The discussion in this paper focuses on how boundary based smooth shape design can be carried out. For this we treat surface generation as a mathematical boundary-value problem. In particular, we utilize elliptic Partial Differential Equations (PDEs) of arbitrary order. Using the methodology outlined here a designer can therefore generate the geometry of shapes satisfying an arbitrary set of boundary conditions. The boundary conditions for the chosen PDE can be specified as curves in 3-space defining the profile geometry of the shape. We show how a compact analytic solution for the chosen arbitrary order PDE can be formulated enabling complex shapes to be designed and manipulated in real time. This solution scheme, although analytic, satisfies exactly, even in the case of general boundary conditions, where the resulting surface has a closed form representation allowing real time shape manipulation. In order to enable users to appreciate the powerful shape design and manipulation capability of the method, we present a set of practical examples

The main result is a new original algorithm: CONDOR ("COnstrained, Non-linear, Direct, parallel Optimization using trust Region method for high-computing load, noisy functions"). The aim of this algorithm is to find the minimum x* of an objective function F(x) (x is a vector whose dimension is between 1 and 150) using the least number of function evaluations of F(x). It is assumed that the dominant computing cost of the optimization process is the time needed to evaluate the objective function F(x) (One evaluation can range from 2 minutes to 2 days). The algorithm will try to minimize the number of evaluations of F(x), at the cost of a huge amount of routine work. CONDOR is a derivate-free optimization tool (i.e., the derivatives of F(x) are not required. The only information needed about the objective function is a simple method (written in Fortran, C++,...) or a program (a Unix, Windows, Solaris,... executable) which can evaluate the objective function F(x) at a given point x. The algorithm has been specially developed to be very robust against noise inside the evaluation of the objective function F(x). This hypotheses are very general, the algorithm can thus be applied on a vast number of situations. CONDOR is able to use several CPU's in a cluster of computers. Different computer architectures can be mixed together and used simultaneously to deliver a huge computing power. The optimizer will make simultaneous evaluations of the objective function F(x) on the available CPU's to speed up the optimization process. The experimental results are very encouraging and validate the quality of the approach: CONDOR outperforms many commercial, high-end optimizer and it might be the fastest optimizer in its category (fastest in terms of number of function evaluations). When several CPU's are used, the performances of CONDOR are currently unmatched (may 2004). CONDOR has been used during the METHOD project to optimize the shape of the blades inside a Centrifugal Compressor (METHOD stands for Achievement Of Maximum Efficiency For Process Centrifugal Compressors THrough New Techniques Of Design). In this project, the objective function is based on a 3D-CFD (computation fluid dynamic) code which simulates the flow of the gas inside the compressor.

The discussion in this paper focuses on how boundary based smooth shape design can be carried out. For this we treat surface generation as a mathematical boundary-value problem. In particular, we utilize elliptic Partial Differential Equations (PDEs) of arbitrary order. Using the methodology outlined here a designer can therefore generate the geometry of shapes satisfying an arbitrary set of boundary conditions. The boundary conditions for the chosen PDE can be specified as curves in 3-space defining the profile geometry of the shape. We show how a compact analytic solution for the chosen arbitrary order PDE can be formulated enabling complex shapes to be designed and manipulated in real time. This solution scheme, although analytic, satisfies exactly, even in the case of general boundary conditions, where the resulting surface has a closed form representation allowing real time shape manipulation. In order to enable users to appreciate the powerful shape design and manipulation capability of the method, we present a set of practical examples.

京都大学 / 0048 / 新制・論文博士 / 博士(工学) / 乙第12950号 / 論工博第4126号 / 新制||工||1628(附属図書館) / 32249 / 京都大学大学院工学研究科化学工学専攻 / (主査)教授 長谷部 伸治, 教授 前 一廣, 教授 吉田 潤一 / 学位規則第4条第2項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

When designing aircraft wings shapes, it is important to ensure that the flight envelope does not overlap with regions of flutter or Limit Cycle Oscillation (LCO). A quick assessment of these dynamic aeroelastic for various design candidates is key to successful design. Flutter based design requires the sensitivity of flutter parameters to be known with the respect of design parameters. Traditionally, frequency domain based methods have been used to predict flutter characteristics and its sensitivity. However, this approach is only applicable for linear or linearized models and cannot be applied to systems undergoing LCO or other nonlinear effects. Though the time accurate approach can be implemented to overcome this problem, it is computationally expensive. Also, the unsteady adjoint formulation for sensitivity analysis, requires the state and adjoint variables to be stored at every time step, which prohibitively increases the memory requirement. In this work, these problems have been overcome by implementing a time spectral method based approach to compute flutter onset, LCOs and their design sensitivities in a computationally efficient manner. The time spectral based formulation approximates the solution as a discrete Fourier series and directly solves for the periodic steady state, leading to a steady formulation. This can lead to the time spectral approach to be faster than the time accurate approach. More importantly, the steady formulation of the time spectral method also eliminates the memory issues faced by the unsteady adjoint formulation. The time spectral based flutter/LCO prediction method was used to predict flutter and LCO characteristics of the AGARD 445.6 wing and pitch/plunge airfoil section with NACA 64A010 airfoil. Furthermore, the adjoint based sensitivity analysis was used to carry out aerodynamic shape optimization, with an objective of maximizing the flutter velocity with and without constraints on the drag coefficient. The resulting designs show significant increase in the flutter velocity and the corresponding LCO velocity profile. The resulting airfoils display a greater sensitivity to the transonic shock which in turn leads to greater aerodynamic damping and hence leading to an increase in flutter velocity. / Doctor of Philosophy / When designing aircrafts, dynamic aeroelastic effects such as flutter onset and Limit Cycle Oscillations need to considered. At low enough flight speeds, any vibrations arising in the aircraft structure are damped out by the airflow. However, beyond a certain flight speed, instead of damping out the vibrations, the airflow accentuates these vibrations. This is known as flutter and it can lead to catastrophic structural failure. Hence, during the aircraft design phase, it must be ensured that the aircraft would not experience flutter during the flight conditions. One of the contribution of this work has been to come up with a fast and accurate method to predict flutter using computational modelling. Depending on the scenario, it is also possible that during flutter, the vibrations in the structure increase to a certain amplitude before leveling off due to interaction of non-linear physics. This condition is known as limit cycle oscillation. While they can arise due to different kinds of non-linearities, in this work the focus has been on aerodynamic non-linearities arising from shocks in transonic flight conditions. While limit cycle oscillations are undesirable as they can cause structural fatigue, they can also save the aircraft from imminent structural fracture and hence it is important to accurately predict them as well. The main advantage of the method developed in this work is that the same method can be used to predict both the flutter onset condition and limit cycle oscillations. This is a novel development as most of the traditional approaches in dynamic aeroelasticity cannot predict both the effects. The developed flutter/LCO prediction method has then been used in design with the goal of achieving superior flutter characteristics. In this study, the shape of the baseline airfoil is changed with the goal of increasing the flutter velocity. This enables the designed system to fly faster without addition of weight. Since the design has been carried out using gradient based optimization approach, an efficient way to compute the gradient needs to be used. Traditional approaches to compute the gradient, such as Finite Difference Method, have computational cost proportional to the number of design variables. This becomes a problem for shape design optimization, where a large number of design variables are required. This has been overcome by developing an adjoint based sensitivity analysis method. The main advantage of the adjoint based sensitivity analysis is that it its computational cost is independent of the number of design variables, and hence a large number of design variables can be accommodated. The developed flutter/LCO prediction and adjoint based sensitivity analysis framework was used to carry out shape design for a pitch/plunge airfoil section. The objective of the design process was to maximize the flutter onset velocity with and without constraints on drag. The resulting optimized airfoils showed significant increase in the flutter velocity.

Shape optimization is an important step in many design processes. With the growing use of Computer Aided Engineering in the design chain, it has become very important to develop robust and efficient shape optimization algorithms. The field of Computer Aided Optimal Shape Design has grown substantially over the recent past. In the early days of its development, the method based on small shape perturbation to probe the parameter space and identify an optimal shape was routinely used. This method is nothing but an educated trial and error method. A key development in the pursuit of good shape optimization algorithms has been the advent of the adjoint method to compute the shape sensitivities more formally and efficiently. While undoubtedly, very attractive, this method relies on very sophisticated and advanced mathematical tools which are an impediment to its wider use in the engineering community. It that spirit, it is the purpose of this thesis to propose a new shape optimization algorithm based on more intuitive engineering principles and numerical procedures. In this thesis, the new shape optimization procedure which is proposed is based on the generation of a body-fitted mesh. This process maps the physical domain into a regular computational domain. Based on simple arguments relating to the use of the chain rule in the mapped domain, it is shown that an explicit expression for the shape sensitivity can be derived. This enables the computation of the shape sensitivity in one single solve, a performance analogous to the adjoint method, the current state-of-the art. The discretization is based on the Finite Difference method, a method chosen for its simplicity and ease of implementation. This algorithm is applied to the Laplace equation in the context of heat transfer problems and potential flows. The applicability of the proposed algorithm is demonstrated on a number of benchmark problems which clearly confirm the validity of the sensitivity analysis, the most important aspect of any shape optimization problem. This thesis also explores the relative merits of different minimization algorithms and proposes a technique to “fix” meshes when inverted element arises as part of the optimization process. While the problems treated are still elementary when compared to complex multiphysics engineering problems, the new methodology presented in this thesis could apply in principle to arbitrary Partial Differential Equations.

A hybrid design optimization method combining the stochastic method based on simultaneous perturbation stochastic approximation (SPSA) and the deterministic method of Broydon-Fletcher-Goldfarb-Shanno (BFGS) is developed in order to take advantage of the high efficiency of the gradient based methods and the global search capabilities of SPSA for applications in the optimal aerodynamic shape design of a three dimensional elliptic nozzle. The performance of this hybrid method is compared with that of SPSA, simulated annealing (SA) and gradient based BFGS method. The objective functions which are minimized are estimated by numerically solving the 3D Euler and Navier-Stokes equations using a TVD approach and a LU implicit scheme. Computed results show that the hybrid optimization method proposed in this study shows a promise of high computational efficiency and global search capabilities. / Singapore-MIT Alliance (SMA)

No description available.

This paper will analyze the effects of using hybrid optimization methods for optimizing objective functions that are determined by computational fluid dynamics solvers for compressible viscous flow for optimal design of airfoils. Previous studies on this topic by the authors had examined the application of deterministic optimization methods and stochastic optimization methods such as Simulated Annealing and Simultaneous Perturbation Stochastic Analysis (SPSA). The studies indicated that SPSA method has a greater or equal efficiency as compared with SA method in reaching optimal airfoil designs for the design problem in question. However, in some situations SPSA method has a tendency to demonstrate an oscillatory behavior in the vicinity of a local optima. To overcome this tendency, a hybrid method designed to take full advantage of SPSA’s high rate of reduction of the objective function at the inception of the design process to drive the design cycles towards the optimal zone at first, and then combining with other methods to perform the final stages of the convergence towards the optimal solutions is considered. SPSA method has been combined with the gradient-based Broydon-Fletcher-Goldfarb-Shanno (BFGS) method as well as Simulated Annealing method for the transonic inverse airfoil design problem that is concerned with the specification of a target airfoil surface pressure distribution and starting from an initial guess of an airfoil shape, the target airfoil shape is reached by way of minimization of a quantity that depends on the difference between the target and current airfoil surface pressure distribution. For a typical transonic flow test case, the effects of using hybrid optimization techniques such as SPSA+BFGS and SPSA+SA as opposed to using SPSA alone can be seen in Figure 1. After 800 design cycles using SPSA, the hybrid SPSA+SA method took 2521 function evaluations of SA while the SPSA+BFGS method took 271 function evaluations to reach similar values which are much better than that reached by using SPSA alone in the entire minimization process. Results indicate that both of the two hybrid methods have capability to find a global optimum more efficiently than the SPSA method. The paper will address issues related to hybridization and its impact on the optimal airfoil shape designs in various contexts. / Singapore-MIT Alliance (SMA)

The master thesis contains a theoretical part with the topic of valve train. It contains measured data and their processing. The processed data are used to create the 1D engine’s simulation. Valve train’s parameters were modified for increased power and torque. Contained two variants of changes can serve as guide for final draft because of next adjustments.