the exception of water and H2S) were determined by dilution of a six gas mixture from Scott Gas (5% CO2, CO, O2, N2, 4% CH4 and H2) using mass flow controllers (MFC5). Water was calibrated at low levels using an in house relative humidity (RH) generator. H 2S and high concentrations of H2 were calibrated by diluting a pure stream of gas with MFCs. Higher concentrations of N2 and 02 were calibrated using Air again diluting with MFCs. There were three modification goals for the GC in EBU2 that would allow this process GC to be used in the field demo for RESOLVE. The first modification was to decrease the weight associated with the GC, this included eliminating the explosion proof case (Figure 1) and replacing it with a lightweight case as well as using an on board COPV tank for the neon carrier gas. The next goal was to add a second oven for the molecular sieve column to allow for dual temperature control during GC operation; the separation of hydrogen and helium is optimum at lower temperatures while the water analysis required higher temperatures creating a competing design requirement. The second oven also allows a lower limit of detection for water quantification and avoids the possibility of water condensing in the GC which could ruin the column characteristics. The final goal was to modify the column arrangement to optimize the system for our specific application. Figure 2 shows the internal details of the module optimized optimized for our field application. The modifications and performance of the gas analysis system will be discussed in detail.

Captain, Janine↗
NASA NTRS1973

DORMAN computer program (study 2.5). Volume 3: Original data bank listing

A number of analyses have been performed using the DORCA program for several NASA-funded Aerospace Corporation studies in the past few years. The data decks containing the input data for these analyses have been compiled and are submitted, under separate cover. A few of the data decks are full (basic) decks containing every data item and are used as reference decks in the data bank. The other data decks were obtained by differencing a full deck with respect to one of the reference decks. Using the DORMAN program, a full deck can be recreated from the modified deck and its reference deck when and if desired. The content and structure of the data bank are described. A description of each of these data decks is presented. Three of the cases that are included in this volume have become so widely recognized and accepted that additional descriptive material has been provided. The three cases are: Case 500 Costs, Case 506 Costs, and Case 403.

Stricker, L. T.↗
NASA NTRS1980

Calculations of transonic flow about an airfoil in a wind tunnel

A combined experimental and numerical study was performed to include wind-tunnel wall interference effects in calculations for airfoil flows at transonic speeds. Pressure-survey-tube and laser-Doppler velocimeter measurements were made in the flow field about an airfoil in the 2- by 2-Foot Transonic Wind Tunnel at Ames Research Center. The results were then used as boundary data in a Navier-Stokes code modified by incorporating a pressure condition on the upper and lower computational boundaries. Comparison of calculated results and experimental data obtained from the surface of the airfoil indicates that the pressure-boundary condition is particularly effective in moving the shock to a position near that observed experimentally when the flow remains attached. For flows with large separation, shock position and viscous-layer properties are not well predicted, principally because of the inadequacies of the algebraic turbulence models employed with the method.

King, L. S.↗
NASA NTRS1983

Mean velocities and Reynolds stresses upstream of a simulated wing-fuselage juncture

Values of three mean velocity components and six turbulence stresses measured in a turbulent shear layer upstream of a simulated wing-fuselage juncture and immediately downstream of the start of the juncture are presented nd discussed. Two single-sensor hot-wire probes were used in the measurements. The separated region just upstream of the wing contains an area of reversed flow near the fuselage surface where the turbulence level is high. Outside of this area the flow skews as it passes around the body, and in this skewed region the magnitude and distribution of the turbulent normal and shear stresses within the shear layer are modified slightly by the skewing and deceleration of the flow. A short distance downstream of the wing leading edge the secondary flow vortext is tightly rolled up and redistributes both mean flow and turbulence in the juncture. The data acquisition technique employed here allows a hot wire to be used in a reversed flow region to indicate flow direction.

Mcmahon, H.↗
NASA NTRS1993

The infrared and microwave spectroscopy of the argon-propyne dimer

Microwave and infrared measurements are reported for the van der Waals complex ArCH3CCH and its isotopomers. The structure is T-shaped with equilibrium center-of-mass separation of 3.73 A and an angle of 82 deg between the molecule symmetry axis and the van der Waals bond. The infrared and microwave spectra are complex due to the effects of a slightly hindered internal rotor. Analysis of the spectral data shows that the dipole moment is almost parallel to the dimer b axis. A modified vibration/rotation Hamiltonian that includes an internal rotor potential is used to show that the barrier to internal rotation is near 10.8/cm.

Blake, Thomas A.↗
NASA NTRS2000

DMFS: A Data Migration File System for NetBSD

I have recently developed DMFS, a Data Migration File System, for NetBSD. This file system provides kernel support for the data migration system being developed by my research group at NASA/Ames. The file system utilizes an underlying file store to provide the file backing, and coordinates user and system access to the files. It stores its internal metadata in a flat file, which resides on a separate file system. This paper will first describe our data migration system to provide a context for DMFS, then it will describe DMFS. It also will describe the changes to NetBSD needed to make DMFS work. Then it will give an overview of the file archival and restoration procedures, and describe how some typical user actions are modified by DMFS. Lastly, the paper will present simple performance measurements which indicate that there is little performance loss due to the use of the DMFS layer.

Studenmund, William↗
NASA NTRS2001

Complete ISOPHOT (C200) Maps of a Nearby Prototypical GMC: W3 (Spring) or NGC7538 (Fall)

We were originally awarded Priority 3 time (approximately 60,000 sec) with Infrared Space Observatory (ISO) to obtain a complete ISOPHOT (PHT32-C200) map of a nearby prototypical giant molecular cloud (GMC). Following the FALL launch and revised estimates for the sensitivity of the ISOPHOT detectors, our program was modified to fit within the time constraints while still carrying out the main science requirements. The revised program requested long strip maps of our FALL target (NGC7538) using sequences of PHT37/38/39 observations with LWS observations of the brightest regions. The large number of AOTs required to cover each GMC required that our observations be spread over four separate proposals (PROP-01, PROP-02, PROP-03, PROP-04) which together comprise a single observing program. Our program was executed in early 1997; nearly 50,000 sec of data were obtained, including all of our requested ISOPHOT C200 observations. None of the LWS data were taken.

Sanders, David B.↗
NASA NTRS2012

Measurement Via Optical Near-Nulling and Subaperture Stitching

A subaperture stitching interferometer system provides near-nulling of a subaperture wavefront reflected from an object of interest over a portion of a surface of the object. A variable optical element located in the radiation path adjustably provides near-nulling to facilitate stitching of subaperture interferograms, creating an interferogram representative of the entire surface of interest. This enables testing of aspheric surfaces without null optics customized for each surface prescription. The surface shapes of objects such as lenses and other precision components are often measured with interferometry. However, interferometers have a limited capture range, and thus the test wavefront cannot be too different from the reference or the interference cannot be analyzed. Furthermore, the performance of the interferometer is usually best when the test and reference wavefronts are nearly identical (referred to as a null condition). Thus, it is necessary when performing such measurements to correct for known variations in shape to ensure that unintended variations are within the capture range of the interferometer and accurately measured. This invention is a system for nearnulling within a subaperture stitching interferometer, although in principle, the concept can be employed by wavefront measuring gauges other than interferometers. The system employs a light source for providing coherent radiation of a subaperture extent. An object of interest is placed to modify the radiation (e.g., to reflect or pass the radiation), and a variable optical element is located to interact with, and nearly null, the affected radiation. A detector or imaging device is situated to obtain interference patterns in the modified radiation. Multiple subaperture interferograms are taken and are stitched, or joined, to provide an interferogram representative of the entire surface of the object of interest. The primary aspect of the invention is the use of adjustable corrective optics in the context of subaperture stitching near-nulling interferometry, wherein a complex surface is analyzed via multiple, separate, overlapping interferograms. For complex surfaces, the problem of managing the identification and placement of corrective optics becomes even more pronounced, to the extent that in most cases the null corrector optics are specific to the particular asphere prescription and no others (i.e. another asphere requires completely different null correction optics). In principle, the near-nulling technique does not require subaperture stitching at all. Building a near-null system that is practically useful relies on two key features: simplicity and universality. If the system is too complex, it will be difficult to calibrate and model its manufacturing errors, rendering it useless as a precision metrology tool and/or prohibitively expensive. If the system is not applicable to a wide range of test parts, then it does not provide significant value over conventional null-correction technology. Subaperture stitching enables simpler and more universal near-null systems to be effective, because a fraction of a surface is necessarily less complex than the whole surface (excepting the extreme case of a fractal surface description). The technique of near-nulling can significantly enhance aspheric subaperture stitching capability by allowing the interferometer to capture a wider range of aspheres. More over, subaperture stitching is essential to a truly effective near-nulling system, since looking at a fraction of the surface keeps the wavefront complexity within the capability of a relatively simple nearnull apparatus. Furthermore, by reducing the subaperture size, the complexity of the measured wavefront can be reduced until it is within the capability of the near-null design.

Forbes, Greg↗
NASA NTRS2019

Reducing The Noise Impact of Unmanned Aerial Vehicles by Flight Control System Augmentation

The aim of this thesis is to explore methods to reduce the noise impact of unmanned aerial vehicles operating within acoustically sensitive environments by flight control systemaugmentation. Two methods are investigated and include: (i) reduction of sound generatedby vehicle speed control while flying along a nominal path and (ii) reduction of acousticexposure by vehicle path control while flying at a nominal speed. Both methods requireincorporation of an acoustic model into the flight control system as an additional controlobjective and an acoustic metric to characterize primary noise sources dependent on vehiclestate. An acoustic model was developed based on Gutin’s work to estimate propeller noise,both to estimate source noise and observer noise using two separate acoustic metrics. Thesemethods can potentially mitigate the noise impact of unmanned aerial systems operatingnear residential communities.The baseline flight control system of a representative aircraft was augmented with acontrol law to reduce propeller noise using feedback control of the commanded flight speeduntil an acoustic target was met, based on the propeller noise model. This control approachfocuses on modifying flight speed only, with no perturbation to the trajectory. Multipleflight simulations were performed and the results showed that integrating an acoustic metricinto the flight control system of an unmanned aerial system is possible and useful. A secondmethod to mitigate the effects of noise on an observer was also pursued to optimize a tra-jectory in order to avoid an acoustically sensitive region during the path planning process.After the propeller noise model was incorporated into the vehicle system, simulations showedthat it is possible to reduce the noise impact on an observer through an optimization of thetrajectory with no perturbation to the flight speed.

Matthew B Galles↗
NASA NTRS2022

Low-Speed Performance Enhancement using Localized Active Flow Control: Program Overview and Summary (1/4)

The Boeing team executed a NASA task order, titled “Low Speed Performance Enhancement using Localized Active Flow Control”, under the BAART framework NNL16AA04B, contract number 80LARC20F0082. The project was executed from Sept. 2020 through April 2022 with the objective to explore localized active flow control (AFC) concepts on a representative commercial aircraft. A detailed literature review solidified the choice of AFC concepts to be explored, including AFC over a deflected aileron, in the leading edge region of the wing, and in the nacelle/pylon/wing region. All three concepts were investigated with numerical tools. The results were used in an integration study to assess the net benefits over the lifetime of a commercial aircraft. The aerodynamic studies concluded that AFC applied over a deflected aileron yields significant net L/D improvements (incl. the penalty of AFC system requirements) of up to potentially 5% L/D. Depending on the configuration, AFC applied in the leading edge region showed improvements in L/D of up to ~1.5%, as well as opportunities for maximum lift and increased lift at fixed angle of attack. AFC applied to the nacelle/pylon/wing region delivered 1.5% increase in L/D during take-off and ~4% increase in CLmax during landing. After considering assumed onboard sources, aspects of system integration and weight penalties, these aerodynamic improvements translate to: ~ 0.5% block fuel reduction potential for AFC over the aileron ~ 0.1% block fuel reduction potential for AFC in the LE slat region ~ 0.2% block fuel reduction potential for AFC in the nacelle/pylon/wing region considering take-off scenarios (potentially larger benefit may be realized considering the landing scenarios which haven’t been fully analyzed yet) The final report is comprised of four separate documents. The current document provides an overview and technical background. The second document details the numerical studies and the third document summarizes the integration and aircraft performance assessment. The fourth document focuses on the Common Research Model, specifically how AFC performs over a deflected aileron and how the NASA 10% scale wind tunnel model may be modified to implement an AFC equipped aileron. Each report offers significantly more detailed summaries and suggestions for future work.

Active flow control↗
NASA NTRS1988

A low-drag nose-body

In the present investigation of the application of convex-curvature viscous drag reduction to an axisymmetric body, the qualification of a modified mixing-length formulation in several complex two-dimensional and axisymmetric turbulent boundary layers that are primarily curved is followed by the design of a low-drag nose-body for low drag at high subsonic speeds. The longitudinal distribution of the cross-sectional area ratio is found to be critical to boundary layer separation, requiring implementation not over a single long span of curvature but over three short spans. The possibility of a 'net' drag reduction is examined in relation to 'equivalent' nose-bodies.

Bandyopadhyay, P. R.↗
NASA NTRS1991

Lunar campsite concept: Space transfer concepts and analysis for exploration missions

The lunar Campsite concept responds to a perceived need to identify early manned science and exploration missions that require minimal initial funding. The Campsite concept defers the build-up of many infrastructure components without escalating total program costs. The lunar Campsite has been sized nominally for four crew for 42 days (1 lunar night and 2 lunar days), but can be modified to span two lunar nights up to 60 days. Total mission fulfillment requires five Earth-to-LEO launches, four (100 mt class launch vehicle) for the two vehicle assemblies and one (PLS or NSTS) for the crew. The lunar Campsite mission mode is tandem direct using a booster stage and a lander stage. The booster is separated from the lander after the TLI burn and is expended into the Earth's atmosphere. In the Campsite mode, the lander lands on the surface not to be returned. In the crew delivery mode, the lander is guided to a precision landing about 500 m from the Campsite, and with enough propellant to return the crew to Earth. The Campsite consists of a habitat and airlock, body mounted radiators with a surface shield, sun tracking solar arrays, and an Earth-tracking high-gain antenna. The CV is very similar to the campsite delivery vehicle. The CV does not, however, have radiators or solar arrays. The vehicle stacks are essentially common in that they utilize the same structure system and engines, the same propellant tanks, the same 'cut-out' in which the CRV and payloads are incorporated, and the same RCS locations. The booster and lander stage propellant tank propellant capacities are identical and have margins which would allow additional fueling for propulsive capture of the boost stage into Earth orbit. This contractual study was performed to identify Campsite and vehicle interfaces and vehicle requirements, and to surface issues related to the integration of the Campsite and LTV's.

Source record↗
NASA NTRS1992

Dynamic Optimization

We distinguish static and dynamic optimization of programs: whereas static optimization modifies a program before runtime and is based only on its syntactical structure, dynamic optimization is based on the statistical properties of the input source and examples of program execution. Explanation-based generalization is a commonly used dynamic optimization method, but its effectiveness as a speedup-learning method is limited, in part because it fails to separate the learning process from the program transformation process. This paper describes a dynamic optimization technique called a learn-optimize cycle that first uses a learning element to uncover predictable patterns in the program execution and then uses an optimization algorithm to map these patterns into beneficial transformations. The technique has been used successfully for dynamic optimization of pure Prolog.

Laird, Philip↗