MATLAB Academic Forum Research and Teaching
Modeling and simulation for climate change, sustainability and electrification

Shubo Chakrabarti, MathWorks (in English)

Open Science with MATLAB: from Open Data to Reproducible Workflows

Science is based on collaboration and the ability to build upon existing data and research findings. However, enabling the re-use and reproducibility of scientific workflows remains a big challenge. This talk will give a broad overview of how scientists and researchers are using MATLAB^® to access open data, use scientific infrastructure like CINECA, create transparent and re-usable scientific workflows and share reproducible content with the community.  

Immacolata Oliva, Sapienza Università di Roma

The concession licensing and the economic and environmental sustainability challenges: theoretical and numerical approaches

Hydrocarbon extraction is still a major source of energy supply for most countries in the world, despite commitments to progressively reduce these operations, to meet the needs of the energy transition. Licensing the exploitation of fossil natural resources is currently a hot topic for both academics and policy makers. In this talk, we present a dynamic stochastic optimization problem within the real options framework, providing the optimal exercise price of the expropriation option, namely, an option in the hands of the public body conceived to protect the social and environmental interests against the over-exploitation of the resource. The flexibility of the model facilitates applications to real(big)data and MATLAB^® plays a crucial role in evaluating and managing the results.

Stefania Gentili, Istituto Nazionale di Oceanografia e di Geofisica Sperimentale - OGS

A MATLAB package for the forecasting of strongest aftershocks: the NESTOREv1.0 software

Forecasting strong subsequent events after a major earthquake is critical for civil defense because it helps to assess the risk of further damage and possible collapse. The NESTORE (Next STrOng Related Earthquake) algorithm was developed specifically for probabilistic forecasting of clusters in which a strong main earthquake is followed by at least one aftershock of comparable magnitude. NESTORE categorizes clusters as Type A or Type B based on the magnitude difference between the main earthquake and its strongest aftershock. This difference is equal or smaller than 1 in the Type A case, larger than 1 in the Type B case. The goal of NESTORE is to estimate the probability that a given cluster belongs to Type A in near real-time.

Previous versions of NESTORE have been successfully used to analyze seismicity in California and Italy and have shown their effectiveness (Gentili & Di Giovambattista, 2017, 2020, 2022). The latest version, NESTOREv1.0, is more robust and reliable due to improved statistical analysis and the ability to be applied to regions with lower seismic network sensitivity. NESTOREv1.0 uses a multi-parameter pattern recognition approach based on one-node decision trees and analyzes nine selected seismicity features at increasing time intervals following a significant mainshock. It has already been successfully applied in California and Greece (Gentili et al., 2023; Anyfanti et al., 2023)

NESTOREv1.0 is now mature enough to be made available to the scientific community for applications and testing in new areas. The software is available on GitHub and on MathWorks. The software includes four modules that correspond to the following goals: (1) identification of clusters from seismic catalogs, (2) determination of appropriate thresholds for features to distinguish cluster types using a training set, (3) evaluation of the performance of the algorithm on an independent test set based on the results of the previous module, and (4) real-time application to clusters while they are occurring.

With the deployment of NESTOREv1.0 software, we aim to extend the impact of the NESTORE algorithm and help advances in the field of strongest aftershockforecasting during seismicity clusters.

Funded by a grant from the Italian Ministry of Foreign Affairs and International Cooperation 

Nicola Acito, Università di Pisa

Atmospheric compensation of satellite hyperspectral images

The talk is focused on the research activity related to the characterization of the atmosphere from satellite remotely sensed hyperspectral images carried out by the Remote Sensing and Image Processing Group at the University of Pisa. Specifically, starting from the physical model of the radiative transfer in the atmosphere, the main tasks required by an atmospheric compensation system and their implementation in the MATLAB environment are summarized. Results concerning the estimation of columnar concentration of water vapor and carbon dioxide in the atmosphere from real satellite data are presented.

Marco Riani, Università degli studi di Parma

Robust data analysis with MATLAB to describe the interaction between soil and atmosphere

The purpose of this talk is to show how we have used MATLAB to address the issues related to climate changes. In particular, this talk will focus on long time series of field experimental data from continuous monitoring at a test site in Oltrepò Pavese (northern Italy): the goal is to provide a statistical model able to describe the interaction between soil and atmosphere. An innovative statistical function is proposed, and its capability of linking the quantities involved in the phenomena, namely soil volumetric water content, soil-water potential, air temperature, rainfalls, solar radiation is shown. The data are treated in the framework of robust statistics by using the combination of robust parametric (LTS and SARIMA) and non-parametric models (SSA). The model has been completely developed inside the MATLAB environment and makes use of the FSDA toolbox. In this talk we show how the fitted models can capture the relevant features present in the experimental data and therefore how it can be used for prediction purposes.  

Lorenzo De Donato, Università degli Studi di Napoli Federico II

Generating synthetic data through RoadRunner to develop AI proofs-of-concept for railway applications

This presentation aims to highlight the potential of RoadRunner in creating simulated scenarios to collect synthetic data for the development of Artificial Intelligence (AI) Proofs-of-Concept (PoCs). When it comes to developing AI systems, data collection is one of the most sensitive phases as it may result to be expensive, time-consuming, or even unsafe. In this context, tools like RoadRunner could introduce several advantages in AI PoCs development as they allow the simulation of various scenarios and the safe collection of multiple data in a short time. This presentation will provide some details on how to generate a scenario in RoadRunner and how AI applications could benefit from it. The case study used to exemplify the generation process and the usage of the synthetic data focuses on Computer Vision applications for railway problems.

Bharat Mishra, INFN - LNS

3D modeling of magnetoplasma electron and ion dynamics via particle-in-cell simulations (in English)

Plasma modeling is a complex task which involves solving coupled equations of motion to describe the state of the plasma. Depending on the density and temperature of the species involved and the external fields, plasma models can be divided into single particle, fluid and kinetic models. Magnetoplasmas such as those used in electron cyclotron resonance ion sources (ECRIS) are better described using kinetic models which are numerically solved through particle-in-cell (PIC) simulations. The plasma physics groups at INFN-LNS and INFN-LNL have developed a 3D PIC code suite in MATLAB to model electron and ion dynamics in a self-consistent manner, capturing the most relevant physics for each (EM wave propagation and damping, stepwise ionization process, self-generated potential dip, atomic excitation etc.) and capable of furnishing space-resolved information on density, energy and atomic level populations. We present here an overview of the status of the senumerical models and their implementation in MATLAB, highlighting the utility and benefits offered by the latter. We also outline some perspectives on space-resolved X-ray emissivity calculations which can pave the way for potential use as a diagnostic tool.

Riccardo Patriarca and Francesco Simone, Sapienza Università di Roma

Studying cyber resilience in a water network: by using Simulink and MATLAB it becomes our cup of… water

Cyber-physical systems (CPS) are becoming increasingly prevalent in various sectors and are commonly found in modern industrial contexts. While CPS offer enhanced efficiency, coordination, and quality, they also introduce new disruption scenarios that jeopardize the integrity and resilience of the systems. In this context, analyzing system performance requires a perspective of cyber resilience, which integrates traditional assessment of physical resilience with the specificities of cybersecurity. Such analysis needs to encompass the system as a whole, considering both system elements and their interconnections. 

In this research work, we present a systematic method for evaluating the cyber resilience of a water treatment plant and its connected distribution system. To accomplish this, a simulation procedure was developed, wherein a Simulink^® model of the plant was integrated with a georeferenced EPANET representation of a water distribution network. Four stochastic cyber resilience metrics were proposed and computed in MATLAB to assess the societal impact of cyber attacks. These metrics expand the physical modeling of the plant by considering the variability of cyber attack duration, water demand, urban distribution, and system response capabilities. 

The results exemplify the benefits of simulations in understanding the behavior of cyber-physical systems, identifying system vulnerabilities, and potentially supporting decision-making processes in emergency situations. 

Marco Laiolo, Università degli Studi di Torino

MIROVA – A satellite-based Near-Real-Time automated system for volcanic hot-spot detection

Amongst natural hazards, volcanic activity may have a great impact on human life and health, economic goods, transports and infrastructures. To reduce, or at least mitigate, the impact of volcanic events, understanding and predicting the behavior of volcanic systems is crucial, reason why it has long been the ultimate goal of the volcanological community. Satellite-retrieved information became essential in modern volcanology, permitting the acquisition of continuous and multiparametric data, while avoiding the risks associated with field surveys. Since 2013, at the Department of Earth Sciences of the University of Turin – in collaboration with the University of Firenze (Italy) and on behalf of the Italian Department of Civil Protection (DPC) – a fully automated volcanic hotspot detection system has been developed. The system, named Middle Infrared Observation of Volcanic Activity (MIROVA), was originally developed to monitor the thermal flux of the Italian volcanoes, and now boasts a global coverage, monitoring ~200 of the most active volcanoes worldwide. The whole system was designed and developed in MATLAB environment in most of its steps which include: i) the data downloading; ii) satellite image processing; iii) local database construction and iv) the outputs of the static figures accessible on the dedicated website. In this decade, the MIROVA system enabled early recognition and/or tracking of various volcanic behaviors such as volcanic unrest or ongoing effusive eruptions. It is currently used by several volcanic observatories worldwide, contributing to better define the periodic activity level of the volcanoes, thus, supporting their daily monitoring, particularly during eruptive crises. Considering the present and future satellite technological developments, the advanced monitoring requests by volcanological communities, and the increasing global digitalization, a powerful, solid, and trustworthy programming and computing platform such as MATLAB will be essential for making MIROVA system ready to face the future challenges. In this perspective, the main MIROVA’s issues are the time-consumption processes which currently limit the number of investigated volcanic targets (more than 1500 worldwide active volcanoes), and the need for a more user-friendly interface from MATLAB to work directly on web-related products and language environments.

Alice Berthuy - MathWorks

MathWorks Startup Program and the Accelerator Program (in English)

In this presentation we will be introducing the MathWorks Startup Program and the Accelerator Program, which provides startups with access to our industry-leading tools. We will outline the key benefits of the program, including access to our comprehensive portfolio of tools, technical support from our engineers, and co-marketing opportunities.

How in 30+ years of activity SATE used MATLAB and Simulink

How in 30+ years of activity SATE used MATLAB, Simulink, also with code generation and some toolboxes to simulate and analyze compressible and incompressible fluid systems and their controls, in particular:  

• internal acoustics by reciprocating machines and flow induced vibrations  
• large compression systems 
• advance power generation systems 
• gas turbine systems 

SATE encourages graduating students to perform their master thesis at SATE, under detailed supervision by their personnel, in order to gain a priming on professional programming with MATLAB & Simulink in a stimulating and innovative working environment, which could become an employment opportunity. 

A recent thesis related to the energy transition trend regarded the study and simulation by Simulink of electric batteries with the purpose of developing algorithms to estimate their health status. 

Marco Aimo-Boot, IVECO

An integrated practice of MATLAB and Simulink in a 2nd level master program

In 2023 Iveco Group, in partnership with Politecnico di Torino, has launched the 2nd Edition Master of Digitalization and Autonomous Commercial Vehicles for Carbon-Free Logistics.  

This biyearly 2nd level Master program aims at providing training for young graduate apprentices. It develops skills for the application of innovative technologies in the automotive sector, with particular focus on autonomous driving and electrification technologies applied to commercial vehicles. 

The selected young engineers come from various universities in Italy and around the world and have degrees in different engineering sectors.  

MATLAB & Simulink tools play a key role in the scheduled training initiatives starting from the academic practice on the different specialized courses moving to the expert use in the daily project running in the product development departments of Iveco Group.   

An integrated approach has been established among Iveco Group, Politecnico di Torino and MathWorks to promote and facilitate the transition of the toolchains use in the multidisciplinary training. 

In the presentation will be emphasized how the integrated practice work properly and the main benefits coming from this approach implementation under the supervision of both the academic and company coordinators. 

Andrea Fontana, Alessandro Forlani - Politecnico di Milano - Formula Student Team

Leveraging Simscape Multibody Suite for Enhanced Vehicle Design

This presentation addresses how Dynamis PRC, the formula student team of Politecnico di Milano, leverages the Simscape Multibody suite to create a digital twin of their race car prototype. The team showcases how this model integrates multiple subsystems, including thermal, electrical, aerodynamics, and torque vectoring, to optimize the design phase.

The use of the Simscape Multibody suite yields significant advantages, facilitating faster and more efficient decision-making processes. The model provides enhanced accuracy and precision, empowering the team to obtain more insightful and reliable results throughout the design process.

The presentation emphasizes the diverse range of simulations performed using the model, such as open-loop simulations, minimal lap-time simulations, and driver-in-the-loop simulations. These simulations enable the team to assess different scenarios and fine-tune the vehicle's performance, ultimately leading to improved overall capabilities.

Moreover, the team demonstrates the integration of the Vehicle Dynamics Blockset Interface for Unreal Engine 4, enabling immersive driver-in-the-loop experiences on a static simulator. This integration provides an interactive environment, facilitating comprehensive analysis and aiding in the design optimization process.

The validation of the model is vital to ensure its fidelity and credibility. Dynamis PRC conducts rigorous track testing sessions and ISO maneuvers to validate the model against real-world performance, ensuring its accurate representation of the car's behavior.

In summary, this presentation highlights how the Dynamis PRC team employs the Simscape Multibody suite to develop a full-car multibody model, integrating multiple subsystems for comprehensive design optimization. The simulations, aided by an Unreal Engine interface, contribute to faster decision-making and enhanced performance. The rigorous validation process using track testing and ISO maneuvers ensures the reliability and accuracy of the model, supporting the team's success in the formula student competition.

Paolo Panarese, MathWorks

MATLAB resources for Teaching

This presentation will provide an overview of MATLAB resources for teaching. MATLAB is a powerful tool for teaching and learning in engineering, mathematics, and other STEM fields. The presentation will cover a range of resources, including MATLAB tutorials, course materials, and teaching examples for different disciplines. Participants will learn how to use these resources to enhance their teaching and engage students in active learning and innovation projects. The presentation will also cover how to integrate MATLAB into existing curricula, including tips for creating assignments and assessments. The goal of this presentation is to provide participants with practical tools and strategies for using MATLAB to enhance their teaching and help students develop critical skills in problem-solving, computational thinking, data analysis, programming, modeling and simulation. 

Aldo Corbellini and Gianluca Morelli, Università degli Studi di Parma

MATLAB Tools for Teaching Statistics and Data Analysis

The purpose of this talk is to show how MATLAB is used in the Department of Economics of the University of Parma and to illustrate in a nutshell the main contents of the book “Data science con MATLAB”. In the first courses of statistics a series of MATLAB GUIs have been written and incorporated into the FSDA toolbox to show to computation of the most important statistical indexes (see for example http://rosa.unipr.it/FSDA/GUIregress.html). In the most advanced course of statistics a set of interactive plots have been written in order to better accompany the students in the different phases of data analysis. For example, in the context of preliminary data analysis, the function mdpattern (see documentation) enables the user to understand the eventual presence of patterns in missing data. In the context of dimension reduction and quantitative variables, the app pcaFS performs the computation of different singular value decompositions and enables to show/hide interactively the different ingredients of this technique. In the context of correspondence analysis, the app coranaAPP enables to highlight the atypical rows and columns in a contingency table. The final part of the talk is dedicated to show how MATLAB has been integrated into our Elly platform (Moodle) by means of MATLAB Grader.  

By coding the solutions in the MATLAB programming language and verifying the accuracy of the suggested solutions, Grader gives the teacher the opportunity to provide the students the opportunity to assess their own levels of grasp of key concepts by having them complete interactively articulated assessments. 

Additionally, Grader provides teachers with a simple method for producing MATLAB scripts that can be used to verify students' solutions and to share these script collections with other teachers. 

Fabio Viola, Università degli Studi di Palermo

Didactic innovation project in Electrical Engineering for e-mobility course in Università di Palermo

The three-year degree course in Electrical Engineering for e-mobility is focused on providing students with the knowledge required for designing, managing, and maintaining electric mobility systems in various transportation sectors. This course stands out as the only three-year engineering program in Europe, while master's courses are available in other universities. 

The course is structured into three distinct periods. The first year covers fundamental knowledge common to industrial engineering, including mathematics, physics, and chemistry. In the second year, students receive a comprehensive engineering education that encompasses mechanics, circuit theory, automatic controls, and electronics. The third year focuses on specialized knowledge in electrical engineering and electric mobility, covering topics such as measurements, machines, and electrical systems. 

To achieve the educational goals, an innovative teaching approach called "Problem-based learning" (PBL) has been adopted. In this technique, students work in groups to define their own objectives and develop solutions for real-world problems related to electric mobility. PBL will be implemented through various laboratories. 

While there is no rigid demarcation between different subjects (for example the definition of the model of an electric vehicle and its autonomy requires knowledge of both physics and chemistry), is necessary to provide students with a unique multipurpose development tool: MATLAB.  

The course includes three annual laboratories: Basic Laboratories, Engineering Laboratories, and Characterizing Laboratories. Each laboratory activity carries three university training credits (ECTS). These laboratory activities integrate the theoretical knowledge acquired in the corresponding semesters and involve the use of software tools such as MATLAB, Microsoft Office, and AutoCAD. 

In the first year, laboratory activities focus on integrating the disciplines from the first semester, such as Mathematical Analysis, Geometry, and Computer Aided Design.  

The second-year laboratory activities build upon the knowledge gained in the first semester and involve developing circuit models for diagnosing faulty behavior of power electronic elements. 

In the third year, students continue to integrate their knowledge from previous semesters with ongoing courses. They are presented with problems specifically requested by industrial clients, allowing them to apply the Problem-based learning technique. 

The course offers students autonomy in choosing how to acquire necessary credits. Students can follow university-level courses or independently pursue courses available in the official MathWorks MATLAB training. These courses cover topics such as MATLAB, Simulink, Artificial Intelligence, and the Internet of Things, as is described here.

Overall, this innovative educational project aims to introduce students to problem-based learning techniques and the MATLAB development environment, providing them with the skills necessary for a successful career in electrical engineering for e-mobility. 

Silvio Simani, Università degli Studi di Ferrara

Teaching Control Engineering Courses: enhancing learning by integrating theory & practice

The talk summarizes some teaching tools developed in the MATLAB and Simulink environments that have highlighted to be particularly effective to enhance the learning activity. In particular, it shown how this teaching activity should be supported by a “learning by doing approach”, which enhances the development of theoretical and practical issues proposed to the student at the same time. On the other hand, the use of “realistic examples” taken from different engineering backgrounds helps to engage students and attract their interest towards difficult theoretical activities. Moreover, the design of proper “manual and semi-automated procedures” that are tailored to the considered application examples drives the students to learn the engineering approach to solve practical problems. A typical problem concerns, for example, the teaching of the design of phase lead and lag control networks by using frequency requirements of the Control System Toolbox, as well as the derivation of transfer function and state-space models from physical models by means of the MATLAB and Simulink Toolboxes.  

Mattia Frasca, Università degli Studi di Catania

Use of MATLAB in "Automation Engineering and Control of Complex Systems" Master at Università di Catania

In this talk, I will briefly show the pervasiveness of MATLAB in the Master Degree in "Automation Engineering and Control of Complex Systems" at the Università di Catania. All the courses of the Master Degree are taught in English and include a theoretical part followed by laboratory and/or practical exercises, many of them are carried out in MATLAB or Simulink. The Master Degree comprises two curricula, one named "Automation" and the other one named "Automation for Biotechnology". The use of MATLAB is mainly devoted to a better understanding of the subjects presented during the lectures, where usually the teacher makes a lot of examples with related code, as well as to develop the homeprojects that the students carry out at the end of the course. Examples include subjects such as "robust control" where MATLAB is used to perform the open loop or closed loop balancing, the model order reduction, the design of the robust controller, the solution of LMI problems; "control of electrical machines", where Simulink-based models of the controllers for electrical machines are developed during the lectures; "robotics" and "biorobotics" where the controllers for manipulators, legged and wheeled autonomous robots are simulated in MATLAB; "nonlinear control systems" where MATLAB is extensively used to simulate all controllers studied in the course as well as for some practical control applications; "biomedical analysis and control systems" where MATLAB is used to analyze signals from the human body; "complex adaptive systems" where numerical solutions of nonlinear systems, bifurcations, networked systems are analyzed and controlled with MATLAB; "process modeling and control" where MATLAB is used for identification and modeling.  

Matteo Corno, Politecnico di Milano

Teaching Autonomous Driving principles to high school students with MATLAB

TechCamp@POLIMI aims at building ‘bridges’ between Politecnico di Milano and secondary schools or professional institutes. TechCamp@POLIMI consists of weekly summer courses in English covering many subjects in the STEM field: green energy, race car dynamics, coding, cybersecurity, robotics, and intelligent autonomous vehicles. In the morning sessions, professors teach the foundations that then students have the possibility to put into practice in the afternoon hands-on sessions. 

This presentation describes how MATLAB is used within the intelligent autonomous vehicles course as tool for students to experiment the features of the main modules of an autonomous navigation stack: control, localization, planning and perception. For each topic, we illustrate how we used MATLAB to design guided from building a simple kinematic model to image and Lidar processing.