24
Electrotechnics
REGGIO DI CALABRIA
Overview
Date/time interval
Syllabus
Course Objectives
Program detail:
Steady-state electical networks (3 ECTS)
Circuit model, field-circuit transition and definition of fundamental electrical quantities; definition of bipole and n-pole; bipole networks; classification and
conventions; external features; graphical methods; simple circuit reductions; Kirchhoff's laws for currents and voltages; conservation theorem of virtual
powers (Tellegen); elements of network topology: directed graph, node, edge, mesh, ring, tree, cotree, cutting set, incidence matrix and related properties,
mesh matrix; fundamental matrices; general methods of solving electrical networks: mesh currents and nodal potentials: matrix formulation of the
fundamental system; electrical power absorbed/delivered and related agreements; network theorems: superposition, equivalent generators (Thévenin and Norton),
non-amplification, reciprocity, compensation; linear and non-linear resistive bipoles: definition and characteristics; passive linear n-poles and n-bipoles:
analysis and synthesis; substitution and equivalence: star-polylateral transformation; double resistive bipoles; characterization of double bipoles; concept of
equivalent two-port for small signals; ideal transformer and gyrator; maximum power transfer theorem; Class and laboratory exercises.
Linear and nonlinear networks in general dynamic conditions (2 ECTS)
Dynamic equations and timedomain solution, state variables, initial value problem; terms transitory anpermanent, free and forced evolution; definition of network response to a given input, step and impulse response, convolution integral; Non-linear bipoles; time-varying bipoles bipoles; linearization; piecewise linear features; piecewise
linear analysis of a nonlinear network; state space; nonlinear and time-varying circuits.
Electrical networks in sinusoidal regime (1 ECTS)
Sinusoidal steady state analysis is a technique used to analyze electrical circuits that are driven by sinusoidal voltage or current sources operating at a single frequency. It allow to determine key characteristics like impedance, power, and voltage-current relationships under steady-state sinusoidal operating conditions. This module will include the use of complex numbers, phasorial analysis, and definition of sinusoidal powers.
Neural Networks (3 ECTS)
The artificial neuron: biological and mathematical models (linear function, weights, bias). Activation functions: Sigmoid, ReLU, and Tanh. Network architecture: input, hidden, and output layers.
Learning types: supervised and unsupervised. Training and optimization: Forward pass (calculating predictions). Loss functions: Mean Squared Error and cross-entropy.
Gradient Descent and backpropagation. Control parameters: learning rate, epochs, and overfitting prevention.
Hebbian and competitive learning. Self-organizing maps.
Practical Lab: Working environment—e.g., an introduction to Python and key libraries like TensorFlow or Keras (or cloud environments such as Google Colab).
Building a feedforward neural network for classification or regression tasks.
Testing and evaluation: analyzing results and improving model performance.
Course Prerequisites
Teaching Methods
Assessment Methods
The exam aims at evaluating the skills developed by the student and the methodological rigor in setting and formulating problems and in demonstrating, in particular, the theorems of electrical circuits. The oral exam aims at verifying the level of knowledge of the proposed topics as well as the ability to present the theoretical contents of the discipline.
The final grade will be awarded considering the result obtained in the written test and the outcome of the oral discussion, according to the following evaluation criteria:
30 - 30 cum laude: complete, in-depth and critical knowledge of the topics, excellent language skills, complete and original interpretative skills, full ability to independently apply knowledge to solve the proposed problems;
27 - 29: complete and in-depth knowledge of the topics, full ownership of language, complete and effective interpretative ability, able to independently apply knowledge to solve the proposed problems;
24 - 26: knowledge of the topics with a good degree of learning, good language skills, correct and safe interpretative skills, ability to correctly apply most of the knowledge to solve the proposed problems;
21 - 23: adequate knowledge of the topics, but lack of mastery of the same, satisfactory language skills, correct interpretative ability, limited ability to independently apply knowledge to solve the proposed problems;
18 - 20: basic knowledge of the main topics, basic knowledge of technical language, sufficient interpretative ability, ability to apply the basic knowledge acquired;
Insufficient: if the student decides to take the oral test, it will first aim at examining the written test to discuss the approach to the proposed exercises and assess whether the overall knowledge can be considered sufficient to pass the exam.
Texts
Renzo Perfetti – Circuiti Elettrici – Ed. Zanichelli, 3a edizione, 2024
Chua, Desoer, Kuh – Linear and nonlinear circuits – McGraw Hill
G. Miano – Lezioni di Elettrotecnica – CUEN Napoli/Springer
S. Haykin - Neural Networks ,a comprehensive foundation, IEEE Press
C.Bishop - Neural Networks for Pattern Recognition, Oxford University Press
Additional material and exercises given during the lectures.