MEEPC105-3 Artificial Neural Network & Fuzzy System M.Tech Model Question Paper : mgu.ac.in

Name of the College : Mahatma Gandhi University
Department : Electrical and Electronics Engineering
Subject Code/Name : MEEPC 105-3/Artificial Neural Network & Fuzzy System
Sem : I
Website : mgu.ac.in
Document Type : Model Question Paper

Download Model/Sample Question Paper : https://www.pdfquestion.in/uploads/mgu.ac.in/5056-MEEPC%20105_3%20ARTIFICIAL%20NEURAL%20NETWORK%20&%20FUZZY%20SYSTEM.doc

Artificial Neural Network & Fuzzy System :

M.TECH Degree Examination :
Model Question Paper
First Semester

Related : MGU MEEPP106-3 High Voltage Dc Transmission M.Tech Model Question Paper : www.pdfquestion.in/5055.html

Branch: Electrical and Electronics Engineering
Specialization: Power Electronics and Control
MEEPC 105-3 Artificial Neural Network & Fuzzy System
(2013 admission onwards)
Time : 3Hrs
1. a. What is ANN? Explain multi layer feed forward model of ANN and describe the function and structure of each unit
b. What is XOR problem? How it can be solved?

c. Consider a simple perceptron model with four inputs. Let the initial weight vector be [1 -1 0.5 0]T. Set of input training vectors are x1=[1 -2 0 -1]T, x2=[0 1.5 -0.5 -1]T and x3=[-1 1 0.5 -1]T .Desired responses for these input vectors are -1, -1, and 1 respectively. The activation function is sign(x). Illustrate perceptron learning process
OR

2. a. What is learning rule? Explain Hebbian Learning and competitive learning.
b. What you meant by pattern recognition? How a neural network perform this operation 10 marks
c. Draw the flow chart of back propagation learning algorithm

3. a. How ANN can be used for the control of an inverted pendulum
b. Explain expert system for diagnosis of a medical disease
OR

4. a. What is the purpose of Adaptive resonance theory 1 network?. Explain its operation
b. Write note on the following (i) ambiguity (ii) Fuzziness (iii) in exactness
5. a. Compare fuzzy set and crisp set

b. What is fuzzy compliment? What are the axioms to be satisfied so that a function can be used as fuzzy compliment?. Check whether the function x+y – x.y can be used as fuzzy union.

c. A linguistic variable x which measures the academic excellence is taken from universe of discourse U={ 1 2 3 4 5 6 7 8 9 10}. The membership functions are defined as follows ??(Excellent)={(8, 0.2) (9, 0.6) (10 1)}, ??(good)={(6 0.1) (7 0.5) (8, 0.9) (9,1) (10 1)} Construct the membership function of Good but not excellent.
OR

6. a. Explain four major steps in fuzzy rule based model
b. How multi valued logic and fuzzy logic are related?. Give brief description of (i) un conditional and unqualified fuzzy proposition and (ii) conditional and unqualified fuzzy proposition

7. a. Explain direct methods of fuzzy construction
b. Explain different steps for designing a fuzzy controller
OR

8. a. With an example, discuss fuzzy individual decision making
b. What is genetic algorithm? Explain different steps of genetic algorithm with a flow chart

Syllabus :
Module 1 :
Pattern classification –Learning and generalisation-structure of neural networks – ADA line and Mada line-perceptrons .Linear separability – Back propagation – XOR function-Backpropagation algorithm-Hopfied and Hamming networks

Kohensen’s network-Boltzmenn machine-in and out star network – Art 1 and Art 2 nets-Neuro adaptive control applications-ART architecture – Comparison layer – Recognition layer – ART classification process – ART implementation – Examples

Module 2 :
Character recognition networks, Neural network control application, connectionist expert systems for medical diagnosis Self organizing maps-Applications of neural algorithms and systems

Character recognition networks, Neural network control application, connectionist expert systems for medical diagnosis -Different faces of imprecision – inexactness, Ambiguity, Undecidability, Fuzziness and certainty, Probability and fuzzy logic, Intelligent systems.

Module 3 :
Fuzzy sets and crisp sets – Intersections of Fuzzy sets, Union of Fuzzy sets, the complement of Fuzzy sets. Fuzzy reasoning – Linguistic variables, Fuzzy propositions, Fuzzy compositional rules of inference- Methods of decompositions, Defuzzification.

Module 4 :
Methodology of fuzzy design – Direct & Indirect methods with single and multiple experts, Adaptive fuzzy control, Rule base design using dynamic response. Fuzzy logic applications to engineering, Fuzzy decision making, Neuro-Fuzzy systems, Fuzzy Genetic Algorithms.

References :
1. Martin T. Hogan , Howard B.Demuth, M, ’Neural network design’
2. Zuroda, J.M.,’Introduction to Artificial Neural Systems’, Jaico publishing house, Bombay, 1994.
3. Zimmermann, H.J., ‘Fuzzy set theory and its applications’, Allied publishers limited, Madras,1966