optimization-for-model-aircrafts

This repository contains code that optimizes an aircraft with SAE Aero 2020 Regular Class problem statement as an example

Model Aircraft Design Optimization

Copyright 2022 The MathWorks® Inc.

Introduction

This demo optimizes an aircraft with SAE Aero Design 2020 Regular Class problem statement as an example. A problem based approach is used to construct the design optimization problem. For SAE Aero Design 2020 Regular Class competition, each team's objective was to maximise their Final Flight Score (FFS) which was the sum of three highest Flight Scores (FS_i) and a Payload Prediction Bonus (PPB).

FFS = FS_1 + FS_2 + FS_3 + PPB

Each individual Flight Score (FS) was calculated as follows

FS = 120\times\frac{2\times N_s + W_{BP}}{b_w + l_{cargo}}

where,

N_s = No. of Spherical Payload

W_{BP} = Weight of Boxed Payload (lbs)

b_w = Wingspan (in)

l_{cargo} = Length of Payload Bay (in)

Therefore, this demo maximizes the following objective as it performs calculations in SI Units. Also, as only 1 spherical payload is considered.

Objective = 120\times \frac{2 + 2.2\times W_{BP}}{39.37\times(b_w + l_{cargo})}

A problem based approach is used to construct the design optimization problem. All expressions evaluated during problem setup are stored as a hierarchy of structure inside the aircraft structure. Four domain specific functions incrementally construct the design problem by modelling domain specific expressions and adding any relevant constraints. Finally, a 12 dimensional optimization problem is obtained with the following optimization variables.

Symbolic Variable	Physical Quantity
b_w	Wing Half Span
cr_w	Wing Root Chord
lambda_w	Wing Taper Ratio
X_w	Wing X Location
b_{ht}	Horizontal Tail Half Span
c_{ht}	Horizontal Tail Chord
b_{vt}	Vertical Tail Half Span
c_{vt}	Vertical Tail Chord
l_f	Length of Fuselage
l_{pb}	Length of Boxed Payload
h_{pb}	Height of Boxed Payload
X_p	Cargo Bay X Location

Following is the representation of optimization variables.

Code Structure

optimizeAircraft.mlx sets up and solves an aircraft design optimization problem. All other live functions model domain specific expressions and constraints and help incrementally setup the optimization problem.

Setup

1. Clone the repository.
2. Open MATLAB® and navigate to the repository.
3. Open and execute the live script optimizeAircraft.mlx

MathWorks Products

Requires MATLAB release R2022a or newer

• Aerospace Blockset™
• Global Optimization Toolbox

License

The license for Model Aircraft Design Optimization is available in the LICENSE.TXT file in this GitHub repository.

For any queries, contact the authors at [e-mail]