Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “System dynamics”

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 235 records · Page 13

Forward variable selection enables fast and accurate dynamic system identification with Karhunen-Loève decomposed Gaussian processes

A promising approach for scalable Gaussian processes (GPs) is the Karhunen-Loève (KL) decomposition, in which the GP kernel is represented by a set of basis functions which are the eigenfunctions of the kernel operator. Such decomposed kernels have the potential to be very fast, and do not depend on the selection of a reduced set of inducing points. However KL decompositions lead to high dimensionality, and variable selection thus becomes paramount. This paper reports a new method of forward variable selection, enabled by the ordered nature of the basis functions in the KL expansion of the Bayesian Smoothing Spline ANOVA kernel (BSS-ANOVA), coupled with fast Gibbs sampling in a fully Bayesian approach. It quickly and effectively limits the number of terms, yielding a method with competitive accuracies, training and inference times for tabular datasets of low feature set dimensionality. Theoretical computational complexities are O ( N P 2 ) in training and O ( P ) per point in inference, where N is the number of instances and P the number of expansion terms. The inference speed and accuracy makes the method especially useful for dynamic systems identification, by modeling the dynamics in the tangent space as a static problem, then integrating the learned dynamics using a high-order scheme. The methods are demonstrated on two dynamic datasets: a ‘Susceptible, Infected, Recovered’ (SIR) toy problem, along with the experimental ‘Cascaded Tanks’ benchmark dataset. Comparisons on the static prediction of time derivatives are made with a random forest (RF), a residual neural network (ResNet), and the Orthogonal Additive Kernel (OAK) inducing points scalable GP, while for the timeseries prediction comparisons are made with LSTM and GRU recurrent neural networks (RNNs) along with the SINDy package.

Hayes, Kyle↗

Variational properties of a pumped dynamical system

We earlier constructed a generalized entropy concept to show the direction of time in an evolution following from a Markov generator. In such a dynamical system the entity found changes in a monotonic way starting from any initial state of the system. In this article, we generalize the treatment to the case when population is pumped into the system from levels not explicitly considered. These populations then pass through the coupled levels and exit by decay to levels outside the system. We derive the form of the equation of motion and relate it to our earlier treatments. It turns out that the formalism can be generalized to the new situation. Its physically relevant features are demonstrated and the behavior obtained is illustrated by a numerical treatment of the standard two-level system with pumping and relaxation included.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Data-driven modeling and control of dynamical systems using Koopman and Perron-Frobenius operators

This dissertation studies the data-driven modeling and control problem of nonlinear systems by exploiting the linear operator theoretic framework involving Koopman and Perro-Frobenius operator. A systematic linear-operator based controller design procedure has been established, which can be used to solve a variety of nonlinear control problems, including feedback stabilization using control Lyapunov functions, optimal quadratic regulation using Koopman eigenfunctions and convex optimization formulation of optimal control problem using P-F and Koopman operator approximation. As the core of data-driven modeling, we first propose a new algorithm for the finite-dimensional approximation of the linear transfer Koopman and Perron-Frobenius operator from time-series data. We argue that the existing approach for the finite-dimensional approximation of these transfer operators such as Dynamic Mode Decomposition (DMD) and Extended Dynamic Mode Decomposition (EDMD) do not capture two important properties of these operators, namely positivity and Markov property. The algorithm we propose preserves these two properties. We call the proposed algorithm as naturally structured DMD (NSDMD) since it retains the inherent properties of these operators. Naturally structured DMD algorithm leads to a better approximation of the steady-state dynamics of the system regarding computing Koopman and Perron- Frobenius operator eigenfunctions and eigenvalues. However, preserving positivity property is critical for capturing the real transient dynamics of the system. This positivity property of the transfer operators and it's finite-dimensional approximation play an important role for controller and estimator design of nonlinear systems. To solve the feedback stabilization problem for nonlinear control systems, we tried to take advantage of the Koopman operator framework. The Koopman operator approach provides a linear representation for a nonlinear dynamical system and a bilinear representation for a nonlinear control system. The problem of feedback stabilization of a nonlinear control system is then transformed to the stabilization of a bilinear control system. We propose a control Lyapunov function (CLF)-based approach for the design of stabilizing feedback controllers for the bilinear system. The search for finding a CLF for the bilinear control system is formulated as a convex optimization problem. This leads to a schematic procedure for designing CLF-based stabilizing feedback controllers for the bilinear system and hence the original nonlinear system. Another advantage of the proposed controller design approach outlined in this dissertation is that it does not require explicit knowledge of system dynamics. In particular, the bilinear representation of a nonlinear control system in the Koopman eigenfunction space can be obtained from time-series data. Next, we study the optimal quadratic regulation problem for nonlinear systems. The linear operator theoretic framework involving the Koopman operator is used to lift the dynamics of nonlinear control system to an infinite-dimensional bilinear system. The optimal quadratic regulation problem for nonlinear system is formulated in terms of the finite-dimensional approximation of the bilinear system. A convex optimization-based approach is proposed for solving the quadratic regulator problem for bilinear system. We applied a variety of examples and compared the simulation results between our framework and conventional LQR control using linearized model. For more general optimal control problems, we provide a density-function based convex formulation for the optimal control problem of the nonlinear system. The convex formulation relies on the duality result in the stability theory of a dynamical system involving density function and Perron-Frobenius operator. The optimal control problem is formulated as an infinite-dimensional convex optimization program. The finite-dimensional approximation of the optimization problem relies on the recent advances made in the data-driven computation of the Koopman operator, which is dual to the Perron-Frobenius operator. Simulation results are presented to demonstrate the application of the developed framework.

Huang, Bowen↗

Efficient sensitivity analysis method for chaotic dynamical systems

The direct differentiation and improved least squares shadowing methods are both developed for accurately and efficiently calculating the sensitivity coefficients of time averaged quantities for chaotic dynamical systems. The key idea is to recast the time averaged integration term in the form of differential equation before applying the sensitivity analysis method. An additional constraint-based equation which forms the augmented equations of motion is proposed to calculate the time averaged integration variable and the sensitivity coefficients are obtained as a result of solving the augmented differential equations. The application of the least squares shadowing formulation to the augmented equations results in an explicit expression for the sensitivity coefficient which is dependent on the final state of the Lagrange multipliers. The LU factorization technique to calculate the Lagrange multipliers leads to a better performance for the convergence problem and the computational expense. Numerical experiments on a set of problems selected from the literature are presented to illustrate the developed methods. The numerical results demonstrate the correctness and effectiveness of the present approaches and some short impulsive sensitivity coefficients are observed by using the direct differentiation sensitivity analysis method.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Using System Dynamics to Define, Study, and Implement Smart Control Strategies on the Electric Power Grid

The United States electric power grid is the most complex and expansive control system in the world. Local generation control occurs at individual units based on response time and unit economics, larger regional control coordinates unit response to error conditions, and high level large-area regional control is ultimately administered by a network of humans guided by economic and resiliency related factors. Under normal operating conditions, the grid is a relatively slow moving entity that exhibits high inertia to outside stimuli, and behaves along repeatable diurnal and seasonal patterns. However, that paradigm is quickly changing because of the increasing implementation of renewable generation sources. Renewable generators by nature cannot be tightly controlled or scheduled. They appear like a negative load to the system with all of the variability associated with load on a larger scale. Also, grid-reactive loads (i.e. smart devices) can alter their consumption based on price or demand rules adding more variability to system behavior. This paper demonstrates how a systems dynamic modeling approach capable of operating over multiple time scales, can provide valuable insight into developing new “smart-grid” control strategies and devices needed to accommodate renewable generation and regulate the frequency of the grid.

14 SOLAR ENERGY↗

Temporal Characterization of Hydrates System Dynamics beneath Seafloor Mounds. Integrating Time-Lapse Electrical Resistivity Methods and In Situ Observations of Multiple Oceanographic Parameters

This study was designed to investigate temporal variations in hydrate system dynamics by measuring changes in volumes of hydrate beneath hydrate-bearing mounds on the continental slope of the northern Gulf of Mexico, the landward extreme of hydrate occurrence in this region. Direct Current Resistivity (DCR) measurements were made contemporaneously with measurements of oceanographic parameters at Woolsey Mound, a carbonate-hydrate complex on the mid-continental slope, where formation and dissociation of hydrates are most vulnerable to variations in oceanographic parameters affected by climate change, and where changes in hydrate stability can readily translate to loss of seafloor stability, impacts to benthic ecosystems, and venting of greenhouse gases to the water-column, and eventually, the atmosphere. We focused our study on hydrate within seafloor mounds because the structurally-focused methane flux at these sites likely causes hydrate formation and dissociation processes to occur at higher rates than at sites where the methane flux is less concentrated and we wanted to maximize our chances of witnessing association/dissociation of hydrates. We selected a particularly well-studied hydrate-bearing seafloor mound near the landward extent of the hydrate stability zone, Woolsey Mound (MC118). This mid-slope site has been studied extensively and the project was able to leverage considerable resources from the team’s research experience at MC118. The site exhibits seafloor features associated with gas expulsion, hydrates have been documented at the seafloor, and changes in the outcropping hydrates have been documented, photographically, to have occurred over a period of months. We conducted observatory-based, in situ measurements to 1) characterize, geophysically, the sub-bottom distribution of hydrate and its temporal variability, and 2) contemporaneously record relevant environmental parameters (temperature, pressure, salinity, turbidity, bottom currents) to detect short-term changes within the hydrates system, identify relationships/impacts of local oceanographic parameters on the hydrates system, and improve our understanding of how seafloor instability is affected by hydrates-driven changes. A 2009 DCR survey of MC118 demonstrated that we could image resistivity anomalies to a depth of 75m below the seafloor in water depths of 1km. We reconfigured this system to operate autonomously on the seafloor in a pre-programmed mode, for periods of months. We designed and built a novel seafloor lander and deployment capability that would allow us to investigate the seafloor at potential deployment sites and deploy instruments only when conditions met our criteria. This lander held the DCR system, controlling computers, and battery power supply, as well as instruments to record oceanographic parameters. During the first of two cruises to the study site, we conducted resistivity surveying, selected a monitoring site, and deployed the instrumented lander and DCR, centered on what appeared to be the most active locations within the site, programmed to collect a DCR profile, weekly. After a 4.5-month residence on the seafloor, the team recovered all equipment. Unfortunately, several equipment failures occurred prior to recovery of the instrument packages. Prior to the failures, however, two resistivity profiles were collected together with oceanographic data. Results show, unequivocally, that significant changes can occur in both hydrate volume and distribution during time periods as brief as one week. Occurrences appear to be controlled by both deep and near-surface structure. Results have been integrated with seismic data from the area and show correspondence in space of hydrate and structures, including faults and gas chimneys.

58 GEOSCIENCES↗

Induced topological pressure for topological dynamical systems

In this paper, inspired by the article [J. Jaerisch et al., Stochastics Dyn. 14, 1350016, pp. 1-30 (2014)], we introduce the induced topological pressure for a topological dynamical system. In particular, we prove a variational principle for the induced topological pressure.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Flow-driven spectral chaos (FSC) method for long-time integration of second-order stochastic dynamical systems

For decades, uncertainty quantification techniques based on the spectral approach have been demonstrated to be computationally more efficient than the Monte Carlo method for a wide variety of problems, particularly when the dimensionality of the probability space is relatively low. The time-dependent generalized polynomial chaos (TD-gPC) is one such technique that uses an evolving orthogonal basis to better represent the stochastic part of the solution space in time. Here in this paper, we present a new numerical method that uses the concept of enriched stochastic flow maps to track the evolution of the stochastic part of the solution space in time. The computational cost of this proposed flow-driven stochastic chaos (FSC) method is an order of magnitude lower than TD-gPC for comparable solution accuracy. This gain in computational cost is realized because, unlike most existing methods, the number of basis vectors required to track the stochastic part of the solution space does not depend upon the dimensionality of the probability space. Four representative numerical examples are presented to demonstrate the performance of the FSC method for long-time integration of second-order stochastic dynamical systems in the context of stochastic dynamics of structures.

FSC↗

Quantum tomography meets dynamical systems and bifurcations theory

A powerful tool for studying geometrical problems in Hilbert spaces is developed. We demonstrate the convergence and robustness of our method in every dimension by considering dynamical systems theory. This method provides numerical solutions to hard problems involving many coupled nonlinear equations in low and high dimensions (e.g., quantum tomography problem, existence and classification of Pauli partners, mutually unbiased bases, complex Hadamard matrices, equiangular tight frames, etc.). Additionally, this tool can be used to find analytical solutions and also to implicitly prove the existence of solutions. Here, we develop the theory for the quantum pure state tomography problem in finite dimensions but this approach is straightforwardly extended to the rest of the problems. We prove that solutions are always attractive fixed points of a nonlinear operator explicitly given. As an application, we show that the statistics collected from three random orthonormal bases is enough to reconstruct pure states from experimental (noisy) data in every dimension d ⩽ 32.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A greedy Galerkin method to efficiently select sensors for linear dynamical systems

A key challenge in inverse problems is the selection of sensors to gather the most effective data. In this paper, we consider the problem of inferring the initial condition to a linear dynamical system and develop an efficient control-theoretical approach for greedily selecting sensors. Our method employs a Galerkin projection to reduce the size of the inverse problem, resulting in a computationally efficient algorithm for sensor selection. As a byproduct of our algorithm, we obtain a preconditioner for the inverse problem that enables the rapid recovery of the initial condition. Here, we analyze the theoretical performance of our greedy sensor selection algorithm as well as the performance of the associated preconditioner. Finally, we verify our theoretical results on various inverse problems involving partial differential equations.

97 MATHEMATICS AND COMPUTING↗

Fast and Scalable FFT-Based GPU-Accelerated Algorithms for Block-Triangular Toeplitz Matrices with Application to Linear Inverse Problems Governed by Autonomous Dynamical Systems

In this work, we present an efficient and scalable algorithm for performing matrix-vector multiplications (matvecs) for block Toeplitz matrices. Such matrices, which are shift-invariant with respect to their blocks, arise in the context of solving inverse problems governed by autonomous systems, and time-invariant systems in particular. In this article, we consider inverse problems that infer unknown parameters from observational data of a linear time-invariant dynamical system given in the form of partial differential equations (PDEs). Matrix-free Newton-conjugate-gradient methods are often the gold standard for solving these inverse problems, but they require numerous actions of the Hessian on a vector. Matrix-free adjoint-based Hessian matvecs require solution of a pair of linearized forward/adjoint PDE solves per Hessian action, which may be prohibitive for large-scale inverse problems. Time invariance of the forward PDE problem leads to a block Toeplitz structure of the discretized parameter-to-observable (p2o) map defining the mapping from inputs (parameters) to outputs (observables) of the PDEs. This block Toeplitz structure enables us to exploit two key properties: (1) compact storage of the p2o map and its adjoint, and (2) efficient fast Fourier transform–based Hessian matvecs. The proposed algorithm is mapped onto large multi-GPU clusters and achieves more than 80% of peak bandwidth on NVIDIA A100 GPUs. Excellent weak scaling is shown for up to 48 A100 GPUs. For the targeted problems, the implementation executes Hessian matvecs within fractions of a second, which is orders of magnitude faster than can be achieved by conventional matrix-free Hessian matvecs via forward/adjoint PDE solves.

97 MATHEMATICS AND COMPUTING↗

Quantifying Information without Entropy: Identifying Intermittent Disturbances in Dynamical Systems

A system’s response to disturbances in an internal or external driving signal can be characterized as performing an implicit computation, where the dynamics of the system are a manifestation of its new state holding some memory about those disturbances. Identifying small disturbances in the response signal requires detailed information about the dynamics of the inputs, which can be challenging. This paper presents a new method called the Information Impulse Function (IIF) for detecting and time-localizing small disturbances in system response data. The novelty of IIF is its ability to measure relative information content without using Boltzmann’s equation by modeling signal transmission as a series of dissipative steps. Since a detailed expression of the informational structure in the signal is achieved with IIF, it is ideal for detecting disturbances in the response signal, i.e., the system dynamics. Those findings are based on numerical studies of the topological structure of the dynamics of a nonlinear system due to perturbated driving signals. The IIF is compared to both the Permutation entropy and Shannon entropy to demonstrate its entropy-like relationship with system state and its degree of sensitivity to perturbations in a driving signal.

42 ENGINEERING↗

Generic solar photovoltaic system dynamic simulation model specification

This document is intended to serve as a specification for generic solar photovoltaic (PV) system positive-sequence dynamic models to be implemented by software developers and approved by the WECC MVWG for use in bulk system dynamic simulations in accordance with NERC MOD standards. Two specific dynamic models are included in the scope of this document. The first, a Central Station PV System model, is intended to capture the most important dynamic characteristics of large scale (> 10 MW) PV systems with a central Point of Interconnection (POI) at the transmission level. The second, a Distributed PV System model, is intended to represent an aggregation of smaller, distribution-connected systems that comprise a portion of a composite load that might be modeled at a transmission load bus.

14 SOLAR ENERGY↗

Data-driven Resilience Characterization of Control Dynamical Systems

In this paper, we define and quantify resiliency of a power network and propose data-driven algorithms for computing the same for the power grid. To do this, we use the Koopman operator framework to lift the controlled dynamical system to an abstract (possibly higher) dimensional space, where the evolution is linear. The linear system representation allows us to relate small time local controllability and observability of a general nonlinear control system to the controllability and observability of the lifted linear system. Finally, we define the resiliency of the underlying power grid in terms of the controllability and observability gramians of the lifted linear system. We illustrate the proposed approach to compute the resiliency metrics on time-series data obtained from a microgrid.

koopman operator, resilience, control↗

Modeling a fast spectrum molten salt reactor in a systems dynamics fuel cycles code

Molten salt reactor (MSR) concepts are being actively developed by several private companies, with some touting the fuel cycle benefits of this technology. Simulating the deployment of MSRs and their effect on fuel cycle transitions is the role of a system dynamics fuel cycles tool. However, many current fuel cycles tools have difficulty modeling MSR isotopic variations because they were designed for modeling existing reactors based on discrete fuel elements with batch fueling schemes (i.e., charge and discharge). This work presents ORION fuel cycle modeling results using a specific high power density fast spectrum MSR design. For code verification, a single-stage MSR fuel cycle was set up in ORION and results were compared with the results from the SCALE-based reactor physics model (ChemTriton). A transition scenario was set up in ORION from the current light water reactor fleet to a future fleet of MSRs to study the characteristics of an MSR that affect its performance during transition. Several transition scenarios were set with holds and delays in place that are representative of a sodium fast reactor as modeled in Evaluation Group 23 for the Fuel Cycle Options Campaign Evaluation and Screening Study. Several MSR characteristics lend to its high performance and ease in transition from the current fuel cycle: fissile material is only loaded at initial startup; all fissile material is available as it is bred due to online fuel processing; and fissile material recirculating through the core is briefly held in tanks for processing and has very minimal out-of-core time (on the order of a minutes to a few days). These characteristics are in contrast to a solid-fueled reactor, which requires discharge, cooling, and fabrication of fuel all of which could take a few months to a few years and repeated loading of fresh fuel assemblies. In conclusion, a solid-fueled reactor only discharges a fraction of a core every year.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Combining Agent Based Modeling and System Dynamics to Investigate the Circularity of Plastics

The United States currently produces about 1 million metric ton of ocean plastic pollution annually. One proposed solution to combat plastic waste is a circular economy (CE), which aims to transition from today's take-make-waste linear pattern of production and consumption to a system where the value of resources is maximized over time. Two key methods in industrial ecology are useful in assessing the viability of CE: (1) System Dynamics (SD) and (2) Agent Based Modeling (ABM). In prior work, the plastic life cycle was modeled with SD and ABM. The two models calculate recycling rates and costs in different ways, making it difficult to pinpoint necessary next steps. We integrate the ABM and SD models - linking the emergent patterns from micro-level human decisions to system level processes - which allows a more comprehensive understanding of feedbacks, costs, and environmental impacts. The integrated model is more accurate, and can be used to visualize recycling rates and human health and environmental impacts over time. The difference between the integrated and original SD model prompts a Sobol sensitivity analysis, which is used to understand which behavioral factors most affect plastic recycling patterns. We find that the habitual component is typically the most influential in promoting positive recycling behavior. Additionally, we utilize the combined model to understand and visualize how various behavioral intervention scenarios, like improved access to recycling programs and cart tagging, influence recycling patterns; these results can guide future policy-making.

agent-based modeling↗

Dynamic System Scaling Applied to Zr-4 Cladding under RIA Conditions

The research and development lifecycle for new fuel designs currently has an ideal 20-25 year timeline. Efforts are currrently underway to accelerate fuel design and development using high performance fuel modeling codes and reduced scale experiments with new instrumentation techniques. However current tools available require a full scale prototype to be designed and tested for fuel qualification. The thermal-hydraulics community has leveraged scaling methodologies to use reduced scale integral effect test facilities (Westinghouse AP600/1000, NuScale Power Module) in lieu of building full scale prototypes for reactor design certification. Dynamic System Scaling (DSS) is a recent development to scaling methodologies that enable the tracking and quantification of dynamic scaling distortions experienced throughout a transient. Recent TREAT experiments performed reactivity initiated accident (RIA) pulses on UO2 fuel rodlets in Zr-4 cladding and were accompanied by BISON predictions for the cladding temperature. DSS is applied to pyrometer data of outer cladding temperatures and two BISON predictions after peak cladding temperature has been reached. One BISON case uses a constant gap conductance value between the fuel and cladding and the other using a conventional gap conductance model. Results from the DSS analysis show that there is good agreement between both BISON predictions and the experimental data and that there is not a significant distortion seen between using either gap conductance model.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗