MECHANICS OF

COMPOSITE MATERIALS
LESSON SIX:
INTRODUCTION TO ENGINEERING OPTIMISATION
Contents
Introduction
◦ Definition
◦ Formulation of optimization problem.
◦ Classification of optimization.

Structural optimisation
◦ Topology optimisation
◦ Size optimisation.
◦ Shape optimisation.

Optimisation in laminated composites

◦ Lamination parameters

References
Introduction
Optimization is the act of obtaining the best result under given circumstances.
The ultimate goal is either to minimize the effort required (e.g. weight, drag,
time, cost … etc.) or to maximize the desired benefit (stiffness, durability,
efficiency, … etc).

There is no single method available for solving all optimization problems

efficiently. The optimum seeking methods are also known as mathematical
programming and are generally studied as a part of operations research.
 Methods of Operations Research

Mathematical programming Stochastic process

Statistical methods
or optimization techniques techniques

Calculus methods Statistical decision theory Regression analysis

Calculus of variations Markov processes Cluster analysis, pattern
Nonlinear programming Queueing theory recognition
Geometric programming Renewal theory Discriminate analysis (factor
Quadratic programming Simulation methods analysis)
Linear programming Reliability theory Design of experiments
Dynamic programming
Integer programming
Stochastic programming
Separable programming

Multiobjective programming

Network methods: CPM and

PERT
Game theory
Modern or nontraditional optimization techniques
Genetic algorithms
Simulated annealing
Ant colony optimization
Particle swarm optimization
Neural networks
Fuzzy optimization
Mathematical programming techniques are useful in finding the minimum of a function
of several variables under a prescribed set of constraints.

Stochastic process techniques can be used to analyze problems described by a set of

random variables having known probability distributions.

Statistical methods enable one to analyze the experimental data and build empirical
models to obtain the most accurate representation of the physical situation.
Formulation of optimization problem
An optimization or a mathematical programming problem can be stated as
follows
minmize f ( X)
subjected to p i (X)  0 i = 1, 2, ..., m
and q j ( X)  0 j = 1, 2, ..., n

Where f(X) is the objective function to be minimised of maximised, pi(X) is

inequality constrains, qi(X) is equality constrains and X is design variable or
design vector.
Classification of optimization
The optimization problems can be classified based on the following categories:
1. Existence of constraints.
2. Nature of the design variables.
3. Physical structure of the problem.
4. Nature of the equations involved.
5. Permissible values of the design variables.
6. Deterministic nature of the variables.
7. Separability of the functions.
8. Number of objective functions.
Example of classifying optimisation
Based on existence of constraints
Constrained Unconstrained
Based on linearity of

Based on number of
Linear a b c d

design variables
One D.V
functions

Nonlinear e f g h
Linear i j k l Two or
more
Nonlinear m n o p D.Vs
Single obj. Multi obj. Single obj. Multi obj.
Based on the number of objective functions
Structural optimisation
Structural optimisation may be categorised into three types
◦ Size optimisation is to find the optimal design by changing the size variables
such as the cross-sectional dimensions of trusses and frames, or the
thicknesses of plates.
◦ Shape optimization is mainly performed on continuum structures by
modifying the predetermined boundaries to achieve the optimal designs.
◦ Topology optimization
◦ For discrete structures, such as trusses and frames, is to search for the optimal spatial order
and connectivity of the bars.
◦ For continuum structures is to find the optimal designs by determining the best locations
and geometries of cavities in the design domains.
Three categories of structural optimization. a) Sizing optimization of a
truss structure, b) shape optimization and c) topology optimization. The
initial problems are shown at the left hand side and the optimal solutions
are shown at the right
Most of commercial FE (finite element) softwares are supported with optimisation modules such
as MSC. PATRAN and NASTRAN, Ansys, Altair Hyperwork, Hypermesh, …etc. Using such
application require a good knowledge in FE methods and analysis and high speed computer
processor and large memory for good performance.
Example
Presented by Fox (1969), a plane two - bar truss of tubular members, the height H and the
diameter d are varied in order to minimize the volume of material. There are two requirements
that are to be satisfied:
1. the member stress should be less than the yield stress, Fy.
2. members should not buckle.
Solution
H = 56 in, d = 3.6 in
The previous problem can be solved using “Solver” tool in Microsoft Excel or using any other
programming language such as Matlab which is accompanied with an optimisation toolbox, in
addition to many other programme modules like Scipy in Python
Optimisation in laminated composites
In composite structures, optimisation may involve minimising weight, maximising stiffness, strength
or frequency. The design variables may also include fiber orientations, thickness, number of plies,
geometry of the structure … etc.

There are several methods to solve these types of problem in laminated composites. One of these
solutions is by using a graphical representation proposed by Miki (later called Miki diagrams), the
solution applied to some problems, for instance, when considering limited number of fiber
orientation or plies.

The main idea is to present for example extensional stiffness elements in terms of invariants, this
solution also known as lamination parameter.
Q11  U1  U 2 cos 2  U 3 cos 4 A11 / h  A11*  U1  U 2V1*  U 3V3*
Q12  U 4  U 3 cos 4 A12 / h  A12*  U 4  U 3V3*
Q22  U1  U 2 cos 2  U 3 cos 4 A22 / h  A22
*
 U1  U 2V1*  U 3V3*
Q66  U 5  U 3 cos 4 A66 / h  A66
*
 U 5  U 3V3*

I I I
V1 A tk
V 
1
*
  cos 2 k  k cos 2 k V   k cos 4 k
3
*

h k 1 h k 1 k 1
I


k 1
k 1
Miki’s lamination parameter diagram
For ±𝜃𝑛 𝑠 laminate
V  cos 2 , V  cos 4
1
*
3
*
V  2V  1
3
*
1
*2

For general laminate

V3*  2V1*2  1
Project
Write a computer programme to plot the effective moduli contours using lamination
parameters. Choose a material and stacking sequence you prefer. Discuss the manner of
represented graph that you created.
For example
U 2 *2
V  U E V *
 E U  U 2
 U 2
V3*  2 1 2 x 1 x 1 1 4

U 3  2U1  2U 4  Ex 
References
Singiresu S. Rao, “Engineering Optimization: Theory and Practice”, 4th ed, John Wiley and Sons,
2009.
William R. Spillers · Keith M. MacBain, “Structural Optimization”, Springer, 2009.
Raphael T. Haftka, Zafer Gurdal, “Elements of Structural Optimization”, 3rd ed, Kluwer Academic
Publishers, 1992.
Miki, M., “A Graphical Method for Designing Fibrous Laminated Composites with Required In-
plane Stiffness”, Trans. JSCM, 9, 2, pp. 51-55, 1983.