Global sensitivity analysis and uncertainty quantification of soot formation in an n-dodecane spray flame

ABSTRACTThis research is part of an ongoing assessment to identify the most promising soot models for applications to laboratory turbulent flames, and ultimately to practical combustion systems. To that end, it is of interest to compare soot model simulation results with experiments over a broad range of conditions. Here five steady, one-dimensional, laminar, premixed ethylene flames have been targeted that cover a range of C/O ratios (0.68 to 0.88), pressures (1 atm to 10 bar), and peak soot levels (0.03 to 3.2 ppm). Seven different chemical mechanisms have been considered. For each mechanism, results from several soot model variants are compared. Two different approaches for soot aerosol dynamics are used: a discrete sectional method (DSM), and a method of moments with interpolative closure (MOMIC). DSM and MOMIC results are also compared with those from a widely used semi-empirical two-equation soot model. The main figure-of-merit is the ratio of computed-to-measured peak soot volume fraction. The computed soot volume fraction is most sensitive to variations in the surface chemistry scheme. Sensitivities to variations in model configurations are reduced at high pressure. The DSM-based models yield slightly lower soot volume fraction and slightly larger particle size compared to the corresponding MOMIC-based models. A modified semi-empirical two-equation model produced a good match to experiment for all five flames. Particle sizes and number densities from the models are discussed for a low-sooting flame where such data are available. The DSM-based models produce bimodal particle size distributions that are qualitatively consistent with those observed experimentally. The best results overall are obtained using an underlying chemical mechanism that incorporates recent understanding of polycyclic aromatic hydrocarbons kinetics. There is no clear advantage of DSM over MOMIC in predicting global quantities, such as soot volume fraction, particle number density, and average particle size, while the computational cost of DSM is significantly higher than that of MOMIC.

ABSTRACTPROSAIL is a combination of the leaf optical properties spectra (PROSPECT) model and the scattering by arbitrarily inclined leaves (SAIL) canopy bidirectional reflectance model. When modelling forest canopy reflectance using the PROSAIL radiative transfer model, the sensitivities of parameters can affect the modelling accuracy. Traditionally, sensitivities have been assessed using local sensitivity analysis (LSA); however, drawbacks to this approach include a lack of consideration for coupled effects between different parameters. In this study, parameter sensitivities in the PROSAIL model were calculated using two global sensitivity analysis (GSA) methods (the Extended Fourier Amplitude Sensitivity Test (EFAST) method and the Morris method), field measurements, and Landsat 5 Thematic Mapper (TM) data for a Moso bamboo forest. The results of GSA were compared with those of LSA in order to identify the key parameters impacting the Moso bamboo forest canopy reflectance, and to provide a reference for model optimization and vegetation canopy inversion improvement. The results showed that: (1) the sensitivities of six major input parameters of the PROSAIL model were generally consistent with the sorting orders of the two GSA methods, but were not in accordance with those from the LSA method, especially in the mid-infrared band; (2) coupled effects among parameters acting on reflectance simulation in visible light bands were greater than those in infrared bands; (3) the simulated canopy reflectance was evaluated using Landsat 5 TM data, and the results simulated based on LSA analysis showed higher error than those based on GSA analysis, because the LSA method ignored the influence of some parameters on canopy reflectance, e.g. leaf mesophyll structure (N), average leaf angle (ALA), leaf water content (Cw), and leaf dry matter content (Cm). However, GSA was able to fully consider the coupled effects among parameters, and thus identified the sensitive parameters impacting on reflectance more accurately.

Copula-based methods for global sensitivity analysis with correlated random variables and stochastic processes under incomplete probability information

Abstract. Integrated hydrological modeling of domains with complex subsurface features requires many highly uncertain parameters. Performing a global uncertainty analysis using an ensemble of model runs can help bring clarity as to which of these parameters really influence system behavior and for which high parameter uncertainty does not result in similarly high uncertainty of model predictions. However, already creating a sufficiently large ensemble of model simulation for the global sensitivity analysis can be challenging, as many combinations of model parameters can lead to unrealistic model behavior. In this work we use the method of active subspaces to perform a global sensitivity analysis. While building up the ensemble, we use the already-existing ensemble members to construct low-order meta-models based on the first two active-subspace dimensions. The meta-models are used to pre-determine whether a random parameter combination in the stochastic sampling is likely to result in unrealistic behavior so that such a parameter combination is excluded without running the computationally expensive full model. An important reason for choosing the active-subspace method is that both the activity score of the global sensitivity analysis and the meta-models can easily be understood and visualized. We test the approach on a subsurface-flow model including uncertain hydraulic parameters, uncertain boundary conditions and uncertain geological structure. We show that sufficiently detailed active subspaces exist for most observations of interest. The pre-selection by the meta-model significantly reduces the number of full-model runs that must be rejected due to unrealistic behavior. An essential but difficult part in active-subspace sampling using complex models is approximating the gradient of the simulated observation with respect to all parameters. We show that this can effectively and meaningfully be done with second-order polynomials.

Due to the growing concern on the issues of global warming, climate change, and energy shortage, more rigorous requirements are imposed on the energy conversion process of combustion. Due to the relative safety and convenience in transportation and storage, liquid fuels are commonly used in many practical combustion systems such as diesel engines, gas turbines and industrial furnaces. In these combustion devices, turbulent spray flows are involved. The character of spray dispersion, evaporation, mixing and combustion of fuel droplets strongly determines the performance of these systems with respect to the combustion efficiency, stability, and emissions. Therefore, an improved understanding of turbulent spray combustion and the development of predictive models are needed for a better design of more efficient combustion systems. The present work focuses on the development of a computational methodology based on the transported joint probability density function (PDF) method for the modeling and simulation of two-phase turbulent spray flows without and with chemical reactions. In the non-reacting situation, the dependent variables of the joint PDF include the gas velocity and the mixture fraction. For the simulation of reacting spray flows, a three-variate joint PDF transport equation is derived and modeled in order to account for the pre-vaporization effects and partially premixed regime in turbulent spray flames. The dependent variables include the gas phase mixture fraction, the reaction progress variable and gas enthalpy. Detailed combustion chemistry is considered through an extended spray flamelet model by including a reaction progress variable in addition to the classical formulation. The dilute spray is simulated using a Lagrangian discrete parcel method for the description of droplet motion, heating and evaporation. The infinite conductivity model with consideration of non-equilibrium effects based on the Langmuir-Knudsen law is considered. The spray evolution and flame structures in the frame of the polydisperse reacting spray flows are investigated. Numerical results are compared with experimental data provided by Prof. Masri at the University of Sydney, Australia, and the test cases include three different turbulent non-reacting acetone spray flows in air and turbulent spray flames with the liquid fuel ethanol. For the acetone spray flows, computational results generally show good agreement with experimental data in terms of the droplet size and mean velocity distribution, as well as the liquid volume flux. The results show that the inflow liquid mass loading hardly affects the droplet diameter distribution, whereas the inlet turbulence level has a pronounced effect. The tendency of droplet accumulation near the jet centerline is found with a somewhat overprediction of liquid volume flux at downstream locations. A more sophisticated turbulence model is expected to eliminate this discrepancy. Moreover, the local joint PDF of the gas velocity and the mixture fraction is analyzed. A linear correlation of the gas velocity and the mixture fraction exists close to the nozzle exit outside the main spray jet, and no regions are found where statistical independence prevails. For the simulation of spray flames, computations with the newly developed spray flamelet/progress variable approach and with the previous standard spray flamelet formulation are carried out and compared with the experiments by Prof. Masri at the University of Sydney. A good agreement between the computations with the new formulation and the experiments for gas temperature and droplet size and velocity is achieved. The major spray and combustion properties are correctly captured using this new formulation, which is compared with an unphysically attached flame near the nozzle exit predicted by the previous model. A partially premixed combustion prevails in this piloted turbulent spray flame. Due to the prevaporization of the ethanol droplets near the nozzle exit, a lean premixed gas mixture is found at the inner side of spray jet. Moving downstream, the lean-sided diffusion flame is promoted towards the inner fuel-rich side by heating up the inner premixed core that is controlled by the droplet evaporation. Additionally, it is observed that in the far-field region, the diffusion flame becomes the dominant combustion mode. In summary, an efficient computational model based on the transported joint PDF method is developed to two-phase turbulent spray flows. The combined transported joint PDF and a newly proposed spray flamelet/progress variable approach shows an improved performance in the prediction of complex turbulent spray flames, and new insights on the local flame structure influenced by evaporating sprays are obtained.

Active subspaces for sensitivity analysis and dimension reduction of an integrated hydrologic model

Simulation of fate and transport mechanisms in the porous systems from contaminant transport models are affected by uncertainty associated with input model parameters. Sensitivity analysis (SA) provides tools to quantify the sources of uncertainty to the variations in the model output metrics. To quantify the uncertainty associated with physical non-equilibrium contaminant transport model, global SA was conducted for the problem mimicking reactive transport in the saturated soil column conditions. Five global SA methods, namely Morris, RSA (Regionalized Sensitivity Analysis), Sobol, FAST (Fourier Amplitude Sensitivity Test), and PAWN, were tested based on the temporal moments of contaminant concentrations for two output metrics: zeroth temporal moment (ZTM) and mean residence time (MRT). The ranking order of ten input model parameters from global SA methods for two output metrics was compared. Morris SA implied that the ZTM at outlet of the soil column is highly sensitive to sorption distribution coefficient in the mobile and immobile regions and less sensitive to dispersion coefficient and degradation rate constant in the immobile region. The mass-transfer coefficient showed highest non-linear interactions with other flow and transport parameters based on the highest value of standard deviation of elementary effects (EEs) for ZTM output metric. The sorption distribution coefficient in the mobile region and mass-transfer coefficient showed highest sensitivity and non-linear effect toward MRT based on the Morris method. The comparison of global SA methods revealed that the top two sensitive parameters affecting ZTM and MRT were the same from all the considered methods. However, large difference in the ranking order of bottom three sensitive parameters was observed. The sorption distribution coefficient of mobile region and mass-transfer coefficient were observed as the most sensitive parameters affecting ZTM and MRT based on the comparison of all five global SA methods. Overall results suggest that the non-linear contaminant transport model should be examined based on multiple sensitivity output metrics via a multi-model approach. The multi-model global SA approach implemented in this study highlighted its significance in quantifying the interplay of non-linear model parameters.

Uncertainty and sensitivity analysis of cooling and heating loads for building energy planning

Physiologically based pharmacokinetic (PBPK) models describe the absorption, distribution, metabolism, and excretion of chemicals. Probabilistic PBPK models can be used to produce distributional estimates of human equivalent doses (HEDs), which are external measures of human exposure predicted to result in a target internal dose, generated with Monte Carlo sampling and reverse dosimetry calculations. Very low HED percentiles represent individuals that are more sensitive to possible adverse effects of chemical exposures and are therefore frequently used in risk evaluation. Details of the parameter distributions used in probabilistic PBPK models impact HED distributions, but obtaining precise distributional estimates for all the parameters would be challenging. Therefore, we sought to determine methods that can identify the extent of model parameter influence on extreme HED percentiles. We first analyzed published PBPK models for dichloromethane and chloroform (for inhalation and oral exposures given 3 internal target levels each) by identifying the overall relative importance of parameters using global sensitivity analysis (GSA) methods. Then, we used a novel yet computationally expensive method to analyze the stability and sensitivity of extreme HED percentiles to input parameter distributions. Applying the traditional GSA methods allowed us to identify subsets of parameters most influential for accurately determining 1st and 99th percentiles, but the specific parameters included in those subsets varied for different models and exposure scenarios. Our results demonstrate that better characterizing PBPK model uncertainty by using precise distributional details for influential parameters informed by GSA methods may improve confidence in estimates of extreme percentiles of HEDs.

Development of a behaviour-pattern based global sensitivity analysis procedure for coupled socioeconomic and environmental models

A regularized deconvolution model for sub-grid dispersion in large eddy simulation of turbulent spray flames

This paper presents an application of the Global Sensitivity Analysis (GSA) method in evaluating performance of impedance measurement algorithms implemented in modern Intelligent Electronic Devices (IEDs) when affected by uncertainty. To be specific, the GSA is applied in testing an algorithm for distance protection of a single transmission line in a system with two lines on the same towers. Factors contributing to uncertainty are discussed in the paper. The GSA method decomposes variance of the estimation error according to the factors and their interactions, and in this way identifies the factors that have the highest impact on the estimation error. Computation of the GSA sensitivity indices relies on the Quasi-Monte Carlo (QMC) technique with Sobol's quasi-random sampling. This is time-consuming for large number of factors therefore, fast pre-screening process based on the Morris method is implemented to remove non-influential parameters and reduce the factor space before running the QMC. The methodology is implemented using fault simulation software DIgSILENT PowerFactory to calculate fault impedance as a function of uncertain parameters (i.e. factors) as well as voltage and current signals that are inputs to an IED under test. In this simulation, factor space is sampled using Sobol's quasi-random sequence generated with the standard GSA software tool called SIMLAB. All simulations of the transmission line faults for various parameter values and calculation of the performance indices (i.e. sensitivities) is automated using the program developed in DIgSILENT Programming Language (DPL). This testing technique is applied to study the fault impedance measurement implemented in SEL-421. (6 pages)

Intelligent sensitivity analysis framework based on machine learning for spacecraft thermal design

Gradient free active subspace construction using Morris screening elementary effects

Groundwater models always involve high‐dimensional parameters, which makes computationally tractable data assimilation using surrogate models very challenging. To address this issue, one common practice is to employ dimension reduction (DR) techniques. Nevertheless, traditional DR methods are usually implemented based on prior parameter statistics, that is, without considering the inherent system dynamics. Here, we show that when significant difference in parameter sensitivity exists, further efficiency can be achieved by adopting a supervised DR method, that is, the active subspace (AS) method. To avoid non‐trivial efforts in calculating the gradient information needed in the standard AS method, a cluster‐based gradient‐free AS (GFAS) method is developed in this study. By combining GFAS with Gaussian process regression, a surrogate model for the CPU‐demanding groundwater model can be adaptively constructed to accelerate data assimilation. Furthermore, a compensation scheme is proposed to cope with uncertainty underestimation caused by DR. The developed approach is tested with numerical experiments and field cases, which illustrated that the new approach is more efficient than the previously developed unsupervised ones by incorporating sensitivity information. Although an iterative ensemble smoother is employed in this study, the proposed method can also be used in other data assimilation approaches, such as Markov chain Monte Carlo and ensemble Kalman filter.