Courses of Academic Year 2020/2021

Controllo ottimo vincolato (Constrained optimal control)

Lecturer: Prof. Paolo Falcone

tentative: Jan 2021


Abstract: Il corso introduce alla teoria del controllo ottimo predittivo (Model Predictive Control), partendo dalla formulazione del problema di controllo ottimo con costo quadratico per sistemi lineari (controllo LQ). I principali risultati di stabilita’ e di persistent feasibility verranno illustrati utilizzando esempi numerici e applicazioni in ambito automotive e Matlab toolbox quali MPT e MPC toolbox.

Contenuti del corso:

- Cenni di controllo ottimo (3h)

- Cenni di ottimizzazione convessa (3h)

- Model predictive control. Formulazione del problema (3h)

- Applicazioni. Controllo della dinamica laterale del veicolo. Cruise control (3h)

Not yet defined

CFD: 3

Power electronics converters for automotive applications: trends and challenges

Lecturer: Prof. Giampaolo Buticchi, The University of Nottingham Ningbo China,

June/July 2021

Module 1 (4h) Multi-level converters, an automotive prospective for high speed drive applications
Module 2 (4h) High performance DC/DC converters for automotive applications
Module 3 (4h) Potential of Active thermal control for reliability improvement of automotive drives.

Not yet defined

CFD: 3

Academic English Workshop II

Lecturer: Silvia Cavalieri

to be defined

The main objective of the workshop is to provide PhD students with the knowledge of the rhetorical and linguistic models that characterize the English academic language. The seminar intends to introduce language and stylistic tools t to write accurate texts. In particular, the AEWII will focus on some specific written genres of the academia, i.e. doctoral theses and posters and Phd students will have extensive opportunities to practice both with text analysis and with their production. Special attention will be given to argumentative techniques of posters ‘ oral presentations.

written assessment: writing a paper introduction; oral assessment: poster presentation.

CFD: 3

Academic English Workshop I

Lecturer: Silvia Cavalieri

to be defined

The workshop aims at giving an overview of the linguistic conventions adhered to by the English-speaking academic community, focussing on aspects such as the structure of research articles, the writing of abstracts and the preparation of conference presentations. More specifically, the workshop will try to raise the participants' awareness of the rhetorical and discourse patterns characteristic of academic English, introducing them to the skills required to produce texts which are accurate both from a grammatical and a stylistic point of view. Ample opportunities will be given for practice, both in the analysis and the production of texts. In particular, special attention will be given to argumentative writing techniques and to special genres such as the abstract and conference presentations.

Written assessment: abstract writing; oral assessment: conference presentation.

CFD: 3

Biosensing with advanced electronic devices

Lecturer: Prof. Muhammad Alam Ashraf (Purdue University, USA)

Professor of Electrical and Computer Engineering
Purdue University
School of Electrical and Computer Engineering
Hall for Discovery & Learning Research
207 S. Martin Jischke Dr.
West Lafayette, Indiana 47907-1971

Muhammad Ashraful Alam is a Professor of Electrical and Computer Engineering where his research and teaching focus on physics, simulation, characterization and technology of classical and emerging electronic devices. From 1995 to 2003, he was with Bell Laboratories, Murray Hill, NJ, where he made important contributions to reliability physics of electronic devices, MOCVD crystal growth, and performance limits of semiconductor lasers. At Purdue, Alam’s research has broadened to include flexible electronics, solar cells, and nanobiosensors. He is a fellow of the AAAS, IEEE, and APS and received the 2006 IEEE Kiyo Tomiyasu Award for contributions to device technology.


Tentative: Summer 2021

Lectures (of 1.5 hours each)

Part 1: Introduction to Nanobiosensors

Lecture 1: What is nanobiosensors, anyway?

Lecture 2: Basic concepts: Biomolecules, Analyte density, diffusion distances

Lecture 3:Basic concepts: Types of biosensors, geometry of biosensing

Part 2: Setting Time

Lecture 4: Response time of classical nanobiosensors

Lecture 5: Response time of complex nanobiosensors

Lecture 6: Beating the diffusion limit by biobarcode sensors

Lecture 7: Beating the diffusion limit by Droplet Spectroscopy

Lecture 8: Beating the diffusion limit by analyte flow

Lecture 9: Settling time vs. first passage time

Part 3: Sensitivity

Lecture 10:Sensitivity and types of biosensors

Lecture 11:Potentiometric biosensors

Lecture 12:On charge screening of cylindrical sensors

Lecture 13:ISFET as a pH-meter

Lecture 14:Origin of charges in a biomolecule

Lecture 15:How to beat screening

Lecture 16:Amperometric Sensor – An introduction to glucose sensors

Lecture 17:Amperometric Sensors – Michaelis-Menton equation

Lecture 18:Beating the diffusion limit in an amperometric DNA sensors

Lecture 19: Elements of an Cantilever sensor

Lecture 20:Cantilever sensors and its nonideal response

Lecture 21:Nonlinear nanobiosensing by a Flexture-FET

Part 4:Selectivity

Lecture 22: Selectivity: energetics of molecular recognition

Lecture 23: Selectivity: Spatial distribution of random sequential absorption

Lecture 24: When all else fail: Tag, filter, and amplify

Lecture 25:An information theory perspective on selectivity

Lecture 26:Physics of Ion-selective fuel-cell based glucose sensors

Lecture 27:Physics of Ion-selective sweat-sensors

Part 5:Putting them together

Lecture 28:Genome sequencing by Ion Torrent – Part 1

Lecture 29:Genome sequencing by Ion Torrent – Part 2

Lecture 30:Introduction to Microfluidic and paper based sensors.

Lecture 31: Conclusions: Looking back and looking forward

The course will provide an in-depth analysis of the origin of extra-ordinary sensitivity, fundamental limits, and operating principles of modern nanobiosensors. The primary focus will be physics of biomolecule detection in terms three elementary concepts: response time, sensitivity, and selectivity. And, we will use potentiometric, amperometric, and cantilever-based mass sensors to illustrate the application of concepts for specific sensor technologies. Roughly speaking Lectures 1- 5, 9, 10-15, 22, 23 cover the fundamental topics of sensing, the others complement the scenario with application examples.

Not yet defined

CFD: 12

The “Wide” Advantage in Power Electronics

Lecturer: Alessandro Chini

spring 2021

Aim of this course is to introduce students to the performance improvement that power electronics circuits can achieve by taking advantage of the “wide-band-gap” semiconductor technologies. Gallium-Nitride and Silicon-Carbide based devices recently entered the market but they are expected to replace in a near future the conventional Silicon-based devices that represent nowadays the mainstream technology. The course will cover the following topics (total of 12hours/3CFU) 1) Role of efficiency in power conversion circuits 2) Operating concept of switching converters 3) Semiconductor switches and their figure of merits 4) Example of applications comparing the performances of Silicon, Gallium-Nitride and Silicon-Carbide based semiconductor devices

to be defined

CFD: 3

Data Science and Machine Learning: basics and applications to Health Care

Lecturer: Dr. Paolo Missier

Tentative: Nov 2020


PART I Fundamentals of Machine Learning methods, overview of Data Science in the Health Care

  1. Introduction to Data Science and the role of Machine Learning: Exploratory vs Predictive Data Analytics (EDA, PDA) ?
  2. EDA: Overview of common techniques with notebook examples ?
  3. Machine Learning: basic techniques, common pitfalls, and how to avoid them ?
  4. Expressive intelligible supervised learning models: General Additive models with pairwise interactions (GA2M). Shaply index ?
  5. Application. Insights from wearable activity trackers: Human Activity Recognition using a public benchmark dataset ?


PART II Health Applications and transition to Deep Learning

  • The UK Biobank: opportunities for research: working with Electronic Health Records ?
  • Genomics: From DNA sequencing to variant calling: a big data processing pipeline. ?
  • Genome-wide Association studies (GWAS). ?
    • GWAS using the Hail platform (Spark): a complete example from the Broad InstituteFrom GWAS to machine learning for Genome-Wide Association studies ?
  • Introduction to Deep Neural Networks (Deep Learning) with applications to health care


Lab activities:  

  1. The course takes a very practical, hands-on approach to illustrate key concepts, with the help of python programs that show popular data analytics and ML libraries at work in detail (Pandas and Scikit-learn). These are implemented as Jupyter notebooks and made available for students to work with throughout the course. Some understanding of python programming for scientific application is desirable, but not essential. ?


  1. At the start of the course, students will have the opportunity (optionally) to embark in a week-long data science experience using high frequency triaxial accelerometers (activity trackers) made available by Newcastle University. They will be able to collect their own activity data and pre-process them using open source third party SW, and then implement their owned hoc analysis algorithms to “make sense” of the activity traces. ?


Main references:

Not yet defined

CFD: 8

Corsi Di Formazione Complementare Per Dottorandi E Assegnisti Ediz. 2020


to be defined


Not yet defined

CFD: 6

Uncertainty Quantification with Applications in Science and Engineering

Lecturer: Prof. Clemens Heitzinger (TU Wien)
Clemens Heitzinger received his master's degree (Dipl.-Ing.) in mathematics and his PhD degree (Dr. techn.) in technical sciences with honors both from TU Vienna. He was a visiting researcher in the Department of Mathematics and Statistics at Arizona State University, a research associate in the School of Electrical and Computer Engineering at Purdue University, and a senior research associate in the Department of Applied Mathematics and Theoretical Physics at the University of Cambridge. In 2015, he returned to TU Vienna as an Associate Professor in the Department of Mathematics. He is also an Adjunct Professor in the School of Mathematical and Statistical Sciences at Arizona State University. He was awarded the START Prize, Austria's most prestigious award for young scientists, by the Austrian Science Fund (FWF) in 2013. His research interests are stochastic partial differential equations, uncertainty quantification, Bayesian PDE inversion, and reinforcement learning with applications for example in nanotechnology and medicine.


Tentative: Fall 2020

In recent years, stochastic aspects have started to play an increasingly important role in the modeling and simulation of physical and engineering applications using partial differential equations (PDE). These aspects, models, and methods are often subsumed under the term uncertainty quantification. Applications include, for example, semiconductor devices and nanoscale sensors, as random effects have become essential at the nanometer scale. However, the ideas and methods can be applied to many other applications. In this lecture series, we start with the theory of deterministic PDE for modeling transport phenomena. The Poisson and Poisson-Boltzmann equations are the foundation for self-consistency. The Boltzmann transport equation provides the basis for charge transport and many-body problems, and other transport equations such as the drift-diffusion-Poisson system can be derived from it. We discuss existence and uniqueness of the solutions and introduce numerical methods. Some deterministic and stochastic homogenization problems stemming from nanotechnological applications will also be discussed. Next, stochastic versions of the transport equations are introduced and the multilevel Monte-Carlo method for the efficient solution of stochastic PDE is explained. Various extensions are also discussed. Optimal methods are also explained; this means that given a prescribed total error, the optimal parameters of the method are determined such that the total computational effort is minimized. In the last part, Bayesian inversion for determining unknown parameters in PDE models is discussed. Bayesian inversion makes it possible to calculate unknown model parameters or parameters functions by comparing the model with measurements. Bayesian inversion has the advantage, compared to other, simpler methods, that it yields probability distributions of the unknown parameters. From the probability distributions, conclusions about how well nonlinear, ill-posed inverse problems can be solved can be drawn immediately and, e.g., confidence intervals are found. The theory of Bayesian PDE inversion is discussed, practical algorithms are explained, and applications to engineering problems are shown. Bayesian inversion is considered an essential new tools for sensor devices. Lecture notes (ca. 300 pages) will be distributed.

Not yet defined

CFD: 2