Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “variance reduction”

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.

318 records · Page 18
All sourcesNASA NTRSDOE OSTI
NASA NTRS2000

Dissociation of muscle sympathetic nerve activity and leg vascular resistance in humans

We examined the hypothesis that the increase in inactive leg vascular resistance during forearm metaboreflex activation is dissociated from muscle sympathetic nerve activity (MSNA). MSNA (microneurography), femoral artery mean blood velocity (FAMBV, Doppler), mean arterial pressure (MAP), and heart rate (HR) were assessed during fatiguing static handgrip exercise (SHG, 2 min) followed by posthandgrip ischemia (PHI, 2 min). Whereas both MAP and MSNA increase during SHG, the transition from SHG to PHI is characterized by a transient reduction in MAP but sustained elevation in MSNA, facilitating separation of these factors in vivo. Femoral artery vascular resistance (FAVR) was calculated (MAP/MBV). MSNA increased by 59 +/- 20% above baseline during SHG (P < 0.05) and was 58 +/- 18 and 78 +/- 18% above baseline at 10 and 20 s of PHI, respectively (P < 0.05 vs. baseline). Compared with baseline, FAVR increased 51 +/- 22% during SHG (P < 0.0001) but returned to baseline levels during the first 30 s of PHI, reflecting the changes in MAP (P < 0.005) and not MSNA. It was concluded that control of leg muscle vascular resistance is sensitive to changes in arterial pressure and can be dissociated from sympathetic factors.

Non-NASA Center↗
NASA NTRS2013

Comparison of Precision of Biomass Estimates in Regional Field Sample Surveys and Airborne LiDAR-Assisted Surveys in Hedmark County, Norway

Airborne scanning LiDAR (Light Detection and Ranging) has emerged as a promising tool to provide auxiliary data for sample surveys aiming at estimation of above-ground tree biomass (AGB), with potential applications in REDD forest monitoring. For larger geographical regions such as counties, states or nations, it is not feasible to collect airborne LiDAR data continuously ("wall-to-wall") over the entire area of interest. Two-stage cluster survey designs have therefore been demonstrated by which LiDAR data are collected along selected individual flight-lines treated as clusters and with ground plots sampled along these LiDAR swaths. Recently, analytical AGB estimators and associated variance estimators that quantify the sampling variability have been proposed. Empirical studies employing these estimators have shown a seemingly equal or even larger uncertainty of the AGB estimates obtained with extensive use of LiDAR data to support the estimation as compared to pure field-based estimates employing estimators appropriate under simple random sampling (SRS). However, comparison of uncertainty estimates under SRS and sophisticated two-stage designs is complicated by large differences in the designs and assumptions. In this study, probability-based principles to estimation and inference were followed. We assumed designs of a field sample and a LiDAR-assisted survey of Hedmark County (HC) (27,390 km2), Norway, considered to be more comparable than those assumed in previous studies. The field sample consisted of 659 systematically distributed National Forest Inventory (NFI) plots and the airborne scanning LiDAR data were collected along 53 parallel flight-lines flown over the NFI plots. We compared AGB estimates based on the field survey only assuming SRS against corresponding estimates assuming two-phase (double) sampling with LiDAR and employing model-assisted estimators. We also compared AGB estimates based on the field survey only assuming two-stage sampling (the NFI plots being grouped in clusters) against corresponding estimates assuming two-stage sampling with the LiDAR and employing model-assisted estimators. For each of the two comparisons, the standard errors of the AGB estimates were consistently lower for the LiDAR-assisted designs. The overall reduction of the standard errors in the LiDAR-assisted estimation was around 40-60% compared to the pure field survey. We conclude that the previously proposed two-stage model-assisted estimators are inappropriate for surveys with unequal lengths of the LiDAR flight-lines and new estimators are needed. Some options for design of LiDAR-assisted sample surveys under REDD are also discussed, which capitalize on the flexibility offered when the field survey is designed as an integrated part of the overall survey design as opposed to previous LiDAR-assisted sample surveys in the boreal and temperate zones which have been restricted by the current design of an existing NFI.

Comparison↗
NASA NTRS2002

Controlled breaks as a fatigue countermeasure on the flight deck

BACKGROUND: A major challenge for flight crews is the need to maintain vigilance during long, highly automated nighttime flights. No system currently exists to assist in managing alertness, and countermeasure options are limited. Surveys reveal many pilots use breaks as an in-flight countermeasure, but there have been no controlled studies of their effectiveness. HYPOTHESIS: We hypothesized that brief, regular breaks could improve alertness and performance during an overnight flight. METHOD: A 6-h, uneventful, nighttime flight in a Boeing 747-400 flight simulator was flown by fourteen two-man crews. The 14 subjects in the treatment group received 5 short breaks spaced hourly during cruise; the 14 subjects in the control group received 1 break in the middle of cruise. Continuous EEG/EOG, subjective sleepiness, and psychomotor vigilance performance data were collected. RESULTS: During the latter part of the night, the treatment group showed significant reductions for 15 min post-break in slow eye movements, theta-band activity, and unintended sleep episodes compared with the control group. The treatment group reported significantly greater subjective alertness for up to 25 min post-break, with strongest effects near the time of the circadian trough. There was no evidence of objective vigilance performance improvement at 15-25 min post-break, with expected performance deterioration occurring due to elevated sleep drive and circadian time. CONCLUSIONS: The physiological and subjective data indicate the breaks reduced nighttime sleepiness for at least 15 min post-break and may have masked sleepiness for up to 25 min, suggesting the potential usefulness of short-duration breaks as an in-flight fatigue countermeasure.

Non-NASA Center↗
NASA NTRS2020

MSL Telecom Automated Anomaly Detection

The Mars Science Laboratory (MSL) Telecom Operations Team at the Jet Propulsion Laboratory (JPL) has implemented a machine learning system in order to automate the anomaly detection process as a part of daily operations. Machine learning enables reliable detection of anomalies in Telecom-related telemetry and automated reporting of Telecom subsystem status, resulting in an 90% reduction in team workload and improved anomaly detection reliability. At present, machine learning methods are used to detect: 1. Anomalous long-term trends in telemetry data 2. Anomalous time-domain evolution of telemetry values Both types of anomalies pose their own unique challenges that are addressed in different ways. In the first case, long term trending of daily minima, maximum, and mean telemetry values in temperatures, currents, voltages, and radio frequency (RF) power levels is used in addition to hard threshold safety checks to look for changes in long-term equipment health and performance. Long-term trending methods allow for ordinary seasonal variations in these quantities caused by temperature changes over the course of the Martian year while allowing operators to determine whether current performance remains in line with historical values from previous years. Changes in long-term trends can provide important insights into the health and status of the rover's on-board systems as well as valuable early warning if subtle degradation begins to take hold. But while trending of daily statistics is valuable, it does not detect anomalies in the short-term time evolution of data over the course of minutes or hours during a day, and this task is handled with short-term shape analysis. Principal components analysis (PCA) has been found to provide robust detection of short-term anomalies, and several examples of the use of PCA to detect actual anomalous events will be provided here. In using PCA, we use both the percentage of explained variance and also a log likelihood test on the PCA expansion coefficients to flag telemetry data for human review. Previous work in the field of spacecraft anomaly detection includes [1] for MSL and [2] for some other JPL missions.

Mukai, Ryan↗
DOE OSTI2023

Exploration of a Novel Technique for Waste Heat Recovery Through Molecular Dynamics: Influence of Wettability and Electric Field on Water and Water-Based Nanofluids

Most of the energy produced globally comes by way of a heat engine. The Carnot principle places a limit as to how thermodynamically efficient a heat engine can be. There is no heat engine that can be 100% thermodynamically efficient and as such a substantial proportion of all heat supplied to a heat engine is lost as waste heat. Waste heat therefore is a large energy source ready to be properly utilized. Herein, a novel approach for converting waste heat to electricity is discussed. It involves the use of the liquid to vapor phase change of a material dielectric (water) or electrolyte (nanofluid) in the embodiment of a capacitor for direct thermal to electrostatic energy conversion. While this method of waste heat recovery could potentially be added to the ever expanding portfolio of energy conversion techniques, a number of aspects must be addressed before it can be brought into practice. Water was seen as an ideal dielectric phase change material given its high relative permittivity ratio when in the liquid form as compared to its vapor form. However, given its short voltage holdoff time the phase change of water would need to occur rapidly. This brings up concerns of explosive boiling. Herein, molecular dynamics analysis into the explosive boiling behavior of thin water films gave more insight into how the interaction between the surface and liquid affected explosive boiling onset time. A Lennard-Jones potential with one interaction site and a Morse potential with three interaction sites between water and solid substrate were used. It was found generally that a stronger interaction between water film and substrate led to faster explosive boiling onset times but an increase in the number of interaction sites delayed explosive boiling, even at the same wettability (contact angle). Understanding changes in the density and enthalpy of vaporization of a liquid dielectric such as water in the presence of an electric field is of importance due to the electrostatic nature of the waste heat conversion method under consideration. Specifically, if both density and enthalpy of vaporization are increased, the thermodynamic efficiency of the waste heat conversion method under consideration is decreased. Electric field effects are explored herein via molecular dynamics using two water models, the TIP4P-Ew and SWM4-NDP. The SWM4-NDP model is polarizable while the TIP4P-Ew model is not, which allows for a determination of the importance of model polarizability (i.e. variation in water model dipole moment) on these two properties of water when subjected to an electric field. Herein it was found that both water models respond similarly in terms of density and vaporization enthalpy variance upon the introduction of an electric field. Comparison was also made to the pressure induced by the electric field (electrostriction pressure) by way of a density comparison and it was found that the predicted electrostriction pressure overestimates the pressure experienced by water. Water by itself has a high enthalpy of vaporization, which limits the efficiency of the newly proposed conversion method. Research both experimental and through simulation has shown that the vaporization enthalpy of nanofluids can be engineered via nanoparticle size and material selection. An avenue less explored is manipulating the enthalpy of vaporization by altering the interaction strength between the nanoparticles and the base fluid. In practice this could be achieved through the addition of coatings to the nanoparticles to alter their wettability to the base fluid. This was explored by using a Lennard-Jones potential and Morse potential to model the interaction between base fluid (water) and the nanoparticle. For nanoparticles 2nm in diameter and at weight percentages up to 6%, the change in vaporization enthalpy due to alterations of the interaction strength between nanoparticle and base fluid was not significant (less than a 1% difference) when compared to the effect of altering the weight percentage of nanoparticles in the nanofluid or introducing an electric field. However, the effect of wettability may still become important at other nanoparticle concentrations and sizes. In all, the studies presented here further the understanding of phase change and thermodynamic properties of water and water based nanofluids under an electrostatic field which will help inform the development of a novel approach to waste heat conversion. The reduction of waste heat will improve energy sustainability outlooks.

30 DIRECT ENERGY CONVERSION↗
NASA NTRS1992

Phase noise in pulsed Doppler lidar and limitations on achievable single-shot velocity accuracy

The smaller sampling volumes afforded by Doppler lidars compared to radars allows for spatial resolutions at and below some sheer and turbulence wind structure scale sizes. This has brought new emphasis on achieving the optimum product of wind velocity and range resolutions. Several recent studies have considered the effects of amplitude noise, reduction algorithms, and possible hardware related signal artifacts on obtainable velocity accuracy. We discuss here the limitation on this accuracy resulting from the incoherent nature and finite temporal extent of backscatter from aerosols. For a lidar return from a hard (or slab) target, the phase of the intermediate frequency (IF) signal is random and the total return energy fluctuates from shot to shot due to speckle; however, the offset from the transmitted frequency is determinable with an accuracy subject only to instrumental effects and the signal to noise ratio (SNR), the noise being determined by the LO power in the shot noise limited regime. This is not the case for a return from a media extending over a range on the order of or greater than the spatial extent of the transmitted pulse, such as from atmospheric aerosols. In this case, the phase of the IF signal will exhibit a temporal random walk like behavior. It will be uncorrelated over times greater than the pulse duration as the transmitted pulse samples non-overlapping volumes of scattering centers. Frequency analysis of the IF signal in a window similar to the transmitted pulse envelope will therefore show shot-to-shot frequency deviations on the order of the inverse pulse duration reflecting the random phase rate variations. Like speckle, these deviations arise from the incoherent nature of the scattering process and diminish if the IF signal is averaged over times greater than a single range resolution cell (here the pulse duration). Apart from limiting the high SNR performance of a Doppler lidar, this shot-to-shot variance in velocity estimates has a practical impact on lidar design parameters. In high SNR operation, for example, a lidar's efficiency in obtaining mean wind measurements is determined by its repetition rate and not pulse energy or average power. In addition, this variance puts a practical limit on the shot-to-shot hard target performance required of a lidar.

Mcnicholl, P.↗
NASA NTRS1998

Cytoskeletal role in the transition from compensated to decompensated hypertrophy during adult canine left ventricular pressure overloading

Increased microtubule density causes cardiocyte contractile dysfunction in right ventricular (RV) pressure-overload hypertrophy, and these linked phenotypic and contractile abnormalities persist and progress during the transition to failure. Although more severe in cells from failing than hypertrophied RVs, the mechanical defects are normalized in each case by microtubule depolymerization. To define the role of increased microtubule density in left ventricular (LV) pressure-overload hypertrophy and failure, in a given LV we examined ventricular mechanics, sarcomere mechanics, and free tubulin and microtubule levels in control dogs and in dogs with aortic stenosis both with LV hypertrophy alone and with initially compensated hypertrophy that had progressed to LV muscle failure. In comparing initial values with those at study 8 weeks later, dogs with hypertrophy alone had a very substantial increase in LV mass but preservation of a normal ejection fraction and mean systolic wall stress. Dogs with hypertrophy and associated failure had a substantial but lesser increase in LV mass and a reduction in ejection fraction, as well as a marked increase in mean systolic wall stress. Cardiocyte contractile function was equivalent, and unaffected by microtubule depolymerization, in cells from control LVs and those with compensated hypertrophy. In contrast, cardiocyte contractile function in cells from failing LVs was quite depressed but was normalized by microtubule depolymerization. Microtubules were increased only in failing LVs. These contractile and cytoskeletal changes, when assayed longitudinally in a given dog by biopsy, appeared in failing ventricles only when wall stress began to increase and function began to decrease. Thus, the microtubule-based cardiocyte contractile dysfunction characteristic of pressure-hypertrophied myocardium, originally described in the RV, obtains equally in the LV but is shown here to have a specific association with increased wall stress.

Non-NASA Center↗
NASA NTRS2001

Time course changes in [Ca2+]i, force, and protein content in hindlimb-suspended mouse soleus muscles

BACKGROUND: Exposure to reduced gravitational forces during spaceflight is associated with significant reductions in skeletal muscle mass and strength. The purpose of this study was to test the hypothesis that increases in resting cytosolic free calcium concentration ([Ca2+]i) would precede reductions in protein content and maximal isometric tetanic force (Po) in mouse soleus muscle after initiation of hindlimb suspension. METHODS: Female ICR mice (n = 42) were hindlimb suspended for 1, 2, 3, 5, or 7 d; weight-matched mice were used as controls. Following the hindlimb suspension, the left soleus muscle was used to determine Po in vitro and the right soleus muscle was used to determine protein content and [Ca2+]i via confocal laser scanning microscopy. RESULTS: Compared with controls, [Ca2+]i was elevated by 38% at 2 d, and 117% at 7 d. Compared with controls, soleus muscle total and myofibrillar protein contents were reduced 27-29% and 30-34%, respectively, at 5-7 d after initiation of hindlimb suspension. Compared with controls, soleus muscle Po was decreased by 24% at 3 d, and 38% at 7 d. CONCLUSION: It appears that resting cytosolic Ca2+ homeostasis is disturbed soon after the initiation of hindlimb suspension, and these elevations in [Ca2+]i may play a role in initiating soleus muscle atrophy.

Non-NASA Center↗
DOE OSTI2023

Multifidelity uncertainty quantification with models based on dissimilar parameters

Multifidelity uncertainty quantification (MF UQ) sampling approaches have been shown to significantly reduce the variance of statistical estimators while preserving the bias of the highest-fidelity model, provided that the low-fidelity models are well correlated. However, maintaining a high level of correlation can be challenging, especially when models depend on different input uncertain parameters, which drastically reduces the correlation. Existing MF UQ approaches do not adequately address this issue. In this work, we propose a new sampling strategy that exploits a shared space to improve the correlation among models with dissimilar parameterization. We achieve this by transforming the original coordinates onto an auxiliary manifold using the adaptive basis (AB) method (Tipireddy and Ghanem, 2014). The AB method has two main benefits: (1) it provides an effective tool to identify the low-dimensional manifold on which each model can be represented, and (2) it enables easy transformation of polynomial chaos representations from high- to low-dimensional spaces. This latter feature is used to identify a shared manifold among models without requiring additional evaluations. Here we present two algorithmic flavors of the new estimator to cover different analysis scenarios, including those with legacy and non-legacy high-fidelity (HF) data. We provide numerical results for analytical examples, a direct field acoustic test, and a finite element model of a nuclear fuel assembly. For all examples, we compare the proposed strategy against both single-fidelity and MF estimators based on the original model parameterization.

42 ENGINEERING↗
NASA NTRS1998

Cardiovascular function following reduced aerobic activity

PURPOSE: The aim of this study was to test the hypothesis that a sustained reduction of physical activity (deconditioning) would alter the cardiovascular regulatory function. METHODS: Nineteen young, healthy volunteers participated in physical deconditioning for a period of 8 wk. Before (pre) and following (post) physical deconditioning, the responses of heart rate (HR), mean arterial pressure (MAP, measured by Finapres), central venous pressure (CVP), stroke volume (SV, Doppler), and forearm blood flow (FBF, plethysmography) were determined during lower body negative pressure (LBNP). The carotid baroreflex (CBR) function was assessed using a train of pulsatile neck pressure (NP) and suction, and the aortic baroreflex control of HR was assessed during steady-state phenylephrine (PE) infusion superimposed by LBNP and NP to counteract the PE increased CVP and carotid sinus pressure, respectively. RESULTS: Active physical deconditioning significantly decreased maximal oxygen uptake (-7%) and LBNP tolerance (-13%) without a change in baseline hemodynamics. Plasma volume (-3% at P = 0.135), determined by Evans Blue dilution, and blood volume (-4% at P = 0.107) were not significantly altered. During LBNP -20 to -50 torr, there was a significantly greater drop of SV per unit decrease in CVP in the post- (14.7 +/- 1.6%/mm Hg) than predeconditioning (11.2 +/- 0.7%/mm Hg) test accompanied by a greater tachycardia. Deconditioning increased the aortic baroreflex sensitivity (pre vs post: -0.61 +/- 0.12 vs -0.84 +/- 0.14 bpm.mm-1 Hg, P = 0.009) and the slope of forearm vascular resistance (calculated from [MAP-CVP]/FBF) to CVP (-2.75 +/- 0.26 vs -4.94 +/- 0.97 PRU/mm Hg, P = 0.086). However, neither the CBR-HR (-0.28 +/- 0.03 VS -0.39 +/- 0.10 bpm.mm-1 Hg) nor the CBR-MAP (-0.37 +/- 0.16 vs -0.25 +/- 0.07 mm Hg/mm Hg) gains were statistically different between pre- and postdeconditioning. CONCLUSIONS: We concluded that the functional modification of the cardiac pressure-volume relationship resulted in the reduced LBNP tolerance, despite the accentuated aortic and cardiopulmonary baroreflex function following deconditioning.

Non-NASA Center↗
DOE OSTI2022

Deep learning to estimate permeability using geophysical data

Time-lapse electrical resistivity tomography (ERT) is a popular geophysical method to estimate three-dimensional (3D) permeability fields from electrical potential difference measurements. Traditional inversion and data assimilation methods are used to ingest this ERT data into hydrogeophysical models to estimate permeability. Due to ill-posedness and the curse of dimensionality, existing inversion strategies provide poor estimates and low resolution of the 3D permeability field. Recent advances in deep learning provide us with powerful algorithms to overcome this challenge. This paper presents a deep learning (DL) framework to estimate the 3D subsurface permeability from time-lapse ERT data. To test the feasibility of the proposed framework, we train DL-enabled inverse models on simulation data. Each measurement in both synthetic and field data is standardized by removing the mean and scaling the time-series to unit variance. This pre-processing step is necessary to bring simulation data closer to field observations. Subsurface process models based on hydrogeophysics are used to generate this synthetic data. Training performed on limited simulation data resulted in the DL model over-fitting. An advanced data augmentation based on mixup is implemented to generate additional training samples to overcome this issue. This mixup technique creates weakly labeled (low-fidelity) samples from strongly labeled (high-fidelity) data. The weakly labeled training data is then used to develop DL-enabled inverse models and reduce over-fitting. As both time-lapse ERT (1133048 features/realization) and 3D permeability (585453 features/realization) data samples are from a high-dimensional space, principal component analysis (PCA) is employed to reduce dimensionality. Encoded ERT and encoded permeability are generated using the trained PCA estimators. A deep neural network is then trained to map the encoded ERT to encoded permeability. This mixup training and unsupervised learning allowed us to build a fast and reasonably accurate DL-based inverse model under limited simulation data. Results show that proposed weak supervised learning can capture salient spatial features in the 3D permeability field. Quantitatively, the average mean squared error (in terms of the natural log) on the strongly labeled training, validation, and test datasets is less than 0.5. The R 2 -score (global metric) is greater than 0.75, and the percent error in each cell (local metric) is less than 10%. Finally, an added benefit in terms of computational cost is that the proposed DL-based inverse model is at least O(10 4 ) times faster than running a forward model once it is trained. Data generation, DL model training, and hyperparameter tuning to identify optimal neural network architectures utilized high-performance computing resources while the DL inference is performed on a standard laptop. Approximately, O(10 5 ) processor hours are used for generating data and DL tuning and training. We acknowledge that the data generation and DL model development are expensive. But once a DL model is trained, it can be re-used for inversion rapidly for the given system, with set physics and domain. Note that traditional inversion may require multiple forward model simulations (e.g., in the order of 10 to 1000), which are very expensive. This computational savings ≈ O(10 5 ) – O(10 7 )) makes the proposed DL-based inverse model attractive for subsurface imaging and real-time ERT monitoring applications due to fast and yet reasonably accurate estimations of permeability field.

58 GEOSCIENCES↗
DOE OSTI2020

Multi-Model and Multi-Scale Global Sensitivity Analysis for Identifying Controlling Processes of Complex Systems

An environmental model consists of multiple process level sub-models, and each sub-model represents a process that is key to the operation of the simulated system. Global sensitivity analysis methods have been widely used to identify important processes for system model development and improvement. The existing methods of global sensitivity analysis only consider parametric uncertainty, and are not capable of handling model uncertainty caused by multiple process models that arise from competing hypotheses about one or more processes. To address this problem, this project develops a new method to probe model output sensitivity to competing process models by integrating model averaging methods with variance-based global sensitivity analysis to address uncertainty in process models and parameters. The new method yields three process sensitivity indices. The first one is called first-order process sensitivity index, and it is derived as a single summary measure of relative process importance. Evaluating the index is computationally expensive, because it relies in a Monte Carlo scheme that requires thousands and even millions of model executions. To reduce computational cost, this project develops a computationally efficient, quasi Monte Carlo method, and this method is presented in Chapter 2 of this report with and a numerical example for demonstration. The numerical example shows that the results of the quasi Monte Carlo method are substantially close to those of the full Monte Carlo method, but the computational cost of the quasi Monte Carlo method is only 0.7% of that of the full Monte Carlo method. The second index is called total-effect process sensitivity index, and it measures interactions between different processes. Therefore, this sensitivity index includes the first-order process sensitivity index, and can be used to identify influential processes. On the other hand, the total-effect process sensitivity index can also be used to screen non-influential processes. This is demonstrated by two numerical examples using the Sobol-G* functions and groundwater flow models that consider recharge process, geological process, and snowmelt process. The numerical examples shows that the total-effect process sensitivity index is more informative than the first-order process sensitivity. The derivation of the process sensitivity index and the numerical examples are discussed in Chapter 3. Chapter 4 presents two computationally efficient methods for screening non-influential processes to exclude them from further investigation. The two methods are the multi-model difference-based sensitivity (MMDS) analysis method, which can be implemented using the Latin Hypercube Sampling. The second one is the implementation of MMDS method using a binning method. The numerical example for the Sobol-G* function indicates the two methods are capable of identifying non-influential models, and the numerical examples for the groundwater flow and reactive transport show that the two methods are effective for groundwater problems. However, it should be noted that the two methods are numerical approximations, and they can only be used for screening non-influential processes, not for ranking importance of system processes. All the sensitivity analysis methods are implemented by developing python codes, and the codes are in a software called SAMMPY: a python package for process sensitivity analysis under multiple models. The SAMMPY design and structure are discussed in Chapter 5, and the package is released to the public for free download.

54 ENVIRONMENTAL SCIENCES↗