PEI103: INTELLIGENT CONTROL TECHNIQUES AND APPLICATIONS
Course Objective: To understand the concepts of Artificial Intelligence and Expert Systems, to enable to design Intelligent Controls.
Overview of Intelligent control techniques and Expert Systems: Intelligent control techniques, Concept of artificial intelligence, General Concepts of Expert System.
Artificial Neural Networks: Artificial Neuron models, Types of activation functions, Neural network architectures, Neural Learning: Correlation, Competitive, Feedback based weight adaptation, Evaluation of networks, Quality of results, Generalizability, Computational resources, Supervised learning: Perceptrons, linear separability, Multilayer networks, Back propagation algorithm and its varianta, Unsupervised learning, Winnertake all networks, Adaptive resonance theory, Self-organizing maps, Hopfield networks, Boltzmann machines, Support Vector Machine, Typical application in identification, Optimization, Pattern recognition. Applications of ANN in Process control, Robotics and other industrial control methods.
Fuzzy Logic: Fuzziness vs probability, Crisp logic vs fuzzy logic, Fuzzy sets and systems, Operations on sets, Fuzzy relations, Membership functions, Fuzzy rule generation, De fuzzy controllers, Type2 Fuzzy Logic Controllers, Multi-layer and other advanced Fuzzy Logic Models, Applications of Fuzzy Logic. Applications in Process control, Robotics and other industrial control methods.
Evolutionary Computation: Introduction to optimization problem, constraints, objective functions, unimodel/ multimodel problems, classical techniques/evolutionary computational techniques Genetic Algorithms and its Operators, variants of Genetic Algorithm and its use in Engineering Process Control.
Experiments around Input and output using Fuzzy logic, Graphical analysis of various control systems using Fuzzy logic, Dynamical and optimal training for neural networks, Algorithms around GA.
1. Case studies related to application of artificial intelligence to process control.
2. Application of neural network to pattern recognition and classification.
3. Application of fuzzy logic to pattern recognition and classification.
4. Application of fuzzy logic to process control.
5. Application of ANN/ fuzzy logic techniques to system identification and control.
6. Application of evolutionary algorithms to controller design.
Course Learning Outcomes (CLO):
After the completion of the course the students will be able to:
1. Apply artificial intelligence and expert system concepts.
2. Apply fuzzy logic control to process.
3. Use evolutionary computation applications.
4. Acquire knowledge about hybrid techniques
5. Apply intelligent techniques in process control, robotics and industrial control systems
1. Narayana, Y., Artificial Neural Networks, PrenticeHall of India (1999).
2. Rich, E., and Knight, K., Artificial intelligence, McGraw Hill (1991).
3. Ross, J. T., Fuzzy Logic with Engineering Applications, John Wiley (2004).