Menù principale
B031774 - COMPLEMENTS OF OPERATION AND CONTROL OF SUSTAINABLE SMART GRIDS
Main information
Teaching Language
Course Content
Suggested readings
Learning Objectives
Prerequisites
Teaching Methods
Further information
Type of Assessment
Course program
Sustainable Development Goals 2030
Academic Year 2023-24
Course year
Second year - First Semester
Belonging Department
Industrial Engineering (DIEF)
Modulo di sola Frequenza of
Scientific Area
ING-IND/31 - ELECTRICAL ENGINEERING
Credits
3
Teaching Hours
24
Teaching Term
11/09/2023 ⇒ 15/12/2023
Attendance required
No
Type of Evaluation
Giudizio Finale
Course Content
show
Course program
show
Lectureship
Teaching Language
English
Course Content
Machine learning applications to smart grids to improve sustainability
Suggested readings (Search our library's catalogue)
S. Russell, P. Norvig, “Artificial Intelligence: a modern approach”, 2/e, Prentice Hall, 2005, ISBN: 88-7192-2228-X;
Igor Aizenberg, “Complex-Valued Neural Networks with Multi-Valued Neurons”, Springer, 2011;
I. Goodfellow, Y. Bengio, A. Courville, “Deep Learning”, MIT Press, 2016;
S. Haykin, “Neural Networks and Learning Machines”, 3/e, Prentice Hall, 2008;
E. Mocanu, “Machine Learning applied to Smart Grids”, Technische Universiteit Eindhoven Press, 2017; free access
Jason Bell, “Machine Learning - Hands-On for Developers and Technical Professionals, Wiley, 2016; free access from unifi
Igor Aizenberg, “Complex-Valued Neural Networks with Multi-Valued Neurons”, Springer, 2011;
I. Goodfellow, Y. Bengio, A. Courville, “Deep Learning”, MIT Press, 2016;
S. Haykin, “Neural Networks and Learning Machines”, 3/e, Prentice Hall, 2008;
E. Mocanu, “Machine Learning applied to Smart Grids”, Technische Universiteit Eindhoven Press, 2017; free access
Jason Bell, “Machine Learning - Hands-On for Developers and Technical Professionals, Wiley, 2016; free access from unifi
Learning Objectives
The course will provide students with a working knowledge of fundamentals of computational intelligence, communication technology and decision support system applied to Smart Grids. Power flow analysis and optimization schemes needed for the generation, transmission, distribution, demand response, and reconfiguration is explained in detail and simulation tools such as Matlab, SapWin and Power World are used.
Prerequisites
Electrical circuits Basics
Teaching Methods
The course is developed on:
10 hours of classroom lectures;
14 hours of computational laboratory;
20 hours of individual study and drafting of project work.
10 hours of classroom lectures;
14 hours of computational laboratory;
20 hours of individual study and drafting of project work.
Further information
More detailed information on the e-learning site of the Course (Moodle platform).
Type of Assessment
technical report discussion
Course program
Machine Learning applied to Smart Grids. Definition of Artificial Intelligence (AI), Computational Intelligence (CI), Machine Learning (ML), Artificial Neural Network (ANN). Aspects shared by CI systems. Functioning principle of the biological neuron. Fields of applications of CI-ML systems. Possible applications of CI-ML systems in the Smart Grid field. The elementary Perceptron neuron. Activation function (af) of an artificial neuron: definition and types of afs used in practice. Topologies of ANNs in relation to stratification and data flow. Definition of Learning (Learning) and types of learning used in practice. MLP multilayer feedforward neural network with backpropagation supervised learning, Universal Approximation Theorem. Feedforward Time Delayed (TDNN) and Recurrent Synchronous (RNN) neural topology. Problems related to the generation of learning sets. Information redundancy, linear and non-linear PCA. Encoding and normalization of input and output data. Correct evaluation of the error. The generalization problem: causes and remedies. Recurrent (Asynchronous) neural networks: the Hopfield network and the Self Organizing Maps. Energy function. Hebb's law of learning. Genetic algorithms and PSO for optimization problems. Applications to several case studies related to Smart Grids and RECs (Renewable Energy Communities) and development of a project.
Sustainable Development Goals 2030
7 - 11 -12