Search NASA⌕ Search

SEARCH · Search NASA

Results for “prediction algorithms”

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 865 records · Page 48

Real-time monitoring and prediction of water quality parameters and algae concentrations using microbial potentiometric sensor signals and machine learning tools

We report the overarching hypothesis of this study was that temporal microbial potentiometric sensor (MPS) signal patterns could be used to predict changes in commonly monitored water quality parameters by using artificial intelligence/machine learning tools. To test this hypothesis, the study first examines a proof of concept by correlating between MPS's signals and high algae concentrations in an algal cultivation pond. Then, the study expanded upon these findings and examined if multiple water quality parameters could be predicted in real surface waters, like irrigation canals. Signals generated between the MPS sensors and other water quality sensors maintained by an Arizona utility company, including algae and chlorophyll, were collected in real time at time intervals of 30 min over a period of 9 months. Data from the MPS system and data collected by the utility company were used to train the ML/AI algorithms and compare the predicted with actual water quality parameters and algae concentrations. Based on the composite signal obtained from the MPS, the ML/AI was used to predict the canal surface water's turbidity, conductivity, chlorophyll, and blue-green algae (BGA), dissolved oxygen (DO), and pH, and predicted values were compared to the measured values. Initial testing in the algal cultivation pond revealed a strong linear correlation (R 2 = 0.87) between mixed liquor suspended solids (MLSS) and the MPSs' composite signals. The Normalized Root Mean Square Error (NRMSE) between the predicted values and measured values were <6.5%, except for the DO, which was 10.45%. The results demonstrate the usefulness of MPSs to predict key surface water quality parameters through a single composite signal, when the ML/AI tools are used conjunctively to disaggregate these signal components. The maintenance-free MPS offers a novel and cost-effective approach to monitor numerous water quality parameters at once with relatively high accuracy.

54 ENVIRONMENTAL SCIENCES↗

Preliminary Enhanced LAW Glass Formulation Algorithm

This report summarizes the Preliminary Enhanced LAW Glass Formulation Algorithm, its background information, and the calculations it performs. The Preliminary Enhanced LAW Glass Formulation Algorithm is a tool that has been developed in MATLAB to formulate glass at a given a waste composition, while attempting to maximize waste loading. It is intended for use at the Hanford Waste Treatment and Immobilization Plant, where nuclear waste will be vitrified into glass for safe, long-term storage. The formulated glass is required to meet several processing and product quality constraints. In addition, calculations must account for associated uncertainties in constraint prediction and measurement. The algorithm also allows for the ability to hold glass compositions closer to tested regions in order to increase confidence in the output. The algorithm must also follow nuclear quality assurance procedures and submit to rigorous validation and verification testing.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Machine learning study of magnetism in uranium-based compounds

Actinide and lanthanide-based materials display exotic properties that originate from the presence of itinerant or localized f electrons and include unconventional superconductivity and magnetism, hidden order, and heavy-fermion behavior. Due to the strongly correlated nature of the 5f electrons, magnetic properties of these compounds depend sensitively on applied magnetic field and pressure, as well as on chemical doping. However, precise connection between the structure and magnetism in actinide-based materials is currently unclear. In this investigation, we established such structure-property links by assembling and mining two datasets that aggregate, respectively, the results of high-throughput density functional theory simulations and experimental measurements for the families of uranium- and neptunium-based binary compounds. Various regression algorithms were utilized to identify correlations among accessible attributes (features or descriptors) of the material systems and predict their cation magnetic moments and general forms of magnetic ordering. Descriptors representing compound structural parameters and cation f-subshell occupation numbers were identified as most important for accurate predictions. The best machine learning model developed employs the random forest regression algorithm. It can predict both spin and orbit moment size with root-mean-square error of 0.17 μ B and 0.19 μ B , respectively. Lastly, the random forest classification algorithm is used to predict the ordering (paramagnetic, ferromagnetic, and antiferromagnetic) of such systems with 76% accuracy.

36 MATERIALS SCIENCE↗

Emerging materials intelligence ecosystems propelled by machine learning

We report that the age of cognitive computing and artificial intelligence (AI) is just dawning. Inspired by its successes and promises, several AI ecosystems are blossoming, many of them within the domain of materials science and engineering. These materials intelligence ecosystems are being shaped by several independent developments. Machine learning (ML) algorithms and extant materials data are utilized to create surrogate models of materials properties and performance predictions. Materials data repositories, which fuel such surrogate model development, are mushrooming. Automated data and knowledge capture from the literature (to populate data repositories) using natural language processing approaches is being explored. The design of materials that meet target property requirements and of synthesis steps to create target materials appear to be within reach, either by closed-loop active-learning strategies or by inverting the prediction pipeline using advanced generative algorithms. AI and ML concepts are also transforming the computational and physical laboratory infrastructural landscapes used to create materials data in the first place. Surrogate models that can outstrip physics-based simulations (on which they are trained) by several orders of magnitude in speed while preserving accuracy are being actively developed. Automation, autonomy and guided high-throughput techniques are imparting enormous efficiencies and eliminating redundancies in materials synthesis and characterization. The integration of the various parts of the burgeoning ML landscape may lead to materials-savvy digital assistants and to a human-machine partnership that could enable dramatic efficiencies, accelerated discoveries and increased productivity. Here, we review these emergent materials intelligence ecosystems and discuss the imminent challenges and opportunities. The materials research landscape is being transformed by the infusion of approaches based on machine learning. This Review discusses the emerging materials intelligence ecosystems and the potential of human-machine partnerships for fast and efficient virtual materials screening, development and discovery.

36 MATERIALS SCIENCE↗

A machine learning approach to photometric metallicities of giant stars

ABSTRACT Despite the advances provided by large-scale photometric surveys, stellar features – such as metallicity – generally remain limited to spectroscopic observations often of bright, nearby low-extinction stars. To rectify this, we present a neural network approach for estimating the metallicities and distances of red giant stars with 8-band photometry and parallaxes from Gaia EDR3 and the 2MASS and WISE surveys. The algorithm accounts for uncertainties in the predictions arising from the range of possible outputs at each input and from the range of models compatible with the training set (through drop-out). A two-stage procedure is adopted where an initial network to estimate photoastrometric parallaxes is trained using a large sample of noisy parallax data from Gaia EDR3 and then a secondary network is trained using spectroscopic metallicities from the APOGEE and LAMOST surveys and an augmented feature space utilizing the first-stage parallax estimates. The algorithm produces metallicity predictions with an average uncertainty of $\pm 0.19\, \mathrm{dex}$. The methodology is applied to stars within the Galactic bar/bulge with particular focus on a sample of 1.69 million objects with Gaia radial velocities. We demonstrate the use and validity of our approach by inspecting both spatial and kinematic gradients with metallicity in the Galactic bar/bulge recovering previous results on the vertical metallicity gradient (−0.528 ± 0.002 dex kpc−1) and the vertex deviation of the bar ($-21.29\pm 2.74\, \mathrm{deg}$).

79 ASTRONOMY AND ASTROPHYSICS↗

Predicting Air Traffic Management Initiatives Using Supervised Learning

Terminal Traffic Management Initiatives (TMIs) such as Ground Stops (GS) and Ground Delay Programs (GDP) are implemented to manage excess demand or lowered capacity at an airport. Air Traffic Flow Management (TFM) specialists identify situations such as aviation constraints, current and forecasted weather conditions, airport demand and capacity, and initiate TMIs for safe and orderly movement of air traffic. In this paper, we outline supervised learning techniques that can be used to predict and recommend TMIs at an airport based on current weather and airport conditions. Our research involves building classic Machine Learning (ML) models such as Logistic Regression, K-Nearest Neighbor, Random Forest and XGBoost, as well as Long short-term memory (LSTM) networks. We trained the models on 3-year historical data (weather, airport demand, capacity and TMIs) from Newark (EWR) airport which was selected based on its higher TMI implementation rates and varied weather conditions. Although Random Forest and XGBoost algorithms are able to predict if a TMI is needed or not, they have difficulty in predicting specific program type. For this purpose, we found that LSTM time-series forecasting models performed better as they also learn from past TMI program type sequences. This study also lays down the foundation for advanced modeling techniques and architectures to predict TMIs in advance for future periods. The ability to predict TMIs in advance will be highly beneficial to the traffic controllers and managers as this will help them to prepare for and manage TMIs more efficiently.

Manoj Agrawal↗

Uncertainty-Guided Prediction Horizon of Phase-Resolved Ocean Wave Forecasting Under Data Sparsity: Experimental and Numerical Evaluation

Accurate short-term wave forecasting is critical for the safe and efficient operation of marine structures that rely on real-time, phase-resolved ocean wave information for control and monitoring purposes (e.g., digital twins). These systems often depend on environmental sensors (e.g., waverider buoys, wave-sensing LIDAR). Challenges arise when upstream sensor data are missing, sparse, or phase-shifted due to drift. This study investigates the performance of two machine learning models, time-series dense encoder (TiDE) and long short-term memory (LSTM), for forecasting phase-resolved ocean surface elevations under varying degrees of data degradation. We introduce the τ-trimming algorithm, which adapts the prediction horizon based on uncertainty thresholds derived from historical forecasts. Numerical wave tank (NWT) and wave basin experiments are used to benchmark model performance under short- and long-term data masking, spatially coarse sensor grids, and upstream phase shifts. Results show under a 50% probability of upstream data loss, the τ-trimmed TiDE model achieves a 46% reduction in error at the most upstream target, compared to 22% for LSTM. Furthermore, phase misalignment in upstream data introduces a near-linear increase in forecast error. Under moderate model settings, a ±3 s misalignment increases the mean absolute error by approximately 0.5 m, while the same error is accumulated at ±4 s using the more conservative approach. These findings inform the design of resilient, uncertainty-aware wave forecasting systems suited for realistic offshore sensing environments.

42 ENGINEERING↗

Evaluation, Analysis, and Application of Internal Strain-Gage Balance Data

Experimental processes, analytical methods, and numerical algorithms are described that may be used to predict the forces and moments of an internal strain–gage balance during a wind tunnel test. First, the control volume model of a strain–gage balance and the concepts of load state, load space, and output space are introduced. These important abstractions provide a better understanding of fundamental characteristics of different balance load prediction approaches. Then, the description of strain–gage balance data and the definition of the primary bridge sensitivity are discussed. Afterwards, basic elements of the calibration of a typical six–component balance are reviewed. Two fundamentally different balance load prediction methods, the processing of check loads, and related topics are also discussed. Three real–world balance data examples are reviewed in great detail to illustrate typical analysis results for a variety of strain–gage balance designs. Finally, important observations are summarized and recommendations are provided. – Additional information and detailed mathematical derivations can be found in the appendices of the document. They include the following topics: balance terminology, definitions of important statistical metrics, balance axis system conventions, balance load transformations, the combined load diagram, electrical output format options, bi–directional output characteristics, determination of the natural zeros, derivation of two balance load prediction methods, description of two tare load iteration algorithms, modeling of balance temperature effects, basics of three–component moment balances, definition of the percent contribution, detection of linear and near–linear dependencies in balance calibration data, a regression model search algorithm, balance interactions, and other related information.

strain-gage balance↗

MODIS Retrieval of Dust Aerosol

The MODerate resolution Imaging Spectroradiometer (MODIS) currently aboard both the Terra and Aqua satellites produces a suite of products designed to characterize global aerosol distribution, optical thickness and particle size. Never before has a space-borne instrument been able to provide such detailed information, operationally, on a nearly global basis every day. The three years of Terra-MODIS data have been validated by comparing with co-located AERONET observations of aerosol optical thickness and derivations of aerosol size parameters. Some 8000 comparison points located at 133 AERONET sites around the globe show that the MODIS aerosol optical thickness retrievals are accurate to within the pre-launch expectations. However, the validation in regions dominated by desert dust is less accurate than in regions dominated by fine mode aerosol or background marine sea salt. The discrepancy is most apparent in retrievals of aerosol size parameters over ocean. In dust situations, the MODIS algorithm tends to under predict particle size because the reflectances at top of atmosphere measured by MODIS exhibit the stronger spectral signature expected by smaller particles. This pattern is consistent with the angular and spectral signature of non-spherical particles. All possible aerosol models in the MODIS Look-Up Tables were constructed from Mie theory, assuming a spherical shape. Using a combination of MODIS and AERONET observations, in regimes dominated by desert dust, we construct phase functions, empirically, with no assumption of particle shape. These new phase functions are introduced into the MODIS algorithm, in lieu of the original options for large dust-like particles. The results will be analyzed and examined.

Remer, Lorraine A.↗

Influence of local capillary trapping on containment system effectiveness

Immobilization of CO 2 injected into deep subsurface storage reservoirs is a critical component of risk assessment for geologic CO 2 storage (GCS). Local capillary trapping (LCT) is a recently established mode of immobilization that arises when CO 2 migrates due to buoyancy through heterogeneous storage reservoirs. This project sought to assess the amount and extent of LCT expected in storage formations under a range of injection conditions, and to confirm the persistence of LCT if the seal overlying the reservoir were to lose its integrity. Numerical simulation using commercial reservoir simulation software was conducted to assess the influence of injection. Laboratory experiments, modeling and numerical simulation were conducted to assess the effect of compromised seal integrity. Bench-scale (0.6 m by 0.6 m by 0.03 m) experiments with surrogate fluids provided the first empirical confirmation of the key concepts underlying LCT: accumulation of buoyant nonwetting phase at above residual saturations beneath capillary barriers in a variety of structures, which remains immobile under normal capillary pressure gradients. Immobilization of above-residual saturations is a critical distinction between LCT and the more familiar “residual saturation trapping.” To estimate the possible extent of LCT in a storage reservoir an algorithm was developed to identify all potential local traps, given the spatial distribution of capillary entry pressure in the reservoir. The algorithm assumes that the driving force for CO 2 migration can be represented as a single value of “critical capillary entry pressure” P c,entry crit , such that cells with capillary entry pressure greater/less than P c,entry crit act as barriers/potential traps during CO 2 migration. At intermediate values of P c,entry crit , the barrier regions become more laterally extensive in the reservoir, approaching a percolation threshold while non-barrier regions remain numerous. The maximum possible extent of LCT thus occurs at P c,entry crit near this threshold. Testing predictions of this simple algorithm against full-physics simulations of buoyancy-driven CO 2 migration support the concept of critical capillary entry pressure. However, further research is needed to determine whether a single value of critical capillary entry pressure always applies and how that value can be determined a priori. Simulations of injection into high-resolution (cells 0.3 m on a side) 2D and 3D heterogeneous domains show two characteristic behaviors. At small gravity numbers (vertical flow velocity much less than horizontal flow velocity) the CO 2 fills local traps as well as regions that would act as local barriers if CO 2 were moving only due to buoyancy. When injection ceases, the CO 2 migrates vertically to establish large saturations within local traps and residual saturation elsewhere. At large gravity numbers, the CO 2 invades a smaller portion of the perforated interval. Within this smaller swept zone the local barriers are not invaded, but local traps are filled to large saturation during injection and remain during post-injection gravity-driven migration. The small gravity number behavior is expected in the region within 100 m of a vertical injection well at anticipated rates of injection for commercial GCS. Simulations of leakage scenarios (through-going region of large permeability imposed in overlying seal) indicate that LCT persists (i.e. CO 2 remains held in a large fraction of the local iv traps) and the persistence is independent of injection rate during storage. Simulations of leakage for the limiting case of CO 2 migrating vertically from an areally extensive emplacement in the lower portion of a reservoir showed similar strong persistence of LCT. This research has two broad implications for GCS. The first is that LCT can retain a significant fraction of the CO 2 stored in a reservoir – above and beyond the residual saturation -- if the overlying seal were to fail. Thus frameworks for risk assessment should be extended to account for LCT. The second implication is that compared to pressure driven flow in reservoirs, CO 2 migration and trapping behave in a qualitatively different manner in heterogeneous reservoirs when buoyancy is the dominant driving force for flow. Thus simulations of GCS that neglect capillary heterogeneity will fail to capture important features of the CO 2 plume. While commercial reservoir simulation software can account for fine scale capillary heterogeneity, it has not been designed to work efficiently with such domains, and no simulators can handle fine-scale resolution throughout the reservoir. A possible way to upscale the migration and trapping is to apply an “effective residual saturation” to coarse-scale grids. While the extent of overall immobilization can be correlated in this way, all coarser grids failed to capture the distance traveled by the migrating CO 2 for large gravity number. Thus it remains unclear how best to account for LCT in the routine simulation work-flow that will be needed for large-scale GCS. Alternatives meriting investigation include streamline methods, reduced-physics proxies (e.g. particle tracking), and biased invasion percolation algorithms, which are based on precisely the capillary heterogeneity essential for LCT.

58 GEOSCIENCES↗

MWm-array sensors for in situ monitoring of high-temperature components in power plants

Utilization of America's substantial coal reserves for energy production has become a national priority. Advanced coal-fired power plants offer an environmentally friendly means to achieve that goal. These power plants, such as ultrasupercritical power plants, will provide high thermal efficiency along with greatly reduced emissions of CO and other pollutants. Life cycle costs for the advanced coal-fired plants can be reduced by enhanced observability in support of condition-based maintenance. The enhanced observability can be achieved by using networks of condition-monitoring sensors that would provide component-level material condition information and through-wall temperature monitoring. This would reduce uncertainties in knowledge of material condition, at the level of individual components, and improve capability to predict remaining life of critical components. One approach being developed under the U.S. Department of Energy Small Business Innovation Research Program is to develop and implement high-temperature versions of the meandering winding magnetometer (HT-MWM) for temperatures up to 1000{sup o}C. These patented sensors, coupled with multivariate inverse methods, would provide superior performance for in situ material condition monitoring (material degradation, flaw detection, stress relaxation, and/or creep monitoring) and through-wall temperature measurement. Networks of HT-MWMs will generate material condition information to be used by adaptive life-management algorithms for remaining life prediction and decision support.

20 FOSSIL-FUELED POWER PLANTS↗

Real-Time Prediction of Power System Frequency in FNET: A State Space Approach

This paper proposes a novel approach to predict power frequency by applying a state-space model to describe the time-varying nature of power systems. It introduces the Expectation maximization (EM) and prediction error minimization (PEM) algorithms to dynamically estimate the parameters of the model. This paper discusses how the proposed models can be used to ensure the efficiency and reliability of power systems in Frequency Monitoring Network (FNET), if serious frequency fluctuation or measurement failure occur at some nodes; this is achieved without requiring the exact model of complex power systems. Our approach leads to an easy online implementation with high precision and short response time that are key to effective frequency control. We randomly pick a set of frequency data for one power station in FNET and use it to estimate and predict the power frequency based on past measurements. Several computer simulations are provided to evaluate the method. Simulation results showed that the proposed technique could achieve good performance regarding the frequency monitoring with very limited measurement input information.

Djouadi, Seddik M↗

Evaluation, Analysis, and Application of Internal Strain-Gage Balance Data

Experimental processes, analytical methods, and numerical algorithms are described that may be used to predict the forces and moments of an internal strain-gage balance during a wind tunnel test. First, the control volume model of a strain-gage balance and the concepts of load state, load space, and output space are introduced. These important abstractions provide a better understanding of fundamental characteristics of different balance load prediction approaches. Then, the description of strain-gage balance data and the definition of the primary gage sensitivity is discussed. Afterwards, basic elements of the calibration of a typical six-component balance are reviewed. Two fundamentally different balance load prediction methods, the processing of check loads, and related topics are also discussed. Three real-world balance data examples are reviewed in great detail to illustrate typical analysis results for a variety of strain-gage balance designs. Finally, important observations are summarized and recommendations are provided. Additional information and detailed mathematical derivations can be found in the appendices of the document. They include the following topics: balance terminology, definitions of important statistical metrics, balance axis system conventions, balance load transformations, the combined load diagram, electrical output format options, bi-directional gage output characteristics, determination of the natural zeros, derivation of two balance load prediction methods, description of two tare load iteration algorithms, modeling of balance temperature effects, basics of three-component moment balances, definition of the percent contribution, detection of linear and near-linear dependencies in balance calibration data, a regression model term selection algorithm, and other related topics.

wind tunnel test↗

Science in 60 – Tabletop Fire Prediction

At the Interagency Fire Center at Los Alamos National Laboratory, experts rely on the state-of-the-art Simtable to help them predict the unpredictable. With algorithms that include camera-based object-tracking and projection developed by the Laboratory, the emergency operations team can simulate a wildland fire spreading across any terrain. The table system, developed and marketed by Simtable, a Santa Fe, N.M., company, takes into account weather, vegetation and fuel conditions. The project also received funding from the LANS Venture Acceleration Fund to improve the user interface.

54 ENVIRONMENTAL SCIENCES↗

Solid phase stability of molybdenum under compression: Sound velocity measurements and first-principles calculations

The high-pressure solid phase stability of molybdenum (Mo) has been the center of a long-standing controversy on its high-pressure melting. In this work, experimental and theoretical researches have been conducted to check its solid phase stability under compression. First, we performed sound velocity measurements from 38 to 160 GPa using the two-stage light gas gun and explosive loading in backward- and forward-impact geometries, along with the high-precision velocity interferometry. From the sound velocities, we found no solid-solid phase transition in Mo before shock melting, which does not support the previous solid-solid phase transition conclusion inferred from the sharp drops of the longitudinal sound velocity [Hixson et al., Phys. Rev. Lett. 62, 637 (1989)]. Then, we searched its structures globally using the multi-algorithm collaborative crystal structure prediction technique combined with the density functional theory. By comparing the enthalpies of body centered cubic structure with those of the metastable structures, we found that bcc is the most stable structure in the range of 0–300 GPa. The present theoretical results together with previous ones greatly support our experimental conclusions.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

S-OPT: A Points Selection Algorithm for Hyper-Reduction in Reduced Order Models

While projection-based reduced order models can reduce the dimension of full order solutions, the resulting reduced models may still contain terms that scale with the full order dimension. Hyper-reduction techniques are sampling-based methods that further reduce this computational complexity by approximating such terms with a much smaller dimension. The goal of this work is to introduce the points selection algorithm developed by Shin and Xiu as a hyper-reduction method. The selection algorithm was originally proposed as a stochastic collocation method for uncertainty quantification. Since the algorithm aims at maximizing a quantity $\mathcal{S}$ that measures both the column orthogonality and the determinant, we refer to the algorithm as S-OPT. Numerical examples are provided to demonstrate the performance of S-OPT and to compare its performance with a gappy proper orthogonal decomposition (POD) algorithm. Here, we found that using the S-OPT algorithm is shown to predict the full order solutions with higher accuracy than gappy POD especially when the number of sampling points is small, although we note that S-OPT shows slow asymptotic convergence with respect to the number of samples for some applications, e.g., Lagrangian hydrodynamics.

97 MATHEMATICS AND COMPUTING↗

TH-E-BRE-02: A Forward Scattering Approximation to Dose Calculation Using the Linear Boltzmann Transport Equation

Purpose: To investigate the use of the linear Boltzmann transport equation as a dose calculation tool which can account for interface effects, while still having faster computation times than Monte Carlo methods. In particular, we introduce a forward scattering approximation, in hopes of improving calculation time without a significant hindrance to accuracy. Methods: Two coupled Boltzmann transport equations were constructed, one representing the fluence of photons within the medium, and the other, the fluence of electrons. We neglect the scattering term within the electron transport equation, resulting in an extreme forward scattering approximation to reduce computational complexity. These equations were then solved using a numerical technique for solving partial differential equations, known as a finite difference scheme, where the fluence at each discrete point in space is calculated based on the fluence at the previous point in the particle's path. Using this scheme, it is possible to develop a solution to the Boltzmann transport equations by beginning with boundary conditions and iterating across the entire medium. The fluence of electrons can then be used to find the dose at any point within the medium. Results: Comparisons with Monte Carlo simulations indicate that even simplistic techniques for solving the linear Boltzmann transport equation yield expected interface effects, which many popular dose calculation algorithms are not capable of predicting. Implementation of a forward scattering approximation does not appear to drastically reduce the accuracy of this algorithm. Conclusion: Optimized implementations of this algorithm have been shown to be very accurate when compared with Monte Carlo simulations, even in build up regions where many models fail. Use of a forward scattering approximation could potentially give a reasonably accurate dose distribution in a shorter amount of time for situations where a completely accurate dose distribution is not required, such as in certain optimization algorithms.

97 MATHEMATICS AND COMPUTING↗

ML-GA

ML-GA is a software that can be used to expedite design optimization process combining machine learning and genetic algorithm approaches. It employs a machine learning (ML) model (any ML algorithm can be incorporated) to predict the quality (merit) of a design from the input parameters. Then, a stochastic global optimization genetic algorithm (GA) is used, with the machine learning model as the objective function, to optimize the input parameters based on the merit function. ML-GA is scalable to higher computational platforms like supercomputers and clusters enabling optimization to be performed in significantly short time frames (e.g., in a day).

PAL, PINAKI↗