SEARCH · Search NASA

Results for “Optimal Control”

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 523 records · Page 29

All sources NASA NTRS DOE OSTI

DOE OSTI2016

Discrete Time McKean–Vlasov Control Problem: A Dynamic Programming Approach

We consider the stochastic optimal control problem of nonlinear mean-field systems in discrete time. We reformulate the problem into a deterministic control problem with marginal distribution as controlled state variable, and prove that dynamic programming principle holds in its general form. We apply our method for solving explicitly the mean-variance portfolio selection and the multivariate linear-quadratic McKean–Vlasov control problem.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Leveraging Multitime Hamilton–Jacobi PDEs for Certain Scientific Machine Learning Problems

Hamilton-Jacobi partial differential equations (HJ PDEs) have deep connections with a wide range of fields, including optimal control, differential games, and imaging sciences. By considering the time variable to be a higher dimensional quantity, HJ PDEs can be extended to the multi-time case. In this paper, we establish a novel theoretical connection between specific optimization problems arising in machine learning and the multi-time Hopf formula, which corresponds to a representation of the solution to certain multi-time HJ PDEs. Through this connection, we increase the interpretability of the training process of certain machine learning applications by showing that when we solve these learning problems, we also solve a multi-time HJ PDE and, by extension, its corresponding optimal control problem. As a first exploration of this connection, we develop the relation between the regularized linear regression problem and the Linear Quadratic Regulator (LQR). We then leverage our theoretical connection to adapt standard LQR solvers (namely, those based on the Riccati ordinary differential equations) to design new training approaches for machine learning. Lastly, we provide some numerical examples that demonstrate the versatility and possible computational advantages of our Riccati-based approach in the context of continual learning, post-training calibration, transfer learning, and sparse dynamics identification.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2016

Identification of the Diffusion Parameter in Nonlocal Steady Diffusion Problems

The problem of identifying the diffusion parameter appearing in a nonlocal steady diffusion equation is considered. The identification problem is formulated as an optimal control problem having a matching functional as the objective of the control and the parameter function as the control variable. The analysis makes use of a nonlocal vector calculus that allows one to define a variational formulation of the nonlocal problem. In a manner analogous to the local partial differential equations counterpart, we demonstrate, for certain kernel functions, the existence of at least one optimal solution in the space of admissible parameters. We introduce a Galerkin finite element discretization of the optimal control problem and derive a priori error estimates for the approximate state and control variables. Using one-dimensional numerical experiments, we illustrate the theoretical results and show that by using nonlocal models it is possible to estimate non-smooth and discontinuous diffusion parameters.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2015

SU-E-T-266: Development of Evaluation System of Optimal Synchrotron Controlling Parameter for Spot Scanning Proton Therapy with Multiple Gate Irradiations in One Operation Cycle

Purpose: We have developed a gated spot scanning proton beam therapy system with real-time tumor-tracking. This system has the ability of multiple-gated irradiation in a single synchrotron operation cycle controlling the wait-time for consecutive gate signals during a flat-top phase so that the decrease in irradiation efficiency induced by irregular variation of gate signal is reduced. Our previous studies have shown that a 200 ms wait-time is appropriate to increase the average irradiation efficiency, but the optimal wait-time can vary patient by patient and day by day. In this research, we have developed an evaluation system of the optimal wait-time in each irradiation based on the log data of the real-time-image gated proton beam therapy (RGPT) system. Methods: The developed system consists of logger for operation of RGPT system and software for evaluation of optimal wait-time. The logger records timing of gate on/off, timing and the dose of delivered beam spots, beam energy and timing of X-ray irradiation. The evaluation software calculates irradiation time in the case of different wait-time by simulating the multiple-gated irradiation operation using several timing information. Actual data preserved in the log data are used for gate on and off time, spot irradiation time, and time moving to the next spot. Design values are used for the acceleration and deceleration times. We applied this system to a patient treated with the RGPT system. Results: The evaluation system found the optimal wait-time of 390 ms that reduced the irradiation time by about 10 %. The irradiation time with actual wait-time used in treatment was reproduced with accuracy of 0.2 ms. Conclusion: For spot scanning proton therapy system with multiple-gated irradiation in one synchrotron operation cycle, an evaluation system of the optimal wait-time in each irradiation based on log data has been developed. Funding Support: Japan Society for the Promotion of Science (JSPS) through the FIRST Program.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2024

A Networked Microgrid Framework and Testbed for Communication, Controls, and Optimization Testing

This paper presents the development and experimental results of a networked AC microgrid testbed located at Oak Ridge National Laboratory. The testbed comprises two, 480V three-phase, four wire microgrids designed to operate standalone, grid-tied, or as a network of microgrids. This testbed represents both the state of the industry, by incorporating grid-assets commonly found in real microgrids, and the state of the art, as it is a platform to evaluate advanced controllers. The main elements of this networked microgrid testbed are presented in this paper including a Scenario Manager, local microgrid controls, and a networked microgrid control. The Scenario Manager has the objective of emulating real-world conditions. The local microgrid controller oversees standalone, grid-tied, or islanded operation. The microgrid control is a higher-level control that coordinates interaction between islanded microgrids. This paper delves into these controllers and validates their operation in the networked microgrid testbed showcasing the operational flexibility and advance control capabilities.

Ferrari Maglia, Max↗

DOE OSTI2014

Using a Ventilation Controller to Optimize Residential Passive Ventilation For Energy and Indoor Air Quality

One way to reduce the energy impact of providing residential ventilation is to use passive and hybrid systems. However, these passive and hybrid (sometimes called mixed-mode) systems must still meet chronic and acute health standards for ventilation. This study uses a computer simulation approach to examine the energy and indoor air quality (IAQ) implications of passive and hybrid ventilation systems, in 16 California climate zones. Both uncontrolled and flow controlled passive stacks are assessed. A new hybrid ventilation system is outlined that uses an intelligent ventilation controller to minimise energy use, while ensuring chronic and acute IAQ standards are met. ASHRAE Standard 62.2-2010 – the United States standard for residential ventilation - is used as the chronic standard, and exposure limits for PM 2.5 , formaldehyde and NO 2 are used as the acute standards. The results show that controlled passive ventilation and hybrid ventilation can be used in homes to provide equivalent IAQ to continuous mechanical ventilation, for less use of energy.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2015

Control strategy optimization of HVAC plants

In this paper we present a methodology to optimize the operating conditions of heating, ventilation and air conditioning (HVAC) plants to achieve a higher energy efficiency in use. Semi-empiric numerical models of the plant components are used to predict their performances as a function of their set-point and the environmental and occupied space conditions. The optimization is performed through a graph-based algorithm that finds the set-points of the system components that minimize energy consumption and/or energy costs, while matching the user energy demands. The resulting model can be used with systems of almost any complexity, featuring both HVAC components and energy systems, and is sufficiently fast to make it applicable to real-time setting.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

DOE OSTI2015

Finite-Dimensional Representations for Controlled Diffusions with Delay

We study stochastic delay differential equations (SDDE) where the coefficients depend on the moving averages of the state process. As a first contribution, we provide sufficient conditions under which the solution of the SDDE and a linear path functional of it admit a finite-dimensional Markovian representation. As a second contribution, we show how approximate finite-dimensional Markovian representations may be constructed when these conditions are not satisfied, and provide an estimate of the error corresponding to these approximations. These results are applied to optimal control and optimal stopping problems for stochastic systems with delay.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2019

Networked Control and Optimization for Widescale Integration of Power Electronic Devices in Residential Homes

Power electronics devices are emerging at residential houses, including electric vehicles, energy storage systems, solar photovoltaic systems, and variable-speed air conditioners. Little work has been done on scalable and end-to-end architectures for integrating and engaging large number of residential behind-the-meter assets to provide grid services. In this paper, a framework for integration of a demand management system, home energy management system, and residential smart inverters is presented. Full measured results of inverter response to control signals is also presented.

Starke, Michael↗

DOE OSTI2024

Implications of stop-and-go traffic on training learning-based car-following control

Learning-based car-following control (LCC) of connected and autonomous vehicles (CAVs) is gaining significant attention with the advancement of computing power and data accessibility. While the flexibility and large model capacity of model-free architecture enable LCC to potentially outperform the model-based car-following (CF) model in improving traffic efficiency and mitigating congestion, the generalizability of LCC for traffic conditions different from the training environment/dataset is not well-understood. Herein, this study seeks to explore the impact of stop-and-go traffic in the training dataset on the generalizability of LCC. It uses the characteristics of lead vehicle trajectories to describe stop-and-go traffic, and links the theory of identifiability (i.e., obtaining a unique parameter estimation result using sensor measurements) to the generalizability of behavior cloning (BC) and policy-based deep reinforcement learning (DRL). Correspondingly, the study shows theoretically that: (i) stop-and-go traffic can enable the property of identifiability and enhance the control performance of BC-based LCC in different traffic conditions; (ii) stop-and-go traffic is not necessary for DRL-based LCC to generalize to different traffic conditions; (iii) DRL-based LCC trained with only constant-speed lead vehicle trajectories (not sufficient to ensure identifiability) can be generalized to different traffic conditions; and (iv) stop-and-go traffic increases variance in the training dataset, which improves the convergence of parameter estimation while negatively impacting the convergence of DRL to the optimal control policy. Numerical experiments validate the above findings, illustrating that BC-based LCC entails comprehensive training datasets for generalizing to different traffic conditions, while DRL-based LCC can achieve generalization with simple free-flow traffic training environments. This further suggests DRL as a more promising and cost-effective LCC approach to reduce operational costs, mitigate traffic congestion, and enhance safety and mobility, which can accelerate the deployment and acceptance of CAVs.

33 ADVANCED PROPULSION SYSTEMS↗

DOE OSTI2018

When Data Analytics Meet Site Operation: Benefits and Challenges

Demand for using data analytics for energy management in buildings is rising. Such analytics are required for advanced measurement and verification, commissioning, automated fault-detection and diagnosis, and optimal control. While novel analytics algorithms continue to be developed, bottlenecks and challenges arise when deploying them for demonstration, for a number of reasons that do not necessarily have to do with the algorithms themselves. It is important for developers of new technologies to be aware of the challenges and potential solutions during demonstration. Therefore, this paper describes a recent deployment of an automated, physical model-based, FDD and optimal control tool, highlighting its design and as-operated benefits that the tool provides. Furthermore, the paper presents challenges faced during deployment and testing along with solutions used to overcome these challenges. The challenges have been grouped into four categories: Data Management, Physical Model Development and Integration, Software Development and Deployment, and Operator Use. The paper concludes by discussing how challenges with this project generalize to common cases, how they could compare to other projects in their severity, and how they may be addressed.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2015

Defining a region of optimization based on engine usage data

Methods and systems for engine control optimization are provided. One or more operating conditions of a vehicle engine are detected. A value for each of a plurality of engine control parameters is determined based on the detected one or more operating conditions of the vehicle engine. A range of the most commonly detected operating conditions of the vehicle engine is identified and a region of optimization is defined based on the range of the most commonly detected operating conditions of the vehicle engine. The engine control optimization routine is initiated when the one or more operating conditions of the vehicle engine are within the defined region of optimization.

33 ADVANCED PROPULSION SYSTEMS↗

DOE OSTI2020

A General Framework for Bounding Approximate Dynamic Programming Schemes

For years, there has been interest in approximation methods for solving dynamic programming problems, because of the inherent complexity in computing optimal solutions characterized by Bellman’s principle of optimality. A wide range of approximate dynamic programming (ADP) methods now exists. It is of great interest to guarantee that the performance of an ADP scheme be at least some known fraction, say ß , of optimal. This letter introduces a general approach to bounding the performance of ADP methods, in this sense, in the stochastic setting. The approach is based on new results for bounding greedy solutions in string optimization problems, where one has to choose a string (ordered set) of actions to maximize an objective function. This bounding technique is inspired by submodularity theory, but submodularity is not required for establishing bounds. Instead, the bounding is based on quantifying certain notions of curvature of string functions; the smaller the curvatures the better the bound. The key insight is that any ADP scheme is a greedy scheme for some surrogate string objective function that coincides in its optimal solution and value with those of the original optimal control problem. The ADP scheme then yields to the bounding technique mentioned above, and the curvatures of the surrogate objective determine the value ß of the bound. The surrogate objective and its curvatures depend on the specific ADP.

discrete event systems↗

DOE OSTI2023

Loose Belt Fault Detection and Virtual Flow Meter Development Using Identified Data-driven Energy Model for Fan Systems

An energy model that correlates fan airflow, head, speed, and system power input is essential to detect device faults and optimize control strategies in fan systems. Since the application of variable-frequency drives (VFDs) makes the motor-efficiency data published by manufacturers inapplicable for VFD–motor–fan systems, the fan efficiency and drive (belt–motor–VFD) efficiency must be identified for each individual system to obtain accurate energy models. The objectives of this paper are to identify an energy model of existing VFD–motor–fan systems using available experimental data and demonstrate its applications in loose belt fault detection and virtual airflow meter development for optimal control. First, an approach is developed to identify the fan head, fan efficiency, and drive-efficiency curves using available fan head, speed, and system power input as well as temporarily measured airflow rate without measuring shaft power. Then, the energy model is identified for an existing VFD–motor–fan system. Finally, the identified model is applied to detect the slipped belt faults and develop the virtual airflow meter. The experiment results reveal that the developed approach can effectively obtain the energy model of VFD–motor–fan systems and the model can be applied to effectively detect slipped belt faults and accurately calculate the fan airflow rate.

42 ENGINEERING↗

DOE OSTI2024

Dislocation half-loop control for optimal V-defect density in GaN-based light emitting diodes

V-defects are morphological defects that typically form on threading dislocations during epitaxial growth of $(0001)$-oriented GaN layers. A V-defect is a hexagonal pyramid-shaped depression with six {$10\bar{1}1$}-oriented sidewalls. These semipolar sidewalls have a lower polarization barrier than the polarization barriers present between the polar c-plane quantum wells and quantum barriers and can laterally inject carriers directly into quantum wells in GaN-based light emitting diodes (LEDs). This is especially important, as the high polarization field in c-plane GaN is a significant factor in the high forward voltage of GaN LEDs. The optimal V-defect density for efficient lateral carrier injection in a GaN LED (∼10 9 cm −2 ) is typically an order of magnitude higher than the threading dislocation density of GaN grown on patterned sapphire substrates (∼10 8 cm −2 ). Pure-edge dislocation loops have been known to exist in GaN, and their formation into large V-defects via low-temperature growth with high Si-doping has recently been studied. Here, in this work, we develop a method for pure-edge threading dislocation half-loop formation and density control via disilane flow, growth temperature, and thickness of the half-loop generation layer. We also develop a method of forming the threading dislocation half-loops into V-defects of comparable size to those originating from substrate threading dislocations.

Crystallographic defects↗

DOE OSTI2025

Design, Optimization, and Control of a 100 kW Electric Traction Motor Meeting or Exceeding DOE 2025 Targets

The overall objective of the electric motor portion of the Electric Drives Technology consortium is to research, develop, and test electric motors for use in electric vehicle applications capable of a peak power greater than 100 kW, power density greater than or equal to 50 kW/l, and a cost less than 3.3 $/kW. To meet the electric traction motor power density and cost targets a number of approaches were pursued simultaneously throughout the course of this project which address all of the major volumetric power density variables. The specific research thrusts at the Illinois Institute of Technology (IIT) are the following: multiphysics design for increased power density through maximum utilization of active materials, synthesis of electric machine windings and PM flux barriers with controlled space harmonics, high slot fill windings for increased current loadings or efficiency, aggressive cooling strategies, and design studies and prototype construction of candidate electric machines.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2020

Interactive Metadata Integration with Brick

Many different digital representations of a building are produced over the course of its lifecycle. While these representations individually contain metadata required to support different stages of the building's lifecycle, they are largely not interoperable due to differences in structure, syntax and semantics. This impedes the development and deployment of data-driven applications providing fault detection and diagnosis, virtual metering or optimal control. [4] introduces a new platform for the continuous curation of a unified, Brick [1]-based metadata model that can be maintained throughout the building lifecycle. In this demonstration, we present a live proof-of-concept implementation of the platform with support for several metadata representations in the context of a simulated building lifecycle.

Fierro, Gabe↗

DOE OSTI2018

Data Driven Temperature Control of Heat Exchangers

The following poster presents a couple of approaches for temperature control at heat exchanger’s exit. One of the approaches is based on using n-step ahead neural network while the other approach utilizes a combination of control + learning module for temperature control. Designing robust controllers for typical heat exchangers is very challenging primarily because of complexities associated with the synthesis of dynamics of industrial heat exchangers. Even if a dynamics model is synthesized, the accuracy of it would be called into question as there can be lot of unmodeled system dynamics and random noise in input parameters. In many situations, the degree of uncertainty in the model of system being controlled limits the utility of optimal control design. Controller can be manually tuned on the field but it is very difficult to determine manual adjustments that result in overall improvement. These practical challenges call for learning based approaches for temperature control that would result in optimal heat exchanger performance. We utilize the existing mathematical model for air-water system synthesized by Underwood and Crawford for testing our algorithms. Underwood and Crawford developed a model of heating coil by fitting a second-order non-linear equations to measurements of air and water temperature and flow rates obtained from the actual coil. The uncertainties in the model include air and water inlet temperatures and air-flow rate. These uncertainties are modeled as random variables that are sampled from a uniform distribution. The control variable is the water flow rate which is controlled for regulating the air-outlet temperature. Neural networks are frequently used in various literatures for learning time-series. We propose a modified version of neural networks known as n-step ahead neural network that take the state variables and output at time t as input to successfully predict output at time t+n. The variable n is a hyper-parameter which can be optimized based on performance constraints. The n-step ahead neural network was then trained on training data generated from experimental model to high training accuracies. This trained model performed appreciably in its objective to regulate outlet temperature of a heat exchanger. The second approach uses a combination of control and learning module for temperature regulation. The control module used is a proportional controller while the learning module used is steady state neural network. The steady state neural network is used to learn the controller output as a function of random variables and temperature set-point. The proportional controller is then used to compensate for transient behavior of linearized system around various operating points. The algorithm works effectively for temperature regulation as neural network captures the non-linear input-output behavior of closed loop system leaving proportional controller to compensate for steady state errors of a linearized plant. The proposed algorithm’s performances were evaluated against the performance of a PI controller. Further, the robustness of the algorithms were also studied in great detail under small perturbations around tuned parameters of the proposed algorithms.

Sundar, Sreenath↗