Neural networks and fuzzy logic systems – MCQs – EE

1. What is the main inspiration behind Artificial Neural Networks (ANN)?
(A) Human brain and biological neurons
(B) Digital logic circuits
(C) Power grids
(D) Boolean algebra

2. A neuron in an ANN performs which basic operation?
(A) Weighted summation followed by activation
(B) Boolean comparison
(C) Arithmetic division
(D) Logical inversion

3. The function used to introduce non-linearity in neural networks is called:
(A) Activation function
(B) Threshold function
(C) Power function
(D) Gain function

4. Which of the following is a commonly used activation function?
(A) Sigmoid
(B) Linear
(C) Boolean
(D) Step only

5. The process of adjusting weights in a neural network is known as:
(A) Training
(B) Forward propagation
(C) Initialization
(D) Scaling

6. The algorithm used for training feedforward neural networks is:
(A) Backpropagation
(B) Fourier Transform
(C) K-Means Clustering
(D) Linear Regression

7. In a multilayer perceptron (MLP), there are:
(A) Input, hidden, and output layers
(B) Only input and output layers
(C) Only hidden layers
(D) One single neuron

8. The learning rate controls:
(A) The speed of weight adjustment during training
(B) The size of the neural network
(C) The number of neurons
(D) The data sampling rate

9. Which neural network is suitable for time-series prediction in electrical systems?
(A) Recurrent Neural Network (RNN)
(B) Feedforward Network
(C) Convolutional Neural Network (CNN)
(D) Hopfield Network

10. Which neural network type is most effective for image-based fault detection in power systems?
(A) Convolutional Neural Network (CNN)
(B) Feedforward Network
(C) Linear Network
(D) SVM

11. The output of a neuron is typically given by:
(A) Activation function applied to weighted input sum
(B) Direct input
(C) Random output
(D) Input minus threshold only

12. Fuzzy logic deals with:
(A) Approximate reasoning and uncertainty
(B) Binary logic only
(C) Crisp sets
(D) Exact values

13. In fuzzy logic, the degree of membership lies between:
(A) 0 and 1
(B) 1 and 1
(C) 0 and 10
(D) 1 and 100

14. The basic components of a fuzzy inference system are:
(A) Fuzzification, inference, and defuzzification
(B) Training, testing, and evaluation
(C) Generation, transmission, and distribution
(D) Input, output, and feedback

15. Fuzzification converts:
(A) Crisp inputs into fuzzy sets
(B) Fuzzy sets into crisp outputs
(C) Data into binary form
(D) Numbers into logic gates

16. Defuzzification converts:
(A) Fuzzy results into crisp values
(B) Binary data into fuzzy values
(C) Inputs into random outputs
(D) Probabilities into logic

17. Which of the following is a common defuzzification method?
(A) Centroid method
(B) Boolean method
(C) Gradient descent
(D) Nearest neighbor

18. A membership function in fuzzy logic represents:
(A) The degree to which a variable belongs to a fuzzy set
(B) A crisp decision boundary
(C) A neural weight
(D) A logic gate

19. Rule-based reasoning in fuzzy systems is expressed using:
(A) IF–THEN rules
(B) Boolean equations
(C) Arithmetic formulas
(D) Transfer functions

20. The main advantage of fuzzy logic in control systems is:
(A) It can handle uncertainty and imprecise data effectively
(B) It requires exact mathematical models
(C) It eliminates feedback
(D) It ignores noise

21. The Sugeno fuzzy model is often used when:
(A) Output is a linear or constant function
(B) Rules are purely linguistic
(C) No defuzzification is required
(D) Inputs are binary

22. The Mamdani fuzzy model is suitable for:
(A) Human-interpretable control systems
(B) High-speed real-time systems
(C) Digital filters
(D) Linear models only

23. Adaptive Neuro-Fuzzy Inference System (ANFIS) combines:
(A) Neural networks and fuzzy logic
(B) Boolean logic and probability
(C) Circuit theory and control
(D) Analog and digital logic

24. The main advantage of ANFIS is:
(A) It combines learning ability of ANN with reasoning of fuzzy logic
(B) It removes uncertainty
(C) It needs no training
(D) It only uses crisp data

25. In control applications, fuzzy logic is often used for:
(A) Speed control of motors
(B) Boolean computations
(C) Circuit simplification
(D) Manual calibration

26. Which layer in a neural network performs feature extraction?
(A) Hidden layer
(B) Input layer
(C) Output layer
(D) All layers equally

27. The bias in a neuron helps to:
(A) Shift the activation function
(B) Scale the weights
(C) Decrease learning rate
(D) Eliminate overfitting

28. Which neural network model is used for pattern recognition?
(A) Hopfield Network
(B) PID Controller
(C) Boolean Network
(D) SVM

29. The training error of a neural network is minimized using:
(A) Gradient descent algorithm
(B) Fourier analysis
(C) Boolean logic
(D) Relay control

30. The combined use of Neural Networks and Fuzzy Logic in EE leads to:
(A) Intelligent, adaptive, and robust control systems
(B) Simplified binary decisions
(C) Manual computation systems
(D) Fixed-parameter controllers