ARTIFICIAL NEURAL NETWORK 

Course Objective:
To introduce the concept of artificial network as an alternative options for solving engineering problems.

1. Working with data (4 hours)
1. Data types; data, information and knowledge; concept of data mining; Dimension reduction of data matrix: Principal component analysis.


2. Introducton of Artifical Neural Network (ANN) (6 hours)
1. Biological Analogy, Historical development; ANN terminology; network structure; basis functions; activation functions; advantages of ANN; application areas of ANN.


3. Learning process & optimization techniques (10 hours)
1. supervised learning: Error correction learning, memory based learning
2. unsupervised learning: Hebian learning, competitive learning
3. learning with critic
4. gradient descent and least mean square
5. Derivative free optimization techniques: advantages of derivative free techniques; genetic algorithm: fundamental of GA and biological background.; GA operators & GA operation.
6. Simulated annealing: theoretical background and algorithm.


4. Supervised network (8 hours)
1. McCullotch and Pitt Neuron; LTUs, simple perceptron and perceptorn learning. Limitation of simple percepron.
2. ADDALINE network and delta rule
3. Multilayer perceptron: Needs of multilayer network, generalized delta rule (error‐backpropagation), effect of momentum term and learning rate
4. Error backpropagation learning of sigmoidal units; drawbacks of error‐backpropagation


5. Unsupervised network (4 hours)
1. competitive network: network structure & working;
2. dissimilarity measures;
3. Self Organizing Map and Kohonen learning;
4. applications


6. Special networks (4 hours)
1. Radial basis function network: structure and working procedure, advantages
2. LVQ network: structure and learning approach
3. Hopefield network
4. Autoassociative memory network: general structure and Purpose, Autocorrelator; Heterocorrelator


7. Application of ANN in Electrical Engineering (8 hours)
1. Fault diagnosis
2. Control application
3. Network planning
4. Forecasting task.
5. State estimation
6. Unit commitment

Practical:

1. Computer simulation of PCA.
2. Computer simulation of perceptron network
3. computer simulation of back propagation network
4. A Shortterm case study demonstrating ANN application for a specific purpose.

References:

1. Simon Hykin, "Neural networks A Comprehensive Foundation", second edition; Pearson Education.