nn

 

www.andhracolleges.com

The Complete Information About Colleges in Andhra Pradesh

  No: N0201 Code

Set No. 1

IV B.Tech I Semester Regular Examinations, Examinations, November 2009 NEURAL NETWORKS AND FUZZY LOGIC (Common to Electrical & Electronics Engineering, Aeronautical Engineering, Electronics & Control Engineering and Instrumentation &Control Engineering)

Time: 3 hours

Max Marks: 80 Answers any Five Questions All Questions carry equal marks *****

1. (a) Explain the architectures for the following artificial neuron models (i) Hodgkin –Huxley model (ii) Integrate and fire neuron model (b) Give in detail about characteristics of ANN.

[8+8]

2. (a) Explain different neuron activation functions and also compare them. (b) Discuss with block diagrams different learning strategies.

[8+8]

3. (a) What is perception? Explain briefly the architecture and algorithm of discrete perceptron network. (b) Discuss in detail about about single layer continuous perceptron networks for linearly Separable classifications. [8+8] 4. (a) State and explain Kolmogorov theorem. (b) Discuss learning difficulties and improvement in multilayer feed forward neural Network. [8+8] 5. (a) Analyze on stability for Hopfield network and also find how to find find capacity of Hopfield network. (b) Discuss Hebbian learning in detail. [8+8]

6. (a) Give and explain the properties of crisp sets (b) Let R,S be defined on the sets sets {1, 3, 5} × {1,3,5}. Let R: {(x, y)|y = x + 2}, S : {(x, y)|x < y}. Using max-min max -min composition .find i)  RoS ii)  SoR

[8+8]

1 of 2 Seminar Topics - Scholarships - Admission/Entrance Exam Notifications USA-UK-Australia-Germany-France-NewZealand Universities List www.andhracolleges.com

Engineering-MBA-MCA-Medical-Pharmacy-B.Ed-Law Colleges Information

 

www.andhracolleges.com

The Complete Information About Colleges in Andhra Pradesh

  Code No: N0201

Set No. 1

~ ~ ~ 7. Let A1 , A 2 , A 3 are three fuzzy sets as shown in figure 1, 2, 3. Find the aggregated ~ ~ ~ fuzzy set of A1 , A 2 , A 3 and find the defuzzification using centroid method. [16]

Figure: 1

~ Figure: 2 A 2  

Figure: 3

~ A3  

8. (a) Explain how ANN is used for process Identification. (b) Explain in detail how classification is done using Fuzzy logic.

[8+8]

2 of 2 Seminar Topics - Scholarships - Admission/Entrance Exam Notifications USA-UK-Australia-Germany-France-NewZealand Universities List www.andhracolleges.com

Engineering-MBA-MCA-Medical-Pharmacy-B.Ed-Law Colleges Information

 

www.andhracolleges.com

The Complete Information About Colleges in Andhra Pradesh

 

Set No. 2

Code No: N0201

IV B.Tech I Semester Regular Examinations, Examinations, November 2009 NEURAL NETWORKS AND FUZZY LOGIC (Common to Electrical & Electronics Engineering, Aeronautical Engineering, Electronics & Control Engineering and Instrumentation &Control Engineering)

Time: 3 hours

Max Marks: 80 Answers any Five Questions All Questions carry equal marks *****

1. (a) Explain briefly about the the spiking neuron model and Mc Culloh Pitts model in detail. (b) Discuss applications of artificial neural networks. [8+8] 2. (a) Discuss in detail operations of Artificial neuron. (b) Explain Widrow-Hoff and delta learning rule in detail.

[8+8]

3. (a) What is hard problem and why it is hard to solve for perceptron network. (b) State and prove perceptron convergence theorem. [8+8] 4. (a) Discuss in detail about credit assignment problem. (b) Which criteria is followed to decide the number of neurons in back propagation Network. [8+8] 5. (a) Explain paradigms of associative memory. (b) Give the aarchitecture rchitecture of Hopfield network for discrete and continuous versions. [8+8] 6. In a computer engineering different logic families families are are often often compared on the basis of their power-delay product. The fuzzy set F is the logic lo gic families F= {NMOS, CMOS, TTL, ECL, JJ}. The range of delay time D= {0.1, 1, 10, 100} in Nano seconds. The power dissipation in micro watts P= {0.01, 0.1, 1, 10, 100} By using max-min composition, obtain a fuzzy relation between delay time and Power dissipation. [16]

1 of 2 Seminar Topics - Scholarships - Admission/Entrance Exam Notifications USA-UK-Australia-Germany-France-NewZealand Universities List www.andhracolleges.com

Engineering-MBA-MCA-Medical-Pharmacy-B.Ed-Law Colleges Information

 

www.andhracolleges.com

The Complete Information About Colleges in Andhra Pradesh

  Code No: N0201

Set No. 2

7 . Let X={a,b,c,d} Y={1,2,3,4} ~ And A  ={ (a,0)(b,0.8)(c,0.6)(d,1)} ~ B = {(1,0.2)(2,1)(3,0.8)(4,0)} ~ C ={(1,0)(2,0.4)(3,1)(4,0.8)} Determine the implication relations ~ ~ (a) IF x is A  THEN y is B   ~ ~ ~ (b) IF x is A THEN y is B  ELSE y is C  

[16] 

8. (a) Explain how ANN is used for process control (b) Give any Application of Fuzzy logic for classification.

[8+8]

2 of 2

Seminar Topics - Scholarships - Admission/Entrance Exam Notifications USA-UK-Australia-Germany-France-NewZealand Universities List www.andhracolleges.com

Engineering-MBA-MCA-Medical-Pharmacy-B.Ed-Law Colleges Information

 

www.andhracolleges.com

The Complete Information About Colleges in Andhra Pradesh

 

Set No. 3

Code No: N0201

IV B.Tech I Semester Regular Examinations, Examinations, November 2009 NEURAL NETWORKS AND FUZZY LOGIC (Common to Electrical & Electronics Engineering, Aeronautical Engineering, Electronics & Control Engineering and Instrumentation &Control Engineering)

Time: 3 hours

Max Marks: 80 Answers any Five Questions All Questions carry equal marks *****

1. (a) Using Mc-Culloh Pitts model implement the following logic functions i) AND gate ii) OR gate (b) Explain in detail about the organization of brain. [8+8] 2. (a) Clearly discuss different architectures of artificial neural networks. (b) Compare supervised and unsupervised learning strategies.

[8+8]

3. (a) Explain the architecture and algorithm of continuous perceptron network (b) Discuss about multi category single layer perceptron network. [8+8] 4. (a) Give derivation for back propagation training. (b) State credit assignment Problem. Explain in detail.

[8+8]

5. (a) Discuss BAM training algorithms in detail. (b) Explain why linear associative network network provides no means of suppression of cross talk noise. [8+8] 6. (a) Ifnumber the fuzzy  = {(x1, 0.4 ) fuzzy 0.4) , (x2, 0.6) , (x3, 0.8)} multiplied with a crisp a = set 0.3.Find the new set formed alongiswith its membership function. (b) Explain the following terms in sets: (i) CON (ii) DIL (iii) Membership function (iv) CRISP. [8+8]

1 of 2 Seminar Topics - Scholarships - Admission/Entrance Exam Notifications USA-UK-Australia-Germany-France-NewZealand Universities List www.andhracolleges.com

Engineering-MBA-MCA-Medical-Pharmacy-B.Ed-Law Colleges Information

 

www.andhracolleges.com

The Complete Information About Colleges in Andhra Pradesh

 

Set No. 3

Code No: N0201

7. (a) Explain with example MOM method for defuzzification. (b) Explain in detail about Fuzzy rule based system with an example.

[8+8]

8. (a) Write short notes on Application of ANN for process fault diagnosis. (b) Explain how Fuzzy logic is used for logic control.

[8+8]

2 of 2

Seminar Topics - Scholarships - Admission/Entrance Exam Notifications USA-UK-Australia-Germany-France-NewZealand Universities List www.andhracolleges.com

Engineering-MBA-MCA-Medical-Pharmacy-B.Ed-Law Colleges Information

 

www.andhracolleges.com

The Complete Information About Colleges in Andhra Pradesh

 

Set No. 4

Code No: N0201

IV B.Tech I Semester Regular Examinations, Examinations, November 2009 NEURAL NETWORKS AND FUZZY LOGIC (Common to Electrical & Electronics Engineering, Aeronautical Engineering, Electronics & Control Engineering and Instrumentation &Control Engineering)

Time: 3 hours

Max Marks: 80 Answers any Five Questions All Questions carry equal marks *****

1. (a) Compare artificial and biological neural networks. (b) Explain in detail about historical development to neural networks.

[8+8]

2. (a) Discuss different learning rules in detail. (b) Give a brief note on neural Dynamics.

[8+8]

3. (a) Discuss different perceptron models. (b) Give applications of single layer feed forward neural networks.

[8+8]

4. (a) Give suggestions to improve and modify back propagation network. (b) Explain generalized delta rule in detail.

[8+8]

5. (a) State and prove BAM theorem. (b) The following unipolar binary vectors must be stored in the recurrent auto associative memory using the outer product method with the nullification of the diagonal. S(1) = [ 1 0 0 1 0 ]t S(2) = [ 0 1 1 0 1 ]t  S(3) = [ 1 1 0 1 0 ]t  (i) Compute matrix W. (ii) Find the analytical expression for the the energy function that the memory is minimizing. [8+8]

1 of 2

Seminar Topics - Scholarships - Admission/Entrance Exam Notifications USA-UK-Australia-Germany-France-NewZealand Universities List www.andhracolleges.com

Engineering-MBA-MCA-Medical-Pharmacy-B.Ed-Law Colleges Information

 

www.andhracolleges.com

The Complete Information About Colleges in Andhra Pradesh

 

Set No. 4

Code No: N0201

6. (a) Consider the fuzzy sets à and defined defined on the interval X=[0,5] of real numbers, by the membership grade functions. x  A~ (x) x - 1    

=

 B~ (x) 2- x    

=

Determine the mathematical formulae and graphs of the membership grade functions of each of the following sets. (i) AC , BC   (ii) A  B (iii) A U B (iv) (A U B)C   (b) What do you mean by CRISP Relations? Explain with an example max-min composition relation. [8+8]

7. (a) Explain in detail about Fuzzy rule based system with an example. (b) Define defuzzification. Explain briefly different methods.

[8+8]

8. (a) Explain how ANN is used for load forecasting. (b) Write short notes on Memory based learning Algorithms.

[8+8]

2 of 2 Seminar Topics - Scholarships - Admission/Entrance Exam Notifications USA-UK-Australia-Germany-France-NewZealand Universities List www.andhracolleges.com

Engineering-MBA-MCA-Medical-Pharmacy-B.Ed-Law Colleges Information