Search NASA⌕ Search

SEARCH · Search NASA

Results for “two-timescale”

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.

10 records
All sourcesNASA NTRSDOE OSTI
NASA NTRS2004

A Two-Timescale Discretization Scheme for Collocation

The development of a two-timescale discretization scheme for collocation is presented. This scheme allows a larger discretization to be utilized for smoothly varying state variables and a second finer discretization to be utilized for state variables having higher frequency dynamics. As such. the discretization scheme can be tailored to the dynamics of the particular state variables. In so doing. the size of the overall Nonlinear Programming (NLP) problem can be reduced significantly. Two two-timescale discretization architecture schemes are described. Comparison of results between the two-timescale method and conventional collocation show very good agreement. Differences of less than 0.5 percent are observed. Consequently. a significant reduction (by two-thirds) in the number of NLP parameters and iterations required for convergence can be achieved without sacrificing solution accuracy.

Desai, Prasun↗
NASA NTRS2005

Six-Degree-of-Freedom Trajectory Optimization Utilizing a Two-Timescale Collocation Architecture

Six-degree-of-freedom (6DOF) trajectory optimization of a reentry vehicle is solved using a two-timescale collocation methodology. This class of 6DOF trajectory problems are characterized by two distinct timescales in their governing equations, where a subset of the states have high-frequency dynamics (the rotational equations of motion) while the remaining states (the translational equations of motion) vary comparatively slowly. With conventional collocation methods, the 6DOF problem size becomes extraordinarily large and difficult to solve. Utilizing the two-timescale collocation architecture, the problem size is reduced significantly. The converged solution shows a realistic landing profile and captures the appropriate high-frequency rotational dynamics. A large reduction in the overall problem size (by 55%) is attained with the two-timescale architecture as compared to the conventional single-timescale collocation method. Consequently, optimum 6DOF trajectory problems can now be solved efficiently using collocation, which was not previously possible for a system with two distinct timescales in the governing states.

Desai, Prasun N.↗
DOE OSTI2024

Mixed-mode oscillations in a three-timescale coupled Morris–Lecar system

Mixed-mode oscillations (MMOs) are complex oscillatory behaviors of multiple-timescale dynamical systems in which there is an alternation of large-amplitude and small-amplitude oscillations. It is well known that MMOs in two-timescale systems can arise either from a canard mechanism associated with folded node singularities or a delayed Andronov–Hopf bifurcation (DHB) of the fast subsystem. While MMOs in two-timescale systems have been extensively studied, less is known regarding MMOs emerging in three-timescale systems. In this work, we examine the mechanisms of MMOs in coupled Morris–Lecar neurons with three distinct timescales. We investigate two kinds of MMOs occurring in the presence of a singularity known as canard-delayed-Hopf (CDH) and in cases where CDH is absent. In both cases, we examine how features and mechanisms of MMOs vary with respect to variations in timescales. Our analysis reveals that MMOs supported by CDH demonstrate significantly stronger robustness than those in its absence. Moreover, we show that the mere presence of CDH does not guarantee the occurrence of MMOs. This work yields important insights into conditions under which the two separate mechanisms in two-timescale context, canard and DHB, can interact in a three-timescale setting and produce more robust MMOs, particularly against timescale variations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Finite-Time Analysis of Whittle Index based Q-Learning for Restless Multi-Armed Bandits with Neural Network Function Approximation

Whittle index policy is a heuristic to the intractable restless multi-armed bandits (RMAB) problem. Although it is provably asymptotically optimal, finding Whittle indices remains difficult. In this paper, we present Neural-Q-Whittle, a Whittle index based Q-learning algorithm for RMAB with neural network function approximation, which is an example of nonlinear two-timescale stochastic approximation with Q-function values updated on a faster timescale and Whittle indices on a slower timescale. Despite the empirical success of deep Q-learning, the non-asymptotic convergence rate of Neural-Q-Whittle, which couples neural networks with two-timescale Q-learning largely remains unclear. This paper provides a finite-time analysis of Neural-Q-Whittle, where data are generated from a Markov chain, and Q-function is approximated by a ReLU neural network. Our analysis leverages a Lyapunov drift approach to capture the evolution of two coupled parameters, and the nonlinearity in value function approximation further requires us to characterize the approximation error. Combing these provide Neural-Q-Whittle with convergence rate, where is the number of iterations.

reinforcement learning, structured learning, conve↗
DOE OSTI2026

Non-Markovian relaxation spectroscopy of fluxonium qubits

Recent studies have shown that parasitic two-level systems (TLS) in superconducting qubits, which are a leading source of decoherence, can have relaxation times longer than the qubits themselves. However, the standard techniques used to characterize qubit relaxation is only valid for measuring T 1 under the Born-Markov approximation and could mask environmental memory effects in practice. Here, we introduce two-timescale relaxometry, a technique to probe the qubit and environment relaxation simultaneously and efficiently. We apply it to high-coherence fluxonium qubits over a frequency range of 0.1-0.4 GHz, and reveal a discrete spectrum of TLS with millisecond lifetimes. Our analysis of the spectrum is consistent with a random distribution of TLS in the aluminum oxide tunnel barrier of the Josephson junction chain of the fluxonium, with a spectral and volumetric density and average electric dipole similar to previous TLS studies at much higher frequencies. Our study suggests that investigating and mitigating TLS in the junction chain is crucial to the development of various types of noise-protected qubits in circuit QED.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2025

A Novel Secondary Frequency Regulation with Optimal Priority Selection of AGC Contributions

Automatic generation control (AGC) is used to maintain acceptable frequencies during operation owing to fluctuations in load and variable resources. In conventional industry applications, the AGC signal is allocated to each generator according to the predispatched frequency regulation capacity or the order of economic efficiency. However, with the increasing integration of inverter-based resources (IBRs), the retirement of conventional synchronous generators (SGs) has posed new challenges to frequency control schemes because fewer of them are optional for AGC regulation. In this paper, we propose a novel model predictive control (MPC)-based frequency regulation model to reduce control cost and ensure stability, by considering different critical dynamic factors when optimally selecting the AGC units. The proposed control model – developed in a general form – comprehensively embeds characteristics such as generator ramping rates, reserve capacity, and operation cost. The model predictive control–based two-timescale AGC scheme enhances the capability of immunizing the power disturbance from types of resources by coordinating the control signals between faster IBRs and slower SGs. The case study’s proposed model is verified to be effective in synergistically enforcing different dynamic properties of AGC units into the frequency regulation scheme.

Jiang, Sufan [The University of North Carolina at ↗
NASA NTRS1992

Closed form solutions of constrained trajectories - Application in optimal ascent of aerospace plane

The present consideration of the flight trajectory of hypersonic aerospace vehicles subject to a class of path constraints notes the constrained dynamics to constitute a natural two-timescale system, so that problems of trajectory optimization and guidance can be dramatically simplified by means of the asymptotic analytical solutions thus obtained. An illustrative application in ascent trajectory optimization for an aerospace vehicle is presented.

Lu, Ping↗
NASA NTRS1995

The limitation and applicability of Musher-Sturman equation to two dimensional lower hybrid wave collapse

The existence of localized regions of intense lower hybrid waves in the auroral ionosphere recently observed by rocket and satellite experiments can be understood by the study of a non-linear two-timescale coupling process. In this Letter, we demonstrate that the leading non-linear term in the standard Musher-Sturman equation vanishes identically in strict two-dimensions (normal to the magnetic field). Instead, the new two-dimensional equation is characterized by a much weaker non-linear term which arises from the ponderomotive force perpendicular to the magnetic field, particularly that due to the ions. The old and new equations are compared by means of time-evolution calculations of wave fields. The results exhibit a remarkable difference in the evolution of the waves as governed by the two equations. Such dissimilar outcomes motivate our investigation of the limitation of Musher-Sturman equation in quasi-two-dimensions. Only within all these limits can Musher-Sturman equation adequately describe the collapse of lower hybrid waves.

Tam, Sunny W. Y.↗
NASA NTRSDate not supplied

Coupled Stratospheric Ozone and Atlantic Meridional Overturning Circulation Feedbacks on the Northern Hemisphere Midlatitude Jet Response to 4xCO(2)

Stratospheric ozone, and its response to anthropogenic forcings, provide an important pathway for the coupling between atmospheric composition and climate. In addition to stratospheric ozone’s radiative impacts, recent studies have shown that changes in the ozone layer due to 4xCO2 have a considerable impact on the Northern Hemisphere (NH) tropospheric circulation inducing an equatorward shift of the North Atlantic jet during boreal winter. Here we show that this equatorward jet shift can induce a more rapid weakening of the Atlantic Meridional Overturning Circulation (AMOC), resulting in a poleward shift of the midlatitude eddy-driven jet on longer timescales. As such, coupled feedbacks from both stratospheric ozone and the AMOC result in a two-timescale response of the NH midlatitude jet to abrupt 4xCO2 forcing: a “fast” response (5-20 years) during which it shifts equatorward and a “total” response (∼100-150 years) during which the jet shifts poleward. The latter is driven by a weakening of the AMOC that develops in response to weaker surface zonal winds, that result in reduced heat fluxes out of the subpolar gyre and reduced North Atlantic Deep Water formation. Our results suggest that stratospheric ozone changes in the lower stratosphere can have a surprisingly powerful effect on the AMOC, independent of other aspects of climate change.

Climate models↗
DOE OSTI2023

Bi-Level Adaptive Storage Expansion Strategy for Microgrids Using Deep Reinforcement Learning

Battery energy storage (BES) is a versatile resource for the secure and economic operation of microgrids (MGs). Prevailing stochastic optimization-based approaches for BES expansion planning for MGs are computationally complicated. This work proposes a data-driven bi-level multi-period BES expansion planning framework to determine the siting, sizing, and timing of BES installations. The proposed planning framework unifies deep reinforcement learning (DRL) and linear programming, thereby decoupling the determinations for the integer and continuous decision variables in two time scales, respectively. In the upper level, a rainbow DRL agent with quantile regression is trained to provide dynamic planning policies to accommodate stochastic renewable energy resources (RESs), load, and battery price changes efficiently. Further, the lower level computes the optimal operation of MGs with frequency constraints to hedge the islanding contingency. The two levels communicate with one another by exchanging storage configuration and operating expenses in order to accomplish the shared goal of minimizing investment and operation costs. Comparative case studies on an MG are carried out to demonstrate the superiority of the proposed DRL-based solution to the mixed-integer linear programming counterpart on efficiency, scalability, and adaptability.

24 POWER TRANSMISSION AND DISTRIBUTION↗