SEARCH · Engineering Papers

Results for “Nonlinear robust 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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 217 records · Page 12

All sources NASA NTRS DOE OSTI

DOE OSTI2026

Real-Time GPU-Accelerated OFDR With an Integrated Auxiliary Interferometer

A GPU-accelerated optical frequency domain reflectometry (OFDR) system with an improved integrated auxiliary interferometer is proposed. Unlike conventional approaches that require separate auxiliary interferometers and multiple detection channels, the proposed OFDR system embeds this functionality directly into the signal via an intentional beat component. This enables self-calibration of laser nonlinearity while maintaining a cost-effective hardware configuration. Building on this simplified configuration, the system leverages GPU acceleration with an NVIDIA RTX 4070 Ti to achieve real-time performance, delivering high-throughput signal processing for continuous OFDR interrogation. The signal processing pipeline comprises signal capture, resampling for nonlinearity compensation, and frequency shift computation, all optimized for parallel execution. Hardware benchmarking demonstrates substantial acceleration over CPU implementations, achieving up to a 45× speedup for resampling and frequency shift computations and enabling processing latencies below 30 ms. Thermal response validation is conducted under two complementary scenarios: localized heating using a water bath and cryogenic-temperature conditions using liquid nitrogen. Under localized heating, the system achieves an accuracy of 0.249 °C with a thermal sensitivity of 5.971 GHz/°C, while cryogenic-temperature validation demonstrates a frequency shift response with a sensitivity of 2.383 GHz/°C and an accuracy of 2.04 °C. The high acceleration of the proposed GPU-accelerated OFDR system and its accuracy are achieved by exploiting CUDA-based stride indexing, enabling efficient parallel segmentation and processing of large datasets without additional memory copies. The benchmarking results confirm the robustness, accuracy, and deployability of the proposed OFDR system across a wide temperature range, establishing it as a practical platform for real-time distributed fiber sensing in structurally dynamic environments.

Harb, Salah [Lawrence Berkeley National Laboratory↗

DOE OSTI2024

Enhancing Distribution System Resilience: A First-Order Meta-RL Algorithm for Critical Load Restoration

The increasing frequency of extreme events and the integration of distributed energy resources (DERs) into modern grids have elevated the need for resilient and efficient critical load restoration strategies in distribution systems. However, the stochastic nature of renewable DERs, limited energy resource availability and the intricate nonlinearities inherent in complex grid control problem make the problem challenging. Although reinforcement learning (RL) and warm-start RL methods have shown promising results, their performance often falls short in rapidly adapting to new, unseen situations and typically requires exhaustive problem-specific tuning. To address these gaps, we propose a First-Order Meta-based RL (FOM-RL) algorithm within an online framework for adaptive and robust critical load restoration. By harnessing local DERs as the enabling technology, FOM-RL allows the RL agent to swiftly adapt to new unseen scenarios by leveraging previously acquired knowledge of different tasks. Experimental results provide evidence that proposed algorithm learns more efficiently and showcases generalization capabilities across diverse set of operational scenarios. Moreover, a rigorous theoretical analysis yields a tight sublinear regret bound, sensitive to temporal variability, with a task-averaged optimality gap bounded by O(VM+D*/(Tsquare root(M))). These results suggest that optimality improves with task similarity and an increased number of tasks M, reaffirming the efficacy and scalability of the proposed approach in addressing the complexities of critical load restoration in distribution systems.

complexity theory↗

DOE OSTI2020

Resistivity phase diagram of cuprates revisited

The phase diagram of the cuprate superconductors has posed a formidable scientific challenge for more than three decades. This challenge is perhaps best exemplified by the need to understand the normal-state charge transport as the system evolves from Mott insulator to Fermi- liquid metal with doping. Here in this paper, we report a detailed analysis of the temperature (T) and doping (p) dependence of the planar resistivity of simple-tetragonal HgBa 2 CuO 4+δ (Hg1201), the single- CuO 2 -layer cuprate with the highest optimal superconducting transition temperature, T c . The data allow us to test a recently proposed phenomenological model for the cuprate phase diagram that combines a universal transport scattering rate with spatially inhomogeneous (de)localization of the Mott-localized hole. We find that the model provides a good description of the data. We then extend this analysis to prior transport results for several other cuprates, including the Hall number in the overdoped part of the phase diagram, and find little compound-to-compound variation in (de)localization gap scale. The results point to a robust, universal structural origin of the inherent gap inhomogeneity that is unrelated to doping-related disorder. They are inconsistent with the notion that much of the phase diagram is controlled by a quantum critical point, and instead indicate that the unusual electronic properties exhibited by the cuprates are fundamentally related to strong nonlinearities associated with subtle nanoscale inhomogeneity.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

DOE OSTI2025

Building a quantum computing architecture using 3D superconducting cavities

Quantum computers promise advantages over classical machines for solving certain complex problems, but building processors that truly deliver this advantage remains a central challenge, particularly due to limited coherence times. Three-dimensional superconducting radio-frequency (SRF) cavities offer an attractive platform due to their exceptionally long lifetimes. However, since these harmonic systems require nonlinear elements, such as transmons, for control, additional losses are often introduced. In this talk, I will present a multimode quantum system based on an elliptical SRF cavity hosting two cavity modes weakly coupled to an ancillary transmon circuit. This architecture is carefully engineered to preserve coherence while enabling efficient control. By optimizing the design to mitigate transmon-induced decoherence, we realize single-photon lifetimes of 20.6 ms and 15.6 ms in the two modes, with pure dephasing times exceeding 40 ms. Using sideband interactions and error-resilient protocols, such as measurement-based correction and post-selection, we demonstrate high-fidelity state control, including preparation of Fock states up to N=20 with fidelities above 95% (to our knowledge, the highest reported to date), as well as high-fidelity two-mode entanglement. These results highlight 3D SRF cavities as a robust foundation for qudit-based quantum information processing, harnessing the large Hilbert space of cavity modes. I will conclude by outlining strategies to further enhance coherence in both cavities and ancilla qubits, and discuss pathways toward scaling this architecture into a larger quantum computing platform.

Roy, Tanay [Fermilab]↗

DOE OSTI2025

Uncertainty quantification of a physics-informed model based on sparse identification of a Thermal Energy Distribution System

Integrated energy systems (IES)s are crucial for enhancing the economy and efficiency of power generation sources (e.g., nuclear energy) necessary to unleash American energy dominance. These systems can be integrated with thermal energy storage (TES) and intermittent renewable energies to optimize overall energy use, peak-load regulation, and demand-side responses. However, the stabilization of energy generation, transport, and utilization introduces operational complexities that exceed the challenges of managing each sub-component individually. Currently, though IESs rely on human operators for efficiency and stability, reducing human error risk and enhancing performance through automation is highly desirable. Recent advances at Idaho National Laboratory have demonstrated successful control of the Thermal Energy Distributed System (TEDS). However, the automatic control system depends on a deterministic Sparse Identification of Nonlinear Dynamics with Control (SINDyC) model, which are trained based on simulation data from physics-based simulations. Because of uncertainties in physics-based simulation, SINDyC model results in large discrepancies against experimental data and cannot be reliably used in automatic control. In this paper, we present an innovative approach to address these discrepancies by quantifying uncertainties and developing a more robust model. We first generated trajectories by using first-principles physics codes to encapsulate the experiment. Next, we trained thousands of models by randomly sampling these trajectories. We then collapsed all those models into one probabilistic SINDyC by fitting a multivariate Gaussian distribution onto the resulting coefficient’s distribution. Despite its simplicity, our approach successfully produced 95% confidence intervals that captured the experimental trajectories. It even did so with a higher probability and better U-pooling score across six of the seven relevant quantities of interest (QoIs), as compared to other classical approaches. In conclusion, ongoing research is focusing on generating new experimental trajectories to validate this approach, and on employing Bayesian calibration to refine parametric uncertainties and guide future model development efforts.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2020

Toward fusion plasma scenario planning for NSTX-U using machine-learning-accelerated models

One of the most promising devices for realizing power production through nuclear fusion is the tokamak. To maximize performance, it is preferable that tokamak reactors achieve advanced operating scenarios characterized by good plasma confinement, improved magnetohydrodynamic (MHD) stability, and a largely non-inductively driven plasma current. Such scenarios could enable steady-state reactor operation with high \emph{fusion gain} --- the ratio of produced fusion power to the external power provided through the plasma boundary. Precise and robust control of the evolution of the plasma boundary shape as well as the spatial distribution of the plasma current, density, temperature, and rotation will be essential to achieving and maintaining such scenarios. The complexity of the evolution of tokamak plasmas, arising due to nonlinearities and coupling between various parameters, motivates the use of model-based control algorithms that can account for the system dynamics. In this work, a learning-based accelerated model trained on data from the National Spherical Torus Experiment Upgrade (NSTX-U) is employed to develop planning and control strategies for regulating the density and temperature profile evolution around desired trajectories. The proposed model combines empirical scaling laws developed across multiple devices with neural networks trained on empirical data from NSTX-U and a database of first-principles-based computationally intensive simulations. The reduced execution time of the accelerated model will enable practical application of optimization algorithms and reinforcement learning approaches for scenario planning and control development. An initial demonstration of applying optimization approaches to the learning-based model is presented, including a strategy for mitigating the effect of leaving the finite validity range of the accelerated model. The approach shows promise for actuator planning between experiments and in real-time.

machine learning↗

NASA NTRS1991

A trajectory planning scheme for spacecraft in the space station environment

Simulated annealing is used to solve a minimum fuel trajectory problem in the space station environment. The environment is special because the space station will define a multivehicle environment in space. The optimization surface is a complex nonlinear function of the initial conditions of the chase and target crafts. Small permutations in the input conditions can result in abrupt changes to the optimization surface. Since no prior knowledge about the number or location of local minima on the surface is available, the optimization must be capable of functioning on a multimodal surface. It was reported in the literature that the simulated annealing algorithm is more effective on such surfaces than descent techniques using random starting points. The simulated annealing optimization was found to be capable of identifying a minimum fuel, two-burn trajectory subject to four constraints which are integrated into the optimization using a barrier method. The computations required to solve the optimization are fast enough that missions could be planned on board the space station. Potential applications for on board planning of missions are numerous. Future research topics may include optimal planning of multi-waypoint maneuvers using a knowledge base to guide the optimization, and a study aimed at developing robust annealing schedules for potential on board missions.

Soller, Jeffrey Alan↗

NASA NTRS2014

Evaluating LSM-Based Water Budgets Over a West African Basin Assisted with a River Routing Scheme

Within the framework of the African Monsoon Multidisciplinary Analysis (AMMA) Land Surface Model Intercomparison Project phase 2 (ALMIP-2), this study evaluates the water balance simulated by the Interactions between Soil, Biosphere, and Atmosphere (ISBA) over the upper Oum River basin, in Benin, using a mesoscale river routing scheme (RRS). The RRS is based on the nonlinear Muskingum Cunge method coupled with two linear reservoirs that simulate the time delay of both surface runoff and base flow that are produced by land surface models. On the basis of the evidence of a deep water-table recharge in that region,a reservoir representing the deep-water infiltration (DWI) is introduced. The hydrological processes of the basin are simulated for the 2005-08 AMMA field campaign period during which rainfall and stream flow data were intensively collected over the study area. Optimal RRS parameter sets were determined for three optimization experiments that were performed using daily stream flow at five gauges within the basin. Results demonstrate that the RRS simulates stream flow at all gauges with relative errors varying from -22% to 3% and Nash-Sutcliffe coefficients varying from 0.62 to 0.90. DWI varies from 24% to 67% of the base flow as a function of the sub-basin. The relatively simple reservoir DWI approach is quite robust, and further improvements would likely necessitate more complex solutions (e.g., considering seasonality and soil type in ISBA); thus, such modifications are recommended for future studies. Although the evaluation shows that the simulated stream flows are generally satisfactory, further field investigations are necessary to confirm some of the model assumptions.

Multidisciplinary↗

NASA NTRS2008

Ares-I Bending Filter Design using a Constrained Optimization Approach

The Ares-I launch vehicle represents a challenging flex-body structural environment for control system design. Software filtering of the inertial sensor output is required to ensure adequate stable response to guidance commands while minimizing trajectory deviations. This paper presents a design methodology employing numerical optimization to develop the Ares-I bending filters. The design objectives include attitude tracking accuracy and robust stability with respect to rigid body dynamics, propellant slosh, and flex. Under the assumption that the Ares-I time-varying dynamics and control system can be frozen over a short period of time, the bending filters are designed to stabilize all the selected frozen-time launch control systems in the presence of parameter uncertainty. To ensure adequate response to guidance command, step response specifications are introduced as constraints in the optimization problem. Imposing these constrains minimizes performance degradation caused by the addition of the bending filters. The first stage bending filter design achieves stability by adding lag to the first structural frequency to phase stabilize the first flex mode while gain stabilizing the higher modes. The upper stage bending filter design gain stabilizes all the flex bending modes. The bending filter designs provided here have been demonstrated to provide stable first and second stage control systems in both Draper Ares Stability Analysis Tool (ASAT) and the MSFC MAVERIC 6DOF nonlinear time domain simulation.

Hall, Charles↗

DOE OSTI2026

Stochastic Microgrid Scheduling With Chance‐Constrained Resilience Consideration

Traditionally, it is assumed that microgrids transition seamlessly from grid‐connected operation to islanded mode in the event of sudden main grid outages. In reality, the islanding process, especially unintentional islanding, is rarely seamless. Instead, it is subject to voltage and frequency fluctuations caused by the instantaneous disconnection of the point of common coupling (PCC) switch, variations in loads and renewable generation output and even the protection tripping of distributed energy resources (DERs). To mitigate these fluctuations and facilitate a smooth islanding process, we propose a stochastic microgrid scheduling model that incorporates chance‐constrained resilience measures. Specifically, the resilience measure is defined as the probability of successful islanding (PSI), that is, the probability that a microgrid can mitigate the generation‐demand imbalance caused by the disconnection of the PCC switch, variations in load and renewable generation and DER tripping. This measure is modelled using chance constraints. Unlike existing reliability and resilience indices, which typically neglect the possibility of microgrid/DER failure under extreme events and assume their survival while primarily focussing on reducing impact duration or magnitude, the proposed PSI‐based framework explicitly addresses microgrid and DER survival during the islanding transition. The formulated nonlinear chance constraints are approximated using a multiinterval approach and equivalently represented as a mixed‐integer linear programming (MILP) formulation. Case study results validate the proposed method, showing that the PSI estimation error is reduced to less than 8%, compared to approximately 28% with existing methods. Various sensitivity analyses on the DER tripping rate and PSI settings were performed to validate the robustness of the proposed method. In particular, the necessity of accounting for DER tripping in the PSI calculation was demonstrated.

chance constrained optimization↗

DOE OSTI2024

Denoising diffusion probabilistic models for generative alloy design

Inverse material design is an extremely challenging optimization task made difficult by, in part, the highly nonlinear relationship linking performance with composition. Quantitative approaches have improved significantly owing to advances in high throughput experimentation and computational thermodynamics. However, existing physics-based tools are mostly forward models; input a chemistry and obtain a prediction. More recently the materials community has leveraged advances in the machine learning community to establish novel inverse design frameworks. Very recently denoising diffusion probabilistic models have been shown to be extremely powerful generators producing synthetic data of various modalities e.g. images, text, audio, tables, etc.. In this work a novel framework for alloy design and optimization is proposed leveraging these class of models. Five key generative tasks are demonstrated (1) unconditional generation (2) composition conditioned generation (3) property conditioned generation (4) multi-feedstock conditioned generation and (5) generative optimization. These methods were tested on three case studies: high entropy alloy design, superalloy binder jet additive manufacturing, and in-situ dual-feedstock wire-arc additive manufacturing. Results indicate that the established models are extremely flexible, expressive, and robust. The architecture’s flexibility and training procedure empower the model to learn complex intra-compositional and composition-property relationships. Furthermore, the probabilistic nature of these models makes them well suited for addressing solution non-uniqueness and tackling uncertainty quantification tasks. While the fidelity and quantity of the underlying training data is paramount, we envision that future alloy design frameworks will make extensive use of these kinds of machine learning models as “search” tools bolstering the utility of experimental and computational approaches.

36 MATERIALS SCIENCE↗

DOE OSTI2020

Multi-Stage and Multi-Timescale Robust Co-Optimization Planning for Reliable and Sustainable Power Systems. Final Report

In this project, Clarkson University, in collaboration with Southern Methodist University and University of Pittsburgh, conducted the research to model, design, and implement a sophisticated generation and transmission co-optimization planning decision tool, called Multi-stage and Multi-timescale robust Co-Optimization Planning (MMCOP). The MMCOP decision tool intends to facilitate generation and transmission co-optimization planning of emerging power systems, while mitigating risks and uncertainties in both short-term operation dynamics and long-term policy and technology changes. Long-term power system planning aims at optimizing asset utilization by investing in a proper mix of various generation technologies and transmission lines to supply the future load growth. In particular, the Clean Power Plan (CPP), which is designed to combat climate change and reduce carbon emissions by setting a national limit on carbon pollution from power plants, may dramatically change the landscape of the power industry by further promoting clean energy and phasing out emissions-intensive generation technologies. In addition, novel non-wire alternatives (e.g., demand response (DR), distributed generation (DG), energy efficiency (EE), and smart grid technologies) and the computational complexity for large-scale systems significantly complicate the system planning procedure even further. However, existing conventional planning approaches neglect short-term variability and uncertainty of renewable energy, hourly chronological operation details, and physical nonlinear characteristics of the alternating current transmission network. As a result, existing conventional planning approaches may not work properly, and power systems reliability could be in jeopardy. In observing the limitations of existing conventional planning approaches and addressing new challenges of emerging power systems, the main scope of this project to develop co-optimization planning models within a multi-stage and multi-timescale framework. In particular, random contingencies, key uncertainty factors, and AC power flows are included to derive expansion plans while considering both long-term reliability and short-term flexibility.

24 POWER TRANSMISSION AND DISTRIBUTION↗

NASA NTRS1996

Cascade Optimization Strategy for Aircraft and Air-Breathing Propulsion System Concepts

Design optimization for subsonic and supersonic aircraft and for air-breathing propulsion engine concepts has been accomplished by soft-coupling the Flight Optimization System (FLOPS) and the NASA Engine Performance Program analyzer (NEPP), to the NASA Lewis multidisciplinary optimization tool COMETBOARDS. Aircraft and engine design problems, with their associated constraints and design variables, were cast as nonlinear optimization problems with aircraft weight and engine thrust as the respective merit functions. Because of the diversity of constraint types and the overall distortion of the design space, the most reliable single optimization algorithm available in COMETBOARDS could not produce a satisfactory feasible optimum solution. Some of COMETBOARDS' unique features, which include a cascade strategy, variable and constraint formulations, and scaling devised especially for difficult multidisciplinary applications, successfully optimized the performance of both aircraft and engines. The cascade method has two principal steps: In the first, the solution initiates from a user-specified design and optimizer, in the second, the optimum design obtained in the first step with some random perturbation is used to begin the next specified optimizer. The second step is repeated for a specified sequence of optimizers or until a successful solution of the problem is achieved. A successful solution should satisfy the specified convergence criteria and have several active constraints but no violated constraints. The cascade strategy available in the combined COMETBOARDS, FLOPS, and NEPP design tool converges to the same global optimum solution even when it starts from different design points. This reliable and robust design tool eliminates manual intervention in the design of aircraft and of air-breathing propulsion engines where it eases the cycle analysis procedures. The combined code is also much easier to use, which is an added benefit. This paper describes COMETBOARDS and its cascade strategy and illustrates the capability of the combined design tool through the optimization of a subsonic aircraft and a high-bypass-turbofan wave-rotor-topped engine.

Patnaik, Surya N.↗

NASA NTRS2010

Ares I Flight Control System Design

The Ares I launch vehicle represents a challenging flex-body structural environment for flight control system design. This paper presents a design methodology for employing numerical optimization to develop the Ares I flight control system. The design objectives include attitude tracking accuracy and robust stability with respect to rigid body dynamics, propellant slosh, and flex. Under the assumption that the Ares I time-varying dynamics and control system can be frozen over a short period of time, the flight controllers are designed to stabilize all selected frozen-time launch control systems in the presence of parametric uncertainty. Flex filters in the flight control system are designed to minimize the flex components in the error signals before they are sent to the attitude controller. To ensure adequate response to guidance command, step response specifications are introduced as constraints in the optimization problem. Imposing these constraints minimizes performance degradation caused by the addition of the flex filters. The first stage bending filter design achieves stability by adding lag to the first structural frequency to phase stabilize the first flex mode while gain stabilizing the higher modes. The upper stage bending filter design gain stabilizes all the flex bending modes. The flight control system designs provided here have been demonstrated to provide stable first and second stage control systems in both Draper Ares Stability Analysis Tool (ASAT) and the MSFC 6DOF nonlinear time domain simulation.

Jang, Jiann-Woei↗

DOE OSTI2024

NeuroSEM: A hybrid framework for simulating multiphysics problems by coupling PINNs and spectral elements

Multiphysics problems that are characterized by complex interactions among fluid dynamics, heat transfer, structural mechanics, and electromagnetics, are inherently challenging due to their coupled nature. While experimental data on certain state variables may be available, integrating these data with numerical solvers remains a significant challenge. Physics-informed neural networks (PINNs) have shown promising results in various engineering disciplines, particularly in handling noisy data and solving inverse problems in partial differential equations (PDEs). However, their effectiveness in forecasting nonlinear phenomena in multiphysics regimes, particularly involving turbulence, is yet to be fully established. Here, this study introduces NeuroSEM, a hybrid framework integrating PINNs with the highfidelity Spectral Element Method (SEM) solver, Nektar++. NeuroSEM leverages the strengths of both PINNs and SEM, providing robust solutions for multiphysics problems. PINNs are trained to assimilate data and model physical phenomena in specific subdomains, which are then integrated into the Nektar++ solver. We demonstrate the efficiency and accuracy of NeuroSEM for thermal convection in cavity flow and flow past a cylinder. The framework effectively handles data assimilation by addressing those subdomains and state variables where the data is available. We applied NeuroSEM to the Rayleigh-B´enard convection system, including cases with missing thermal boundary conditions and noisy datasets. Finally, we applied the proposed NeuroSEM framework to real particle image velocimetry (PIV) data to capture flow patterns characterized by horseshoe vortical structures. Our results indicate that NeuroSEM accurately models the physical phenomena and assimilates the data within the specified subdomains. The framework’s plug-and-play nature facilitates its extension to other multiphysics or multiscale problems. Furthermore, NeuroSEM is optimized for efficient execution on emerging integrated GPU-CPU architectures. This hybrid approach enhances the accuracy and efficiency of simulations, making it a powerful tool for tackling complex engineering challenges in various scientific domains.

42 ENGINEERING↗

NASA NTRS1995

Dissipative rendering and neural network control system design

Model-based control system designs are limited by the accuracy of the models of the plant, plant uncertainty, and exogenous signals. Although better models can be obtained with system identification, the models and control designs still have limitations. One approach to reduce the dependency on particular models is to design a set of compensators that will guarantee robust stability to a set of plants. Optimization over the compensator parameters can then be used to get the desired performance. Conservativeness of this approach can be reduced by integrating fundamental properties of the plant models. This is the approach of dissipative control design. Dissipative control designs are based on several variations of the Passivity Theorem, which have been proven for nonlinear/linear and continuous-time/discrete-time systems. These theorems depend not on a specific model of a plant, but on its general dissipative properties. Dissipative control design has found wide applicability in flexible space structures and robotic systems that can be configured to be dissipative. Currently, there is ongoing research to improve the performance of dissipative control designs. For aircraft systems that are not dissipative active control may be used to make them dissipative and then a dissipative control design technique can be used. It is also possible that rendering a system dissipative and dissipative control design may be combined into one step. Furthermore, the transformation of a non-dissipative system to dissipative can be done robustly. One sequential design procedure for finite dimensional linear time-invariant systems has been developed. For nonlinear plants that cannot be controlled adequately with a single linear controller, model-based techniques have additional problems. Nonlinear system identification is still a research topic. Lacking analytical models for model-based design, artificial neural network algorithms have recently received considerable attention. Using their universal approximation property, neural networks have been introduced into nonlinear control designs in several ways. Unfortunately, little work has appeared that analyzes neural network control systems and establishes margins for stability and performance. One approach for this analysis is to set up neural network control systems in the framework presented above. For example, one neural network could be used to render a system to be dissipative, a second strictly dissipative neural network controller could be used to guarantee robust stability.

Gonzalez, Oscar R.↗

DOE OSTI2025

Autonomous Nanoparticle Synthesis Guided by In Situ Multiscale Structural Characterization

Autonomous synthesis platforms promise rapid exploration of vast parameter spaces; yet, integrating in situ structural characterization in closed-loop synthesis optimization remains challenging. We demonstrate a realization of such a closed-loop platform coupled with a droplet-flow microreactor, in situ X-ray scattering methods (SAXS/WAXS), and Gaussian process optimization to synthesize citrate-reduced Au nanoparticles with targeted characteristics. The system efficiently explored ∼19,000 synthesis recipes through 365 experiments, achieving precise control over size (4–60 nm) and polydispersity (σ < 0.11) across large citrate/gold ratios, exceeding traditional synthesis boundaries (1–10). Beyond confirming classical Turkevich–Frens trends, partial-dependence analysis revealed strong nonlinear coupling among precursor, citrate, and pH effects. Combining quantitative SAXS/WAXS analysis with electron microscopy characterization, we uncovered that crystallite size (d c ) and particle size (d) follow d c = 0.18d + β, where synthesis chemistry controls the intercept β while maintaining a universal slope. This parallel-band structure enables independent tuning of crystallite domain size at fixed particle diameter through a combination of chloride, gold precursor, citrate, and pH contributions (cross-validated Spearman ρ = 0.7 ± 0.1). High-resolution electron microscopy shows multiple lattice-fringe orientations within single particles, directly confirming polycrystalline domains and the ability to tune d c at the fixed d. The platform’s validation includes indistinguishable static versus flowing measurements, stable droplet transport at 100 °C, and <5% run-to-run variation, establishing a robust framework for mapping and controlling multiscale nanoparticle structure across expansive chemical spaces. In conclusion, the developed closed-loop platform can be applied to a borad range of nanosyntheis processes.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

DOE OSTI2025

Constrained or unconstrained? Neural-network-based equation discovery from data

Throughout many fields, practitioners often rely on differential equations to model systems. Yet, for many applications, the theoretical derivation of such equations and/or the accurate resolution of their solutions may be intractable. Instead, recently developed methods, including those based on parameter estimation, operator subset selection, and neural networks, allow for the data-driven discovery of both ordinary and partial differential equations (PDEs), on a spectrum of interpretability. The success of these strategies is often contingent upon the correct identification of representative equations from noisy observations of state variables and, as importantly and intertwined with that, the mathematical strategies utilized to enforce those equations. Specifically, the latter has been commonly addressed via unconstrained optimization strategies. Representing the PDE as a neural network, we propose to discover the PDE (or the associated operator) by solving a constrained optimization problem and using an intermediate state representation similar to a physics-informed neural network (PINN). The objective function of this constrained optimization problem promotes matching the data, while the constraints require that the discovered PDE is satisfied at a number of spatial collocation points. We present a penalty method and a widely used trust-region barrier method to solve this constrained optimization problem, and we compare these methods on numerical examples. Our results on several example problems demonstrate that the latter constrained method outperforms the penalty method, particularly for higher noise levels or fewer collocation points. This work motivates further exploration into using sophisticated constrained optimization methods in scientific machine learning, as opposed to their commonly used, penalty-method or unconstrained counterparts. For both of these methods, we solve these discovered neural network PDEs with classical methods, such as finite difference methods, as opposed to PINNs-type methods relying on automatic differentiation. Here, we briefly highlight how simultaneously fitting the data while discovering the PDE improves the robustness to noise and other small, yet crucial, implementation details.

Data-driven discovery↗