EX-802-D Application of AI in Electrical/Electronics Engineering - May 2023 Question Paper

Under each question, How to answer lists the points an answer should cover, outlined by CampusPrep from the syllabus. Check the details against your textbook.

1. 7 Marks
   aExploit the error correction learning in artificial intelligence with an appropriate example.
   How to answer

   • Define error-correction learning rule and error signal e = desired - actual
   • Explain weight update delta-w = eta * e * x with learning rate
   • Describe iterative correction over epochs with example e.g. perceptron/ADALINE
   • State convergence condition and role of example
2. 7 Marks
   bCompare Hebbian and Boltzmann learning and conclude which one is better and why?
   How to answer

   • Define Hebbian learning principle and its local correlation rule
   • Define Boltzmann learning based on stochastic units and energy minimization
   • Compare on supervision, architecture, learning rule and stochasticity in table
   • Conclude which is better for what condition and why
3. 7 Marks
   aExploit radial basis function network with its architecture.
   How to answer

   • Define RBF network and its three-layer structure
   • Explain input, hidden radial basis and output linear layers with Gaussian function
   • Draw architecture diagram showing centres and weights
   • Mention training: centres selection and linear weight learning
4. 7 Marks
   bDiscuss the functional limit network with its architecture.
   How to answer

   • Define functional link network and functional expansion concept
   • Explain architecture with enhanced inputs using basis functions
   • Draw diagram showing input expansion to single-layer net
   • State advantages: reduced complexity, non-linear separability
5. 7 Marks
   aDiscuss the operations on fuzzy relations.
   How to answer

   • Define fuzzy relation and membership matrix
   • Explain union, intersection, complement operations
   • Explain composition: max-min and max-product composition
   • Give small matrix example of operations
6. 7 Marks
   bExplain the Fuzzy logic with an appropriate example.
   How to answer

   • Define fuzzy logic vs crisp logic and linguistic variables
   • Explain fuzzification, inference, defuzzification steps
   • Give appropriate example e.g. temperature / fan speed control
   • Show membership functions and rule evaluation
7. 7 Marks
   aDiscuss various types of genetic operator with suitable examples.
   How to answer

   • Define genetic operators: selection, crossover, mutation
   • Explain types of selection and crossover with examples
   • Illustrate single-point, two-point, uniform crossover on strings
   • State role of each operator in exploration vs exploitation
8. 7 Marks
   bExplain Mutation operator and its application.
   How to answer

   • Define mutation operator and its purpose to maintain diversity
   • Explain types e.g. bit-flip, swap, Gaussian mutation with example
   • Describe mutation probability and procedure steps
   • List applications in optimization, power system, scheduling
9. 7 Marks
   aExplain the economic load dispatch with suitable example.
   How to answer

   • Define economic load dispatch objective and cost functions
   • State equality and inequality constraints with power balance
   • Explain solution using lambda-iteration / coordination equation with example
   • Mention significance for fuel cost minimization
10. 7 Marks
    bDiscuss load frequency control. Also exploit on what parameters the load frequency control depends.
    How to answer

    • Define load frequency control need and primary vs secondary control
    • Explain single-area LFC block diagram and governor-turbine dynamics
    • List parameters: load change, inertia, droop, tie-line bias
    • Explain dependence on system inertia, damping and regulation
11. 7 Marks
    aExplain how the stability for small signal system is achieved?
    How to answer

    • Define small-signal stability and linearized model
    • Explain eigenvalue analysis of state matrix
    • State conditions for stability: negative real parts, damping
    • Explain role of PSS / excitation control in achieving stability
12. 7 Marks
    bExplain the fuzzy rule based system.
    How to answer

    • Define fuzzy rule based system and IF-THEN rules
    • Explain components: fuzzifier, knowledge base, inference engine, defuzzifier
    • Draw block diagram of fuzzy rule based system
    • Give example of rule base evaluation
13. 7 Marks
    aDiscuss self-organizing maps with their appropriate examples and diagrams.
    How to answer

    • Define SOM as Kohonen unsupervised competitive network
    • Explain architecture: input to 2D lattice and winner-takes-all
    • Draw SOM topology and neighbourhood update with example
    • Mention applications e.g. clustering, visualization
14. 7 Marks
    bWhat is meant by Load Forecasting? Explain it.
    How to answer

    • Define load forecasting and its need in power system
    • Explain types: short, medium, long term forecasting
    • Describe factors and methods e.g. regression, ANN
    • State uses in generation planning and ELD
15. 7 Marks
    aDiscuss the single and two area system.
    How to answer

    • Define single-area vs two-area interconnected system
    • Explain components and tie-line power exchange
    • Draw block diagrams for single and two-area LFC
    • Compare frequency control and ACE in both cases
16. 7 Marks
    bWrite a short notes on any two of the following:

    1. i)Competitive learning
    2. ii)Supervised learning
    3. iii)Learning tasks
    How to answer

    • Select any two topics from the three options
    • Define each with key principle and learning rule/diagram
    • Give suitable example for each selected topic
    • Conclude with applications