Search NASA⌕ Search

SEARCH · Search NASA

Results for “penalty 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 235 records · Page 13
All sourcesNASA NTRSDOE OSTI
NASA NTRS1991

A summary of NASA research on effects of heavy rain on airfoils

Results from a broad NASA research program are presented. The intention of the program was to obtain fundamental aerodynamic information regarding the effect of heavy rain on airplane performance. The take-off and landing characteristics are of particular concern, and the aim of the program is to understand the physical phenomena associated with any aerodynamic performance penalty that may occur during a rain encounter. This overview includes results of recent attempts to measure high-intensity, short-duration rainfall, a discussion of some of the earlier analytical investigations of rain effects on airfoils, a review of some promising experimental methods for evaluating rain effects, and some important scaling considerations for extrapolating model data.

Dana J. Dunham↗
NASA NTRS2021

An Information Theory Approach to Physical Domain Discovery

The project of physics discovery is often equivalent to finding the most concise description of a physical system. The description with optimum predictive capability for a dataset generated by a physical system is one that minimizes both predictive error on the dataset and the complexity of the description. The discovery of the governing physics of a system can therefore be viewed as a mathematical optimization problem. We outline here a method to optimize the description of arbitrarily complex physical systems by minimizing the entropy of the description of the system. The Recursive Domain Partitioning (RDP) procedure finds the optimum partitioning of each physical domain into subdomains, and the optimum predictive function within each subdomain. Penalty functions are introduced to limit the complexity of the predictive function within each domain. Examples are shown in 1D and 2D. In 1D, the technique effectively discovers the elastic and plastic regions within a stress-strain curve generated by simulations of amorphous carbon material, while in 2D the technique discovers the free-flow region and the inertially-obstructed flow region in the simulation of fluid flow across a plate.

Daniel Shea↗
NASA NTRS1993

A new approach to mixed H2/H infinity controller synthesis using gradient-based parameter optimization methods

In the past few years, the mixed H(sub 2)/H-infinity control problem has been the object of much research interest since it allows the incorporation of robust stability into the LQG framework. The general mixed H(sub 2)/H-infinity design problem has yet to be solved analytically. Numerous schemes have considered upper bounds for the H(sub 2)-performance criterion and/or imposed restrictive constraints on the class of systems under investigation. Furthermore, many modern control applications rely on dynamic models obtained from finite-element analysis and thus involve high-order plant models. Hence the capability to design low-order (fixed-order) controllers is of great importance. In this research a new design method was developed that optimizes the exact H(sub 2)-norm of a certain subsystem subject to robust stability in terms of H-infinity constraints and a minimal number of system assumptions. The derived algorithm is based on a differentiable scalar time-domain penalty function to represent the H-infinity constraints in the overall optimization. The scheme is capable of handling multiple plant conditions and hence multiple performance criteria and H-infinity constraints and incorporates additional constraints such as fixed-order and/or fixed structure controllers. The defined penalty function is applicable to any constraint that is expressible in form of a real symmetric matrix-inequity.

Ly, Uy-Loi↗
NASA NTRSDate not supplied

Multiple-Effector Control Allocation: Theory and Practice

Redundant controls are ubiquitous in many aerospace domains, including aircraft, spacecraft, and launch vehicles. Multiple control effectors having different capabilities (e.g., reaction controls, aerosurfaces, and vectored engines) must be optimally allocated to achieve high-performance and robust vehicle control, while considering different constraints, cost, or performance metrics associated with various control mixing schemes. These can include force or deflection limits, rate limits, propellant usage or drag penalties, cross-axis coupling, uncertainties, or even servoelastic interactions. This presentation introduces the fundamentals of control allocation theory, based on linear algebra and convex sets, using a handful of examples taken from the air and space vehicle domains. It is shown that even simple methods can provide verifiable and quantifiable performance benefits over ad-hoc approaches, and can support enhanced functions like fault tolerance and control reconfiguration.

Bill Benson↗
NASA NTRS1990

Multi-disciplinary optimization of aeroservoelastic systems

Efficient analytical and computational tools for simultaneous optimal design of the structural and control components of aeroservoelastic systems are presented. The optimization objective is to achieve aircraft performance requirements and sufficient flutter and control stability margins with a minimal weight penalty and without violating the design constraints. Analytical sensitivity derivatives facilitate an efficient optimization process which allows a relatively large number of design variables. Standard finite element and unsteady aerodynamic routines are used to construct a modal data base. Minimum State aerodynamic approximations and dynamic residualization methods are used to construct a high accuracy, low order aeroservoelastic model. Sensitivity derivatives of flutter dynamic pressure, control stability margins and control effectiveness with respect to structural and control design variables are presented. The performance requirements are utilized by equality constraints which affect the sensitivity derivatives. A gradient-based optimization algorithm is used to minimize an overall cost function. A realistic numerical example of a composite wing with four controls is used to demonstrate the modeling technique, the optimization process, and their accuracy and efficiency.

Karpel, Mordechay↗
NASA NTRS1992

Multidisciplinary optimization of aeroservoelastic systems using reduced-size models

Efficient analytical and computational tools for simultaneous optimal design of the structural and control components of aeroservoelastic systems are presented. The optimization objective is to achieve aircraft performance requirements and sufficient flutter and control stability margins with a minimal weight penalty and without violating the design constraints. Analytical sensitivity derivatives facilitate an efficient optimization process which allows a relatively large number of design variables. Standard finite element and unsteady aerodynamic routines are used to construct a modal data base. Minimum State aerodynamic approximations and dynamic residualization methods are used to construct a high accuracy, low order aeroservoelastic model. Sensitivity derivatives of flutter dynamic pressure, control stability margins and control effectiveness with respect to structural and control design variables are presented. The performance requirements are utilized by equality constraints which affect the sensitivity derivatives. A gradient-based optimization algorithm is used to minimize an overall cost function. A realistic numerical example of a composite wing with four controls is used to demonstrate the modeling technique, the optimization process, and their accuracy and efficiency.

Karpel, Mordechay↗
NASA NTRS1973

Optimal detumbling of a large manned spacecraft using an internal moving mass

The use of a movable mass control system to stabilize an arbitrarily tumbling asymmetric vehicle about the maximum inertia axis is considered. A first-order gradient optimization technique is used to minimize angular velocity components along the intermediate and minimum inertia axes. This method permits a wide range of initial guesses for mass position history. Motion of the control mass is along a linear track fixed in the vehicle. The control variable is taken as mass acceleration with respect to body coordinates. Motion is limited to defined quantities and a penalty function is used to insure a given range of positions. Numerical solutions of the optimization equations verify that minimum time detumbling is achieved with the largest permissible movable mass, length of linear track, and positions of the mass on the two coordinates perpendicular to the linear motion.

Kunciw, B. G.↗
NASA NTRS1995

Lead-Lag Control for Helicopter Vibration and Noise Reduction

As a helicopter transitions from hover to forward flight, the main rotor blades experience an asymmetry in flow field around the azimuth, with the blade section tangential velocities increasing on the advancing side and decreasing on the retreating side. To compensate for the reduced dynamic pressure on the retreating side, the blade pitch angles over this part of the rotor disk are increased. Eventually, a high enough forward speed is attained to produce compressibility effects on the advancing side of the rotor disk and stall on the retreating side. The onset of these two phenomena drastically increases the rotor vibratory loads and power requirements, thereby effectively establishing a limit on the maximum achievable forward speed. The alleviation of compressibility and stall (and the associated decrease in vibratory loads and power) would potentially result in an increased maximum forward speed. In the past, several methods have been examined and implemented to reduce the vibratory hub loads. Some of these methods are aimed specifically at alleviating vibration at very high flight speeds and increasing the maximum flight speed, while others focus on vibration reduction within the conventional flight envelope. Among the later are several types passive as well as active schemes. Passive schemes include a variety of vibration absorbers such as mechanical springs, pendulums, and bifilar absorbers. These mechanism are easy to design and maintain, but incur significant weight and drag penalties. Among the popular active control schemes in consideration are Higher Harmonic Control (HHC) and Individual Blade Control (IBC). HHC uses a conventional swash plate to generate a multi-cyclic pitch input to the blade. This requires actuators capable of sufficiently high power and bandwidth, increasing the cost and weight of the aircraft. IBC places actuators in the rotating reference frame, requiring the use of slip rings capable of transferring enough power to the actuators. Both schemes cause an increase in pitch link loads. Trailing Edge Flap (TEF) deployment can also used to generate unsteady aerodynamic forces and moments that counter the original vibratory loads, and thereby reduce rotor vibrations. While the vibrations absorbers, HHC, IBC, and TEF concepts discussed above attempt to reduce the vibratory loads, they do not specifically address the phenomena causing the vibrations at high advance ratios. One passive method that attempts to directly alleviate compressibility and stall, instead of reducing the ensuing vibrations, is the use of advanced tip designs. Taper, sweep, anhedral, and the manipulation of other geometric properties of the blade tips can reduce the severity of stall and compressibility effects , as well as reduce rotor power. A completely different approach to solve these problems is the tiltrotor configuration. As the forward velocity of the aircraft increases, the rotors, in this case, are tilted forward until they are perpendicular to the flow and act as propellers. This eliminates the edgewise flow encountered by conventional rotors and circumvents all the problems associated with flow asymmetry. However, the success involves a tremendous increase in cost and complexity of the aircraft. Another possible approach that has been proposed for the alleviation of vibratory loads at high forward flight speeds involves the use of controlled lead-lag motions to reduce the asymmetry in flow. A correctly phased 1/rev controlled lag motion could be introduced such that it produces a backward velocity on the advancing side and a forward velocity on the retreating side, to delay compressibility effects and stall to a higher advance ratio. Using a large enough lead-lag amplitude, the tip velocities could be reduced to levels encountered in hover. This concept was examined by two groups in the 1950's and early 1960's. In the United States, the Research Labs Division of United Aircraft developed a large lead-lag motion rotor, meant to achieve lag motion amplitudes up to 45 degrees. In order to reduce the required actuation force, the blade hinges were moved to 40% of the blade radius to increase the rotating lag frequency to approximately 1/rev. The blade hinges were redesigned to produce a flap-lag coupling so the large flapwise aerodynamic loads could be exploited to actuate the blades in the lag direction. A wind tunnel test of this rotor concept revealed actuation and blade motion scheduling problems. The project was eventually discontinued due to these problems and high blade stresses. Around the same time, at Boelkow in Germany, a similar lead-lag rotor program was conducted under the leadership of Hans Derschmidt. Here, too, the blade hinges were moved outboard to 34% radius to reduce the actuation loads. The main difference between this and the United Aircraft program was the use of a mechanical actuation scheme with maximum lead-lag motions of 400. This program was also discontinued for unclear reasons. The present study is directed toward conducting a comprehensive analytical examination to evaluate the effectiveness of controlled lead-lag motions in reducing vibratory hub loads and increasing maximum flight speed. Since both previous studies on this subject were purely experimental, only a limited data set and physical understanding of the problem was obtained. With the currently available analytical models and computational resources, the present effort is geared toward developing an in-depth physical understanding of the precise underlying mechanisms by which vibration reduction may be achieved. Additionally, in recognition of the fact that large amplitude lead-lag motions would - (i) be difficult to implement, and (ii) produce very large blade stresses; the present study examines the potential of only moderate-to-small lead-lag motions for reduction of vibratory hub loads. Using such an approach, the emphasis is not on eliminating the periodic variations in tangential velocity at the blade tip, but at best reducing these variations slightly so that compressibility and stall are delayed to slightly higher advance ratios. This study was conducted in two steps. In the first step, a hingeless helicopter rotor was modeled using rigid blades undergoing flap-lag-torsion rotations about spring restrained hinges and bearings. This model was then modified by separating the lead-lag degree of freedom into two components, a free and a prescribed motion. Using this model, a parametric study of the effect of phase and amplitude of a prescribed lead-lag motion on hub vibration was conducted. The data gathered was analyzed to obtain an understanding of the basic physics of the problem and show the capability of this method to reduce vibration and expand the flight envelope. In the second half of the study, the similar analysis was conducted using an elastic blade model to confirm the effects predicted by the simpler model.

Gandhi, Farhan↗
DOE OSTI2024

Enhancing the cooling performance of thermocouples: a power-constrained topology optimization procedure

Abstract Heat pumping through thermoelectric devices has many advantages over traditional cooling. However, their current efficiency is a limiting factor in their implementation. In this paper, we approach the non-convex topology optimization of thermoelectrical elements for cooling applications through the method of moving asymptotes (MMA) to improve their cooling capabilities per watt usage. The optimization problem is defined for a given power budget, aiming for the minimum temperature with a known heat pumping need. The introduction of power as a constraint justifies the introduction of the voltage gradient across the thermocouple as a design variable to maintain the thermoelectrical device in its optimum power-to-heat extraction ratio. To better understand the convergence of this non-convex problem, we present a two-variable analytical thermoelectric optimization model. This example provides information on how to select the penalty parameters used to scale the three material coefficients involved in the problem to obtain lower objective values and better convergence using MMA. The analytical model shows the non-convexity of the problem and provides the recommendation to use penalization coefficients of the form $$p_k=p_{\sigma }>p_{\alpha }=1$$ p k = p σ > p α = 1 for the thermal conductivity, electrical conductivity, and Seebeck coefficients. We tested these penalization coefficients through optimizations of a model based on the 1MC10-031 commercial thermoelectric-cooler (TEC) using the finite element method (FEM). These penalization coefficients provided local minima without the need for volume constraints. With this procedure, we found designs that provided temperatures close to 10 degrees lower using 60% less semiconductor material volume compared to the initial design.

Gutiérrez, G. Reales↗
NASA NTRS2000

Flexible Wing Model for Structural Sizing and Multidisciplinary Design Optimization of a Strut-Braced Wing

This paper describes a structural and aeroelastic model for wing sizing and weight calculation of a strut-braced wing. The wing weight is calculated using a newly developed structural weight analysis module considering the special nature of strut-braced wings. A specially developed aeroelastic model enables one to consider wing flexibility and spanload redistribution during in-flight maneuvers. The structural model uses a hexagonal wing-box featuring skin panels, stringers, and spar caps, whereas the aerodynamics part employs a linearized transonic vortex lattice method. Thus, the wing weight may be calculated from the rigid or flexible wing spanload. The calculations reveal the significant influence of the strut on the bending material weight of the wing. The use of a strut enables one to design a wing with thin airfoils without weight penalty. The strut also influences wing spanload and deformations. Weight savings are not only possible by calculation and iterative resizing of the wing structure according to the actual design loads. Moreover, as an advantage over the cantilever wing, employment of the strut twist moment for further load alleviation leads to increased savings in structural weight.

Gern, Frank H.↗
NASA NTRS1974

An automated water iodinating subsystem for manned space flight

Controlling microbial growth by injecting iodine (l2) into water supplies is a widely acceptable technique, but requires a specialized injection method for space flight. An electrochemical l2 injection method and l2 level monitor are discussed in this paper, which also describe iodination practices previously used in the manned space program and major l2 biocidal characteristics. The development and design of the injector and monitor are described, and results of subsequent experiments are presented. Also presented are expected vehicle penalties for utilizing the l2 injector in certain space missions, especially the Space Shuttle, and possible injector failure modes and their criticality.

Houck, O. K.↗
NASA NTRS1974

Optimal detumbling of a large manned spacecraft using an internal moving mass

This investigation deals with the use of a movable mass control system to stabilize a tumbling asymmetric spacecraft about the maximum inertia axis. A first-order gradient optimization technique is used to minimize angular velocity components along the intermediate and minimum inertia axes, thus, permitting a wide range of initial guesses for mass position history. Motion of the control mass is along a linear track fixed in the vehicle. The control variable is taken as mass acceleration with respect to body coordinates. Motion is limited to defined quantities and a penalty function is used to insure a given range of positions. Numerical solutions of the optimization equations verify that minimum time detumbling is achieved with the largest permissible movable mass, length of linear track, and positions of the mass on the two coordinates perpendicular to the linear motion. The optimal method permits detumbling in about one-fourth the time when compared to a force control law formulation available in the literature.

Kunciw, B. G.↗
NASA NTRS1991

Design optimization of space structures

The topology-shape-size optimization of space structures is investigated through Kikuchi's homogenization method. The method starts from a 'design domain block,' which is a region of space into which the structure is to materialize. This domain is initially filled with a finite element mesh, typically regular. Force and displacement boundary conditions corresponding to applied loads and supports are applied at specific points in the domain. An optimal structure is to be 'carved out' of the design under two conditions: (1) a cost function is to be minimized, and (2) equality or inequality constraints are to be satisfied. The 'carving' process is accomplished by letting microstructure holes develop and grow in elements during the optimization process. These holes have a rectangular shape in two dimensions and a cubical shape in three dimensions, and may also rotate with respect to the reference axes. The properties of the perforated element are obtained through an homogenization procedure. Once a hole reaches the volume of the element, that element effectively disappears. The project has two phases. In the first phase the method was implemented as the combination of two computer programs: a finite element module, and an optimization driver. In the second part, focus is on the application of this technique to planetary structures. The finite element part of the method was programmed for the two-dimensional case using four-node quadrilateral elements to cover the design domain. An element homogenization technique different from that of Kikuchi and coworkers was implemented. The optimization driver is based on an augmented Lagrangian optimizer, with the volume constraint treated as a Courant penalty function. The optimizer has to be especially tuned to this type of optimization because the number of design variables can reach into the thousands. The driver is presently under development.

Felippa, Carlos↗
NASA NTRS1975

Primer vector theory and applications

A method developed to compute two-body, optimal, N-impulse trajectories was presented. The necessary conditions established define the gradient structure of the primer vector and its derivative for any set of boundary conditions and any number of impulses. Inequality constraints, a conjugate gradient iterator technique, and the use of a penalty function were also discussed.

Jezewski, D. J.↗
NASA NTRS1988

Optimal placement of tuning masses for vibration reduction in helicopter rotor blades

Described are methods for reducing vibration in helicopter rotor blades by determining optimum sizes and locations of tuning masses through formal mathematical optimization techniques. An optimization procedure is developed which employs the tuning masses and corresponding locations as design variables which are systematically changed to achieve low values of shear without a large mass penalty. The finite-element structural analysis of the blade and the optimization formulation require development of discretized expressions for two performance parameters: modal shaping parameter and modal shear amplitude. Matrix expressions for both quantities and their sensitivity derivatives are developed. Three optimization strategies are developed and tested. The first is based on minimizing the modal shaping parameter which indirectly reduces the modal shear amplitudes corresponding to each harmonic of airload. The second strategy reduces these amplitudes directly, and the third strategy reduces the shear as a function of time during a revolution of the blade. The first strategy works well for reducing the shear for one mode responding to a single harmonic of the airload, but has been found in some cases to be ineffective for more than one mode. The second and third strategies give similar results and show excellent reduction of the shear with a low mass penalty.

Pritchard, Jocelyn I.↗
NASA NTRS2006

A LEAST ABSOLUTE SHRINKAGE AND SELECTION OPERATOR (LASSO) FOR NONLINEAR SYSTEM IDENTIFICATION

Identification of parametric nonlinear models involves estimating unknown parameters and detecting its underlying structure. Structure computation is concerned with selecting a subset of parameters to give a parsimonious description of the system which may afford greater insight into the functionality of the system or a simpler controller design. In this study, a least absolute shrinkage and selection operator (LASSO) technique is investigated for computing efficient model descriptions of nonlinear systems. The LASSO minimises the residual sum of squares by the addition of a 1 penalty term on the parameter vector of the traditional 2 minimisation problem. Its use for structure detection is a natural extension of this constrained minimisation approach to pseudolinear regression problems which produces some model parameters that are exactly zero and, therefore, yields a parsimonious system description. The performance of this LASSO structure detection method was evaluated by using it to estimate the structure of a nonlinear polynomial model. Applicability of the method to more complex systems such as those encountered in aerospace applications was shown by identifying a parsimonious system description of the F/A-18 Active Aeroelastic Wing using flight test data.

Kukreja, Sunil L.↗
NASA NTRS1992

A variational method for finite element stress recovery: Applications in one-dimension

It is well-known that stresses (and strains) calculated by a displacement-based finite element analysis are generally not as accurate as the displacements. In addition, the calculated stress field is typically discontinuous at element interfaces. Because the stresses are typically of more interest than the displacements, several procedures have been proposed to obtain a smooth stress field, given the finite element stresses, and to improve the accuracy. Hinton and Irons introduced global least squares smoothing of discrete data defined on a plane using a finite element formulation. Tessler and co-workers recently developed a conceptually similar formulation for smoothing of two-dimensional data based on a discrete least square approximation with a penalty constraint. The penalty constraint results in a stress field which is C(exp 1)-continuous, a result not previously obtained. The approach requires additional, 'smoothing' finite element analysis and for their two-dimensional application, they used a conforming C(exp 0)-continuous triangular finite element based on a conforming plate element. This paper presents the results of a detailed investigation into the application of Tessler's smoothing procedure to the smoothing of finite element stresses from one-dimensional problems. Although the one-dimensional formulation has some practical applicability, such as in truss, beam, axisymmetric mechanics, and one-dimensional heat conduction, the primary motivation for developing the one-dimensional smoothing case is to explore the characteristics of the general smoothing strategy. In particular, it is used to describe the behavior of the method and to explore the suitability of criteria proposed for the smoothing analysis. Prior to presenting numerical results, the variational formulation of the smoothing strategy is presented and a criterion for the smoothing analysis is described.

Riggs, H. Ronald↗
DOE OSTI2024

Precision Calcination Mechanism of CaCO 3 to High-Porosity Nanoscale CaO CO 2 Sorbent Revealed by Direct In Situ Observations

Deploying energy storage and carbon capture at scale is hindered by the substantial endothermic penalty of decomposing CaCO 3 to CaO and CO 2 , and the rapid loss of CO 2 absorption capacity by CaO sorbent particles due to sintering at the high requisite decomposition temperatures. The decomposition reaction mechanism underlying sorbent deactivation remains unclear at the atomic level and nanoscale due to past reliance on postmortem characterization methods with insufficient spatial and temporal resolution. Thus, elucidating the important CaCO 3 decomposition reaction pathway requires direct observation by time-resolved (sub-)nanoscale methods. Here, chemical and structural dynamics during the decomposition of CaCO 3 nanoparticles to nanoporous CaO particles comprising high-surface-area CaO nanocrystallites are examined. Comparing in situ transmission electron microscopy (TEM) and synchrotron X-ray diffraction experiments gives key insights into the dynamics of nanoparticle calcination, involving anisotropic CaCO 3 thermal distortion before conversion to thermally dilated energetically stable CaO crystallites. Time-resolved TEM uncovered a novel CaO formation mechanism involving heterogeneous nucleation at extended CaCO 3 defects followed by sweeping reaction front motion across the initial CaCO 3 particle. These observations clarify longstanding, yet incomplete, reaction mechanisms and kinetic models lacking accurate information about (sub-)nanoscale dynamics, while also demonstrating calcination of CaCO 3 without sintering through rapid heating and precise temperature control.

36 MATERIALS SCIENCE↗