# Pune University BE Electronics Soft Computing Tools

# Question Papers

**B.E. (Electronics) SOFT COMPUTING TOOLS (2008 Pattern) (Elective – III) (Sem. – II)**

Time :3 Hours] [Max. Marks :100

Instructions to the candidates :

** 1) **Answer 3 questions from Section -1 and 3 questions from Section – II.

** 2) **Answers to the two sections should be written in separate answer books.

** 3) **Neat diagrams must be drawn wherever necessary.

** 4) **Pigures to the right indicate full marks.

** 5) **Use of logarithmic tables, slide rule, Mollier charts, electronic pocket calculator and steam tables is allowed.

** 6) **Assume suitable data, if necessary.

SECTION – I

QIA a) Define soft computing and explain its constituents along with conventional artificial intelligence in detail. [9]

b) Explain Neurofuzzy and soft computing characteristics. [9]

OR

Q2) Consider two fuzzy sets A 8 B. [18]

. f 1 0.3 0.5 0.21

A = <-+—+—+—\

0 2 4 6 8 j

„ f 0.5 0.6 0.1 11

B=\—+—+—+-\

1 2 4 6 8j

Perform the following operations on fuzzy sets.

a) A u B

b) A n B

c) Complement of fuzzy set A.

d) Difference (A/B)

e) A u B

f) A u B

g) algebraic sum of the given fuzzy sets

h) bounded sum of the given fuzzy set

i) algebraic product of the given fuzzy sets.

Q3) a) Describe in detail the process of defuzzification. Explain any three defuzzification schemes in detail. [8]

b) Explain the extension principle for fuzzy set with suitable example. [8]

OR

Q4) a) Let A = {(x_{p} 0.2), (x_{2}, 0.3) (x_{Q}, 0.4)} 8 B = {(y_{1}, 0.3), (y_{2}, 0.6)} be two fuzzy sets defined on the universes of discourse X = {x_{1}, x_{2}, x_{Q}} and Y = (y_{1}, y_{2}} respectively.

i) F ind the fuzzy relation R resulting out of the fuzzy cartesian product. A x B.

ii) If fuzzy set C = {(z_{1}, 0.3), (z_{2}, 0.4) (z_{Q}, 0.6)} then find relation between R and C using max-min composition. [12]

b) Define and explain following terms. [4]

i) Support ii) core

iii) convex set iv) fuzzy singleton

Q5) a) What are the steps involved in designing a fuzzy logic controller. [8]

b) With a neat block diagram, explain the architecture of a fuzzy logic controller. [8]

OR

Q6) Write notes on. [IT]

a) Mamdani inference system / model.

b) Sugeno Model.

c) Tsukamoto model.

d) Synthesis and validation of fuzzy controller.

SECTION – II

Q7) a) Explain the architecture of multilayered network. Give the steps used in training a multilayered network using back propogation algorithm. [10]

b) List and explain various Neural network learning methods. [8]

OR

Q8) a) Explain the architecture of SOM given by kohonen. Describe the sequence of training steps involved in training of SOM. [8]

b) Train a perceptron network to recognize two input AND gate patterns. Assume initial weights W_{1} and W_{2} equal to 0.2 with a bias weight b_{1} = 0.1, The learning rate of 0.1 is to be considered. Train the network for four iterations. Find the error in each iteration. [10]

Q9) a) Explain how neural network principles are useful for a texture classification problem. [8]

b) Explain application of Neural Network in communication field. [8]

OR

QIO)Discuss various problems involved in recognizing handwritten characters. Explain the convolutional network architecture’s usefulness in recognizing handwritten digits. [16]

Qll)a) Explain the concept of adaptive network based fuzzy inference system (ANFIS) with architecture. [10]

b) Write notes on : [6]

i) Hybrid learning algorithm.

ii) ANN for process control.

OR

Ql2)a) What are Radial Basis function networks. Explain RBF training steps. [8]

b) Explain the equivalence between. ANFIS and RBFN with conditions. [8]