Checking date: 03/04/2019


Course: 2019/2020

Intelligent Control
(14046)
Study: Bachelor in Industrial Electronics and Automation Engineering (223)


Coordinating teacher: MORENO LORENTE, LUIS ENRIQUE

Department assigned to the subject: Department of Systems Engineering and Automation

Type: Electives
ECTS Credits: 6.0 ECTS

Course:
Semester:




Competences and skills that will be acquired and learning results. Further information on this link
The aim of this course is that the student knows the basics needed to use the techniques of intelligent control for both modeling and identification of systems and control systems. the concept of fuzzy set and fuzzy operations to further define the concepts of fuzzy relations and fuzzy rules will be introduced. From these concepts a basic fuzzy controller is introduced and will be to identify and control systems from these fuzzy regulators. Then neural networks will be addressed, starting with the notion of artificial neuron, layers of neurons, neural networks and learning strategies in neural networks. the most common neural networks will be introduced and will see how to use them for identification and control systems. Subsequently different systems optimization techniques, both derivative and derivative such as single point, multipoint will be studied. genetic algorithms, differential evolution techniques and PSO among others will be introduced. To achieve these objectives, the student must acquire a range of knowledge and skills. As regards knowledge, the end of the course the student will be able to: 1. Design basic fuzzy controllers for dynamic systems. 2. Approximating a non-linear system using a fuzzy system. 3. Use fuzzy systems for adaptive control schemes. 4. Approximating a nonlinear system using a neural network. 5. To approximate a nonlinear dynamical system using a neural network. 6. Design a system based on neural networks for systems duinámicos control. 7. Using optimization methods based on genetic algorithms. 8. Using optimization methods based on differential evolution algorithms. 9. Using optimization methods based on PSO algorithms. As for general skills or skills during the course they will work: ¿Overview of the problem of identification and control of a nonlinear dynamical system with the techniques discussed. ¿Ability to design controllers for nonlinear dynamical systems, as well as to analyze and interpret the results. This capability is especially work in the labs and in the resolution and discussion of case studies. Capacity for team work cooperatively, critical and respectful of the solutions proposed by others, creative and responsible as a member of a team, to make the designs considered, distributing the workload to tackle complex problems. This capability will work in both practices in laboratory, to be held in equipment, such as in solving exercises, discussions and tutorials that may also have group character. ¿Recognition of the need for continuous learning and the ability to obtain and apply required lainformación accessing technical literature related to the scope of the subject both inEnglish and English. Ability to access the information required to know the details of a particular configuration. ¿Ability to communicate effectively both orally, written or graphic in both Spanish and English throughout the development of the activities proposed in the course (exercises, discussions, practices, etc.).
Description of contents: programme
The program is broken down as follows: 1. Fundamentals of fuzzy or blurred logic. 1.1. Basics of fuzzy logic. Imprecision and uncertainty. 1.2. fuzzy sets. 1.3. Membership functions. 1.4. Operations on fuzzy sets. 1.5. fuzzy relations. 1.6 Operations with fuzzy relations. 1.7. Approximate reasoning. Linguistic variables. 1.8. fuzzy propositions. 1.9. Operations with fuzzy propositions. 1.10. Fuzzy if-then rules. 1.11. Operators involvement. fuzzy inference. 1.12. Controller design based on fuzzy logic rules. 1.13. Models Takagi-Sugeno Mandani and-Kang. 2. Modeling and identification systems using fuzzy techniques. 2.1. Fuzzy function approximation. 2.2. Fuzzy modeling systems. 2.3. Model types. 2.4. Fuzzy model state of a dynamic system. 2.5. Models Takagi-Sugeno Mandani and-Kang. 2.6. Mandani and TSK fuzzy models equivalent of a classic controller. 2.7. Identification of fuzzy models. Methods. 2.8. Identification of the structure. 2.9. Parameter estimation. 3. Design of fuzzy controllers. 3.1. Design of fuzzy controllers without model. 3.2. PID fuzzy controllers. 3.3. Design of fuzzy model based controllers. Adaptive Methods. Direct synthesis methods. Optimization methods online. 3.4. Fuzzy controller design with matlab. 4. Fundamentals of neural networks. 4.1. Concept artificial neuron. Layers of neurons. Concept of neural network. 4.2. Multilayer networks. recurrent networks. 4.3. basic neural networks. Network linear flow: Perceptron and Adaline. Recurrent networks: Hopfield and Hamming. Learning methods. 4.4. feedforward networks. Learning backpropagation. 4.5. Radial basis functions. Probabilistic networks and networks generalized regression. 4.6. Neural networks in matlab. 5. Identification of neural network systems 5.1. Function approximation with neural networks. 5.2. Types of system models. 5.3. Modeling systems with neural networks. NN-FIR. NN-ARX. NN-ARMAX, OE-NN, NN-SSIF. hybrid models. 5.4. Types of networks used in modeling. Networks with delay in inner layers. Backpropagation in dynamic systems. 5.5. Identification of dynamic systems. 6. Control systems with neural networks. 6.1. Direct control schemes. reverse direct control. Internal model control. Feedback linearization. feedforward control. 6.2. Indirect control schemes. 7. Fundamentals of optimization and evolutionary algorithms. 7.1 Methods single point optimization. 7.2 Methods based on the derivative: maximum slope, Newton-Raphson, Quasi-Newton, conjugate gradient. 7.3 Non-derivative methods: brute force, random walk, Hooke-Jeeves, Simulated Annealing-. 7.4 Multipoint optimization methods. 7.5 Derivative Methods: MultiStart and clustering. 7.6 Non-derivative methods: Nelder-Mead, CRS, Genetic Algorithms, Differential Evolution, PSO.
Learning activities and methodology
The activities carried out in the teaching of the subject are: ¿Lectures. Presentation of the main concepts. Discussion and clarification of doubts about the concepts. It will work on transparencies that will be given to students to facilitate learning in addition to a text or basic reference texts on the subject required. ¿Classes of practical exercises. Sessions in which problems arise and let students into groups to raise their solutions. ¿Laboratories. The students (in teams of 2 or 3) will solve a practical case studies, should study and then make the simulation data and analysis. The knowledge about the topics covered in lectures and practical classes in the subject will be used. A previous study will, will work in the laboratory and then a written report with the results and proposed solutions will be delivered.
Assessment System
  • % end-of-term-examination 0
  • % of continuous assessment (assigments, laboratory, practicals...) 100
Basic Bibliography
  • Kriesel, D.. Neural Networks. http://www.dkriesel.com/en/science/neural_networks.
  • Oliver Nelles. Nonlinear System Identification: from classical approaches to Neural Networks and Fuzzy Models, . Springer Verlag. 2001
  • Spall, J.C. . Introduction to stochastic search and optimization. Ed Wiley-Interscience..
  • Zhang, H. and Liu, P. . Fuzzy modelling and control.. Ed Birkhauser..
Additional Bibliography
  • A. Eiben and J. Smith. Introduction to evolutionary computing, . Springer. 2003

The course syllabus and the academic weekly planning may change due academic events or other reasons.