Subject: ✈ MDOlab Newsletter

View this email online if it doesn't display correctly
MDOlab Newsletter—Fall 2015
Dear Colleague,
It has been over a year since the last MDOlab Newsletter, and much has happened since then. In this newsletter you will find a selection of our new papers and software updates. 
You can watch a summary of our high-fidelity wing design optimization work in a recent presentation [PDF] [Video].
Happy holidays, and best wishes for 2016!
Latest publications
Wing-body-tail aerodynamic shape optimization
RANS-optimized wing
We have extended previous work on wing aerodynamic shape optimization to consider the full Common Research Model configuration, including a rotating horizontal tail for trimming.
We perform a series of optimizations to study the trim drag trade-offs, and develop a surrogate model for the trim drag to replace the effect of the tail when it is not included in the CFD model.
Aerodynamic shape optimization of a morphing trailing-edge wing
BWB 3D-printed models
Since aircraft operate at different flight conditions, they should be designed using a multipoint approach. However, if we are able to change the shape for the different conditions through an adaptive morphing trailing edge, the performance can be improved throughout the flight envelope. This paper quantifies the gains of such a morphing technology.

Wing aerostructural optimization considering buffet onset
Optimized CRM
Aerodynamics is not enough in wing design optimization, as we must account for wing flexibility and aerostructural tradeoffs, as we have shown in previous work. In this paper, we optimize the undeformed Common Research Model (uCRM) configuration using RANS CFD coupled to a detailed wingbox structural model, and we develop a new formulation to prevent buffet onset. 
                             [ Paper ]
Hydrofoil shape optimization with cavitation constraints
TACS wingbox model
While aerodynamic shape optimization with respect to hundreds of shape variables has been applied to wing design for decades, hydrofoil design has not benefited from the same design optimization methods. In this paper, we perform the hydrodynamic shape optimization of a 3D foil and address some additional challenges that relate to the different fluid physics, including the addition of a constraint that prevents cavitation.
Computing stability derivatives for aerodynamic shape optimization
TACS wingbox model
Aerodynamic shape optimization of aircraft configurations often ignores stability considerations. To address this, we develop a method for the computation of static, dynamic, and transient aircraft stability derivatives and their sensitivities for use in gradient-based optimization.

Surrogate models for aircraft performance
TACS wingbox model
Accurate evaluation of aircraft fuel burn over a complete mission is computationally expensive, requiring a large number of high-fidelity evaluations. Thus, it is advantageous to use surrogate models to represent the drag polar. In this work we develop a mixtures of experts approach that can accurately handle the nonlinearities of transonic drag polars.

                          [ Paper ]     [ Preprint ]
Wing design optimization for nonspecialists

TACS wingbox model
This short article starts with an introduction to wing design optimization for nonspecialists, written for the applied mathematics community. This note was written as a followup to my presentation at the SIAM Conference on Optimization.

Open-source software

TACS wingbox model
We have been collaborating with NASA's OpenMDAO team to implement adjoint methods for computing coupled derivatives. The latest version of OpenMDAO implements the coupled adjoint method, using ideas from our previous work on unifying derivative computations, and adjoint-based multidisciplinary optimization. OpenMDAO facilitates the coupling of disciplines for both gradient-based and gradient-free optimization, and provides an interface to various optimizers.
   [ Paper ]    [ Website ]
pyOptSparse: A common Python interface for optimization

TACS wingbox model
Each time you want to use a new numerical optimization package, you need to study yet another manual and programming interface. pyOptSparse solves this issue by providing a common Python interface and wrappers to a variety of numerical optimization algorithms. This software is open source and includes OptView, an interactive viewer of your optimization history (objective, design variables, and constraints).

                         [ Code repository ]
Online optimization game

TACS wingbox model
This online game serves to demonstrate the challenge of finding the minimum of a function when only a few function evaluations are available, even in two dimensions.

(compatible with Firefox or Chrome browsers)   
You may unsubscribe or change your contact details at any time.