Sathyabama Institute of Science and Technology B.E. - Electronics and Instrumentation Engineering SEIA3004 Soft Computing Techniques Syllabus

SATHYABAMA INSTITUTE OF SCIENCE AND TECHNOLOGY

SEIA3004 SOFT COMPUTING TECHNIQUES

UNIT 1 INTRODUCTION TO SOFT COMPUTING AND ARTIFICIAL NEURAL NETWORKS 9 Hrs.
Evolution of Computing - Soft Computing Constituents - From Conventional AI to Computational Intelligence - Machine
Learning Basics, Fundamentals of ANN - Biological Neurons and Their Artificial Models - Types of ANN - Properties -
Different Learning Rules - Types of Activation Functions - Training of ANN - Hebb learning - Perceptron Model (Both Single
& Multi Layer) - Training Algorithm - Problems Solving Using Learning Rules and Algorithms - Linear Separability –
Limitation.

UNIT 2 DETERMINISTIC AND STATISTICAL NETWORKS 9 Hrs.
Back Propagation Training Algorithm - Practical Difficulties - Counter Propagation Network - Structure & Operation -
Training of Kohonen and Grossberg Layer - Applications of BPN & CPN - Statistical Method – Training Application -
Boltzman Training - Cauchy Training - Hop Field Network and Boltzman Machine - Speed Energy Function -
Network Capacity - RBF Network, BAM, Architecture of SOM, ANN based water level controller.

UNIT 3 FUZZY LOGIC 9 Hrs.
Introduction to Fuzzy Set Theory - Basic Concepts of Fuzzy Sets - Classical Set Vs Fuzzy Set - Properties of Fuzzy Set -
Fuzzy Logic Operation on Fuzzy Sets - Fuzzy Logic Control Principles - Fuzzy Relations - Fuzzy Rules - Defuzzification -
Fuzzy Inference Systems - Fuzzy Expert Systems - Fuzzy Decision Making.

UNIT 4 FUZZY LOGIC CONTROLLER AND ITS APPLICATION 9 Hrs.
Fuzzy Logic Controller - Fuzzification Interface - Knowledge Base- Decision Making Logic – Defuzzification Interface-
Application of Fuzzy Logic to Water Level Controller - Temperature Controller - Control of Blood Pressure during
Anaesthesia. Introduction to Neuro - Fuzzy Systems - Fuzzy System Design Procedures - Fuzzy Sets and Logic
Background - Fuzzy / ANN Design and Implementation.

UNIT 5 GENETIC ALGORITHMS 9 Hrs.
Introduction - Robustness of Traditional Optimization and Search Techniques - The goals of optimization - Survival of the
Fittest - Fitness Computations - Cross over - Mutation -Reproduction- Rank method- Rank space method.
Max. 45 Hrs.

COURSE OUTCOMES
On completion of the course, student will be able to
CO1 - Understand the concept of ANN, Fuzzy Logic, Genetic Algorithm.
CO2 - Illustrate the types of ANN, fuzzification and defuzzifiction techniques.
CO3 - Select the appropriate techniques for applications.
CO4 - Compare the different techniques.
CO5 - Judge the various techniques in different types of applications.
CO6 - Plan for implementation of ANN, FL, GA based control.

TEXT / REFERENCE BOOKS
1. Laurene Fausett, " Fundamentals of Neural Networks: Architectures, Algorithms and Applications", 2008.
2. Timothy J. Ross , “Fuzzy Logic with Engineering Applications”, McGraw - Hill International Editions, 2004.
3. Jang J.S.R., Sun C.T. and Mizutani E, "Neuro-Fuzzy and soft computing", Pearson Education, 2003.
4. Rajasekaran. S, Pai. G.A.V. “Neural Networks, Fuzzy Logic and Genetic Algorithms”, Prentice Hall of India, 2003.

END SEMESTER EXAMINATION QUESTION PAPER PATTERN
Max. Marks: 100 Exam Duration: 3 Hrs.
PART A: 10 Questions of 2 marks each-No choice 20 Marks
PART B: 2 Questions from each unit of internal choice, each carrying 16 marks 80 Marks