You are on page 1of 15

Engineering optimization-

EEE1020
Text books
• Text Books:
• Chong and Zak, “Introduction to Optimization",
John Wiley & Sons, Inc., 4th Ed., 2013.
• S S Rao, "Engineering Optimization, Theory and
Practice", John Wiley & Sons, Inc., 4th Ed., 2009.
Module -1
• Classical Optimization basics:
• Taylor’s series, Single-variable optimization,
Multivariable optimization without and with
equality and inequality constraints,
Definiteness of matrices, Sylvester’s criterion,
Convex programming problem.
Module-2
1-D search methods
Golden Section Search, Fibonacci Search, Inexact
line search.
Module-3
• Gradient based optimization
Gradient descent method, method of steepest
descent, Newton’s Method, Levenberg-
Marquardt algorithm.
Module-4
• Conjugate Direction Methods:
Conjugate directions and conjugate gradient
method, Fletcher-Reeves formula. Convergence
analysis of all algorithms.
Module-5
• Miscellaneous topics
• Dynamic programming. Dynamic optimization.
Sample applications of gradient based and
gradient free methods in engineering.
Module-6
• Application of optimization methods to
neural networks
NN basics, capabilities and limitations of single
perceptron, multilayer perceptron. Training by
gradient based and gradient free methods.
Module-7
• Gradient-free Optimization
Direct and indirect methods, Limitations of
gradient based methods, metaheuristic
algorithms, Introduction to the genetic
algorithm, particle swarm optimization.
Simulated annealing.
Module-8
• Recent Advances and contemporary issues by
industry experts
List of Sample Projects (Indicative) SLO:
1. Design of a transformer for maximum efficiency
2. Determination of load for maximum power delivery from source
3. Solution of MPPT problem by golden section search
4. Solution of MPPT problem by Fibonacci search
5. Solution of MPPT problem by gradient descent method
6. Solution of MPPT problem by steepest descent method 1, 2, 17
7. Solution of MPPT problem by Newton's method
8. Training a NN by gradient descent method
9. Training a NN by steepest descent method
10. Training a NN by Newton's method
11. Training a NN by Levenberg-Marquardt algorithm
12. Training a NN by conjugate gradient method
13. Function optimization by genetic algorithm
14. Function optimization by particle swarm optimization
15. Function optimization by simulated annealing.
Statement of an Optimization Problem
SINGLE-VARIABLE OPTIMIZATION
• A single-variable optimization problem is one
in which the value of x = x∗ is to be found in
the interval [a, b] such that x∗ minimizes f (x).
Applications of Optimization in
Engineering
Optimization in its broadest sense can be applied to solve
any engineering problem,
• design of aircraft for minimum weight;
• optimal (minimum time) trajectories for space missions;
• minimum weight design of structures for earthquake;
• optimal design of electric networks;
• optimal production planning, resources allocation,
scheduling;
• shortest route;
• design of optimum pipeline networks;
• minimum processing time in production systems;
• optimal control.

You might also like