B.E./B.Tech.DEGREE EXAMINATION,APRIL/MAY 2010
Eight Semester
Electrical and Electronics Engineering
IC1403- Neural Network And Fuzzy Logic Control
(Common to Seventh Semester B.E.Instrumentation and Control Engineering and
Electronics and Instrumentation Engineering)
(Regulation 2004)
Time:Three Hours Maximum:100 marks
Answer ALL questions.
PART A-(10*2=20 MARKS)
1. What is Artificial Neuron?
2. Distinguish between supervised and unsupervised learning.
3. Comment on the suitability of Artificial neural Network for control application.
4. what is the use of feedback network?
5. Define the terms membership function and universe of discourse with respect to fuzzy sets.
6. What is a fuzzy relation?
7. State the three forms of fuzzy statement with suitable examples.
8. Given low pressure = {1/50+0.8/52+0.6/53+0.4/54+0.2/55+0/56} find the membership function for pressure not very low.
9. compare the fuzzy logic control and classical control system.
10. What is the need for ANN in Fuzzy Logic Controller design during fuzzification?
PART B-(5*16=80 MARKS)
11.(a)What is a Back Propogation Network? Derive the expression for weight updation in a multilayer feed forward neural network using standard back propogation learning.(16)
or
(b)(i)Perform three training steps of simple network with 3 inputs using the delta learning rule for ?=1,C=0.25.Train the network using the following data pair,
X1= [2 ],d1=-1, X2=[2 ],d2=1, X2=[-1],d3=1
[0 ] [-2] [1 ]
[-1] [-1] [-1]
[1 ] [0.5] [0.5]
Assume bipolar continuous activation function with initial weight vector W^1=[1,0,-0.5,1].(8)
(ii) Explain in detail the types of activation functions used in artificial neuron.(8)
12.(a)(i) Explain ANN configuration for forward plant identification and plant inverse identification.(10)
(ii) Write down the algorithmic steps for Discrete hopfield network(6)
or
(b)(i) Discuss the application of ANN for inverted pendulam control.(10)
(ii) Discuss any one of the neuro control schemes in detail.(6)
13.(a) Discuss in detail the methods of defuzzification used in fuzzy logic with examples.(16)
or
(b)(i) Fuzzy sets A,B,C are defined in the interval X=[0,10] of real numbers by the membership function µA(X)=X/X+2,µB(X)=2^-X,µC(X)=1/1+10(X-2)^2.
Determine
(1) A UNION B
(2) B UNION C
(3) B INTERSECTION C
(4) -
A(4)
(5)SCALAR CARDINALITY AND FUZZY CARDINALITY OF FUZZY SET A(2)
(6)a-CUT FOR THE FUZZY SET A FOR a=0.2,0.4(2)
(ii)Let the fuzzy set A={0.1/x1 + 0.9/x2 + 0/x3}, B= {0/Y1 + 1/Y2 + 0/Y3} find the relation R=AxB using a cartesian product,let C be another fuzzy set C={0.3/x1 + 1.0/x2 + 0.0/x3}.usimg max-min composition find S=C?R and using max-product composition find S=C?R .(8)
14.(a)With a neat block diagram, explain FLC system and discuss the steps involved in designing a FLC. (16)
Or
(b)(i)Explain the concept of adaptive fuzzy system with suitable example(8)
(ii)Explain the steps involved in designing simple Genetic Algorithm by explaining p,ANNA UNIVERSITY MAIL,ANNA UNIVERSITY QUESTION PAPERS,ANNA UNIVERSITY COUNSELLING DATES,ANNA UNIVERSITY RE-EVALUATION RESULTS
Eight Semester
Electrical and Electronics Engineering
IC1403- Neural Network And Fuzzy Logic Control
(Common to Seventh Semester B.E.Instrumentation and Control Engineering and
Electronics and Instrumentation Engineering)
(Regulation 2004)
Time:Three Hours Maximum:100 marks
Answer ALL questions.
PART A-(10*2=20 MARKS)
1. What is Artificial Neuron?
2. Distinguish between supervised and unsupervised learning.
3. Comment on the suitability of Artificial neural Network for control application.
4. what is the use of feedback network?
5. Define the terms membership function and universe of discourse with respect to fuzzy sets.
6. What is a fuzzy relation?
7. State the three forms of fuzzy statement with suitable examples.
8. Given low pressure = {1/50+0.8/52+0.6/53+0.4/54+0.2/55+0/56} find the membership function for pressure not very low.
9. compare the fuzzy logic control and classical control system.
10. What is the need for ANN in Fuzzy Logic Controller design during fuzzification?
PART B-(5*16=80 MARKS)
11.(a)What is a Back Propogation Network? Derive the expression for weight updation in a multilayer feed forward neural network using standard back propogation learning.(16)
or
(b)(i)Perform three training steps of simple network with 3 inputs using the delta learning rule for ?=1,C=0.25.Train the network using the following data pair,
X1= [2 ],d1=-1, X2=[2 ],d2=1, X2=[-1],d3=1
[0 ] [-2] [1 ]
[-1] [-1] [-1]
[1 ] [0.5] [0.5]
Assume bipolar continuous activation function with initial weight vector W^1=[1,0,-0.5,1].(8)
(ii) Explain in detail the types of activation functions used in artificial neuron.(8)
12.(a)(i) Explain ANN configuration for forward plant identification and plant inverse identification.(10)
(ii) Write down the algorithmic steps for Discrete hopfield network(6)
or
(b)(i) Discuss the application of ANN for inverted pendulam control.(10)
(ii) Discuss any one of the neuro control schemes in detail.(6)
13.(a) Discuss in detail the methods of defuzzification used in fuzzy logic with examples.(16)
or
(b)(i) Fuzzy sets A,B,C are defined in the interval X=[0,10] of real numbers by the membership function µA(X)=X/X+2,µB(X)=2^-X,µC(X)=1/1+10(X-2)^2.
Determine
(1) A UNION B
(2) B UNION C
(3) B INTERSECTION C
(4) -
A(4)
(5)SCALAR CARDINALITY AND FUZZY CARDINALITY OF FUZZY SET A(2)
(6)a-CUT FOR THE FUZZY SET A FOR a=0.2,0.4(2)
(ii)Let the fuzzy set A={0.1/x1 + 0.9/x2 + 0/x3}, B= {0/Y1 + 1/Y2 + 0/Y3} find the relation R=AxB using a cartesian product,let C be another fuzzy set C={0.3/x1 + 1.0/x2 + 0.0/x3}.usimg max-min composition find S=C?R and using max-product composition find S=C?R .(8)
14.(a)With a neat block diagram, explain FLC system and discuss the steps involved in designing a FLC. (16)
Or
(b)(i)Explain the concept of adaptive fuzzy system with suitable example(8)
(ii)Explain the steps involved in designing simple Genetic Algorithm by explaining p,ANNA UNIVERSITY MAIL,ANNA UNIVERSITY QUESTION PAPERS,ANNA UNIVERSITY COUNSELLING DATES,ANNA UNIVERSITY RE-EVALUATION RESULTS