Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Stochastic optimization”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

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

At least 937 records · Page 52

Constrained Deep Reinforcement Learning for Energy Sustainable Multi-UAV Based Random Access IoT Networks With NOMA

In this paper, we apply the Non-Orthogonal Multiple Access (NOMA) technique to improve the massive channel access of a wireless IoT network where solar-powered Unmanned Aerial Vehicles (UAVs) relay data from IoT devices to remote servers. Specifically, IoT devices contend for accessing the shared wireless channel using an adaptive p-persistent slotted Aloha protocol; and the solar-powered UAVs adopt Successive Interference Cancellation (SIC) to decode multiple received data from IoT devices to improve access efficiency. To enable an energy-sustainable capacity-optimal network, we study the joint problem of dynamic multi-UAV altitude control and multi-cell wireless channel access management of IoT devices as a stochastic control problem with multiple energy constraints. We first formulate this problem as a Constrained Markov Decision Process (CMDP), and propose an online model-free Constrained Deep Reinforcement Learning (CDRL) algorithm based on Lagrangian primal-dual policy optimization to solve the CMDP. Extensive simulations demonstrate that our proposed algorithm learns a cooperative policy in which the altitude of UAVs and channel access probability of IoT devices are dynamically controlled to attain the maximal long-term network capacity while ensuring energy sustainability of UAVs, outperforming baseline schemes. The proposed CDRL agent can be trained on a small network, yet the learned policy can efficiently manage networks with a massive number of IoT devices and varying initial states, which can amortize the cost of training the CDRL agent.

42 ENGINEERING↗

The Policy Iteration Algorithm for Average Continuous Control of Piecewise Deterministic Markov Processes

The main goal of this paper is to apply the so-called policy iteration algorithm (PIA) for the long run average continuous control problem of piecewise deterministic Markov processes (PDMP's) taking values in a general Borel space and with compact action space depending on the state variable. In order to do that we first derive some important properties for a pseudo-Poisson equation associated to the problem. In the sequence it is shown that the convergence of the PIA to a solution satisfying the optimality equation holds under some classical hypotheses and that this optimal solution yields to an optimal control strategy for the average control problem for the continuous-time PDMP in a feedback form.

97 MATHEMATICS AND COMPUTING↗

Snell Envelope with Small Probability Criteria

We present a new algorithm to compute the Snell envelope in the specific case where the criteria to optimize is associated with a small probability or a rare event. This new approach combines the Stochastic Mesh approach of Broadie and Glasserman with a particle approximation scheme based on a specific change of measure designed to concentrate the computational effort in regions pointed out by the criteria. The theoretical analysis of this new algorithm provides non asymptotic convergence estimates. Finally, the numerical tests confirm the practical interest of this approach.

97 MATHEMATICS AND COMPUTING↗

Jensen–Shannon divergence based novel loss functions for Bayesian neural networks

Bayesian neural networks (BNNs) are state-of-the-art machine learning methods that can naturally regularize and systematically quantify uncertainties using their stochastic parameters. Kullback–Leibler (KL) divergence-based variational inference used in BNNs suffer from unstable optimization and challenges in approximating light-tailed posteriors due to the unbounded nature of the KL divergence. To resolve these issues, we formulate a novel loss function for BNNs based on a new modification to the generalized Jensen–Shannon (JS) divergence, which is bounded. In addition, we propose a Geometric JS divergence-based loss, which is computationally efficient since it can be evaluated analytically. We found that the JS divergence-based variational inference is intractable, and hence employed a constrained optimization framework to formulate these losses. Our theoretical analysis and empirical experiments on multiple regression and classification data sets suggest that the proposed losses perform better than the KL divergence-based loss, especially when the data sets are noisy or biased. Specifically, there are approximately 5% and 8% improvements in accuracy for a noise-added CIFAR-10 dataset and a regression dataset, respectively. There is about 13% reduction in false negative predictions of a biased histopathology dataset. Additionally, we quantify and compare the uncertainty metrics for the regression and classification tasks.

97 MATHEMATICS AND COMPUTING↗

Extracting Grain Orientations from EBSD Patterns of Polycrystalline Materials Using Convolutional Neural Networks

Abstract We present a deep learning approach to the indexing of electron backscatter diffraction (EBSD) patterns. We design and implement a deep convolutional neural network architecture to predict crystal orientation from the EBSD patterns. We design a differentiable approximation to the disorientation function between the predicted crystal orientation and the ground truth; the deep learning model optimizes for the mean disorientation error between the predicted crystal orientation and the ground truth using stochastic gradient descent. The deep learning model is trained using 374,852 EBSD patterns of polycrystalline nickel from simulation and evaluated using 1,000 experimental EBSD patterns of polycrystalline nickel. The deep learning model results in a mean disorientation error of 0.548° compared to 0.652° using dictionary based indexing.

Materials Science↗

Radionuclide gas transport through nuclear explosion-generated fracture networks

Underground nuclear weapon testing produces radionuclide gases which may seep to the surface. Barometric pumping of gas through explosion-fractured rock is investigated using a new sequentially-coupled hydrodynamic rock damage/gas transport model. Fracture networks are produced for two rock types (granite and tuff) and three depths of burial. The fracture networks are integrated into a flow and transport numerical model driven by surface pressure signals of differing amplitude and variability. There are major differences between predictions using a realistic fracture network and prior results that used a simplified geometry. Matrix porosity and maximum fracture aperture have the greatest impact on gas breakthrough time and window of opportunity for detection, with different effects between granite and tuff simulations highlighting the importance of accurately simulating the fracture network. In particular, maximum fracture aperture has an opposite effect on tuff and granite, due to different damage patterns and their effect on the barometric pumping process. From stochastic simulations using randomly generated hydrogeologic parameters, normalized detection curves are presented to show differences in optimal sampling time for granite and tuff simulations. In conclusion, seasonal and location-based effects on breakthrough, which occur due to differences in barometric forcing, are stronger where the barometric signal is highly variable.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

A shared-mobility-based framework for evacuation planning and operations under forecast uncertainty

To meet evacuation needs from carless populations who need personalized assistance to evacuate safely, in this article we propose a ridesharing-based evacuation program that recruits volunteer drivers before a disaster strikes, and then matches volunteer drivers with evacuees once demand is realized. Here we optimize resource planning and evacuation operations under uncertain spatiotemporal demand, and construct a two-stage stochastic mixed-integer program to ensure high demand fulfillment rates. We consider three formulations to improve the number of evacuees served, by minimizing an expected penalty cost, imposing a probabilistic constraint, and enforcing a constraint on the conditional value at risk of the total number of unserved evacuees, respectively. We discuss the benefits and disadvantages of the different risk measures used in the three formulations, given certain carless population sizes and the variety of evacuation modes available. We also develop a heuristic approach to provide quick, dynamic and conservative solutions. We demonstrate the performance of our approaches using five different networks of varying sizes based on regions of Charleston County, South Carolina, an area that experienced a mandatory evacuation order during Hurricane Florence, and utilize real demographic data and hourly traffic count data to estimate the demand distribution.

97 MATHEMATICS AND COMPUTING↗

Aperiodic signals processing via parameter-tuning stochastic resonance in a photorefractive ring cavity

Based on solving numerically the generalized nonlinear Langevin equation describing the nonlinear dynamics of stochastic resonance by Fourth-order Runge-Kutta method, an aperiodic stochastic resonance based on an optical bistable system is numerically investigated. The numerical results show that a parameter-tuning stochastic resonance system can be realized by choosing the appropriate optical bistable parameters, which performs well in reconstructing aperiodic signals from a very high level of noise background. The influences of optical bistable parameters on the stochastic resonance effect are numerically analyzed via cross-correlation, and a maximum cross-correlation gain of 8 is obtained by optimizing optical bistable parameters. This provides a prospective method for reconstructing noise-hidden weak signals in all-optical signal processing systems.

97 MATHEMATICS AND COMPUTING↗

Charging Decision Making for Automated Electric Vehicles

Abstract: Overview/Significance With the emerging of autonomous driving and its application in electric vehicles, the charging decision making should be taken over by the vehicles. Autonomous electric vehicles will require the capability to make charging decisions according to the battery energy state, the travel demand and also the available charging infrastructure. The introduction of autonomous driving technology would remove the challenge of co-locating charging infrastructure with driver destinations and presents a driver-free method for EVs to reach nearby charging stations. This will significantly change the charging behavior of electric vehicles. EV driver will no longer need to be present at charging stations for charging actions. Automated EVs can drive to nearby charging stations to perform charging actions by themselves when necessary. Meanwhile, connected vehicles technology makes electric vehicles have the capability to sense and obtain pertinent information from nearby charging station networks and then calculate the corresponding costs and availability for charging. All these promising capabilities bring new opportunities to construct an automatic charging decision making system and improve energy efficiency of charging actions for automatic electric vehicles. Abstract: Methodology This research introduces optimization tools and case studies on charging decision making for automated electric vehicles under personal usage scenario. The designed optimization tools aim to provide charging strategies, i.e. the optimal choice of charging station outside home and the amount of charged energy, by considering constraints from personal daily itineraries and existing charging infrastructure. Abundant real world EV data is utilized to construct stochastic energy consumption prediction model under realistic driving conditions. Advanced dynamic programming technology is utilized to model and solve the potential multi-stage decision making process by considering realistic uncertainties of EV energy cost and dynamics of itinerary information. Large-scale simulation studies using Chicago itinerary data are demonstrated to illustrate energy benefits from automatic charging decision making for future autonomous EVs. Abstract: Results/Conclusions Optimal automatic charging decision making and its energy benefits have been studied for personal automated electric vehicles. A data-driven method based on large amount of real-world data and the corresponding real-time update algorithm have been proposed to construct multi-channel stochastic energy consumption prediction model. Based on the energy cost prediction, dynamic multi-stage charging decision-making models are established for optimal strategies during a daily itinerary. Derived charging strategies can minimize the monetary and energy cost of charging actions for autonomous vehicles. Simulations and case studies show the potential ability of autonomous electric vehicles to reduce the range anxiety and charging infrastructure dependency. Automatic charging decision making can mitigate the pain from electric vehicle charging necessity outside home. In conclusion, automated electric vehicles equipped with an appropriate automatic charging decision making system will help EV driver to relieve from heavy time cost of recharging, reduce the energy cost of performing charging actions and improve the energy efficiency in future electrified transportation system.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Control Improvement for Jump-Diffusion Processes with Applications to Finance

We consider stochastic control problems with jump-diffusion processes and formulate an algorithm which produces, starting from a given admissible control {pi}, a new control with a better value. If no improvement is possible, then {pi} is optimal. Such an algorithm is well-known for discrete-time Markov Decision Problems under the name Howard's policy improvement algorithm. The idea can be traced back to Bellman. Here we show with the help of martingale techniques that such an algorithm can also be formulated for stochastic control problems with jump-diffusion processes. As an application we derive some interesting results in financial portfolio optimization.

97 MATHEMATICS AND COMPUTING↗

Stochastic noise characteristics in matrix inversion tomosynthesis (MITS)

Matrix inversion tomosynthesis (MITS) uses known imaging geometry and linear systems theory to deterministically separate in-plane detail from residual tomographic blur in a set of conventional tomosynthesis (''shift-and-add'') planes. A previous investigation explored the effect of scan angle (ANG), number of projections (N), and number of reconstructed planes (NP) on the MITS impulse response and modulation transfer function characteristics, and concluded that ANG=20 deg., N=71, and NP=69 is the optimal MITS imaging technique for chest imaging on our prototype tomosynthesis system. This article examines the effect of ANG, N, and NP on the MITS exposure-normalized noise power spectra (ENNPS) and seeks to confirm that the imaging parameters selected previously by an analysis of the MITS impulse response also yield reasonable stochastic properties in MITS reconstructed planes. ENNPS curves were generated for experimentally acquired mean-subtracted projection images, conventional tomosynthesis planes, and MITS planes with varying combinations of the parameters ANG, N, and NP. Image data were collected using a prototype tomosynthesis system, with 11.4 cm acrylic placed near the image receptor to produce lung-equivalent beam hardening and scattered radiation. Ten identically acquired tomosynthesis data sets (realizations) were collected for each selected technique and used to generate ensemble mean images that were subtracted from individual image realizations prior to noise power spectra (NPS) estimation. NPS curves were normalized to account for differences in entrance exposure (as measured with an ion chamber), yielding estimates of the ENNPS for each technique. Results suggest that mid- and high-frequency noise in MITS planes is fairly equivalent in magnitude to noise in conventional tomosynthesis planes, but low-frequency noise is amplified in the most anterior and posterior reconstruction planes. Selecting the largest available number of projections (N=71) does not incur any appreciable additive electronic noise penalty compared to using fewer projections for roughly equivalent cumulative exposure. Stochastic noise is minimized by maximizing N and NP but increases with increasing ANG. The noise trend results for NP and ANG are contrary to what would be predicted by simply considering the MITS matrix conditioning and likely result from the interplay between noise correlation and the polarity of the MITS filters. From this study, the authors conclude that the previously determined optimal MITS imaging strategy based on impulse response considerations produces somewhat suboptimal stochastic noise characteristics, but is probably still the best technique for MITS imaging of the chest.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Learning and meta-learning of stochastic advection–diffusion–reaction systems from sparse measurements

Physics-informed neural networks (PINNs) were recently proposed in [18] as an alternative way to solve partial differential equations (PDEs). A neural network (NN) represents the solution, while a PDE-induced NN is coupled to the solution NN, and all differential operators are treated using automatic differentiation. Here, we first employ the standard PINN and a stochastic version, sPINN, to solve forward and inverse problems governed by a non-linear advection–diffusion–reaction (ADR) equation, assuming we have some sparse measurements of the concentration field at random or pre-selected locations. Subsequently, we attempt to optimise the hyper-parameters of sPINN by using the Bayesian optimisation method (meta-learning) and compare the results with the empirically selected hyper-parameters of sPINN. In particular, for the first part in solving the inverse deterministic ADR, we assume that we only have a few high-fidelity measurements, whereas the rest of the data is of lower fidelity. Hence, the PINN is trained using a composite multi-fidelity network, first introduced in [12], that learns the correlations between the multi-fidelity data and predicts the unknown values of diffusivity, transport velocity and two reaction constants as well as the concentration field. For the stochastic ADR, we employ a Karhunen–Loève (KL) expansion to represent the stochastic diffusivity, and arbitrary polynomial chaos (aPC) to represent the stochastic solution. Correspondingly, we design multiple NNs to represent the mean of the solution and learn each aPC mode separately, whereas we employ a separate NN to represent the mean of diffusivity and another NN to learn all modes of the KL expansion. For the inverse problem, in addition to stochastic diffusivity and concentration fields, we also aim to obtain the (unknown) deterministic values of transport velocity and reaction constants. The available data correspond to 7spatial points for the diffusivity and 20 space–time points for the solution, both sampled 2000 times. We obtain good accuracy for the deterministic parameters of the order of 1–2% and excellent accuracy for the mean and variance of the stochastic fields, better than three digits of accuracy. In the second part, we consider the previous stochastic inverse problem, and we use Bayesian optimisation to find five hyper-parameters of sPINN, namely the width, depth and learning rate of two NNs for learning the modes. Here, we obtain much deeper and wider optimal NNs compared to the manual tuning, leading to even better accuracy, i.e., errors less than 1% for the deterministic values, and about an order of magnitude less for the stochastic fields.

97 MATHEMATICS AND COMPUTING↗

Surrogate Modeling for Capacity Planning of Charging Station Equipped with PV and Hydropneumatic Energy Storage

Due to the promising potential for environmental sustainability, there has been a significant increase of electric vehicles (EVs) and plug-in hybrid electric vehicles (PHEV) in the market. To support this increasing demand for EVs and PHEVs, challenges related to capacity planning and investment costs of public charging infrastructure must be addressed. Hence, in this paper, a capacity planning problem for EV charging stations is developed and aims to balance current capital investment costs and future operational revenue. The charging station considered in this work is assumed to be equipped with solar photovoltaic panel (PV) and an energy storage system which could be electric battery or the recently invented hydro-pneumatic energy storage (GLIDES, Ground-Level Integrated Diverse Energy Storage) system. A co-optimization model that minimizes investment and operation cost is established to determine the global optimal solution while combining the capacity and operational decision making. The operational decision making considers EV mobility which is modeled as an Erlang-loss system. Meanwhile, stochastic programming is adopted to capture uncertainties from solar radiation and charging demand of the EV fleet. To provide a more general and computationally efficient model, main configuration parameters are sampled in the design space and then fixed in solving the co-optimization model. The model can be used to provide insights for charging station placement in different practical situations. The sampled parameters include: the total number of EV charging slots, the PV area, the maximum capacity of the energy storage system, and daily mean EV arrival number in the Erlang-loss system. Based on the sampled parameter combinations and its responses, black-box mappings are then constructed using surrogate models (RBF, Kriging etc). The effectiveness of proposed surrogate modeling approach is demonstrated in the numerical experiments.

Chen, Yang↗

Stochastic Thermo-Hydro Modeling and Neural Network Surrogate Development for Thermal Resource Assessment of the Galleries-to-Calories Geobattery

The Galleries-to-Calories Geobattery concept explores the use of abandoned coal mine workings for large-scale thermal energy transport and storage. The system involves injecting waste heat from a supercomputing facility into flooded mine galleries, where groundwater flow can store and transport thermal energy for potential recovery in downgradient district heating and cooling applications. To evaluate the feasibility and performance of the Geobattery under geological and operational uncertainty, we developed a suite of stochastic thermo-hydrological (TH) simulations using Monte Carlo sampling of key uncertain parameters (e.g., permeability, porosity, thermal conductivity, specific heat capacity) and operating conditions (e.g., injection rate, injection temperature). Results identified injection rate and temperature as the most influential parameters governing thermal front propagation, while the geometry of the room-and-pillar structure played a critical role in directing the extent and orientation of thermal advancement. Optimal combinations of material properties for maximizing heat recovery were also determined. To address the high computational cost of coupled-process stochastic modeling, we trained a neural network surrogate model on 24,000 physics-based realizations, achieving an R² > 0.99 and MAE < 0.1 for temperature predictions at monitoring locations. This surrogate enabled an additional 100,000 realizations for global sensitivity analysis and probabilistic thermal resource assessment. The integrated stochastic physics–surrogate modeling framework offers a computationally efficient tool for quantifying uncertainty, identifying key drivers, and informing early-stage design decisions for Geobattery systems.

15 - GEOTHERMAL ENERGY↗

The American Foreign Exchange Option in Time-Dependent One-Dimensional Diffusion Model for Exchange Rate

The classical Garman-Kohlhagen model for the currency exchange assumes that the domestic and foreign currency risk-free interest rates are constant and the exchange rate follows a log-normal diffusion process.In this paper we consider the general case, when exchange rate evolves according to arbitrary one-dimensional diffusion process with local volatility that is the function of time and the current exchange rate and where the domestic and foreign currency risk-free interest rates may be arbitrary continuous functions of time. First non-trivial problem we encounter in time-dependent case is the continuity in time argument of the value function of the American put option and the regularity properties of the optimal exercise boundary. We establish these properties based on systematic use of the monotonicity in volatility for the value functions of the American as well as European options with convex payoffs together with the Dynamic Programming Principle and we obtain certain type of comparison result for the value functions and corresponding exercise boundaries for the American puts with different strikes, maturities and volatilities.Starting from the latter fact that the optimal exercise boundary curve is left continuous with right-hand limits we give a mathematically rigorous and transparent derivation of the significant early exercise premium representation for the value function of the American foreign exchange put option as the sum of the European put option value function and the early exercise premium.The proof essentially relies on the particular property of the stochastic integral with respect to arbitrary continuous semimartingale over the predictable subsets of its zeros. We derive from the latter the nonlinear integral equation for the optimal exercise boundary which can be studied by numerical methods.

97 MATHEMATICS AND COMPUTING↗

Enhancing capillary trapping effectiveness through proper time scheduling of injection of supercritical CO 2 in heterogeneous formations

The impact of geologic heterogeneity on capillary trapping of supercritical CO 2 (scCO 2 ) has been recognized and appraised through laboratory experimentation and modeling. However, how different injection strategies can be optimized to improve capillary trapping has not received adequate attention. Here, we present a study based on stochastic analysis to show the impact of injection scheduling on capillary trapping of scCO 2 in heterogeneous geological formations. Improvement of trapping efficiency consists of maximizing the trapped volume within a predefined secure zone of the storage formation with the goal of avoiding plume intersection with potential leakage pathways. Using knowledge acquired from physical experiments in intermediate-scale tanks with controlled heterogeneity, we conduct numerical simulations following a stochastic approach to extend the observed outcomes to a series of equally probable permeability scenarios. Our simulations involve the same combination of surrogate fluids that are used in the experiments to mimic viscosity and density contrasts between scCO 2 and brine. To account for uncertainty of formation heterogeneity, several permeability fields with three different variances and two horizontal correlation lengths are generated. The same volume of scCO 2 is emplaced using four different modes of injection scheduling. Results suggest that injection strategies aimed at enhancing capillary trapping of scCO 2 and increasing the probability of constraining the plume within a secure zone are strictly related to the specific heterogeneity of the reservoir. Finally, this preliminary theoretical finding suggests the potential for maximizing secure capillary trapping through the proper selection of injection schedule to fit the formation heterogeneity.

58 GEOSCIENCES↗

The triple decomposition of a fluctuating velocity field in a multiscale flow

A new method for the triple decomposition of a multiscale flow, which is based on the novel optimal mode decomposition (OMD) technique, is presented. OMD provides low order linear dynamics, which fits a given data set in an optimal way and is used to distinguish between a coherent (periodic) part of a flow and a stochastic fluctuation. The method needs no external phase indication since this information, separate for coherent structures associated with each length scale introduced into the flow, appears as the output. The proposed technique is compared against two traditional methods of the triple decomposition, i.e., bin averaging and proper orthogonal decomposition. This is done with particle image velocimetry data documenting the near wake of a multiscale bar array. It is shown that both traditional methods are unable to provide a reliable estimation for the coherent fluctuation while the proposed technique performs very well. The crucial result is that the coherence peaks are not observed within the spectral properties of the stochastic fluctuation derived with the proposed method; however, these properties remain unaltered at the residual frequencies. This proves the method’s capability of making a distinction between both types of fluctuations. The sensitivity to some prescribed parameters is checked revealing the technique’s robustness. Additionally, an example of the method application for analysis of a multiscale flow is given, i.e., the phase conditioned transverse integral length is investigated in the near wake region of the multiscale object array.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

HackAttack: Game-Theoretic Analysis of Realistic Cyber Conflicts

Game theory is appropriate for studying cyber conflict because it allows for an intelligent and goal-driven adversary. Applications of game theory have led to a number of results regarding optimal attack and defense strategies. However, the overwhelming majority of applications explore overly simplistic games, often ones in which each participant s actions are visible to every other participant. These simplifications strip away the fundamental properties of real cyber conflicts: probabilistic alerting, hidden actions, unknown opponent capabilities. In this paper, we demonstrate that it is possible to analyze a more realistic game, one in which different resources have different weaknesses, players have different exploits, and moves occur in secrecy, but they can be detected. Certainly, more advanced and complex games are possible, but the game presented here is more realistic than any other game we know of in the scientific literature. While optimal strategies can be found for simpler games using calculus, case-by-case analysis, or, for stochastic games, Q-learning, our more complex game is more naturally analyzed using the same methods used to study other complex games, such as checkers and chess. We define a simple evaluation function and employ multi-step searches to create strategies. We show that such scenarios can be analyzed, and find that in cases of extreme uncertainty, it is often better to ignore one s opponent s possible moves. Furthermore, we show that a simple evaluation function in a complex game can lead to interesting and nuanced strategies.

botnets↗