Control co-design of commercial building chiller plant using Bayesian optimization
Not Available
Bhattacharya, Arnab↗
SEARCH · Engineering Papers
Results for “parameter 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.
Implicit neural representations for experimental steering of advanced experiments
Scattering measurements using electrons, neutrons, or photons are essential for obtaining microscopic insights into materials. However, limited facility availability and high-dimensional scattering data necessitate more efficient experimental steering techniques. Here, we report a machine learning method that guides scattering data collection and facilitates real-time estimation of model parameters, given a reliable forward model to simulate experimental signals. We employ implicit neural representations as efficient surrogates that link model parameters with simulated spectroscopies. This enables a Bayesian optimal experimental design framework to estimate the probability distributions of parameters from high-dimensional scattering data. We demonstrate the proposed method using inelastic neutron scattering with simulated and real experimental data, highlighting the method’s ability to provide real-time parameter estimation with quantified uncertainties and to deliver informed experimental guidance that reduces experimental time while maximizing scientific output. This approach paves the way for accelerated discoveries in condensed matter through scattering measurements.
36 MATERIALS SCIENCE↗
Machine learning for reparameterization of multi-scale closures
Scientific machine learning (ML) is becoming increasingly useful in learning closure models for multi-scale physics problems; however, many ML approaches require a vast array of training data and can struggle with generalization and interpretability. Here, rather than learning an entire closure operator, we adopt an existing reduced-dimension model of the microphysics and learn an optimal re-parameterization of the solver. We demonstrate two approaches for training the reduced dimension closure model (1) an a priori method that optimizes the closure parameterization and the neural network parameters separately and (2) an a posteriori method that simultaneously optimizes both. Using the simulation of biomass pyrolysis as a motivating example, we show that the a posteriori method achieves better target losses and is less dependent on training dataset size for generalizability. We then demonstrate the impact that implementing this reparameterization has at the macroscale, showing improved predictive performance with no modification to the underlying macroscale solvers.
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
OPTIMIZATION OF COOLING DISTRIBUTION OF THE EIC SHC COOLER ERL
The Electron-Ion Collider (EIC) Hadron Storage Ring (HSR) will use strong hadron cooling to maintain the beam brightness and high luminosity during long collision ex?periments. An Energy Recovery Linac is used to deliver the high-current high-brightness electron beam for cooling. For the best cooling effect, the electron beam requires low emittance, small energy spread, and uniform longitudinal distribution. In this work, we simulate and optimize the longitudinal laser-beam distribution shaping at the photo?cathode, modeling space charge forces accurately. Machine parameters such as RF cavity phases are optimized in con?junction with the beam distribution using a genetic optimizer. We demonstrate improvement of the cooling distribution in key parameters.
Wang, N.↗
Tracking renewable carbon in bio-oil/crude co-processing with VGO through 13C/12C ratio analysis
Not Available
Li, Zheng-Hua (ORCID:000000030946500X)↗
Coupling localized Noah-MP-Crop model with the WRF model improved dynamic crop growth simulation across Northeast China
Croplands play a critical role in regulating the energy and moisture exchanges between the land surface and atmosphere. However, the interactions between cropland and climate are usually poorly represented due to a lack of detailed representation in crop types and field management. Here, we coupled the Noah-MP-Crop model with the state-of-the-art Weather Research and Forecasting (WRF) model to explore and evaluate the crop growth dynamics in response to climate variations across Northeast China. The default parameters of the crop model were not exactly suitable for the agricultural ecosystems in Northeast China. The detailed cropland distribution, and crop phenology parameters including growing degree days (GDD) and planting (harvesting) date were first created using multi-source remote sensing products and reanalysis data, and was then successfully used to simulate the growth and yield for corn and soybean and associated energy exchanges. We also optimized and calibrated other crop parameters using the time-series of the Moderate Resolution Imaging Spectroradiometer (MODIS) land surface products. The modified crop model substantially improved the simulation of crop growth, plant physiology, and biomass accumulation for both corn and soybean. Coupling the localized dynamic crop model into the WRF led to considerable decreases in the simulated mean-absolute-errors (MAEs) and biases of the leaf area index, evapotranspiration, and gross primary production compared with the MODIS observed values. Compared with the statistical yield from each province, the modified crop model underestimated the corn yield from 11.1% to 48.6%, whereas overestimated the soybean yield from 16.5% to 162.6%.
54 ENVIRONMENTAL SCIENCES↗
A Novel Active Optimization Approach for Rapid and Efficient Design Space Exploration Using Ensemble Machine Learning
In this work, a novel design optimization technique based on active learning, which involves dynamic exploration and exploitation of the design space of interest using an ensemble of machine learning algorithms, is presented. In this approach, a hybrid methodology incorporating an explorative weak learner (regularized basis function model) that fits high-level information about the response surface and an exploitative strong learner (based on committee machine) that fits finer details around promising regions identified by the weak learner is employed. For each design iteration, an aristocratic approach is used to select a set of nominees, where points that meet a threshold merit value as predicted by the weak learner are selected for evaluation. In addition to these points, the global optimum as predicted by the strong learner is also evaluated to enable rapid convergence to the actual global optimum once the most promising region has been identified by the optimizer. Additionally, this methodology is first tested by applying it to the optimization of a two-dimensional multi-modal surface and, subsequently, to a complex internal combustion (IC) engine combustion optimization case with nine control parameters related to fuel injection, initial thermodynamic conditions, and in-cylinder flow. It is found that the new approach significantly lowers the number of function evaluations that are needed to reach the optimum design configuration (by up to 80%) when compared to conventional optimization techniques, such as particle swarm and genetic algorithm-based optimization techniques.
24 POWER TRANSMISSION AND DISTRIBUTION↗
Optimization Layers for Pyomo [SWR-25-132]
Optimization Layers for Pyomo solves an optimization problem using Pyomo and IPOPT during the forward pass. It computes the gradient of the optimal solution with respect to the parameters based on the KKT conditions in the backward pass. It is a Python library for constructing differentiable optimization layers in PyTorch from Pyomo optimization models.
Chen, Kejun [National Laboratory of the Rockies (N↗
Robust and Simple ADMM Penalty Parameter Selection
We present a new method for online selection of the penalty parameter for the alternating direction method of multipliers (ADMM) algorithm. ADMM is a widely used method for solving a range of optimization problems, including those that arise in signal and image processing. In its standard form, ADMM includes a scalar hyperparameter, known as the penalty parameter, which usually has to be tuned to achieve satisfactory empirical convergence. In this work, we develop a framework for analyzing the ADMM algorithm applied to a quadratic problem as an affine fixed point iteration. Using this framework, we develop a new method for automatically tuning the penalty parameter by detecting when it has become too large or small. We analyze this and several other methods with respect to their theoretical properties, i.e., robustness to problem transformations, and empirical performance on several optimization problems. Our proposed algorithm is based on a theoretical framework with clear, explicit assumptions and approximations, is theoretically covariant/invariant to problem transformations, is simple to implement, and exhibits competitive empirical performance.
42 ENGINEERING↗
xesn: Echo state networks powered by Xarray and Dask
Xesn is a Python package that allows scientists to easily design Echo State Networks (ESNs) for forecasting problems. ESNs are a Recurrent Neural Network architecture introduced by Jaeger (2001) that are part of a class of techniques termed Reservoir Computing. One defining characteristic of these techniques is that all internal weights are determined by a handful of global, scalar parameters, thereby avoiding problems during backpropagation and reducing training time significantly. Because this architecture is conceptually simple, many scientists implement ESNs from scratch, leading to questions about computational performance. Xesn offers a straightforward, standard implementation of ESNs that operates efficiently on CPU and GPU hardware. The package leverages optimization tools to automate the parameter selection process, so that scientists can reduce the time finding a good architecture and focus on using ESNs for their domain application. Importantly, the package flexibly handles forecasting tasks for out-of-core, multi-dimensional datasets, eliminating the need to write parallel programming code. Xesn was initially developed to handle the problem of forecasting weather dynamics, and so it integrates naturally with Python packages that have become familiar to weather and climate scientists such as Xarray (Hoyer & Hamman, 2017). However, the software is ultimately general enough to be utilized in other domains where ESNs have been useful, such as in signal processing (Jaeger & Haas, 2004).
97 MATHEMATICS AND COMPUTING↗
Optimizing spin dressing sensitivity for the nEDMSF experiment
nEDMSF aims to measure the neutron electric dipole moment (d n ) with unprecedented precision. In this paper we explore the experiment's sensitivity when operating with an implementation of the critical dressing method in which the angle between the neutron and Helium-3 spins (ϕ 3n ) is subjected to a square modulation by an amount ϕ d (the “dressing angle”). Several parameters can be tuned to optimize sensitivity. We find roughly 10% improvement over a previous estimate, resulting primarily from the addition of a waiting period between the π/2 pulse that initiates d n -driven ϕ 3n growth and the start of ϕ3n modulation. We find negligible further improvement by allowing ϕ d to vary continuously over the course of a run, and no degradation resulting from the addition of an in situ background measurement into each ϕ3n modulation sequence. A complete simulation confirms a 300 live-day sensitivity ofσ = 1.45×10 -28 e ·cm. At this level of sensitivity, σ ϕ3n0 = 1 mrad precision on the initial n/ 3 He angle difference is not negligible.
47 OTHER INSTRUMENTATION↗
A novel active optimization approach for rapid and efficient design space exploration using ensemble machine learning
In this work, a novel design optimization technique based on active learning, which involves dynamic exploration and exploitation of the design space of interest using an ensemble of machine learning algorithms, is presented. In this approach, a hybrid methodology incorporating an explorative weak learner (regularized basis function model) which fits high-level information about the response surface, and an exploitative strong learner (based on committee machine) that fits finer details around promising regions identified by the weak learner, is employed. For each design iteration, an aristocratic approach is used to select a set of nominees, where points that meet a threshold merit value as predicted by the weak learner are selected to be evaluated using expensive function evaluation. In addition to these points, the global optimum as predicted by the strong learner is also evaluated to enable rapid convergence to the actual global optimum once the most promising region has been identified by the optimizer. This methodology is first tested by applying it to the optimization of a two-dimensional multi-modal surface. The performance of the new active learning approach is compared with traditional global optimization methods, namely micro-genetic algorithm (pGA) and particle swarm optimization (PSO). It is demonstrated that the new optimizer is able to reach the global optimum much faster with a significantly fewer number of function evaluations. Subsequently, the new optimizer is also applied to a complex internal combustion (IC) engine combustion optimization case with nine control parameters related to fuel injection, initial thermodynamic conditions, and in-cylinder flow. It is again found that the new approach significantly lowers the number of function evaluations that are needed to reach the optimum design configuration (by up to 80%) when compared to particle swarm and genetic algorithm-based optimization techniques.
Machine learning, active learning, design optimiza↗
Development of Corrosion- and Erosion-Resistant Coatings for Advanced Ultra-Supercritical Materials
This final report summarized the research efforts and major findings of the Phase I Project “Development of Corrosion- and Erosion-Resistant Coatings for Advanced Ultra-Supercritical Materials”, for the period of October 1, 2019 – Sept. 30, 2021. This project is a collaborative endeavor between Tennessee Tech University, Purdue University, Oak Ridge National Laboratory, Siemens Corporation, and Eastern Plating, LLC, aiming at improving the durability and lifetime of high-pressure (HP) steam turbine blades in advanced ultra-supercritical (A-USC) coal-fired power plants through the development of corrosion/erosion-resistant coatings manufactured via a low-cost electrolytic codeposition process. While Tribaloy alloy T-400C was identified by the U.S. A-USC Materials Consortium as a promising coating composition, further composition optimization is needed to enhance its corrosion and erosion resistance for protecting the A-USC Ni-base turbine components. An integrated computational and experimental approach was employed to optimize coating composition/microstructure and processing parameters. In order to identify candidate coating compositions that could offer balanced properties, thermodynamic calculations were performed to explore the γ+Laves composition space in the Co-Ni-Cr-Mo-Si system with different alloying additions at 600-800 °C. Guided by the calculation results, experimental assessment of selected alloys led to the development of a new generation of Tribaloy compositions with the optimal levels of Cr, Mo and Si, reactive element (e.g., 0.4-0.6 wt.% Y) and other alloying additions. The low-cost and non-line-of-sight electro-codeposition process was employed to deposit a Ni(Co)-CrMoSiY composite coating on commercial Haynes 282 (H282) Ni-base alloy. A diffusion treatment was subsequently applied to convert the composite to the Tribaloy-type coating. Both the codeposition parameters and heat treatment conditions were varied to achieve the desired coating composition, microstructure and phase constituents. In addition, since additive manufacturing (AM) may be an alternative cost-saving option for potential A-USC turbine repair, laser direct deposition was explored to fabricate the H282 alloy with minimal defects. The electro-codeposited Tribaloy coating was also applied to the AM H282 substrate to demonstrate the viability of the coating process in improving the surface finish of AM alloys. Both high-temperature oxidation performance and solid particle erosion (SPE) resistance of model alloys and electro-codeposited coatings were evaluated. About 25 model alloys with various Cr/Mo ratios and reactive element levels, as well as partial substitution of Mo with Nb were evaluated with regard to their oxidation resistance in both air and pure steam at 760-800°C. Compositions based on Ni(Co)-20Cr-18Mo-2.6Si-0.6Y (wt.%) showed significantly improved oxidation resistance over the baseline T400-C. Based on the alloy development results, three generations of new Tribaloy coatings (Gen-1, Gen-2, and Gen-3) with various Mo/Cr contents and Y levels were prepared via electro-codeposition and their microstructure/performance were evaluated. Outstanding air and steam oxidation resistance was achieved for the Gen-2 and Gen-3 coatings. Furthermore, while the SPE resistance of the coatings depended on many factors such as temperature, environment, erodent, velocity and impact angle, the coated samples exhibited similar or better SPE resistance compared to the H282 substrate when magnetite was used as erodent (which is a realistic erodent in A-USC steam turbines). The new Tribaloy coatings also had good long-term compatibility with the H282 alloy substrate. The two large-sized rotating barrels were designed, constructed, and employed to coat dummy HP blades. Uniform coating thickness and microstructure were achieved at various blade locations. Also, a preliminary techno-economic analysis of the proposed coating process was conducted to quantify the cost-effectiveness and to assess the commercial viability of the corrosion- and erosion-resistant coatings. It is estimated that a cost reduction of ~30% could be achieved with the electro-codeposition coating process over the state-of-the-art high velocity oxygen fuel (HVOF) thermal spray. Compared to the leading Tribaloy coating technologies such as HVOF and plasma transfer alloying, electro-codeposition based process has advantages such as low-cost process equipment, uniform deposition even for complex shapes, low levels of contaminants/porosities, reduction of powder waste, and potentially better surface finish and longer turbine service life. The Phase 1 study has demonstrated that it is feasible to develop an electro-codeposited Tribaloy coating with balanced corrosion and erosion properties, even though additional research efforts such as further coating process scale-up and longer-term performance evaluation under realistic A-USC conditions are clearly needed.
20 FOSSIL-FUELED POWER PLANTS↗
Once-Through Steam Generator Model Analysis Using Python and Advanced Optimization Tools (Summer Internship Report)
This study focuses on the parametric analysis of design parameters for a once-through steam generator (OTSG) model, using python and advanced optimization tools to facilitate applications such as the flowing autoclave steam generator (FASG) test cases. Building on previous research involving another OTSG with a different design, this project aims to enhance our understanding of how steam generators (SGs) behave and how their outputs are influenced by changes in design. The reason for this design change is to allow for more precise modeling and optimization of SG performance, to provide a comparative analysis between the two designs, and to set up the model for integration with the FASG test case. The OTSG python-model is a mathematical representation (including fluid flow and heat transfer equations/models/correlations) of a steam-generating unit in a pressurized water reactor-type small modular reactor system. Design studies involve changing the model’s input design parameters to observe the resulting effects on the output of the system. By using advanced optimization tools, such as the Risk Analysis Virtual Environment (RAVEN) developed at Idaho National Laboratory, detailed design parametric studies and model optimization were performed. Six input parameters—pressure, temperature, and mass flow rate for the inlet of the primary-side (hot fluid) and secondary-side (cold fluid), respectively, of the SG—were randomly perturbed via RAVEN’s Monte Carlo Sampler module, using uniform distributions (i.e., ±1%, ±5% and ±10% relative changes) for 600 samples. The analysis provides valuable insights into SG optimization and can be used for sensor placement optimization to effectively monitor and obtain experimental data in other tests.
20 FOSSIL-FUELED POWER PLANTS↗
Machine learning-guided design of direct methanol fuel cells with a platinum group metal-free cathode
Direct methanol fuel cells (DMFCs) offer a promising solution for clean electricity generation, particularly in small electronics and remote auxiliary power units. However, optimizing their efficiency and performance is challenging due to the complex interactions between various factors. Here, we present a novel approach that integrates experiments with machine learning to model and predict the performance of these fuel cells using atomically dispersed platinum group metal (PGM)-free catalysts at the cathode. Further, our machine learning models, trained on diverse input parameters, allow for the comprehensive optimization of DMFC performance prior to fabrication and testing. Through extensive experimental validation, we demonstrate that this data-driven approach accurately predicts key performance metrics, such as maximum power output and polarization curves. By combining our models with interpretable game-theory methods, we provide deep insights into the factors governing fuel cell performance, ultimately paving the way for the design of scalable and efficient DMFC technologies.
25 ENERGY STORAGE↗
Remarkable enhancement in thermal conductivity of stainless-steel leveraging metal composite via laser powder bed fusion: 316L-Cu composite
This work aimed to significantly enhance the thermal conductivity of the SS matrix by fabricating a metal composite SS-Cu system utilizing a laser powder bed fusion process. The impact of process parameters on the microstructure of SS-Cu samples is investigated. Here, the processing and post-processing factors that influence the effective thermal conductivity of the metal composite SS-Cu system are discussed. Then, the effective thermal conductivity and mechanical properties of the SS-Cu composite are measured at room temperature and operating temperature of 300 °C. The optimized laser powder bed fusion parameters led to low Marangoni convection, high cooling rate, greater macro-segregation, and higher thermal conductivity values albeit at the cost of ductility in SS-Cu metal composite likely due to the presence of the unmelted powder particles. As-printed SS-60 vol % Cu and annealed SS-60 vol % Cu had 2.5X and 6.6X thermal conductivity enhancement, respectively. During annealing, phase separation, partial melting of Cu, and formation of a continuous network of Cu surrounding SS play a key role in enhancing the thermal conductivity of the SS-Cu metal composite structure.
36 MATERIALS SCIENCE↗
Active Reaction Control of Cu Redox State Based on Real-Time Feedback from In Situ Synchrotron Measurements
Abstract not provided
Rakita, Yevgeny↗
Inverse design of pore wall chemistry and topology through active learning of surface group interactions
Design of next-generation membranes requires a nanoscopic understanding of the effect of biologically inspired heterogeneous surface chemistries and topologies (roughness) on local water and solute behavior. In particular, the rejection of small, neutral solutes, such as boric acid, poses a heretofore unsolved challenge. In prior work, a computational inverse design technique using an evolutionary optimization successfully uncovered new surface design strategies for optimized transport of water over solutes in smooth, model pores consisting of two surface chemistries. However, extending such an approach to more complex (and realistic) scenarios involving many surface chemistries as well as surface roughness is challenging due to the expanded design space. In this work, we develop a new approach that uses active learning to optimize in a reduced feature space of surface group interactions, finding parameters that lead to their assembly into ordered, optimal patterns. This approach rapidly identifies novel surface functionalizations that maximize the difference in water and boric acid transport through the nanopore. Moreover, we find that the roughness of the nanopore wall, independent of its chemistry, can be leveraged to enhance transport selectivity: oscillations in the pore wall diameter optimally inhibit boric acid transport by creating energetic wells from which the solute must escape to transport down the pore. Furthermore, this proof-of-concept demonstrates the potential for active learning strategies, in concert with molecular simulations, to rapidly navigate complex design spaces of aqueous interfaces and is promising as a tool for engineering water-mediated surface interactions for a broad range of applications.
36 MATERIALS SCIENCE↗