SEARCH · Search NASA

Results for “tuning technique”

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.

SourceResearch connectionFrom yearThrough year

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

At least 199 records · Page 11

All sources NASA NTRS DOE OSTI

NASA NTRS2018

New Neural Network Cloud Mask Algorithm Based on Radiative Transfer Simulations

Cloud detection and screening constitute critically important first steps required to derive many satellite data products. Traditional threshold-based cloud mask algorithms require a complicated design process and fine tuning for each sensor, and they have difficulties over areas partially covered with snow/ice. Exploiting advances in machine learning techniques and radiative transfer modeling of coupled environmental systems, we have developed a new, threshold-free cloud mask algorithm based on a neural network classifier driven by extensive radiative transfer simulations. Statistical validation results obtained by using collocated CALIOP and MODIS data show that its performance is consistent over different ecosystems and significantly better than the MODIS Cloud Mask (MOD35 C6) during the winter seasons over snow-covered areas in the mid-latitudes. Simulations using a reduced number of satellite channels also show satisfactory results, indicating its flexibility to be configured for different sensors. Comparedto threshold-based methods and previous machine-learning approaches, this new cloud mask (i) does not rely on thresholds, (ii) needs fewer satellite channels, (iii) has superior performance during winter seasons in mid-latitude areas, and (iv) can easily be applied to different sensors.

cloud mask algorithms↗

NASA NTRSDate not supplied

Towards an Aviation Large Language Model by Fine-tuning and Evaluating Transformers

In the aviation domain, there are many applications for machine learning and artificial intelligence tools that utilize natural language. For example, there is a desire to know the commonalities in written safety reports such as voluntary post incidents reports or aerial wildfire operations reports to better understand the risks present. Another use-case is the possibility of extracting airspace procedures and constraints currently written in documents such as Letters of Agreement. These applications can benefit from the use of state-of-the-art natural language processing techniques when adapted to the language/phraseology specific to the aviation domain. This paper evaluates the viability of adaptation of NLP tools to the aviation domain by fine-tuning transformer based models using aviation data sets. In 2018, a novel language model based on neural units (also called transformers) was created and became known as “Bidirectional Encoder Representations from Transformers” or BERT. This architecture combined with large amounts of English training data and innovative semi-supervised training tasks set the standard for what would later emerge as Large Language Models. The performance of these models was further improved by hyperparameter tuning and refinement of the semi-supervised training task and resulted in “Robustly Optimized BERT Pre-training Approach through hyperparameter tuning” or RoBERTa models. These pre-trained Large Language Models proved to be useful for a wide variety of natural language processing tasks such as text classification and question answering through a process called fine-tuning. The transformer architecture with pre-trained weights served as the basis with the last few layers replaced with layers fine-tuned to perform a new task e.g., a layer that provides a label for the entire input text. This process of fine-tuning can also be used to adapt the models to new domains; e.g., BioBERT started with the pre-trained BERT model and was completed by additional fine-tuning and training on biomedical documents. Transformer-based architectures can also be used to create rich representations of text called embeddings which can serve as the input to other machine learning models. This allows simpler algorithms such as logistic regression to use context-rich representations of the text while still remaining quick to train and evaluate. In the world of aviation, there is a growing demand for natural language processing and understanding but the domain presents unique challenges. Due to the technical content (and specialized language) of most aviation documents, fine-tuning pre-trained Large Language Models to specific tasks has not met the benchmark on natural language processing tasks set by simpler models trained from scratch on the data. To address this deficiency, this paper evaluates the improvements from fine-tuning a Large Language Model on a large set of aviation documents using the original semi-supervised training tasks before performing specific natural language tasks. In fine-tuning, a domain-specific dataset is used on the original training task but with the pre-trained Large Language Model instead of starting from a random initialization. This approach allows the model to be adapted to the specific domain language without discarding the information gained from training on general English data. This paper utilized two major dataset types to train and assess the RoBERTa fine-tuning performance. The first are 7,057 Letters of Agreement which are Federal Aviation Administration (FAA) documents that formalize airspace operations across the national airspace system. They contain many examples of ‘aviation English’ using domain specific terminology and phrasing which serves as a representative basis to perform the semi-supervised fine-tuning. The second type is the 494 document classification labels to be used for evaluation. This down-stream evaluation aims to show the performance of the fine-tuned model, better understand how much data is needed for an effective fine-tuning, and how fine-tuning can be adapted for different applications in-the domain. After semi-supervised training, evaluation begins by encoding the documents for classification using the fine-tuned RoBERTa model. Then a logistic regression classifier is trained to label the document type and compared against our ground truth labels. This currently leads to a 82.8% accuracy on 10-fold cross validation showing improvement over baseline RoBERTa which achieved 81.0%. We plan to measure the improvements on additional tasks and it is expected that these improvements will lead to more robust models that can tackle the natural language processing challenges present in aviation datasets.

ATM↗

NASA NTRS2011

Performance of the GLAS Space Lidar Receiver Through Its Seven-Year Space Mission

NASA s Ice, Cloud, and land Elevation Satellite (ICESat) mission [1,2] carrying the Geoscience Laser Altimeter System (GLAS) Instrument, was launched on January 12, 2003. The three lasers on ICESat have made a total of 1.98 billion laser shot measurements of the Earth s surface and atmosphere during its 17 science data collection campaigns over its seven year operating lifetime. ICESat completed its science mission after the last laser stopped operating in October 2009. The spacecraft was de-orbited on August 30, 2010. The GLAS instrument carried 3 diode-pumped Q-switched Nd:YAG lasers, which emitted 6-nsec wide pulses at 1064 and 532 nm at a 40-Hz rate. There are three lidar receiver channels, a 1064 nm surface altimetry channel, a 1064 nm cloud backscattering lidar channel, and a 532 nm cloud and aerosol backscattering lidar channel. The altimetry and cloud backscatter channels used Si avalanche photodiode (APD) operated in analog mode as in the Mars Global Surveyor s Mars Orbital Laser Altimeter [3,4]. GLAS also utilized a number of new technologies and techniques for space lidar, including passively Q-switched diode-pumped Nd:YAG lasers, a 1-m diameter telescope, a temperature tuned etalon optical bandpass filter, Si APD single photon counting detectors, 1 Gsample/sec waveform digitizers, ultra stable clock oscillators, and digital signal processing and detection algorithms [5]. A global position system (GPS) receiver was used to provide the spacecraft position and epoch times. The ICESat mission provided a unique opportunity to monitor the lidar component performance in the space environment over a multi-year time period. We performed a number of engineering tests periodically to monitor the lidar receiver performance, including receiver sensitivity, timing precision, detector dark noise, etc. A series of engineering tests were also performed after the end of the science mission to evaluate the performance of the spare detector, oscillator, waveform digitizer, and GPS receiver. An experiment was conducted which pointed GLAS to Venus to test the receiver sensitivity to star light and to verify GLAS bore sight with respect to the spacecraft coordinate system. These tests provided unique data to assess the degradation and the rate of change of these key lidar components due to space radiation and aging. They also helped to validate new techniques to operate and calibrate future space lidars.

Sun, Xiaoli↗

NASA NTRSDate not supplied

Development of an Additively Manufactured Turbine Blade Tuned Mass Absorber

An innovative concept using a tuned mass absorber for resonant response reduction of turbine blades that is integrally fabricated into the blades using additive manufactured has been developed. Avoiding high resonant response and resulting high cycle fatigue failure is a major concern of the $100B worldwide turbine industry, which encompasses power generation, jet engines and rocket engine turbomachinery. Blade failure is potentially catastrophic, and although preventative measures using dampers do exist, these solutions are extremely expensive to develop and are not always effective. In addition, many of the existing techniques do not work for newer integrally bladed disks (or blisks), where there is no inherent damping in the assembly. The proposed concept, in which a tuned mass absorber re-arranges the structural dynamics and moves energy from the blade to the absorber, is designed specifically for blisks, and is intentionally linear, enabling accurate response prediction during design. Single blade prototypes have been analyzed, fabricated, and tested, and different absorber concepts have shown a reduction in response of up to 60%. After an extensive commercialization and prior art search, MSFC determined the innovation merited the generation and submission of a patent application in August ’21. Funding is now being sought to fabricate a 20-bladed blisk which would be spun at operational speeds with substantially higher and more realistic pressure loads in a special spin facility. Successful completion of this program would raise the TRL from 3/4 to 6/7, significantly improving the viability for adoption of the concept not only in rocket engines but other industrial applications as well.

Turbine Blades↗

NASA NTRS1999

Optimal Frequency-Domain System Realization with Weighting

Several approaches are presented to identify an experimental system model directly from frequency response data. The formulation uses a matrix-fraction description as the model structure. Frequency weighting such as exponential weighting is introduced to solve a weighted least-squares problem to obtain the coefficient matrices for the matrix-fraction description. A multi-variable state-space model can then be formed using the coefficient matrices of the matrix-fraction description. Three different approaches are introduced to fine-tune the model using nonlinear programming methods to minimize the desired cost function. The first method uses an eigenvalue assignment technique to reassign a subset of system poles to improve the identified model. The second method deals with the model in the real Schur or modal form, reassigns a subset of system poles, and adjusts the columns (rows) of the input (output) influence matrix using a nonlinear optimizer. The third method also optimizes a subset of poles, but the input and output influence matrices are refined at every optimization step through least-squares procedures.

Juang, Jer-Nan↗

NASA NTRS2006

Studies in Software Cost Model Behavior: Do We Really Understand Cost Model Performance?

While there exists extensive literature on software cost estimation techniques, industry practice continues to rely upon standard regression-based algorithms. These software effort models are typically calibrated or tuned to local conditions using local data. This paper cautions that current approaches to model calibration often produce sub-optimal models because of the large variance problem inherent in cost data and by including far more effort multipliers than the data supports. Building optimal models requires that a wider range of models be considered while correctly calibrating these models requires rejection rules that prune variables and records and use multiple criteria for evaluating model performance. The main contribution of this paper is to document a standard method that integrates formal model identification, estimation, and validation. It also documents what we call the large variance problem that is a leading cause of cost model brittleness or instability.

data mining↗

NASA NTRS2014

Stability Assessment and Tuning of an Adaptively Augmented Classical Controller for Launch Vehicle Flight Control

Recently, a robust and practical adaptive control scheme for launch vehicles [ [1] has been introduced. It augments a classical controller with a real-time loop-gain adaptation, and it is therefore called Adaptive Augmentation Control (AAC). The loop-gain will be increased from the nominal design when the tracking error between the (filtered) output and the (filtered) command trajectory is large; whereas it will be decreased when excitation of flex or sloshing modes are detected. There is a need to determine the range and rate of the loop-gain adaptation in order to retain (exponential) stability, which is critical in vehicle operation, and to develop some theoretically based heuristic tuning methods for the adaptive law gain parameters. The classical launch vehicle flight controller design technics are based on gain-scheduling, whereby the launch vehicle dynamics model is linearized at selected operating points along the nominal tracking command trajectory, and Linear Time-Invariant (LTI) controller design techniques are employed to ensure asymptotic stability of the tracking error dynamics, typically by meeting some prescribed Gain Margin (GM) and Phase Margin (PM) specifications. The controller gains at the design points are then scheduled, tuned and sometimes interpolated to achieve good performance and stability robustness under external disturbances (e.g. winds) and structural perturbations (e.g. vehicle modeling errors). While the GM does give a bound for loop-gain variation without losing stability, it is for constant dispersions of the loop-gain because the GM is based on frequency-domain analysis, which is applicable only for LTI systems. The real-time adaptive loop-gain variation of the AAC effectively renders the closed-loop system a time-varying system, for which it is well-known that the LTI system stability criterion is neither necessary nor sufficient when applying to a Linear Time-Varying (LTV) system in a frozen-time fashion. Therefore, a generalized stability metric for time-varying loop=gain perturbations is needed for the AAC.

VanZwieten, Tannen↗

NASA NTRS1995

Laser diagnostics for microgravity droplet studies

An instrument has been designed, built, and tested for performing laser diagnostic measurements of droplet combustion in low-gravity-flight aircraft. Nonintrusive measurements are of particular importance for droplet combustion (the simplest example of non-premixed combustion) and transport in microgravity environments, where physical contact would introduce an unacceptable level of perturbations. The resolution of these diagnostics can also isolate transport to length scales much smaller than the droplet diameter. These techniques can be configured to instantaneously map an entire flow field in two and three dimensions, providing either qualitative or quantitative information on the distribution of a desired scalar or vector quantity. Detailing the gas-phase flow field and position of the flame front can be achieved using planar laser-induced fluorescence (PLIF) of OH or another flame front marker. An alternative approach is to obtain LIF from a diagnostic seed included in the liquid phase fuel; it would be consumed at the flame front. The main advantage to this approach is that it is easier to choose the wavelength of the molecular absorption which coincides with convenient laser wavelengths rather than finding lasers which can be configured to access OH. Our present method uses a nitrogen-pumped dye laser tuned to a sodium absorption and addition of small concentrations of NaCl to the fuel. Particle image velocimetry (PIV) is a laser-based technique which has recently had its practicality greatly enhanced by the development of high-resolution CCD cameras and the increase in speed and capacity of computer systems. With this technique, a seeded flow is illuminated with a double-pulsed laser sheet to generate a double exposure image on a film or CCD camera. Computer analysis of the image is used to determine the particle velocity vectors and, thus, the gas velocity within the plane of the laser sheet. Our current experiment uses PIV for measuring relative droplet gas-phase velocities in a parabolic flight environment. The results show successful application of PLIF relative to luminosity and radical emission near 308 nm. PIV results are less successful due to gas thermal expansion deflecting the seed particles away from the measurement volume. PIV is applicable for measurements in the absence of combustion.

Winter, Michael↗

NASA NTRS2012

An Integrated Approach for Aircraft Engine Performance Estimation and Fault Diagnostics

A Kalman filter-based approach for integrated on-line aircraft engine performance estimation and gas path fault diagnostics is presented. This technique is specifically designed for underdetermined estimation problems where there are more unknown system parameters representing deterioration and faults than available sensor measurements. A previously developed methodology is applied to optimally design a Kalman filter to estimate a vector of tuning parameters, appropriately sized to enable estimation. The estimated tuning parameters can then be transformed into a larger vector of health parameters representing system performance deterioration and fault effects. The results of this study show that basing fault isolation decisions solely on the estimated health parameter vector does not provide ideal results. Furthermore, expanding the number of the health parameters to address additional gas path faults causes a decrease in the estimation accuracy of those health parameters representative of turbomachinery performance deterioration. However, improved fault isolation performance is demonstrated through direct analysis of the estimated tuning parameters produced by the Kalman filter. This was found to provide equivalent or superior accuracy compared to the conventional fault isolation approach based on the analysis of sensed engine outputs, while simplifying online implementation requirements. Results from the application of these techniques to an aircraft engine simulation are presented and discussed.

imon, Donald L.↗

NASA NTRS1987

Passive stabilization for large space systems

The optimal tuning of multiple tuned-mass dampers for the transient vibration damping of large space structures is investigated. A multidisciplinary approach is used. Structural dynamic techniques are applied to gain physical insight into absorber/structure interaction and to optimize specific cases. Modern control theory and parameter optimization techniques are applied to the general optimization problem. A design procedure for multi-absorber multi-DOF vibration damping problems is presented. Classical dynamic models are extended to investigate the effects of absorber placement, existing structural damping, and absorber cross-coupling on the optimal design synthesis. The control design process for the general optimization problem is formulated as a linear output feedback control problem via the development of a feedback control canonical form. The techniques are applied to sample micro-g and pointing problems on the NASA dual keel space station.

Sesak, J. R.↗

NASA NTRS1986

Inversion of SAGE II measurements

Stratospheric aerosol data acquired by the seven channels of the SAGE II passive radiometer system are first converted to atmospheric slant path transmission (SPT) values before deriving inversion values. The sun-pointing configuration permits bypassing calibration equipment required by other instruments and also furnishes high resolution vertical profile data. The retrieval techniques for quantifying the amounts of aerosol, water vapor, NO2 and O3, the species to which the SAGE II sensors are tuned, are summarized, along with methods for estimating the uncertainty in the calculations.

Chu, W. P.↗

NASA NTRS1989

Periodic-disturbance accommodating control of the Space Station for asymptotic momentum management

Periodic-disturbance accommodating control is investigated for asymptotic momentum management of control moment gyros used as primary actuating devices for the Space Station. The proposed controller utilizes the concepts of quaternion feedback control and periodic-disturbance accommodation to achieve oscillations about the constant torque equilibrium attitude, while minimizing the control effort required. Three-axis coupled equations of motion, written in terms of quaternions, are derived for roll/yaw controller design and stability analysis. The quaternion feedback controller designed using the linear-quadratic regulator synthesis technique is shown to be robust for a wide range of pitch angles. It is also shown that the proposed controller tunes the open-loop unstable vehicle to a stable oscillatory motion which minimizes the control effort needed for steady-state operations.

Warren, Wayne↗

NASA NTRS1989

Periodic-disturbance accommodating control of the space station for asymptotic momentum management

Warren, Wayne↗

NASA NTRS2001

Functionalization of Carbon Nanotubes using Atomic Hydrogen

We have investigated the irradiation of multi walled and single walled carbon nanotubes (SWNTs) with atomic hydrogen. After irradiating the SWNT sample, a band at 2940/cm (3.4 microns) that is characteristic of the C-H stretching mode is observed using Fourier transform infrared (FTIR) spectroscopy. Additional confirmation of SWNT functionalization is tested by irradiating with atomic deuterium. A weak band in the region 1940/cm (5.2 micron) to 2450/cm (4.1 micron) corresponding to C-D stretching mode is also observed in the FTIR spectrum. This technique provides a clean gas phase process for the functionalization of SWNTs, which could lead to further chemical manipulation and/or the tuning of the electronic properties of SWNTs for nanodevice applications.

Khare, Bishun N.↗

NASA NTRS2020

Fast Neutron Spectroscopy with Organic Scintillation Detectors in a High-Radiation Environment

Organic scintillators with pulse shape discrimination capability are widely used in both research and practical applications of neutron detection. The neutron and gamma-ray identification performance of the detector depends on the classification algorithms, noise filters and pileup rejection criteria in a high-flux bremsstrahlung radiation environment. In this paper, a technique has been developed and implemented for the neutron detection with multiple filter and discrimination steps, which to a high confidence level eliminates counting of gamma-ray pulses. Such a technique is merited when making measurements in a high flux bremsstrahlung and secondary fluorescence environment. The EJ-309 and stilbene detectors coupled to the digital data acquisition system were used for the calibration assessments with standard gamma-ray and neutron sources such as 137Cs, 60Co, 252Cf, and Am-Be. The MCNPX-PoliMi and GEANT4 toolkits were used to simulate the light output and the optical photon transport in the scintillators and create detector response functions for each type of detector. The neutron spectrum unfolding algorithm, GRAVELW, was used to recreate and calibrate with the Am-Be as final step before applying the neutron detection system to extract fusion neutron spectra generated in an intense bremsstrahlung radiation environment. This new technique described offers the user the ability to measure neutron spectra in a high-flux gamma-ray field and tune the parameters to meet required filtering needs.

Neutron Detector, Organic Scintillator, PulseShape↗

NASA NTRS2012

Intelligent Systems Approach for Automated Identification of Individual Control Behavior of a Human Operator

Results have been obtained using conventional techniques to model the generic human operator?s control behavior, however little research has been done to identify an individual based on control behavior. The hypothesis investigated is that different operators exhibit different control behavior when performing a given control task. Two enhancements to existing human operator models, which allow personalization of the modeled control behavior, are presented. One enhancement accounts for the testing control signals, which are introduced by an operator for more accurate control of the system and/or to adjust the control strategy. This uses the Artificial Neural Network which can be fine-tuned to model the testing control. Another enhancement takes the form of an equiripple filter which conditions the control system power spectrum. A novel automated parameter identification technique was developed to facilitate the identification process of the parameters of the selected models. This utilizes a Genetic Algorithm based optimization engine called the Bit-Climbing Algorithm. Enhancements were validated using experimental data obtained from three different sources: the Manual Control Laboratory software experiments, Unmanned Aerial Vehicle simulation, and NASA Langley Research Center Visual Motion Simulator studies. This manuscript also addresses applying human operator models to evaluate the effectiveness of motion feedback when simulating actual pilot control behavior in a flight simulator.

Zaychik, Kirill B.↗

NASA NTRS2000

Incremental Parallelization of Non-Data-Parallel Programs Using the Charon Message-Passing Library

Message passing is among the most popular techniques for parallelizing scientific programs on distributed-memory architectures. The reasons for its success are wide availability (MPI), efficiency, and full tuning control provided to the programmer. A major drawback, however, is that incremental parallelization, as offered by compiler directives, is not generally possible, because all data structures have to be changed throughout the program simultaneously. Charon remedies this situation through mappings between distributed and non-distributed data. It allows breaking up the parallelization into small steps, guaranteeing correctness at every stage. Several tools are available to help convert legacy codes into high-performance message-passing programs. They usually target data-parallel applications, whose loops carrying most of the work can be distributed among all processors without much dependency analysis. Others do a full dependency analysis and then convert the code virtually automatically. Even more toolkits are available that aid construction from scratch of message passing programs. None, however, allows piecemeal translation of codes with complex data dependencies (i.e. non-data-parallel programs) into message passing codes. The Charon library (available in both C and Fortran) provides incremental parallelization capabilities by linking legacy code arrays with distributed arrays. During the conversion process, non-distributed and distributed arrays exist side by side, and simple mapping functions allow the programmer to switch between the two in any location in the program. Charon also provides wrapper functions that leave the structure of the legacy code intact, but that allow execution on truly distributed data. Finally, the library provides a rich set of communication functions that support virtually all patterns of remote data demands in realistic structured grid scientific programs, including transposition, nearest-neighbor communication, pipelining, gather/scatter, and redistribution. At the end of the conversion process most intermediate Charon function calls will have been removed, the non-distributed arrays will have been deleted, and virtually the only remaining Charon functions calls are the high-level, highly optimized communications. Distribution of the data is under complete control of the programmer, although a wide range of useful distributions is easily available through predefined functions. A crucial aspect of the library is that it does not allocate space for distributed arrays, but accepts programmer-specified memory. This has two major consequences. First, codes parallelized using Charon do not suffer from encapsulation; user data is always directly accessible. This provides high efficiency, and also retains the possibility of using message passing directly for highly irregular communications. Second, non-distributed arrays can be interpreted as (trivial) distributions in the Charon sense, which allows them to be mapped to truly distributed arrays, and vice versa. This is the mechanism that enables incremental parallelization. In this paper we provide a brief introduction of the library and then focus on the actual steps in the parallelization process, using some representative examples from, among others, the NAS Parallel Benchmarks. We show how a complicated two-dimensional pipeline-the prototypical non-data-parallel algorithm- can be constructed with ease. To demonstrate the flexibility of the library, we give examples of the stepwise, efficient parallel implementation of nonlocal boundary conditions common in aircraft simulations, as well as the construction of the sequence of grids required for multigrid.

VanderWijngaart, Rob F.↗

NASA NTRS2005

Attitude and position estimation on the Mars Exploration Rovers

NASA/JPL 's Mars Exploration Rovers acquire their attitude upon command and autonomously propagate their attitude and position. The rovers use accelerometers and images of the sun to acquire attitude, autonomously searching the sky for the sun with a pointable camera. To propagate the attitude and position the rovers use either accelerometer and gyro readings or gyro readings and wheel odometiy, depending on the nature of the movement ground operators are commanding. Where necessary, visual odometry is performed on images to fine tune the position updates, particularly in high slip environments. The capability also exists for visual odometry attitude updates. This paper describes the techniques used by the rovers to acquire and maintain attitude and position knowledge, the accuracy which is obtainable, and lessons learned after more than one year in operation.

attitude estimation↗