Search NASA⌕ Search

SEARCH · Search NASA

Results for “Nonlinear problems”

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 955 records · Page 53

Higher-order LaSDI: Reduced order modeling with multiple time derivatives

Solving complex partial differential equations (PDEs) is essential across scientific disciplines but often requires numerical models that can be prohibitively expensive in time-sensitive applications. Reduced-order models (ROMs) address this challenge by exploiting low-dimensional structure to create fast approximations. The Latent Space Dynamics Identification (LaSDI) framework has demonstrated success in learning ROMs for parameterized PDE families, but remains limited to first-order systems. Here, in this paper, we propose Higher-Order LaSDI (HLaSDI), which extends the LaSDI framework to PDEs with arbitrary order of time derivatives. This generalization significantly expands the applicability of LaSDI-based methods to systems previously outside their scope, including hyperbolic PDEs. We demonstrate HLaSDI’s accuracy and efficiency on several linear and nonlinear benchmark problems.

97 MATHEMATICS AND COMPUTING↗

Autoignited laminar lifted flames of propane in coflow jets with tribrachial edge and mild combustion

Characteristics of laminar lifted flames have been investigated experimentally by varying the initial temperature of coflow air over 800 K in the non-premixed jets of propane diluted with nitrogen. The result showed that the lifted flame with the initial temperature below 860 K maintained the typical tribrachial structure at the leading edge, which was stabilized by the balance mechanism between the propagation speed of tribrachial flame and the local flow velocity. For the temperature above 860 K, the flame was autoignited without having any external ignition source. The autoignited lifted flames were categorized in two regimes. In the case with tribrachial edge structure, the liftoff height increased nonlinearly with jet velocity. Especially, for the critical condition near blowout, the lifted flame showed a repetitive behavior of extinction and reignition. In such a case, the autoignition was controlled by the non-adiabatic ignition delay time considering heat loss such that the autoignition height was correlated with the square of the adiabatic ignition delay time. In the case with mild combustion regime at excessively diluted conditions, the liftoff height increased linearly with jet velocity and was correlated well with the square of the adiabatic ignition delay time. (author)

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Response dynamics of bluff-body stabilized conical premixed turbulent flames with spatial mixture gradients

Response of bluff-body stabilized conical turbulent premixed flames was experimentally studied for a range of excitation frequencies (10-400 Hz), mean flow velocities (5, 10 and 15 m/s) and three different spatial mixture distributions (uniform, inner and outer enrichment). Upstream excitation was provided by a loudspeaker producing velocity oscillation amplitudes of about 8% of the mean flow velocity. Flame response was detected by a photomultiplier observing the CH{sup *} emission from the flame. The studied turbulent flames exhibited transfer function characteristics of a low-pass filter with a cutoff Strouhal number between 0.08 and 0.12. The amplification factors at low frequencies ranged from 2 to 20 and generally increased for mean flow velocities from 5 to 15 m/s. The highest levels of amplification were found for the outer mixture enrichment followed in decreasing order by uniform and inner mixture gradient cases. The high levels of flame response for the outer enrichment case were attributed to the enhanced flame-vortex interaction in outer jet shear layer. At high excitation levels (u{sup '}/U{sub m}{approx}0.3) for U{sub m}=5 m/ s where non-linear flame response is expected, the flame exhibited a reduced amplitude response in the frequency range between 40 and 100 Hz for the uniform and outer equivalence ratio gradient cases and no discernible effect for the inner equivalence ratio gradient. In all cases, transfer function phase was found to vary linearly with excitation frequency. Finally, a relationship between the amplitude characteristics of the bluff-body wake transfer function and flame blowoff equivalence ratio was presented. (author)

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Visible emission of hydrogen flames

The common misconception that hydrogen flames are not visible is examined. Examples are presented of clearly visible emissions from typical hydrogen flames. It is shown that while visible emissions from these flames are considerably weaker than those from comparable hydrocarbon flames, they are indeed visible, albeit at reduced light levels in most cases. Detailed flame spectra are presented to characterize flame emission bands in the ultraviolet, visible and infrared regions of the spectrum that result in a visible hydrogen flame. The visible blue emission is emphasized, and recorded spectra indicate that fine spectral structure is superimposed on a broadband continuum extending from the ultraviolet into the visible region. Tests were performed to show that this emission does not arise from carbon or nitrogen chemistry resulting from carbon-containing impurities (hydrocarbons) in the hydrogen fuel or from CO{sub 2} or N{sub 2} entrainment from the surrounding air. The spectral structure, however, is also observed in methane flames. The magnitude of the broadband emission increases with flame temperature in a highly nonlinear manner while the finer spectral structure is insensitive to temperature. A comparison of diffusion and premixed H{sub 2} flames shows that the fine scale structure is comparable in both flames. (author)

08 HYDROGEN↗

Computational and experimental study of the effects of adding dimethyl ether and ethanol to nonpremixed ethylene/air flames

Two sets of axisymmetric laminar coflow flames, each consisting of ethylene/air nonpremixed flames with various amounts (up to 10%) of either dimethyl ether (CH{sub 3}-O-CH{sub 3}) or ethanol (CH{sub 3}-CH{sub 2}-OH) added to the fuel stream, have been examined both computationally and experimentally. Computationally, the local rectangular refinement method, which incorporates Newton's method, is used to solve the fully coupled nonlinear conservation equations on solution-adaptive grids for each flame in two spatial dimensions. The numerical model includes C6 chemical kinetic mechanisms with up to 59 species, detailed transport, and an optically thin radiation submodel. Experimentally, thermocouples are used to measure gas temperatures, and mass spectrometry is used to determine concentrations of over 35 species along the flame centerline. Computational results are examined throughout each flame, and validation of the model occurs through comparison with centerline measurements. Very good agreement is observed for temperature, major species, and several minor species. As the level of additive is increased, temperatures, some major species (CO{sub 2}, C{sub 2}H{sub 2}), flame lengths, and residence times are essentially unchanged. However, peak centerline concentrations of benzene (C{sub 6}H{sub 6}) increase, and this increase is largest when dimethyl ether is the additive. Computational and experimental results support the hypothesis that the dominant pathway to C{sub 6}H{sub 6} formation begins with the oxygenates decomposing into methyl radical (CH{sub 3}), which combines with C2 species to form propargyl (C{sub 3}H{sub 3}), which reacts with itself to form C{sub 6}H{sub 6}. (author)

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Nonlinear effects in the extraction of laminar flame speeds from expanding spherical flames

Various factors affecting the determination of laminar flames speeds from outwardly propagating spherical flames in a constant-pressure combustion chamber were considered, with emphasis on the nonlinear variation of the stretched flame speed to the flame stretch rate, and the associated need to nonlinearly extrapolate the stretched flame speed to yield an accurate determination of the laminar flame speed and Markstein length. Experiments were conducted for lean and rich n-butane/air flames at 1atm initial pressure, demonstrating the complex and nonlinear nature of the dynamics of flame evolution, and the strong influences of the ignition transient and chamber confinement during the initial and final periods of the flame propagation, respectively. These experimental data were analyzed using the nonlinear relation between the stretched flame speed and stretch rate, yielding laminar flame speeds that agree well with data determined from alternate flame configurations. It is further suggested that the fidelity in the extraction of the laminar flame speed from expanding spherical flames can be facilitated by using small ignition energy and a large combustion chamber. (author)

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Thermoacoustic limit cycles in a premixed laboratory combustor with open and choked exits

This paper presents an experimental and theoretical investigation of the response of a turbulent premixed flame during thermoacoustic limit cycle in a simple, laboratory combustor. The flame dynamics are examined using high-speed pressure transducers and CH* chemiluminescence. The so-called 'interaction index' and time delay between the acoustic velocity fluctuations at the flame holder and the flame's overall heat release fluctuations are then determined. A wide range of operating conditions, traversing the combustor's flammability limits in Mach number and equivalence ratio, are studied for four different combustor exits, including one where the exit is choked. In all cases the time delay correlates very well with the amplitude of the velocity fluctuations. There is also some correlation between the interaction index and these velocity fluctuations, but this is less clear. These results suggest a novel, nonlinear flame model, derived entirely empirically. An existing low-order thermoacoustic model is then extended to include convection and dispersion of entropy fluctuations downstream of the flame, enabling the effect of the choked nozzle to be examined. The novel nonlinear flame model is integrated into the low-order thermoacoustic model, and used to investigate the experimentally observed thermoacoustic limit cycles. The model correctly simulates the observed switch to a low-frequency, entropically driven instability observed when the combustor exit is choked. (author)

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Propagation and extinction of premixed C{sub 5}-C{sub 12}n-alkane flames

Laminar flame speeds and extinction strain rates of premixed C{sub 5}-C{sub 12}n-alkane flames were determined at atmospheric pressure and elevated unburned mixture temperatures, over a wide range of equivalence ratios. Experiments were performed in the counterflow configuration and flow velocities were measured using Laser Doppler Velocimetry. The laminar flame speeds were obtained using a non-linear extrapolation technique utilizing numerical simulations of the counterflow experiments with detailed descriptions of chemical kinetics and molecular transport. Compared to linearly extrapolated values, the laminar flame speeds obtained using non-linear extrapolations were found to be 1-4 cm/s lower depending on the equivalence ratio. It was determined that the laminar flame speeds of all n-alkane/air mixtures considered in this investigation are similar to each other and sensitive largely to the H{sub 2}/CO and C{sub 1}-C{sub 4} hydrocarbon kinetics. Additionally, the resistance to extinction decreases as the fuel molecular weight increases. Simulations of the experiments were performed using the recently developed JetSurF 0.2 reaction model consisting of 194 species and 1459 reactions. The laminar flame speeds were predicted with good accuracy for all the n-alkane-air mixtures considered. The experimental extinction strain rates are well predicted by the model for fuel-lean mixtures. For stoichiometric and fuel-rich mixtures, the predicted extinction strain rates are approximately 10% lower than the experimental values. Insights into the physical and chemical processes that control the response of n-alkane flames are provided through detailed sensitivity analyses on both reaction rates and binary diffusion coefficients. (author)

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Linear and non-linear forced response of a conical, ducted, laminar premixed flame

This paper presents an experimental study on the dynamics of a ducted, conical, laminar premixed flame subjected to acoustic excitation of varying amplitudes. The flame transfer function is measured over a range of forcing frequencies and equivalence ratios. In keeping with previous works, the measured flame transfer function is in good agreement with that predicted by linear kinematic theory at low amplitudes of acoustic velocity excitation. However, a systematic departure from linear behaviour is observed as the amplitude of the velocity forcing upstream of the flame increases. This non-linearity is mostly in the phase of the transfer function and manifests itself as a roughly constant phase at high forcing amplitude. Nonetheless, as predicted by non-linear kinematic arguments, the response always remains close to linear at low forcing frequencies, regardless of the forcing amplitude. The origin of this phase behaviour is then sought through optical data post-processing. (author)

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

On droplet combustion of biodiesel fuel mixed with diesel/alkanes in microgravity condition

The burning characteristics of a biodiesel droplet mixed with diesel or alkanes such as dodecane and hexadecane were experimentally studied in a reduced-gravity environment so as to create a spherically symmetrical flame without the influence of natural convection due to buoyancy. Small droplets on the order of 500 {mu}m in diameter were initially injected via a piezoelectric technique onto the cross point intersected by two thin carbon fibers; these were prepared inside a combustion chamber that was housed in a drag shield, which was freely dropped onto a foam cushion. It was found that, for single component droplets, the tendency to form a rigid soot shell was relatively small for biodiesel fuel as compared to that exhibited by the other tested fuels. The soot created drifted away readily, showing a puffing phenomenon; this could be related to the distinct molecular structure of biodiesel leading to unique soot layers that were more vulnerable to oxidative reactivity as compared to the soot generated by diesel or alkanes. The addition of biodiesel to these more traditional fuels also presented better performance with respect to annihilating the soot shell, particularly for diesel. The burning rate generally follows that of multi-component fuels, by some means in terms of a lever rule, whereas the mixture of biodiesel and dodecane exhibits a somewhat nonlinear relation with the added fraction of dodecane. This might be related to the formation of a soot shell. (author)

09 BIOMASS FUELS↗

Dual-pump CARS temperature and major species concentration measurements in counter-flow methane flames using narrowband pump and broadband Stokes lasers

Dual-pump coherent anti-Stokes Raman scattering (CARS) is used to measure temperature and species profiles in representative non-premixed and partially-premixed CH{sub 4}/O{sub 2}/N{sub 2} flames. A new laser system has been developed to generate a tunable single-frequency beam for the second pump beam in the dual-pump N{sub 2}-CO{sub 2} CARS process. The second harmonic output ({proportional_to}532 nm) from an injection-seeded Nd:YAG laser is used as one of the narrowband pump beams. The second single-longitudinal-mode pump beam centered near 561 nm is generated using an injection-seeded optical parametric oscillator, consisting of two non-linear {beta}-BBO crystals, pumped using the third harmonic output ({proportional_to}355 nm) of the same Nd:YAG laser. A broadband dye laser (BBDL), pumped using the second harmonic output of an unseeded Nd:YAG laser, is employed to produce the Stokes beam centered near 607 nm with full-width-at-half-maximum of {proportional_to}250 cm{sup -1}. The three beams are focused between two opposing nozzles of a counter-flow burner facility to measure temperature and major species concentrations in a variety of CH{sub 4}/O{sub 2}/N{sub 2} non-premixed and partially-premixed flames stabilized at a global strain rate of 20 s{sup -1} at atmospheric-pressure. For the non-premixed flames, excellent agreement is observed between the measured profiles of temperature and CO{sub 2}/N{sub 2} concentration ratios with those calculated using an opposed-flow flame code with detailed chemistry and molecular transport submodels. For partially-premixed flames, with the rich side premixing level beyond the stable premixed flame limit, the calculations overestimate the distance between the premixed and the non-premixed flamefronts. Consequently, the calculated temperatures near the rich, premixed flame are higher than those measured. Accurate prediction of the distance between the premixed and the non-premixed flames provides an interesting challenge for future computations. (author)

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

A comparative experimental and computational study of methanol, ethanol, and n-butanol flames

Laminar flame speeds and extinction strain rates of premixed methanol, ethanol, and n-butanol flames were determined experimentally in the counterflow configuration at atmospheric pressure and elevated unburned mixture temperatures. Additional measurements were conducted also to determine the laminar flame speeds of their n-alkane/air counterparts, namely methane, ethane, and n-butane in order to compare the effect of alkane and alcohol molecular structures on high-temperature flame kinetics. For both propagation and extinction experiments the flow velocities were determined using the digital particle image velocimetry method. Laminar flame speeds were derived through a non-linear extrapolation approach based on direct numerical simulations of the experiments. Two recently developed detailed kinetics models of n-butanol oxidation were used to simulate the experiments. The experimental results revealed that laminar flame speeds of ethanol/air and n-butanol/air flames are similar to those of their n-alkane/air counterparts, and that methane/air flames have consistently lower laminar flame speeds than methanol/air flames. The laminar flame speeds of methanol/air flames are considerably higher compared to both ethanol/air and n-butanol/air flames under fuel-rich conditions. Numerical simulations of n-butanol/air freely propagating flames, revealed discrepancies between the two kinetic models regarding the consumption pathways of n-butanol and its intermediates. (author)

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Response of partially premixed flames to acoustic velocity and equivalence ratio perturbations

This article describes an experimental investigation of the forced response of a swirl-stabilized partially premixed flame when it is subjected to acoustic velocity and equivalence ratio fluctuations. The flame's response is analyzed using phase-resolved CH{sup *} chemiluminescence images and flame transfer function (FTF) measurements, and compared with the response of a perfectly premixed flame under acoustic perturbations. The nonlinear response of the partially premixed flame is manifested by a partial extinction of the reaction zone, leading to rapid reduction of flame surface area. This nonlinearity, however, is observed only when the phase difference between the acoustic velocity and the equivalence ratio at the combustor inlet is close to zero. The condition, {delta}{phi}{sub {phi}}'-V'{approx}0 , indicates that reactant mixtures with high equivalence ratio impinge on the flame front with high velocity, inducing large fluctuations of the rate of heat release. It is found that the phase difference between the acoustic velocity and equivalence ratio nonuniformities is a key parameter governing the linear/nonlinear response of a partially premixed flame, and it is a function of modulation frequency, inlet velocity, fuel injection location, and fuel injector impedance. The results presented in this article will provide insight into the response of a partially premixed flame, which has not been well explored to date. (author)

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Nonlinear effects of stretch on the flame front propagation

In all experimental configurations, the flames are affected by stretch (curvature and/or strain rate). To obtain the unstretched flame speed, independent of the experimental configuration, the measured flame speed needs to be corrected. Usually, a linear relationship linking the flame speed to stretch is used. However, this linear relation is the result of several assumptions, which may be incorrected. The present study aims at evaluating the error in the laminar burning speed evaluation induced by using the traditional linear methodology. Experiments were performed in a closed vessel at atmospheric pressure for two different mixtures: methane/air and iso-octane/air. The initial temperatures were respectively 300 K and 400 K for methane and iso-octane. Both methodologies (linear and nonlinear) are applied and results in terms of laminar speed and burned gas Markstein length are compared. Methane and iso-octane were chosen because they present opposite evolutions in their Markstein length when the equivalence ratio is increased. The error induced by the linear methodology is evaluated, taking the nonlinear methodology as the reference. It is observed that the use of the linear methodology starts to induce substantial errors after an equivalence ratio of 1.1 for methane/air mixtures and before an equivalence ratio of 1 for iso-octane/air mixtures. One solution to increase the accuracy of the linear methodology for these critical cases consists in reducing the number of points used in the linear methodology by increasing the initial flame radius used. (author)

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

On the extraction of laminar flame speed and Markstein length from outwardly propagating spherical flames

Large discrepancies among the laminar flame speeds and Markstein lengths of methane/air mixtures measured by different researchers using the same constant-pressure spherical flame method are observed. As an effort to reduce these discrepancies, one linear model (LM, the stretched flame speed changes linearly with the stretch rate) and two non-linear models (NM I and NM II, the stretched flame speed changes non-linearly with the stretch rate) for extracting the laminar flame speed and Markstein length from propagating spherical flames are investigated. The accuracy and performance of the LM, NM I, and NM II are found to strongly depend on the Lewis number. It is demonstrated that NM I is the most accurate for mixtures with large Lewis number (positive Markstein length) while NM II is the most accurate for mixtures with small Lewis number (negative Markstein length). Therefore, in order to get accurate laminar flame speed and Markstein length from spherical flame experiments, different non-linear models should be used for different mixtures. The validity of the theoretical results is further demonstrated by numerical and experimental studies. The results of this study can be used directly in spherical flame experiments measuring the laminar flame speed and Markstein length. (author)

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Does the high–tech industry consistently reduce CO{sub 2} emissions? Results from nonparametric additive regression model

China is currently the world's largest carbon dioxide (CO{sub 2}) emitter. Moreover, total energy consumption and CO{sub 2} emissions in China will continue to increase due to the rapid growth of industrialization and urbanization. Therefore, vigorously developing the high–tech industry becomes an inevitable choice to reduce CO{sub 2} emissions at the moment or in the future. However, ignoring the existing nonlinear links between economic variables, most scholars use traditional linear models to explore the impact of the high–tech industry on CO{sub 2} emissions from an aggregate perspective. Few studies have focused on nonlinear relationships and regional differences in China. Based on panel data of 1998–2014, this study uses the nonparametric additive regression model to explore the nonlinear effect of the high–tech industry from a regional perspective. The estimated results show that the residual sum of squares (SSR) of the nonparametric additive regression model in the eastern, central and western regions are 0.693, 0.054 and 0.085 respectively, which are much less those that of the traditional linear regression model (3.158, 4.227 and 7.196). This verifies that the nonparametric additive regression model has a better fitting effect. Specifically, the high–tech industry produces an inverted “U–shaped” nonlinear impact on CO{sub 2} emissions in the eastern region, but a positive “U–shaped” nonlinear effect in the central and western regions. Therefore, the nonlinear impact of the high–tech industry on CO{sub 2} emissions in the three regions should be given adequate attention in developing effective abatement policies. - Highlights: • The nonlinear effect of the high–tech industry on CO{sub 2} emissions was investigated. • The high–tech industry yields an inverted “U–shaped” effect in the eastern region. • The high–tech industry has a positive “U–shaped” nonlinear effect in other regions. • The linear impact of the high–tech industry in the eastern region is the strongest.

54 ENVIRONMENTAL SCIENCES↗

Mid- and long-term runoff predictions by an improved phase-space reconstruction model

In recent years, the phase-space reconstruction method has usually been used for mid- and long-term runoff predictions. However, the traditional phase-space reconstruction method is still needs to be improved. Using the genetic algorithm to improve the phase-space reconstruction method, a new nonlinear model of monthly runoff is constructed. The new model does not rely heavily on embedding dimensions. Recognizing that the rainfall–runoff process is complex, affected by a number of factors, more variables (e.g. temperature and rainfall) are incorporated in the model. In order to detect the possible presence of chaos in the runoff dynamics, chaotic characteristics of the model are also analyzed, which shows the model can represent the nonlinear and chaotic characteristics of the runoff. The model is tested for its forecasting performance in four types of experiments using data from six hydrological stations on the Yellow River and the Yangtze River. Results show that the medium-and long-term runoff is satisfactorily forecasted at the hydrological stations. Not only is the forecasting trend accurate, but also the mean absolute percentage error is no more than 15%. Moreover, the forecast results of wet years and dry years are both good, which means that the improved model can overcome the traditional ‘‘wet years and dry years predictability barrier,’’ to some extent. The model forecasts for different regions are all good, showing the universality of the approach. Compared with selected conceptual and empirical methods, the model exhibits greater reliability and stability in the long-term runoff prediction. Our study provides a new thinking for research on the association between the monthly runoff and other hydrological factors, and also provides a new method for the prediction of the monthly runoff. - Highlights: • The improved phase-space reconstruction model of monthly runoff is established. • Two variables (temperature and rainfall) are incorporated in the model. • Chaotic characteristics of the model are also analyzed. • The forecast results of the mid and long-term runoff in six stations are accurate.

97 MATHEMATICS AND COMPUTING↗