Search NASA⌕ Search

SEARCH · Search NASA

Results for “process control and optimization”

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 991 records · Page 55

beluga

beluga is an open-source optimal control software that uses publicly available, indirect methods from GitHub. beluga's indirect optimization engine has been reworked to follow a lazy functional paradigm. This improves the software's flexibility and enables users to create their own optimization processes without modification of the source code. beluga is general-purpose and therefore very broad in scope and application. SAND2020-12847 M Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Sparapany, Michael↗

Package Equivalent Reactor Networks as Reduced Order Models for Use with CAPE-OPEN Compliant Simulation

Engineering simulations of coal gasifiers are typically performed using computational fluid dynamics (CFD) software, where a 3-D representation of the gasifier equipment is used to model the fluid flow in the gasifier and source terms from the coal gasification process are captured using discrete-phase model source terms. Simulations using this approach can be very time consuming, making it difficult to imbed such models into overall system simulations for plant design and optimization. For such system-level designs, process flowsheet software is typically used, such as Aspen Plus®, where each component where each component is modeled using a reduced-order model. For advanced power-generation systems, such as integrated gasifier/gas-turbine combined-cycle systems (IGCC), the critical components determining overall process efficiency and emissions are usually the gasifier and combustor. Providing more accurate and more computationally efficient reduced-order models for these components, then, enables much more effective plant-level design optimization and design for control.

01 COAL, LIGNITE, AND PEAT↗

Biofilm mitigation in hybrid chemical-biological upcycling of waste polymers

Accumulation of plastic waste in the environment is a serious global issue. To deal with this, there is a need for improved and more efficient methods for plastic waste recycling. One approach is to depolymerize plastic using pyrolysis or chemical deconstruction followed by microbial-upcycling of the monomers into more valuable products. Microbial consortia may be able to increase stability in response to process perturbations and adapt to diverse carbon sources, but may be more likely to form biofilms that foul process equipment, increasing the challenge of harvesting the cell biomass. To better understand the relationship between bioprocess conditions, biofilm formation, and ecology within the bioreactor, in this study a previously-enriched microbial consortium (LS1_Calumet) was grown on (1) ammonium hydroxide-depolymerized polyethylene terephthalate (PET) monomers and (2) the pyrolysis products of polyethylene (PE) and polypropylene (PP). Bioreactor temperature, pH, agitation speed, and aeration were varied to determine the conditions that led to the highest production of planktonic biomass and minimal formation of biofilm. The community makeup and diversity in the planktonic and biofilm states were evaluated using 16S rRNA gene amplicon sequencing. Results showed that there was very little microbial growth on the liquid product from pyrolysis under all fermentation conditions. When grown on the chemically-deconstructed PET the highest cell density (0.69 g/L) with minimal biofilm formation was produced at 30°C, pH 7, 100 rpm agitation, and 10 sL/hr airflow. Results from 16S rRNAsequencing showed that the planktonic phase had higher observed diversity than the biofilm, and that Rhodococcus, Paracoccus, and Chelatococcus were the most abundant genera for all process conditions. Biofilm formation by Rhodococcus sp. And Paracoccus sp. Isolates was typically lower than the full microbial community and varied based on the carbon source. Ultimately, the results indicate that biofilm formation within the bioreactor can be significantly reduced by optimizing process conditions and using pure cultures or a less diverse community, while maintaining high biomass productivity. The results of this study provide insight into methods for upcycling plastic waste and how process conditions can be used to control the formation of biofilm in bioreactors.

36 MATERIALS SCIENCE↗

Laser Powder Directed Energy Deposition of Steels for Nuclear Applications

This comprehensive investigation examines the structure–property relationships in two nuclear alloy systems—Alloy 709 (A709) austenitic stainless steel and Grade 92 (G-92) ferritic/martensitic (F/M) steel—manufactured via directed energy deposition (DED) for sodium-cooled fast reactor applications. This study establishes the fundamental mechanisms for controlling microstructures for optimizing the mechanical performance of additively manufactured nuclear materials through systematic heat treatment optimization and multiscale characterization. As-deposited A709 steel develops a complex multiscale strengthening architecture consisting of a fine cellular solidification structure with diameter of 2-3 µm within10–50 µm grains, elevated dislocation densities from rapid thermal cycling, and grain boundary precipitates that activate concurrent Hall–Petch, dislocation, and precipitation hardening mechanisms to achieve exceptional properties [yield strength (YS): 603 MPa, ultimate tensile strength (UTS): 844 MPa, Vickers hardness: 220 HV] that achieve a 44% superior strength compared to that of the wrought material. Heat treatments produce different results. Solution annealing (SA) dissolves the cellular structure and reduces the hardness to 190 HV. Precipitation treatment (PT) keeps the cellular structure but adds carbides, allowing the hardness to reach 205 HV. The best approach combines both treatments (SA+PT) and creates uniform precipitate distributions with M 23 C 6 carbides at the grain boundaries and MX carbonitrides in the matrix, achieving a hardness of 195 HV. However, directional differences persist, with a 12%–15% strength variation between orientations due to the inherited layered microstructural architecture that survives aggressive heat treatment. While tensile testing at 550°C demonstrates 40%–50% thermal softening with dynamic strain aging, DED A709 steel still maintains a 71% higher YS than that of the wrought material. Ion irradiation studies (100–400 dpa) of DED A709 steel reveal progressive radiation damage with increasing void density and radiation-induced segregation causing nickel enrichment and chromium depletion, which will ultimately compromise mechanical properties. As-deposited G-92 exhibits exceptional strength (UTS: 1650–1700 MPa, 430 HV) through a complex microstructure containing both ferrite and martensite phases, a high geometrically necessary dislocation (GND) density (17.04×10 14 /m 2 ), and fine carbides. Heat treatments create distinct changes. Normalizing produces fresh martensite with the highest hardness (460 HV) and an increased GND density (20.23×10 14 /m 2 ). Tempering develops dual precipitation systems and reduces the hardness to 290 HV. The optimal approach uses sequential normalizing plus tempering, achieving balanced properties with the lowest hardness (250 HV) and a reduced GND density (11.01×10 14 /m 2 ). A processing-dependent anisotropy is observed: horizontal specimens achieve superior ductile behavior, while vertical specimens exhibit brittle failure. A tempering heat treatment successfully mitigates this anisotropic behavior by transforming the hard martensitic as-deposited structure into tempered martensite enabling both horizontal and vertical specimens to exhibit similar stress–strain characteristics with visible necking behavior. Remarkably, testing at 550°C reveals a reversal in the anisotropy, where as-deposited specimens achieve near isotropy with superior thermal stability (a 15%–20% strength reduction), while tempered specimens develop an orientation dependence with a 25%–30% strength reduction. Both alloy systems demonstrate that DED processing creates specimens with a superior strength through refined microstructural features, though with distinct strengthening mechanisms—austenitic through cellular structures and precipitates versus F/M through phase transformations and precipitates. Heat treatment optimization requires alloy-specific approaches, with A709 benefiting from controlled precipitation while G-92 requires careful phase transformation control. The results show that DED manufacturing can produce nuclear materials with exceptional performance, but directional effects and temperature-dependent behavior must be carefully considered for reactor component design and qualification.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Building thermal model development of typical house in U.S. for virtual storage control of aggregated building loads based on limited available information

A building thermal model is an essential component for achieving optimal control of a building’s heating, ventilation, and air conditioning (HVAC). A self-learning grey-box model has been developed by implementing data-driven techniques and utilizing limited available information about the building. In previous research, detailed foreknown knowledge/information, e.g. physical and thermal properties of building materials, as well as sufficient number of observation points, e.g. indoor temperature sensors in different zones, are available for making the decision of model structure and parameters searching range easier. The availability of measured data and a narrowed searching range of model parameters, e.g. resistance and capacity for each wall, made the adopted algorithms quickly achieve near optimal values, which closely approximate the actual heat transfer of a building. Meanwhile, the details of pre-processing of the key inputs were seldom explained. Compared to the previous research, in this study, the available information is assumed to be limited which can fill the requirement for large scale virtual storage control of real residential community in near future, where it may be impractically or over time-consuming to acquire the detailed information for each single house. The developed model is validated by comparing the results to measured data in a recently-built home typical of the southeastern United States.

Cui, Borui↗

In-situ Condition Monitoring of Components in Small Modular Reactors Using Process and Electrical Signature Analysis. Final report, volume 1. Development of experimental flow control loop, data analysis and plant monitoring

The research and development under this project was focused on the following three major objectives: Objective 1: Identification of critical in-vessel SMR components for remote monitoring and development of their low-order dynamic models, along with a simulation model of an integral pressurized water reactor (iPWR). Objective 2: Development of an experimental flow control loop with motor-driven valves and pumps, incorporating data acquisition and on-line monitoring interface. Objective 3: Development of stationary and transient signal processing methods for electrical signatures, machinery vibration, and for characterizing process variables for equipment monitoring. This objective includes the development of a data analysis toolbox. The following is a summary of the technical accomplishments under this project: - A detailed literature review of various SMR types and electrical signature analysis of motor-driven systems was completed. A bibliography of literature is provided at the end of this report. Assistance was provided by ORNL in identifying some key references. - A review of literature on pump-motor modeling and digital signal processing methods was performed. - An existing flow control loop was upgraded with new instrumentation, data acquisition hardware and software. The upgrading of the experimental loop included the installation of a new submersible pump driven by a three-phase induction motor. All the sensors were calibrated before full-scale experimental runs were performed. - MATLAB-Simulink model of a three-phase induction motor and pump system was completed. The model was used to simulate normal operation and fault conditions in the motor-pump system, and to identify changes in the electrical signatures. - A simulation model of an integral PWR (iPWR) was updated and the MATLAB-Simulink model was validated for known transients. The pump-motor model was interfaced with the iPWR model for testing the impact of primary flow perturbations (upsets) on plant parameters and the pump electrical signatures. Additionally, the reactor simulation is being used to generate normal operation data and data with instrumentation faults and process anomalies. A frequency controller was interfaced with the motor power supply in order to vary the electrical supply frequency. The experimental flow control loop was used to generate operational data under varying motor performance characteristics. Coolant leakage events were simulated by varying the bypass loop flow rate. The accuracy of motor power calculation was improved by incorporating the power factor, computed from motor current and voltage in each phase of the induction motor.- A variety of experimental runs were made for steady-state and transient pump operating conditions. Process, vibration, and electrical signatures were measured using a submersible pump with variable supply frequency. High correlation was seen between motor current and pump discharge pressure signal; similar high correlation was exhibited between pump motor power and flow rate. Wide-band analysis indicated high coherence (in the frequency domain) between motor current and vibration signals. - Wide-band operational data from a PWR were acquired from AMS Corporation and used to develop time-series models, and to estimate signal spectrum and sensor time constant. All the data were from different pressure transmitters in the system, including primary and secondary loops. These signals were pre-processed using the wavelet transform for filtering both low-frequency and high-frequency bands. This technique of signal pre-processing provides minimum distortion of the data, and results in a more optimal estimation of time constants of plant sensors using time-series modeling techniques.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Piston geometry and stroke optimization for high efficiency propane spark ignition engines

Propane has unique properties and offers interesting characteristics for high-efficiency spark ignition engines. Its high volatility reduces or completely eliminates fuel-wall wetting and facilitates fuel air mixing. Furthermore, propane has a research octane number of 112 and a high octane sensitivity of 15. Finally, its laminar flame speed is on the same order as that of conventional gasoline, and it exhibits high dilution tolerance. Modern spark ignition internal combustion engines rely on fast combustion rates and high dilution to achieve high brake thermal efficiencies. To accomplish this, high stroke-to-bore ratios and high geometric compression ratios have been used in new engine designs. Therefore, propane’s relatively high laminar flame speeds, high knock resistance, and dilution tolerance make it an excellent candidate fuel for modern spark ignition engines. The objective of this work is to co-optimize the piston geometry and the engine stroke to maximize the efficiency of a spark-ignition engine fueled with propane. 3D computational fluid dynamics (CFD) simulations employing the extended coherent flamelet model were used to study the parametric effects of piston shape and stroke length. A piston geometry based on high performing pistons was parameterized using four controlling parameters. The piston geometry and engine stroke design space was explored using deterministic and quasi-random sampling techniques. In conclusion, a Gaussian process regression model was built using the simulation data to explain the results observed.

33 ADVANCED PROPULSION SYSTEMS↗

Intrinsic geometry of quantum adiabatic evolution and quantum phase transitions

We elucidate the geometry of quantum adiabatic evolution. By minimizing the deviation from adiabaticity, we find a Riemannian metric tensor underlying adiabatic evolution. Equipped with this tensor, we identify a unified geometric description of quantum adiabatic evolution and quantum phase transitions that generalizes previous treatments to allow for degeneracy. The same structure is relevant for applications in quantum information processing, including adiabatic and holonomic quantum computing, where geodesics over the manifold of control parameters correspond to paths which minimize errors. We illustrate this geometric structure with examples, for which we explicitly find adiabatic geodesics. By solving the geodesic equations in the vicinity of a quantum critical point, we identify universal characteristics of optimal adiabatic passage through a quantum phase transition. In particular, we show that in the vicinity of a critical point describing a second-order quantum phase transition, the geodesic exhibits power-law scaling with an exponent given by twice the inverse of the product of the spatial and scaling dimensions.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Biomass for thermochemical conversion: targets and challenges

Bioenergy will be one component of a suite of alternatives to fossil fuels. Effective conversion of biomass to energy will require the careful pairing of advanced conversion technologies with biomass feedstocks optimized for the purpose. Lignocellulosic biomass can be converted to useful energy products via two distinct pathways: enzymatic or thermochemical conversion. The thermochemical pathways are reviewed and potential biotechnology or breeding targets to improve feedstocks for pyrolysis, gasification, and combustion are identified. Biomass traits influencing the effectiveness of the thermochemical process (cell wall composition, mineral and moisture content) differ from those important for enzymatic conversion and so properties are discussed in the language of biologists (biochemical analysis) as well as that of engineers (proximate and ultimate analysis). We discuss the genetic control, potential environmental influence, and consequences of modification of these traits. Improving feedstocks for thermochemical conversion can be accomplished by the optimization of lignin levels, and the reduction of ash and moisture content. We suggest that ultimate analysis and associated properties such as H:C, O:C, and heating value might be more amenable than traditional biochemical analysis to the high-throughput necessary for the phenotyping of large plant populations. Expanding our knowledge of these biomass traits will play a critical role in the utilization of biomass for energy production globally, and add to our understanding of how plants tailor their composition with their environment.

54 ENVIRONMENTAL SCIENCES↗

Harnessing nanoscale densification for controlling gas selectivity in flexible zeolitic imidazolate frameworks

Flexible metal-organic frameworks (MOFs) are promising for gas separation due to their molecular recognition capabilities. However, gate-opening behavior in certain flexible MOFs can reduce gas selectivity, enabling simultaneous adsorption of multiple gases. We introduce a binder-free, pressure-free densification process to enhance gas selectivity in flexible zeolitic imidazolate frameworks (ZIF-7 and ZIF-9) by optimizing nanoparticle packing. Unlike powders, the densified ZIF-7 and ZIF-9 retain rhombohedral crystal structures after solvent removal, confirmed via synchrotron powder X-ray diffraction. This process modifies adsorption behavior by introducing a diffusion barrier, enabling diffusivity-driven kinetic separation and enhanced CO2/hydrocarbon selectivity. At 0.5 bar, ZIF-7 mono shows 2×, 4×, and 9.5× higher CO 2 selectivity over C 2 H 2 , C 2 H 4 , and C 2 H 6 , respectively, versus ZIF-7 pwd . ZIF-9 mono similarly shows 6.5×, 3×, and 39× improvements over ZIF-9 pwd . Both monoliths outperform powders in ternary gas mixtures. This study provides a novel strategy to control gate opening and enhance gas selectivity in flexible MOFs for practical separation applications.

CO2/hydrocarbon separation↗

Resonances in Compton-Like scattering processes in an external magnetized medium

We consider the possible resonance effects in the tree-level two-vertex amplitudes for the transitions jf → j'f' in a constant uniform magnetic field and in the presence of a magnetized plasma consisting of charged fermions for various combinations of scalar, pseudo-scalar, vector, and axial-vector vertices. As an application of the results obtained, we have investigated the scattering of a photon by magnetized-plasma electrons, γe → γe, in the resonance region and calculated the photon absorption coefficient in this reaction. The cross section for this process has been calculated and compared with the available published results.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

The Discounted Method and Equivalence of Average Criteria for Risk-Sensitive Markov Decision Processes on Borel Spaces

This note concerns discrete-time controlled Markov chains with Borel state and action spaces. Given a nonnegative cost function, the performance of a control policy is measured by the superior limit risk-sensitive average criterion associated with a constant and positive risk sensitivity coefficient. Within such a framework, the discounted approach is used (a) to establish the existence of solutions for the corresponding optimality inequality, and (b) to show that, under mild conditions on the cost function, the optimal value functions corresponding to the superior and inferior limit average criteria coincide on a certain subset of the state space. The approach of the paper relies on standard dynamic programming ideas and on a simple analytical derivation of a Tauberian relation.

97 MATHEMATICS AND COMPUTING↗

Optimisation of spontaneous four-wave mixing in a ring microcavity

A theory of spontaneous four-wave mixing in a ring microcavity is developed. The rate of emission of biphotons for pulsed and monochromatic pumping with allowance for the dispersion of group velocities is analytically calculated. In the first case, pulses in the form of an increasing exponential are considered, which are optimal for excitation of an individual resonator mode. The behaviour of the group velocity dispersion as a function of the width and height of the waveguide is studied for a specific case of a ring microcavity made of silicon nitride. The results of the numerical calculation are in good agreement with the experimental data. The ring microcavity is made of two types of waveguides: completely etched and half etched. It is found that the latter allow for better control over the parameters in the manufacturing process, making them more predictable. (nonlinear optical phenomena)

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

GrainNN: A neighbor-aware long short-term memory network for predicting microstructure evolution during polycrystalline grain formation

High fidelity simulations of grain formation in alloys are an indispensable tool for process-to-mechanical-properties characterization. Such simulations, however, can be computationally expensive as they require fine spatial and temporal discretizations. Their cost becomes an obstacle to parametric studies and ensemble runs and ultimately makes downstream tasks like optimal control and uncertainty quantification challenging. To enable such downstream tasks, we introduce GrainNN, an efficient and accurate reduced-order model for epitaxial grain growth in additive manufacturing conditions. GrainNN is a sequence-to-sequence long-short-term-memory (LSTM) deep neural network that evolves the dynamics of manually crafted features. Its innovations are (1) an attention mechanism with grain-microstructure-specific transformer architecture; and (2) an overlapping combination of several clones of the network to generalize to grain configurations that are different from those used for training. This design enables GrainNN to predict grain formation for unseen physical parameters, grain number, domain size and geometry. Furthermore, GrainNN not only reconstructs the quantities of interest but also can be pointwise accurate. In our numerical experiments, we use a polycrystalline phase field method to both generate the training data and assess GrainNN. For multiparametric, ensemble simulations with many grains, GrainNN can be orders of magnitude faster than phase field simulations, while delivering 5%–15% pointwise error. Additionally, this speedup includes the cost of the phase field simulations for generating training data.

36 MATERIALS SCIENCE↗

Reinforcement learning building control approach harnessing imitation learning

Reinforcement learning (RL) has shown significant success in sequential decision making in fields like autonomous vehicles, robotics, marketing and gaming industries. This success has attracted the attention to the RL control approach for building energy systems which are becoming complicated due to the need to optimize for multiple, potentially conflicting, goals like occupant comfort, energy use and grid interactivity. However, for real world applications, RL has several drawbacks like requiring large training data and time, and unstable control behavior during the early exploration process making it infeasible for an application directly to building control tasks. To address these issues, an imitation learning approach is utilized herein where the RL agents starts with a policy transferred from accepted rule based policies and heuristic policies. This approach is successful in reducing the training time, preventing the unstable early exploration behavior and improving upon an accepted rule-based policy - all of these make RL a more practical control approach for real world applications in the domain of building controls.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Programming Amphiphilic Peptoid Oligomers for Hierarchical Assembly and Inorganic Crystallization

Natural organisms make a wide variety of exquisitely complex, nano-, micro-, and macroscale structured materials in an energy-efficient and highly reproducible manner. During these processes, the information-carrying biomolecules (e.g., proteins, peptides, and carbohydrates) enable (1) hierarchical organization to assemble scaffold materials and execute high-level functions and (2) exquisite control over inorganic materials synthesis, generating biominerals whose properties are optimized for their functions. Inspired by nature, significant efforts have been devoted to developing functional materials that can rival those natural molecules by mimicking in vivo functions using engineered proteins, peptides, DNAs, sequence-defined synthetic molecules (e.g., peptoids), and other biomimetic polymers. Among them, peptoids, a new type of synthetic mimetics of peptides and proteins, have received particular attention because they combine the merits of both synthetic polymers (e.g., high chemical stability and efficient synthesis) and biomolecules (e.g., sequence programmability and biocompatibility). The lack of both chirality and hydrogen bonds in their backbone results in a highly designable peptoid-based system with reduced structural complexity and side chain-chemistry-dominated properties. Here in this Account, we present our recent efforts in this field by programming amphiphilic peptoid sequences for (1) the controlled self-assembly into different hierarchically structured nanomaterials with favorable properties and (2) manipulating inorganic (nano)crystal nucleation, growth, and assembly into superstructures. First, we designed a series of amphiphilic peptoids with controlled side chain chemistries that self-assembled into 1D highly stiff and dynamic nanotubes, 2D membrane-mimetic nanosheets, hexagonally patterned nanoribbons, and 3D nanoflowers. These crystalline nanostructures exhibited sequence-dependent properties and showed promise for different applications. The corresponding peptoid self-assembly pathways and mechanisms were also investigated by leveraging in situ atomic force microscopy studies and molecular dynamics simulations, which showed precise sequence dependency. Second, inspired by peptide- and protein-controlled formation of hierarchical inorganic nanostructures in nature, we developed peptoid-based biomimetic approaches for controlled synthesis of inorganic materials (e.g., noble metals and calcite), in which we took advantage of the substantial side chain chemistry of peptoids and investigated the relationship between the peptoid sequences and the morphology and growth kinetics of inorganic materials. For example, to overcome the challenges (e.g., complexity of protein- and peptide-folding, poor thermal and chemical stabilities) facing the area of protein- and peptide-controlled synthesis of inorganic materials, we recently reported the design of sequence-defined peptoids for controlled synthesis of highly branched plasmonic gold particles. Moreover, we developed a rule of thumb for designing peptoids that predictively enabled the morphological evolution from spherical to coral-shaped gold nanoparticles (NPs). With this Account, we hope to stimulate the research interest of chemists and materials scientists and promote the predictive synthesis of functional and robust materials through the design of sequence-defined synthetic molecules.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Assessment of wafer scale MoS 2 atomic layers grown by metal–organic chemical vapor deposition using organo-metal, organo-sulfide, and H 2 S precursors

Transition Metal Dichalcogenides (TMDs) are a unique class of materials that exhibit attractive electrical and optical properties which have generated significant interest for applications in microelectronics, optoelectronics, energy storage, and sensing. Considering the potential of these materials to impact such applications, it is crucial to develop a reliable and scalable synthesis process that is compatible with modern industrial manufacturing methods. Metal–organic chemical vapor deposition (MOCVD) offers an ideal solution to produce TMDs, due to its compatibility with large-scale production, precise layer control, and high material purity. Optimization of MOCVD protocols is necessary for effective TMD synthesis and integration into mainstream technologies. Additionally, improvements in metrology are necessary to measure the quality of the fabricated samples more accurately. In this work, we study MOCVD of wafer-scale molybdenum disulfide (MoS 2 ) utilizing two common chalcogen precursors, H 2 S and DTBS. We then develop a metrology platform for wafer scale samples quality assessment. For this, the coalesced films were characterized using Raman spectroscopy, atomic force microscopy, transmission electron microscopy, X-ray photoelectron spectroscopy, and Kelvin probe force microscopy. We then correlate the structural analysis of these grown films with electrical performance by using aerosol jet printing to fabricate van der Pauw test structures and assess sheet resistance.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

AN AUTOMATED MACHINE LEARNING-GENETIC ALGORITHM FRAMEWORK WITH ACTIVE LEARNING FOR DESIGN OPTIMIZATION

The use of machine learning (ML)-based surrogate models is a promising technique to significantly accelerate simulation-driven design optimization of internal combustion (IC) engines, due to the high computational cost of running computational fluid dynamics (CFD) simulations. However, training the ML models requires hyperparameter selection, which is often done using trial-and-error and domain expertise. Another challenge is that the data required to train these models are often unknown a priori. In this work, we present an automated hyperparameter selection technique coupled with an active learning approach to address these challenges. The technique presented in this study involves the use of a Bayesian approach to optimize the hyperparameters of the base learners that make up a super learner model. In addition to performing hyperparameter optimization (HPO), an active learning approach is employed, where the process of data generation using simulations, ML training, and surrogate optimization is performed repeatedly to refine the solution in the vicinity of the predicted optimum. The proposed approach is applied to the optimization of a compression ignition engine with control parameters relating to fuel injection, in-cylinder flow, and thermodynamic conditions. It is demonstrated that by automatically selecting the best values of the hyperparameters, a 1.6% improvement in merit value is obtained, compared to an improvement of 1.0% with default hyperparameters. Overall, the framework introduced in this study reduces the need for technical expertise in training ML models for optimization while also reducing the number of simulations needed for performing surrogate-based design optimization.

Owoyele, Opeoluwa↗