Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “optimization methods”

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 505 records · Page 28
All sourcesNASA NTRSDOE OSTI
DOE OSTI2015

SU-E-T-07: 4DCT Robust Optimization for Esophageal Cancer Using Intensity Modulated Proton Therapy

Purpose: To develop a 4DCT robust optimization method to reduce the dosimetric impact from respiratory motion in intensity modulated proton therapy (IMPT) for esophageal cancer. Methods: Four esophageal cancer patients were selected for this study. The different phases of CT from a set of 4DCT were incorporated into the worst-case dose distribution robust optimization algorithm. 4DCT robust treatment plans were designed and compared with the conventional non-robust plans. Result doses were calculated on the average and maximum inhale/exhale phases of 4DCT. Dose volume histogram (DVH) band graphic and ΔD95%, ΔD98%, ΔD5%, ΔD2% of CTV between different phases were used to evaluate the robustness of the plans. Results: Compare to the IMPT plans optimized using conventional methods, the 4DCT robust IMPT plans can achieve the same quality in nominal cases, while yield a better robustness to breathing motion. The mean ΔD95%, ΔD98%, ΔD5% and ΔD2% of CTV are 6%, 3.2%, 0.9% and 1% for the robustly optimized plans vs. 16.2%, 11.8%, 1.6% and 3.3% from the conventional non-robust plans. Conclusion: A 4DCT robust optimization method was proposed for esophageal cancer using IMPT. We demonstrate that the 4DCT robust optimization can mitigate the dose deviation caused by the diaphragm motion.

60 APPLIED LIFE SCIENCES↗
NASA NTRS1997

Algorithms for Multiple Fault Diagnosis With Unreliable Tests

In this paper, we consider the problem of constructing optimal and near-optimal multiple fault diagnosis (MFD) in bipartite systems with unreliable (imperfect) tests. It is known that exact computation of conditional probabilities for multiple fault diagnosis is NP-hard. The novel feature of our diagnostic algorithms is the use of Lagrangian relaxation and subgradient optimization methods to provide: (1) near optimal solutions for the MFD problem, and (2) upper bounds for an optimal branch-and-bound algorithm. The proposed method is illustrated using several examples. Computational results indicate that: (1) our algorithm has superior computational performance to the existing algorithms (approximately three orders of magnitude improvement), (2) the near optimal algorithm generates the most likely candidates with a very high accuracy, and (3) our algorithm can find the most likely candidates in systems with as many as 1000 faults.

Shakeri, Mojdeh↗
DOE OSTI2016

SU-F-J-06: Optimized Patient Inclusion for NaF PET Response-Based Biopsies

Purpose: A method to guide mid-treatment biopsies using quantitative [F-18]NaF PET/CT response is being investigated in a clinical trial. This study aims to develop methodology to identify patients amenable to mid-treatment biopsy based on pre-treatment imaging characteristics. Methods: 35 metastatic prostate cancer patients had NaF PET/CT scans taken prior to the start of treatment and 9–12 weeks into treatment. For mid-treatment biopsy targeting, lesions must be at least 1.5 cm{sup 3} and located in a clinically feasible region (lumbar/sacral spine, pelvis, humerus, or femur). Three methods were developed based on number of lesions present prior to treatment: a feasibility-restricted method, a location-restricted method, and an unrestricted method. The feasibility restricted method only utilizes information from lesions meeting biopsy requirements in the pre-treatment scan. The unrestricted method accounts for all lesions present in the pre-treatment scan. For each method, optimized classification cutoffs for candidate patients were determined. Results: 13 of the 35 patients had enough lesions at the mid-treatment for biopsy candidacy. Of 1749 lesions identified in all 35 patients at mid-treatment, only 9.8% were amenable to biopsy. Optimizing the feasibility-restricted method required 4 lesions at pre-treatment meeting volume and region requirements for biopsy, resulting patient identification sensitivity of 0.8 and specificity of 0.7. Of 6 false positive patients, only one patient lacked lesions for biopsy. Restricting for location alone showed poor results (sensitivity 0.2 and specificity 0.3). The optimized unrestricted method required patients have at least 37 lesions in pretreatment scan, resulting in a sensitivity of 0.8 and specificity of 0.8. There were 5 false positives, only one lacked lesions for biopsy. Conclusion: Incorporating the overall pre-treatment number of NaF PET/CT identified lesions provided best prediction for identifying candidate patients for mid-treatment biopsy. This study provides validity for prediction-based inclusion criteria that can be extended to various clinical trial scenarios. Funded by Prostate Cancer Foundation.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2018

A filled function method for global optimization with inequality constraints

In this paper, we propose a new filled function method for finding a global minimizer of global optimization with inequality constraints. The proposed filled function is a continuously differentiable function with only one parameter. Then, we can use classical local optimization methods to find a better minimizer of the proposed filled function with a few parameter adjustment. The numerical experiments are made and the results show that the proposed filled function method is effective.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Robust Operation of Soft Open Points in Active Distribution Networks with High Penetration of Photovoltaic Integration

Distributed generators (DGs) including photovoltaic panels (PVs) have been integrated dramatically in active distribution networks (ADNs). Due to the strong volatility and uncertainty, the high penetration of PV generation immensely exacerbates the conditions of voltage violation in ADNs. However, the emerging flexible interconnection technology based on soft open points (SOPs) provides increased controllability and flexibility to the system operation. For fully exploiting the regulation ability of SOPs to address the problems caused by PV, this paper proposes a robust optimization method to achieve the robust optimal operation of SOPs in ADNs. A two-stage adjustable robust optimization model is built to tackle the uncertainties of PV outputs, in which robust operation strategies of SOPs are generated to eliminate the voltage violations and reduce the power losses of ADNs. A column-and-constraint generation (C&CG) algorithm is developed to solve the proposed robust optimization model, which are formulated as second-order cone program (SOCP) to facilitate the accuracy and computation efficiency. Here, case studies on the modified IEEE 33-node system and comparisons with the deterministic optimization approach are conducted to verify the effectiveness and robustness of the proposed method.

14 SOLAR ENERGY↗
NASA NTRS1999

High Speed Civil Transport Design Using Collaborative Optimization and Approximate Models

The design of supersonic aircraft requires complex analysis in multiple disciplines, posing, a challenge for optimization methods. In this thesis, collaborative optimization, a design architecture developed to solve large-scale multidisciplinary design problems, is applied to the design of supersonic transport concepts. Collaborative optimization takes advantage of natural disciplinary segmentation to facilitate parallel execution of design tasks. Discipline-specific design optimization proceeds while a coordinating mechanism ensures progress toward an optimum and compatibility between disciplinary designs. Two concepts for supersonic aircraft are investigated: a conventional delta-wing design and a natural laminar flow concept that achieves improved performance by exploiting properties of supersonic flow to delay boundary layer transition. The work involves the development of aerodynamics and structural analyses, and integration within a collaborative optimization framework. It represents the most extensive application of the method to date.

Manning, Valerie Michelle↗
NASA NTRS2005

Integrated System-Level Optimization for Concurrent Engineering With Parametric Subsystem Modeling

The introduction of concurrent design practices to the aerospace industry has greatly increased the productivity of engineers and teams during design sessions as demonstrated by JPL's Team X. Simultaneously, advances in computing power have given rise to a host of potent numerical optimization methods capable of solving complex multidisciplinary optimization problems containing hundreds of variables, constraints, and governing equations. Unfortunately, such methods are tedious to set up and require significant amounts of time and processor power to execute, thus making them unsuitable for rapid concurrent engineering use. This paper proposes a framework for Integration of System-Level Optimization with Concurrent Engineering (ISLOCE). It uses parametric neural-network approximations of the subsystem models. These approximations are then linked to a system-level optimizer that is capable of reaching a solution quickly due to the reduced complexity of the approximations. The integration structure is described in detail and applied to the multiobjective design of a simplified Space Shuttle external fuel tank model. Further, a comparison is made between the new framework and traditional concurrent engineering (without system optimization) through an experimental trial with two groups of engineers. Each method is evaluated in terms of optimizer accuracy, time to solution, and ease of use. The results suggest that system-level optimization, running as a background process during integrated concurrent engineering sessions, is potentially advantageous as long as it is judiciously implemented.

Schuman, Todd↗
NASA NTRS1990

Differential equation based method for accurate approximations in optimization

A method to efficiently and accurately approximate the effect of design changes on structural response is described. The key to this method is to interpret sensitivity equations as differential equations that may be solved explicitly for closed form approximations, hence, the method is denoted the Differential Equation Based (DEB) method. Approximations were developed for vibration frequencies, mode shapes and static displacements. The DEB approximation method was applied to a cantilever beam and results compared with the commonly-used linear Taylor series approximations and exact solutions. The test calculations involved perturbing the height, width, cross-sectional area, tip mass, and bending inertia of the beam. The DEB method proved to be very accurate, and in most cases, was more accurate than the linear Taylor series approximation. The method is applicable to simultaneous perturbation of several design variables. Also, the approximations may be used to calculate other system response quantities. For example, the approximations for displacements are used to approximate bending stresses.

Pritchard, Jocelyn I.↗
NASA NTRS2003

Imparting Desired Attributes by Optimization in Structural Design

Commonly available optimization methods typically produce a single optimal design as a Constrained minimum of a particular objective function. However, in engineering design practice it is quite often important to explore as much of the design space as possible with respect to many attributes to find out what behaviors are possible and not possible within the initially adopted design concept. The paper shows that the very simple method of the sum of objectives is useful for such exploration. By geometrical argument it is demonstrated that if every weighting coefficient is allowed to change its magnitude and its sign then the method returns a set of designs that are all feasible, diverse in their attributes, and include the Pareto and non-Pareto solutions, at least for convex cases. Numerical examples in the paper include a case of an aircraft wing structural box with thousands of degrees of freedom and constraints, and over 100 design variables, whose attributes are structural mass, volume, displacement, and frequency. The method is inherently suitable for parallel, coarse-grained implementation that enables exploration of the design space in the elapsed time of a single structural optimization.

Sobieszczanski-Sobieski, Jaroslaw↗
DOE OSTI2016

SU-F-T-198: Dosimetric Comparison of Carbon and Proton Radiotherapy for Recurrent Nasopharynx Carcinoma

Purpose: Various radiotherapy planning methods for locally recurrent nasopharynx carcinoma (R-NPC) have been proposed. The purpose of this study was to compare carbon and proton therapy for the treatment of R-NPC in terms of dose coverage for target volume and sparing for organs at risk (OARs). Methods: Six patients who were suffering from R-NPC and treated using carbon therapy were selected for this study. Treatment plans with a total dose of 57.5Gy (RBE) in 23 fractions were made using SIEMENS Syngo V11. An intensity-modulated radiotherapy optimization method was chosen for carbon plans (IMCT) while for proton plans both intensity-modulated radiotherapy (IMPT) and single beam optimization (proton-SBO) methods were chosen. Dose distributions, dose volume parameters, and selected dosimetric indices for target volumes and OARs were compared for all treatment plans. Results: All plans provided comparable PTV coverage. The volume covered by 95% of the prescribed dose was comparable for all three plans. The average values were 96.11%, 96.24% and 96.11% for IMCT, IMPT, and proton-SBO respectively. A significant reduction of the 80% and 50% dose volumes were observed for the IMCT plans compared to the IMPT and proton-SBO plans. Critical organs lateral to the target such as brain stem and spinal cord were better spared by IMPT than by proton-SBO, while IMCT spared those organs best. For organs in the beam path, such as parotid glands, the mean dose results were similar for all three plans. Conclusion: Carbon plans yielded better dose conformity than proton plans. They provided similar or better target coverage while significantly lowering the dose for normal tissues. Dose sparing for critical organs in IMPT plans was better than proton-SBO, however, IMPT is known to be more sensitive to range uncertainties. For proton plans it is essential to find a balance between the two optimization methods.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2021

Enhanced Data Efficiency Using Deep Neural Networks and Gaussian Processes for Aerodynamic Design Optimization

Adjoint-based optimization methods are attractive for aerodynamic shape design primarily due to their computational costs being independent of the dimensionality of the input space and their ability to generate high-fidelity gradients that can then be used in a gradient-based optimizer. This makes them very well suited for high-fidelity simulation based aerodynamic shape optimization of highly parametrized geometries such as aircraft wings. However, the development of adjoint-based solvers involve careful mathematical treatment and their implementation require detailed software development. Furthermore, they can become prohibitively expensive when multiple optimization problems are being solved, each requiring multiple restarts to circumvent local optima. In this work, we propose a machine learning enabled, surrogate-based framework that replaces the expensive adjoint solver, without compromising on predicting predictive accuracy. Specifically, we first train a deep neural network (DNN) from training data generated from evaluating the high-fidelity simulation model on a model-agnostic design of experiments on the geometry shape parameters. The optimum shape may then be computed by using a gradient-based optimizer coupled with the trained DNN. Subsequently, we also perform a gradient-free Bayesian optimization, where the trained DNN is used as the prior mean. We observe that the latter framework (DNN-BO) improves upon the DNN-only based optimization strategy for the same computational cost. Overall, this framework predicts the true optimum with very high accuracy, while requiring far fewer high-fidelity function calls compared to the adjoint-based method. Furthermore, we show that multiple optimization problems can be solved with the same machine learning model with high accuracy, to amortize the offline costs associated with constructing our models. Our methodology finds applications in the early stages of aerospace design. (C) 2021 Published by Elsevier Masson SAS.

Renganathan, S. Ashwin↗
NASA NTRS2001

Simultaneous Aerodynamic Analysis and Design Optimization (SAADO) for a 3-D Flexible Wing

The formulation and implementation of an optimization method called Simultaneous Aerodynamic Analysis and Design Optimization (SAADO) are extended from single discipline analysis (aerodynamics only) to multidisciplinary analysis - in this case, static aero-structural analysis - and applied to a simple 3-D wing problem. The method aims to reduce the computational expense incurred in performing shape optimization using state-of-the-art Computational Fluid Dynamics (CFD) flow analysis, Finite Element Method (FEM) structural analysis and sensitivity analysis tools. Results for this small problem show that the method reaches the same local optimum as conventional optimization. However, unlike its application to the win,, (single discipline analysis), the method. as I implemented here, may not show significant reduction in the computational cost. Similar reductions were seen in the two-design-variable (DV) problem results but not in the 8-DV results given here.

Gumbert, Clyde R.↗
DOE OSTI2022

Turbine scale and siting considerations in wind plant layout optimization and implications for capacity density

Improvements in wind energy technology, reduced costs, and ambitious clean energy goals have led to projections of high wind contribution in coming years. Developing methodologies to design wind plants with a variety of siting constraints and turbine sizes helps enable high wind penetration, and gain a better understanding of how wind plants are sensitive to setback constraints and turbine design. In this paper, we present a two-step optimization method to simultaneously determine the optimal number of turbines and their locations in a wind plant domain divided into many small, discrete parcels. We present the optimized performance metrics of a wind plant optimized with different turbine sizes and ratings, and with different siting restrictions within the wind plant. Our results indicate that taller and larger turbines are more sensitive to increasing siting constraints. We also compare the optimal wind plant layouts and performance for wind plants optimized for minimum COE and maximum profit. Wind plants optimized for profit had 130%-190% of the capacity of plants optimized for COE, which demonstrates that the optimal results are greatly affected by the objective function, which should be carefully considered. Finally, in this paper we demonstrate the effect of increasing siting constraints on wind plant capacity density, and how the results change when different land areas are used to calculate capacity density. When using the entire wind plant boundary area to determine capacity density, increasing siting constraints decreases the capacity density. However, when we only use the available area (the area left after removing the siting constraints) to calculate the capacity density, increasing the siting constraints increases capacity density. This is a critical insight because of how capacity density is typically defined and used in research, and has important implications for assessment of technical potential and capacity expansion modeling, as well as future wind deployment potential.

17 WIND ENERGY↗
DOE OSTI2015

SU-E-T-527: Is CTV-Based Robust Optimized IMPT in Non-Small-Cell Lung Cancer Robust Against Respiratory Motion?

Purpose: To determine which proton planning technique on average-CT is more vulnerable to respiratory motion induced density changes and interplay effect among (a) IMPT of CTV-based minimax robust optimization with 5mm set-up error considered, (b, c) IMPT/SFUD of 5mm-expanded PTV optimization. Methods: Three planning techniques were optimized in Raystation with a prescription of 60/25 (Gy/fractions) and almost the same OAR constraints/objectives for each of 10 NSCLC patients. 4D dose without/with interplay effect was recalculated on eight 4D-CT phases and accumulated after deforming the dose of each phase to a reference (exhalation phase). The change of D98% of each CTV caused by density changes and interplay was determined. In addition, evaluation of the DVH information vector (D99%, D98%, D95%, Dave, D50%, D2%, D1%) which compares the whole DVH by η score = (cosine similarity × Pearson correlation coefficient − 0.9) × 1000 quantified the degree of DVH change: score below 100 indicates changed DVH. Results: Three 3D plans of each technique satisfied our clinical goals. D98% shift mean±SD (Gy) due to density changes was largest in (c): −0.78±1.1 while (a): −0.11±0.65 and (b): − 0.59±0.93. Also the shift due to interplay effect most was (c): −.54±0.70 whereas (a): −0.25±0.93 and (b): −0.12±0.13. Moreover lowest η score caused by density change was also (c): 69, while (a) and (b) kept around 90. η score also indicated less effect of interplay than density changes. Note that generally the changed DVH were still acceptable clinically. Paired T-tests showed a significantly smaller density change effect in (a) (p<0.05) than in (b) or (c) and no significant difference in interplay effect. Conclusion: CTV-based robust optimized IMPT was more robust against respiratory motion induced density changes than PTV-based IMPT and SFUD. The interplay effect was smaller than the effect of density changes and similar among the three techniques. The JSPS Core-to-Core Program (No. 23003), Japan Society for the Promotion of Science Grant-in-Aid for Scientific Research (No. 23390300), Grant-in-Aid for Young Scientists (B) (No. 21791194) and Grant-in-Aid for Cancer Research (H22-3rd Term Cancer Control-General-043)

60 APPLIED LIFE SCIENCES↗
DOE OSTI2025

A Comparison of Machine Learning Methods of Association Tested on Dense Nodal Arrays

The association of phase picks to form events is one of the fundamental components of seismology. Large and dense sensor networks, such as >1000 geophone arrays (and distributed acoustic sensing), offer unique challenges in association due to the vast numbers of observations and high likelihood of errant picks. In addition, the large number of stations can greatly increase the time it takes to perform the association. For this reason, machine learning (ML) methods might provide a more optimal method of association for such networks. In this work, we examine how well ML methods (e.g., Gaussian mixture model association, PhaseLink, and Graph Earthquake Neural Interpretation Engine) can incorporate dense seismic arrays into regional networks and how well they handle the increasing numbers of stations. Here, we test their capabilities on two dense seismic deployments, one within Rock Valley Nevada (52 nodes and a 9-station sparse local network), and the LArge-n Seismic Survey in Oklahoma dense nodal array (>1800 vertical-component geophones). Processing data from these two different styles of dense seismic deployments allows testing of how the ML algorithms can merge array data with a broader regional network, how they deal with poorly picked phases, and how they handle anthropogenic noise. We compare the ML-associated bulletins to those obtained using the Rapid Earthquake Association and Location algorithm, a more traditional method of association. We find that there are very small differences in results between the methods for small networks (<100 stations) with low pick rates. For large networks (>1000), there are enough errant picks that some of the ML methods start to create false events out of noise. We also find that the ML methods vary in computation time significantly but are all faster than the traditional method tested here.

58 GEOSCIENCES↗
NASA NTRS1987

An overview of the mathematical and statistical analysis component of RICIS

Mathematical and statistical analysis components of RICIS (Research Institute for Computing and Information Systems) can be used in the following problem areas: (1) quantification and measurement of software reliability; (2) assessment of changes in software reliability over time (reliability growth); (3) analysis of software-failure data; and (4) decision logic for whether to continue or stop testing software. Other areas of interest to NASA/JSC where mathematical and statistical analysis can be successfully employed include: math modeling of physical systems, simulation, statistical data reduction, evaluation methods, optimization, algorithm development, and mathematical methods in signal processing.

Hallum, Cecil R.↗
NASA NTRS1993

Nonlinear aerodynamics and the design of wing tips

This contract began in April 1990 with studies on the effect of wingtip shape on induced drag. That work considered problems of determining induced drag from computational aerodynamic methods and examined effects of wake roll up on the induced drag of elliptical and crescent wings. The research contract was continued in April 1991 and again in April 1992 with scope augmented to include considerations other than induced drag. The idea was to develop methods for comparing wing tip shapes on the basis of their effects on total drag, structural weight, and high lift performance. To accomplish this, work was done to improve methods for multidisciplinary analysis and optimization of wings. This report describes results of research conducted from April 1992 through March 1993. The general objective was to improve an existing wing optimization method and apply the method to specific problems of interest. The method, while a valuable tool for wing tip design studies, can be applied to more general problems, and has been applied to some of these other problems during its development. Specific goals that were accomplished are listed below and are explained in more detail in attached reports: analysis methods were improved in the area of structural modeling; small improvements to the high lift model were made, and ideas were collected on how clean wing c(sub l(sub max)) should vary with sweep - these ideas should provide a guide for future examination of high lift with experiment or CFD; and a calculation for drag due to the loss of leading edge suction was added to the wing analysis.

Kroo, Ilan↗
NASA NTRS2001

Simultaneous Aerodynamic Analysis and Design Optimization (SAADO) for a 3-D Flexible Wing

The formulation and implementation of an optimization method called Simultaneous Aerodynamic Analysis and Design Optimization (SAADO) are extended from single discipline analysis (aerodynamics only) to multidisciplinary analysis - in this case, static aero-structural analysis and applied to a simple 3-D wing problem. The method aims to reduce the computational expense incurred in performing shape optimization using state-of-the-art Computational Fluid Dynamics (CFD) flow analysis, Finite Element Method (FEM) structural analysis and sensitivity analysis tools. Results for this small problem show that the method reaches the same local optimum as conventional optimization. However, unlike its application to the rigid wing (single discipline analysis), the method, as implemented here, may not show significant reduction in the computational cost. Similar reductions were seen in the two-design-variable (DV) problem results but not in the 8-DV results given here.

Gumbert, Clyde R.↗