Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Parameter Dimensionality Reduction”

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 361 records · Page 20

Interface and photoluminescence characteristics of graphene-(GaN/InGaN){sub n} multiple quantum wells hybrid structure

The effects of graphene on the optical properties of active system, e.g., the InGaN/GaN multiple quantum wells, are thoroughly investigated and clarified. Here, we have investigated the mechanisms accounting for the photoluminescence reduction for the graphene covered GaN/InGaN multiple quantum wells hybrid structure. Compared to the bare multiple quantum wells, the photoluminescence intensity of graphene covered multiple quantum wells showed a 39% decrease after excluding the graphene absorption losses. The responsible mechanisms have been identified with the following factors: (1) the graphene two dimensional hole gas intensifies the polarization field in multiple quantum wells, thus steepening the quantum well band profile and causing hole-electron pairs to further separate; (2) a lower affinity of graphene compared to air leading to a weaker capability to confine the excited hot electrons in multiple quantum wells; and (3) exciton transfer through non-radiative energy transfer process. These factors are theoretically analysed based on advanced physical models of semiconductor devices calculations and experimentally verified by varying structural parameters, such as the indium fraction in multiple quantum wells and the thickness of the last GaN quantum barrier spacer layer.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Cylindrical ion-acoustic solitary waves in electronegative plasmas with superthermal electrons

By using the standard reductive perturbation technique, a three-dimensional cylindrical Kadomtsev-Petviashvili equation (CKPE), which governs the dynamics of ion acoustic solitary waves (IASWs), is derived for small but finite amplitude ion-acoustic waves in cylindrical geometry in a collisionless unmagnetized plasma with kappa distributed electrons, thermal positrons, and cold ions. The generalized expansion method is used to solve analytically the CKPE. The existence regions of localized pulses are investigated. It is found that the solution of the CKPE supports only compressive solitary waves. Furthermore, the effects of superthermal electrons, the ratio of the electron temperature to positron temperature, the ratio of the positron density to electron density and direction cosine of the wave propagation on the profiles of the amplitudes, and widths of the solitary structures are examined numerically. It is shown these parameters play a vital role in the formation of ion acoustic solitary waves.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Evaluating the Effects of Missing Values and Mixed Data Types on Social Sequence Clustering Using t-SNE Visualization

© 2019 Association for Computing Machinery. The goal of this work is to investigate the impact of missing values in clustering joint categorical social sequences. Identifying patterns in sociodemographic longitudinal data is important in a number of social science settings. However, performing analytical operations, such as clustering on life course trajectories, is challenging due to the categorical and multidimensional nature of the data, their mixed data types, and corruption by missing and inconsistent values. Data quality issues were investigated previously on single variable sequences. To understand their effects on multivariate sequence analysis, we employ a dataset of mixed data types and missing values, a dissimilarity measure designed for joint categorical sequence data, together with dimensionality reduction methodologies in a systematic design of sequence clustering experiments. Given the categorical nature of our data, we employ an “edit” distance using optimal matching. Because each data record has multiple variables of different types, we investigate the impact of mixing these variables in a single dissimilarity measure. Between variables with binary values and those with multiple nominal values, we find that the ability to overcome missing data problems is more difficult in the nominal domain than in the binary domain. Additionally, alignment of leading missing values can result in systematic biases in dissimilarity matrices and subsequently introduce both artificial clusters and unrealistic interpretations of associated data domains. We demonstrate the usage of t-distributed stochastic neighborhood embedding to visually guide mitigation of such biases by tuning the missing value substitution cost parameter or determining an optimal sequence span.

Lazar, Alina↗

Geometry of the electron local observables

We show that the complete set of the second power identities for the electron local observables consists of 36 equations. The identities connect the products of the electron bilinear forms and, being considered as geometrically meaningful equations in 3D Euclidean space, are separated over the groups of equations for the scalar, vector and tensor quantities. Considering the complete set of identities as a set of the second power equations, we solve the equations and find the irreducible representation for the electron local observables. The representation defines the 16 electron local observables as functions of 7 basic parameters and can be formulated in 6 various forms. The basic parameters include scalar and pseudoscalar, the time components of a 4-vector and a 4-pseudovector, and three Euler angles which define the angular position of a local 3D frame with respect to the 3D laboratory frame. These 7 parameters completely define the space components of the 4-vector and the 4-pseudovector, as well as the polar and axial vectors. The developed representation shows that the analysis of the any electron wave packet can be considerably simplified by the reduction of the number of analyzed real functions from 16 to 7. As an example, we present the structure of the local observables defined by the irreducible representation in a case of a traveling electron wave. - Highlights: > Complete set of bilinear identities for the electron observables is derived. > Irreducible representation for the electron observables is found. > The representation defines the electron observables as functions of 7 parameters.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

SU-F-T-147: An Alternative Parameterization of Scatter Behavior Allows Significant Reduction of Beam Characterization for Pencil Beam Proton Therapy

Purpose: 1) To describe the characteristics of pencil beam proton dose deposition kernels in a homogenous medium using a novel parameterization. 2) To propose a method utilizing this novel parametrization to reduce the measurements and pre-computation required in commissioning a pencil beam proton therapy system. Methods: Using beam data from a clinical, pencil beam proton therapy center, Monte Carlo simulations were performed to characterize the dose depositions at a range of energies from 100.32 to 226.08 MeV in 3.6MeV steps. At each energy, the beam is defined at the surface of the phantom by a two-dimensional Normal distribution. Using FLUKA, the in-medium dose distribution is calculated in 200×200×350 mm cube with 1 mm{sup 3} tally volumes. The calculated dose distribution in each 200×200 slice perpendicular to the beam axis is then characterized using a symmetric alpha-stable distribution centered on the beam axis. This results in two parameters, α and γ, that completely describe shape of the distribution. In addition, the total dose deposited on each slice is calculated. The alpha-stable parameters are plotted as function of the depth in-medium, providing a representation of dose deposition along the pencil beam. We observed that these graphs are isometric through a scaling of both abscissa and ordinate map the curves. Results: Using interpolation of the scaling factors of two source curves representative of different beam energies, we predicted the parameters of a third curve at an intermediate energy. The errors are quantified by the maximal difference and provide a fit better than previous methods. The maximal energy difference between the source curves generating identical curves was 21.14MeV. Conclusion: We have introduced a novel method to parameterize the in-phantom properties of pencil beam proton dose depositions. For the case of the Knoxville IBA system, no more than nine pencil beams have to be fully characterized.

60 APPLIED LIFE SCIENCES↗

AEOLUS: Advances in Experimental Design, Optimal Control, and Learning for Uncertain Complex Systems

Sustained advances in the mathematics of modeling and simulation have resulted in the capability today for routine simulation of a number of large scale complex DOE-relevant systems. As remarkable as this capability for solving the so-called forward problem is, it is typically only the first step-an inner loop within an outer loop that explores the simulation model's parameter space and decision space to characterize uncertainty in the model's predictions, learn unknown model parameters from data, design the most informative experiments, determine optimal control strategies, and create optimal designs. Broadly, what unifies all of these outer loop problems is that they are, in one form or another, optimization problems over parameter/control/design space that are constrained by complex uncertain models. To fully realize the power of scientific simulation as a basis for scientific discovery, technological innovation, and rational decision-making, it is imperative to move beyond simulation to tackle the outer loop of optimization for learning from data, experimental design, and control with complex uncertain models. When the models under consideration are large-scale and complex, and when the optimization variable and uncertain parameter spaces are high (or infinite) dimensional, this constitutes a grand challenge of the highest order, and is intractable with conventional methods. To overcome these challenges, the AEOLUS Center was established to develop a unified mathematical, computational, and statistical framework for (1) Learning predictive models from complex data via Bayesian inference and optimization, and (2) Optimizing experiments, processes, and designs using the resulting uncertain models. These problems are intractable with conventional methods, for several reasons: (1) The simulation problems that govern the inner loops of the optimization problems are expensive to execute (due to severe nonlinearity, heterogeneity, multiphysics/multiscale coupling); (2) The optimization variable and uncertain parameter spaces are high dimensional, often stemming from discretizations of infinite dimensional fields such as initial conditions, sources, or material properties. We argue that the key to overcoming these challenges is to develop new mathematical, computational, and statistical methods that exploit the structure of the Bayesian inference and optimization problems mediated by their underlying complex uncertain models. This structure includes the regularity, sparsity, geometry, low intrinsic dimensionality, and multifidelity nature of the maps from uncertain parameter/optimization variable spaces to the specific objectives targeted: Bayesian inference, optimal experimental design, and optimal control design. Black box methods developed as generic tools are incapable of exploiting this structure. To be successful, we must create, integrate, and cross-fertilize ideas across multiple areas of applied math--including approximation theory, Bayesian inference, data science, experimental design, information theory, machine learning, model reduction, optimal control theory, parallel algorithms, PDE-constrained optimization, randomized algorithms, stochastic optimization, and uncertainty quantification--all while exploiting the structure of the problems at hand. With this goal in mind, we have marshaled a team of leading authorities in these areas. While the methods we develop will be broadly applicable across a wide spectrum of DOE problems in which experiments inform models and the systems those models describe must be optimized under uncertainty, we have chosen a specific area, advanced manufacturing and materials, to drive our work. AMM is characterized by complex models across multiple scales, and is a rich source of challenging problems in inference, experimental design, and optimal control, requiring multifaceted and integrated advances in applied mathematics. As such, AMM serves as an excellent vehicle to motivate and demonstrate the advances in applied mathematics developed by our center.

97 MATHEMATICS AND COMPUTING↗

Simulation of Thermal Stratification in BWR Suppression Pools with One Dimensional Modeling Method

The suppression pool in a boiling water reactor (BWR) plant not only is the major heat sink within the containment system, but also provides the major emergency cooling water for the reactor core. In several accident scenarios, such as a loss-of-coolant accident and extended station blackout, thermal stratification tends to form in the pool after the initial rapid venting stage. Accurately predicting the pool stratification phenomenon is important because it affects the peak containment pressure; the pool temperature distribution also affects the NPSHa (available net positive suction head) and therefore the performance of the Emergency Core Cooling System and Reactor Core Isolation Cooling System pumps that draw cooling water back to the core. Current safety analysis codes use zero dimensional (0-D) lumped parameter models to calculate the energy and mass balance in the pool; therefore, they have large uncertainties in the prediction of scenarios in which stratification and mixing are important. While three-dimensional (3-D) computational fluid dynamics (CFD) methods can be used to analyze realistic 3-D configurations, these methods normally require very fine grid resolution to resolve thin substructures such as jets and wall boundaries, resulting in a long simulation time. For mixing in stably stratified large enclosures, the BMIX++ code (Berkeley mechanistic MIXing code in C++) has been developed to implement a highly efficient analysis method for stratification where the ambient fluid volume is represented by one-dimensional (1-D) transient partial differential equations and substructures (such as free or wall jets) are modeled with 1-D integral models. This allows very large reductions in computational effort compared to multi-dimensional CFD modeling. One heat-up experiment performed at the Finland POOLEX facility, which was designed to study phenomena relevant to Nordic design BWR suppression pool including thermal stratification and mixing, is used for validation. Comparisons between the BMIX++, GOTHIC, and CFD calculations against the POOLEX experimental data are discussed in detail.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Explainable AI for Multivariate Time Series Pattern Exploration: Latent Space Visual Analytics With Temporal Fusion Transformer and Variational Autoencoders in Power Grid Event Diagnosis

Detecting and analyzing complex patterns in multivariate time-series data is crucial for decision-making in urban and environmental system operations. However, challenges arise from the high dimensionality, intricate complexity, and interconnected nature of complex patterns, which hinder the understanding of their underlying physical processes. Existing AI methods often face limitations in interpretability, computational efficiency, and scalability, reducing their applicability in real-world scenarios. This paper proposes a novel visual analytics framework that integrates two generative AI models, Temporal Fusion Transformer (TFT) and Variational Autoencoders (VAEs), to reduce complex patterns into lower-dimensional latent spaces and visualize them in 2D using dimensionality reduction techniques such as PCA, t-SNE, and UMAP with DBSCAN. These visualizations, presented through coordinated and interactive views and tailored glyphs, enable intuitive exploration of complex multivariate temporal patterns, identifying patterns’ similarities and uncover their potential correlations for a better interpretability of the AI outputs. The framework is demonstrated through a case study on power grid signal data, where it identifies multi-label grid event signatures, including faults and anomalies with diverse root causes. Additionally, novel metrics and visualizations are introduced to validate the models and assess the performance, efficiency, and consistency of latent maps generated by VAE, which have been utilized in prior studies for latent space cartography and used as a benchmark in this study, and the emerging TFT architecture under various configurations. These analyses provide actionable insights for model parameter tuning and reliability improvements. Comparative results highlight that TFT achieves shorter run times and superior scalability to diverse time-series data shapes compared to VAE. This work advances fault diagnosis in multivariate time series, fostering explainable AI to support critical system operations.

Explainable AI↗

Modeling the high-energy electronic state manifold of adenine: Calibration for nonlinear electronic spectroscopy

Pump-probe electronic spectroscopy using femtosecond laser pulses has evolved into a standard tool for tracking ultrafast excited state dynamics. Its two-dimensional (2D) counterpart is becoming an increasingly available and promising technique for resolving many of the limitations of pump-probe caused by spectral congestion. The ability to simulate pump-probe and 2D spectra from ab initio computations would allow one to link mechanistic observables like molecular motions and the making/breaking of chemical bonds to experimental observables like excited state lifetimes and quantum yields. From a theoretical standpoint, the characterization of the electronic transitions in the visible (Vis)/ultraviolet (UV), which are excited via the interaction of a molecular system with the incoming pump/probe pulses, translates into the determination of a computationally challenging number of excited states (going over 100) even for small/medium sized systems. A protocol is therefore required to evaluate the fluctuations of spectral properties like transition energies and dipole moments as a function of the computational parameters and to estimate the effect of these fluctuations on the transient spectral appearance. In the present contribution such a protocol is presented within the framework of complete and restricted active space self-consistent field theory and its second-order perturbation theory extensions. The electronic excited states of adenine have been carefully characterized through a previously presented computational recipe [Nenov et al., Comput. Theor. Chem. 1040–1041, 295-303 (2014)]. A wise reduction of the level of theory has then been performed in order to obtain a computationally less demanding approach that is still able to reproduce the characteristic features of the reference data. Foreseeing the potentiality of 2D electronic spectroscopy to track polynucleotide ground and excited state dynamics, and in particular its expected ability to provide conformational dependent fingerprints in dimeric systems, the performances of the selected reduced level of calculations have been tested in the construction of 2D electronic spectra for the in vacuo adenine monomer and the unstacked adenine homodimer, thereby exciting the L{sub b}/L{sub a} transitions with the pump pulse pair and probing in the Vis to near ultraviolet spectral window.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Evaluating the benefits of kinetic Monte Carlo and microkinetic modeling for catalyst design studies in the presence of lateral interactions

Popular computational catalyst design strategies rely on the identification of reactivity descriptors, which can be used along with Brønsted–Evans–Polanyi (BEP) and scaling relations as input to a microkinetic model (MKM) to make predictions for activity or selectivity trends. The main benefit of this approach is related to the inherent dimensionality reduction of the large material space to just a few catalyst descriptors. Conversely, it is well documented that a small set of descriptors is insufficient to capture the intricacies and complexities of a real catalytic system. The inclusion of coverage effects through lateral adsorbate-adsorbate interactions can narrow the gap between simplified descriptor predictions and real systems, but mean-field MKMs cannot properly account for local coverage effects. This shortcoming of the mean-field approximation can be rectified by switching to a lattice-based kinetic Monte Carlo (kMC) method using cluster expansion representation of adsorbate–adsorbate lateral interactions. Using the prototypical CO oxidation reaction as an example, we critically evaluate the benefits of kMC over MKM in terms of trend predictions and computational cost when using only a small set of input parameters. After confirming that in the absence of lateral interactions the kMC and MKM approaches yield identical trends and mechanistic information, we observed substantial differences between the two kinetic models when lateral interactions were introduced. Here, the mean-field implementation applies coverage corrections directly to the descriptors, causing an artificial overprediction of the activity of strongly binding metals. In contrast, the cluster expansion in kMC implementation can differentiate among the highly active metals but it is very sensitive to the set of included interaction parameters. Considering that computational screening relies on a minimal set of descriptors, for which MKM makes reasonable trend predictions at a ca. three orders of magnitude lower computational cost than kMC, the MKM approach does provide a better entry point for computational catalyst design.

36 MATERIALS SCIENCE↗

QAARM: quasi-anharmonic autoregressive model reveals molecular recognition pathways in ubiquitin

Motivation: Molecular dynamics (MD) simulations have dramatically improved the atomistic understanding of protein motions, energetics and function. These growing datasets have necessitated a corresponding emphasis on trajectory analysis methods for characterizing simulation data, particularly since functional protein motions and transitions are often rare and/or intricate events. Observing that such events give rise to long-tailed spatial distributions, we recently developed a higher-order statistics based dimensionality reduction method, called quasi-anharmonic analysis (QAA), for identifying biophysically-relevant reaction coordinates and substates within MD simulations. Further characterization of conformation space should consider the temporal dynamics specific to each identified substate. Results: Our model uses hierarchical clustering to learn energetically coherent substates and dynamic modes of motion from a 0.5µs ubiqutin simulation. Autoregressive (AR) modeling within and between states enables a compact and generative description of the conformational landscape as it relates to functional transitions between binding poses. Lacking a predictive component, QAA is extended here within a general AR model appreciative of the trajectory’s temporal dependencies and the specific, local dynamics accessible to a protein within identified energy wells. These metastable states and their transition rates are extracted within a QAA-derived subspace using hierarchical Markov clustering to provide parameter sets for the second-order AR model. We show the learned model can be extrapolated to synthesize trajectories of arbitrary length.

59 BASIC BIOLOGICAL SCIENCES↗

SU-E-T-630: Predictive Modeling of Mortality, Tumor Control, and Normal Tissue Complications After Stereotactic Body Radiotherapy for Stage I Non-Small Cell Lung Cancer

Purpose: While rates of local control have been well characterized after stereotactic body radiotherapy (SBRT) for stage I non-small cell lung cancer (NSCLC), less data are available characterizing survival and normal tissue toxicities, and no validated models exist assessing these parameters after SBRT. We evaluate the reliability of various machine learning techniques when applied to radiation oncology datasets to create predictive models of mortality, tumor control, and normal tissue complications. Methods: A dataset of 204 consecutive patients with stage I non-small cell lung cancer (NSCLC) treated with stereotactic body radiotherapy (SBRT) at the University of Pennsylvania between 2009 and 2013 was used to create predictive models of tumor control, normal tissue complications, and mortality in this IRB-approved study. Nearly 200 data fields of detailed patient- and tumor-specific information, radiotherapy dosimetric measurements, and clinical outcomes data were collected. Predictive models were created for local tumor control, 1- and 3-year overall survival, and nodal failure using 60% of the data (leaving the remainder as a test set). After applying feature selection and dimensionality reduction, nonlinear support vector classification was applied to the resulting features. Models were evaluated for accuracy and area under ROC curve on the 81-patient test set. Results: Models for common events in the dataset (such as mortality at one year) had the highest predictive power (AUC = .67, p < 0.05). For rare occurrences such as radiation pneumonitis and local failure (each occurring in less than 10% of patients), too few events were present to create reliable models. Conclusion: Although this study demonstrates the validity of predictive analytics using information extracted from patient medical records and can most reliably predict for survival after SBRT, larger sample sizes are needed to develop predictive models for normal tissue toxicities and more advanced machine learning methodologies need be consider in the future.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Mobility of the Two-Dimensional Electron Gas in DA-pHEMT Heterostructures with Various δ–n-Layer Profile Widths

The effect of the silicon-atom distribution profile in donor δ-layers of AlGaAs/InGaAs/AlGaAs heterostructures with donor–acceptor doping on the mobility of the two-dimensional electron gas is studied. The parameters of the δ-layer profiles are determined using the normal approximation of the spatial distributions of silicon atoms, measured by secondary-ion mass spectroscopy. It is shown that the standard deviation σ of the δ-layer profile can be reduced from 3.4 to 2.5 nm by the proper selection of growth conditions. Measurements of the magnetic-field dependences of the Hall effect and conductivity show that such a decrease in σ allowed an increase in the mobility of the two-dimensional electron gas in heterostructures by 4000 cm{sup 2}/(V s) at 77 K and 600 cm{sup 2}/(V s) at 300 K. The mobility calculation taking into account filling of the first two size-quantization subbands shows that an increase in the mobility is well explained by a reduction in the Coulomb scattering at ionized donors due to an increase in the effective thickness of the spacer layer with decreasing σ of the δ-layer profile.

36 MATERIALS SCIENCE↗

QAARM: Quasi-anharmonic auto-regressive model reveals molecular recognition pathways in ubiquitin

Molecular dynamics (MD) simulations have dramatically improved the atomistic understanding of protein motions, energetics and function. These growing datasets have necessitated a corresponding emphasis on trajectory analysis methods for characterizing simulation data, particularly since functional protein motions and transitions are often rare and/or intricate events. Observing that such events give rise to long-tailed spatial distributions, we recently developed a higher-order statistics based dimensionality reduction method, called quasi-anharmonic analysis (QAA), for identifying biophysically-relevant reaction coordinates and substates within MD simulations. Further characterization of conformation space should consider the temporal dynamics specific to each identified substate. Our model uses hierarchical clustering to learn energetically coherent substates and dynamic modes of motion from a 0.5 {mu}s ubiqutin simulation. Autoregressive (AR) modeling within and between states enables a compact and generative description of the conformational landscape as it relates to functional transitions between binding poses. Lacking a predictive component, QAA is extended here within a general AR model appreciative of the trajectory's temporal dependencies and the specific, local dynamics accessible to a protein within identified energy wells. These metastable states and their transition rates are extracted within a QAA-derived subspace using hierarchical Markov clustering to provide parameter sets for the second-order AR model. We show the learned model can be extrapolated to synthesize trajectories of arbitrary length.

59 BASIC BIOLOGICAL SCIENCES↗

General Bayesian Inference over the Stiefel Manifold via the Givens Transform

We introduce here an approach based on the Givens representation for posterior inference in statistical models with orthogonal matrix parameters, such as factor models and probabilistic principal component analysis (PPCA). We show how the Givens representation can be used to develop practical methods for transforming densities over the Stiefel manifold into densities over subsets of Euclidean space. We demonstrate how to deal with issues arising from the topology of the Stiefel manifold and how to inexpensively compute the change-of-measure terms. We introduce an auxiliary parameter approach that limits the impact of topological issues. We provide both analysis of our methods and numerical examples demonstrating the effectiveness of the approach. We also discuss how our Givens representation can be used to define general classes of distributions over the space of orthogonal matrices. We then give demonstrations on several examples showing how the Givens approach performs in practice in comparison with other methods.

97 MATHEMATICS AND COMPUTING↗

Model reduction for hypersonic aerodynamics via conservative LSPG projection and hyper-reduction

High-speed aerospace engineering applications rely heavily on computational fluid dynamics (CFD) models for design and analysis due to the expense and difficulty of flight tests and experiments. This reliance on CFD models necessitates performing accurate and reliable uncertainty quantification (UQ) of the CFD models. However, it is very computationally expensive to run CFD for hypersonic flows due to the fine grid resolution required to capture the strong shocks and large gradients that are typically present. Furthermore, UQ approaches are “many-query” problems requiring many runs with a wide range of input parameters. One way to enable computationally expensive models to be used in such many-query problems is to employ projection-based reduced-order models (ROMs) in lieu of the (high-fidelity) full-order model. In particular, the least-squares Petrov–Galerkin (LSPG) ROM (equipped with hyper-reduction) has demonstrated the ability to significantly reduce simulation costs while retaining high levels of accuracy on a range of problems including subsonic CFD applications. This allows computationally inexpensive LSPG ROM simulations to replace the full-order model simulations in UQ studies, which makes this many-query task tractable, even for large-scale CFD models. This work presents the first application of LSPG to a hypersonic CFD application. In particular, we present results for LSPG ROMs of the HIFiRE-1 in a three-dimensional, turbulent Mach 7.1 flow, showcasing the ability of the ROM to significantly reduce computational costs while maintaining high levels of accuracy in computed quantities of interest.

Fluid Dynamics↗

An efficient cost reduction procedure for bounded-control LQR problems

A novel approach has been developed for approximating the solution to the constrained LQR problem, based on updating the final state and costate of a related regular problem, and on slightly shifting the switching times (the instants when the control meets the constraints). The main result is the expression of a suboptimal control in feedback form using the solution of some compatible Riccati equation. The gradient method is applied to reduce the cost via explicit algebraic formula for its partial derivatives with respect to the hidden final state/costate of the related regular problem and to the switching times. The numerical method is termed efficient because it does not involve integrations of states or cost trajectories, and reduces to its minimum the dimension of the unknown parameters at the final condition. All the relevant objects are calculated from a few auxiliary matrices, which are computed only once. The scheme is here applied to two case studies whose optimal solutions are known. The first example is a two-dimensional model of the ‘cheapest stop of a train’ problem. The second one refers to the temperature control of a metallic strip leaving a multi-stand rolling mill, a problem with a high-dimensional state.

97 MATHEMATICS AND COMPUTING↗

Radiotherapy for Early Mediastinal Hodgkin Lymphoma According to the German Hodgkin Study Group (GHSG): The Roles of Intensity-Modulated Radiotherapy and Involved-Node Radiotherapy

Purpose: Cure rates of early Hodgkin lymphoma (HL) are high, and avoidance of late complications and second malignancies have become increasingly important. This comparative treatment planning study analyzes to what extent target volume reduction to involved-node (IN) and intensity-modulated (IM) radiotherapy (RT), compared with involved-field (IF) and three-dimensional (3D) RT, can reduce doses to organs at risk (OAR). Methods and Materials: Based on 20 computed tomography (CT) datasets of patients with early unfavorable mediastinal HL, we created treatment plans for 3D-RT and IMRT for both the IF and IN according to the guidelines of the German Hodgkin Study Group (GHSG). As OAR, we defined heart, lung, breasts, and spinal cord. Dose-volume histograms (DVHs) were evaluated for planning target volumes (PTVs) and OAR. Results: Average IF-PTV and IN-PTV were 1705 cm{sup 3} and 1015 cm{sup 3}, respectively. Mean doses to the PTVs were almost identical for all plans. For IF-PTV/IN-PTV, conformity was better with IMRT and homogeneity was better with 3D-RT. Mean doses to the heart (17.94/9.19 Gy for 3D-RT and 13.76/7.42 Gy for IMRT) and spinal cord (23.93/13.78 Gy for 3D-RT and 19.16/11.55 Gy for IMRT) were reduced by IMRT, whereas mean doses to lung (10.62/8.57 Gy for 3D-RT and 12.77/9.64 Gy for IMRT) and breasts (left 4.37/3.42 Gy for 3D-RT and 6.04/4.59 Gy for IMRT, and right 2.30/1.63 Gy for 3D-RT and 5.37/3.53 Gy for IMRT) were increased. Volume exposed to high doses was smaller for IMRT, whereas volume exposed to low doses was smaller for 3D-RT. Pronounced benefits of IMRT were observed for patients with lymph nodes anterior to the heart. IN-RT achieved substantially better values than IF-RT for almost all OAR parameters, i.e., dose reduction of 20% to 50%, regardless of radiation technique. Conclusions: Reduction of target volume to IN most effectively improves OAR sparing, but is still considered investigational. For the time being, IMRT should be considered for large PTVs especially when the anterior mediastinum is involved.

62 RADIOLOGY AND NUCLEAR MEDICINE↗