Participants of the Industrial Research School in Complex Systems

Project Title: Adaptive Process Control of Zero Defect Products

Brief description: Industry 4.0 has allowed the strategies of Zero Defect Manufacturing (ZDM) to prosper, and with that, new bursting technologies. With the growing focus of mass customization in Industry 4.0, the small and medium sized enterprizes (SME) are under harder competition than before and need to adjust. This thesis focuses on how Industry 4.0, and ZDM principles can be utilized in SMEs in Norway. Starting with the larger, holistic view including roadmapping and templating technology stack for ZDM. Thereafter narrowing down, looking specifically at defect detection in wood with machine learning among other methods, using Norwegian manufacturig businesses as use cases.

Background: Stine is a PhD candidate at the Department of Mechanical and Industrial Engineering at the Norwegian University of Science and Technology (NTNU). She graduated with a MSc in Physics and Mathematics at NTNU in 2013, where the focus was within Industrial mathematics.  Before starting her PhD, she worked in academia with interdisciplinary and virtual teamwork, and teaching mathematics. Her academic interests focus on real-world applications, and spans industry 4.0, machine learning, algorithms and so on.

Project title: ​​​​​​Human Systems Integration in unmanned systems through conceptual modeling and data sense-making

Brief description: Systems are becoming increasingly complex. Having human operators integrated with autonomous solutions imposes challenges for engineers developing such socio-technical systems. This research examines how the industry can utilize conceptual models and data sense-making techniques during a human systems integration approach in the early product development phase. Through conceptual modeling, we create static and dynamic models with an abstraction of complex systems. In data sense-making, we utilize technical, organizational, and human systems data to explore and support the models. The combination can enhance the exploration and insight in a format that is understandable and shareable with key stakeholders. This project uses action research methodology with case studies and industry as the laboratory in collaboration with companies working towards manned-unmanned-teaming. There is a need for the industry to understand how their unmanned systems interact with human operators, how to model the complexity, and how to utilize data suitably.

Background: Tommy is a Ph.D. candidate at the University of South-Eastern Norway (USN), campus Kongsberg. He holds a Master of Science in Systems Engineering with Industrial Economics and a bachelor's in mechanical engineering with Product Development from USN. He has several years of experience in the Subsea Oil & Gas and the Defence industry, working from early concept to testing of complex systems.

Project title: Cyber-physical convergent information communication technologies (ICT) for virtualised protection and control functionalities in energy systems

Brief description: Comparison between traditional and proposed approaches. (a) Traditional control and protection functionalities are applied to a power converter-based resource. (b) Proposed omnipresence cyber-physical convergent-purpose (PhD project) is applied to power converter-based resources. The physical system sends the measurement via sensors to cyber system, then cyber system with 3C (communication, control, computation) features will return feedback signal to physical system. The objectives of the study is to:

• Create a methodology for virtualisation of the control and protection functionalities in the energy system.
• Create a methodology that can have convergence possibilities of independent technologies (protection/controllers, power electronics devices, digital communication technologies, etc.)
• Carry out experiments using laboratory-based real-time simulations, to test and validate the most promising solution and generate the project outcomes that can be disseminated and exploited.

Research methodologies: 

- OBJ1: The methodology of constructing SCADA system is applied. Digital twin, which is a virtual representation of a real-world physical system or product, can be applied in this objective.

- OBJ2: The object-oriented modelling approach is used. With this method, the control and protection functionalities or more can be merged into one-for-all purposes that can be inherited advantages and eliminate disadvantages of these technologies.

 - OBJ3: The real-time simulation and the use of Hardware-in-the-loop (HIL) can fulfil the requirements. The execution of the simulator should have the small time steps in accordance to the real-time constraints of the physical target. The testing experiments for the proposed method are performed mainly at the real-time hardware-in-the-loop laboratory of DIgEnSys-Lab [1] (Digital Energy Systems and Lab) at University of South-Eastern Norway.

Completed work: 

1. Non-directional Overcurrent Protection Relay Testing Using Virtual Hardware-in-the-Loop Device (Book chapter – submitted).
2. Distance Protection Relay Testing Using Virtual Hardware-in-the-Loop Device (Book chapter – submitted).
3. A Short-circuit Analysis in CIGRE European Medium Voltage Distribution Network (Journal – under Prof’s review).
4. Exploring Cyber-Physical Energy and Power Systems: Applications, Challenges and Simulation Approaches (Review paper – under review).

Completed work: 

1. Real-time cyber-physical system for controlling reactive power using IEC 61850 communication protocol. (Target journal)
2. Ethics in using AI

Background: ​​​​​Le Nam Hai Pham or Lee is currently PhD candidate at University of South-Eastern Norway, campus Porsgrunn.

I was educated at University of Technology in Ho Chi Minh city, Vietnam and graduated a bachelor’s degree in electrical and Electronics Engineering before spending nearly 4 years serving in technology consultant company in Vietnam.

In 2022, I received Master’s degree in Electrical Power Engineering at University of South-Eastern Norway and is recruited to become PhD Research Fellow, current position.

My interest is smart grids, cyber-physical systems, renewable energy, static and dynamic power grids.

Project title: Use of Automation Processes for Detection of Emergent Behavior during Systems Integration Testing

Brief description: A potential for improvement regarding test coverage for the company developed products is what triggered this PhD project. My research is about use of automation processes for detection of emergent behavior during systems integration testing. I will conduct action research using the industry-as-laboratory. The goal of my PhD project is to establish a set of "best practices" for detection og weak emergent behavior in engineered complicated systems using case studies as "proof of concepts". These guidelines can help an observer to a better understanding of the system-of-interest, reducing the perceived emergent behavior and complexity. Emergent behavior and complexity are related terms, scaled according to the difficulty of understanding the behavior of the system and the system itself. Design of experiments and automation will be the areas explored in different ongoing company projects. The value for the company will be to ensure more robust products through smarter testing, discovering more inherent undesired system behavior at an earlier stage, facilitating cheaper mitigation efforts. The timeframe of my PhD project is four years (01.01.2021-31.12.2024), where 75% of my time is allocated to research related work and the remaining 25% is anything related to company projects.

Background: Rune is an industrial-PhD candidate at the University of South-Eastern Norway (USN), campus Kongsberg. He holds a Master in Systems Engineering from USN and a Bachelor in Engineering within System Design from USN. He holds a position as a Senior System Engineer in the company Kongsberg Defence and Aerospace (KDA) with many years of experience within product development (system design and system test). Prior to his career in engineering, Rune was in operational service with the Royal Norwegian Air Force for six years, including graduation from the Air Force Academy.

Project title: Real-time estimation of the energy-mix-limit for the secure operation of converter dominated power system

Brief description: Most countries have created measures to increase the implication of renewable energy by incorporating a new form of renewable energy resources (RES) into the electricity grid. Power electronic converter (PEC) based technologies are quickly changing the generation, transmission, distribution, and utilization levels of the modern power system. Decreased rotational inertia immediately affects system frequency and operational security. A system with low inertia can cause unnecessary blackouts due to frequency variability. A decrease in system inertia raises the RoCoF and the nadir frequency. Voltage stability, rotor angle stability, and frequency control approaches have all been extensively studied in the past. However, real-time control stability has received little attention, despite being the principal cause of recent blackouts. The presented investigations do not produce tangible outputs with sufficient validation and have various limitations. There are still many research gaps in this field, which will be filled soon. Hence, this PhD project aims to develop a novel methodology for estimating real-time indicators and ensuring short-term frequency stability of the PEC-integrated power system in normal and emergency situations. It will include developing and testing a method to discover the optimal mix of energy resources and characteristics for a secure power system. The historical and off-line data will be used to construct an algorithm that can be tested in a real-time simulation environment. The tasks will be performed using deep learning, reinforcement learning, and probability programming. The final goal of this PhD thesis is to test and validate the proposed methodology and information model in a laboratory setting. Co-simulation with hardware in the loop will be used to assess the methodology's appropriateness. The system parameters will be evaluated using various standards and grid codes (especially focused on the Nordic grid).

Biography: Ashish Shrestha received a Bachelor's degree in Electrical and Electronics Engineering from School of Engineering, Pokhara University, Nepal, and a Master's degree in Planning and Operation of Energy System from School of Engineering, Kathmandu University, Nepal. He was also an Erasmus Mundus candidate at the Department of Electrical Engineering, Frederick University, Cyprus, funded by European Union. Currently, he is doing his PhD at the Department of Electrical Engineering, Information Technology and Cybernetics, University of South-Eastern Norway, Porsgrunn Campus, Norway. Before his PhD, he was working as a Lecturer at the Department of Electrical and Electronics Engineering, and a Researcher (Activity Leader) at Center for Electric Power Engineering (CEPE), Kathmandu University, Nepal for three and half years. He was also involved in the problem-based-learning project funded by Erasmus+ program of EU, and was leading a project under the funding of Ministry of Foreign Affairs (MFA), Norway, as the Project Co-Principal Investigator. Till today, he published 48 peer-reviewed journal articles and international conference papers and was assigned as a reviewer for numerous international conferences and peer-reviewed journals from IEEE, Springer, Elsevier, IET and so on. His research interests include Power System Dynamics, Distributed Generation Resources, Planning and Operation of Energy System and so on.

Project title: Process Technology in real time data verification and reconciliation for optimal oil production under the presence of uncertainties

Brief description: The purpose of this research project is to improve existing oil well models by using collected data and use them for dynamic data reconciliation and gross error detection. Data for this study can be collected from both simulation and petroleum industry. This study follows the case­study design for models of artificial lifting oil wells, with in­depth analysis of the uncertainty in the dynamic model. By employing these first­principles models, process model constraints are adopted for data validation and data reconciliation

Background: Zhe is a PhD candidate at the University of South-Eastern Norway, campus Porsgrunn. She holds a Master in  Advanced Control and Systems Engineering from The University of Mancheste and a Bachelor in  Automation at Tianjin University of Commerce. Before her PhD, she previously worked for several years in the robotics field and electrical and electronics engineering. Her interest is data analysis, modelling and simulation, especially data processing, data-driven physics-informed modelling considering uncertainty, gross error detection and data reconciliation.

Project title: Towards digital twin enabled thermo-mechanical processing of post-consumer recycled aluminium.

Brief description:  The demand for aluminium in automotive industry is currently increasing significantly, most likely by a factor 3 to 5 from 2020 to 2050. Since the energy consumed in aluminium recycling is only 5-10% of primary production, increasing the use of secondary aluminium is an effective means of limiting the industry's correspondingly increasing CO2 footprint. The use of recycled aluminium in manufacturing is, however, currently limited by the accumulation of grain refiners, impurities and alloying elements. This leads to a downgrading which is today mostly directed towards casting alloys, and so there is a need to increase the ability to keep secondary aluminium within wrought applications. My PhD will center around process planning and control of aluminium extrusion processes based on post-consumer scrap (PCS), aiming to remedy the loss of process control and product characteristics by extracting information from industrial process data with machine learning. It will create knowledge about model implementation, system characteristics, and system integration to successfully utilize this technology in the aluminium extrusion industry.

Project Title: Improvement and development of new equipment for automatic sampling and processing of environmental DNA (eDNA).

Brief Description: Micro Total Analysis Systems (μTAS) are devices that automate and include all necessary steps for a chemical analysis of a sample. These miniaturized fluidic systems or lab-on-a-chip (LOC) platforms can perform laboratory operations (preparation, separation, detection) on a single device. One of their beneficiaries is small size and channel dimensions of the order of tens of micrometers, μTAS platforms feature negligible sample consumption, reduced cost of the process, and short analysis time.

The focused research will be on micro- and nanotechnologies to improve and develop new equipment for automatic sampling and processing of environmental DNA (eDNA). Many different micropumps, microvalves, optical systems, microchannels, microreactors, micromechanical devices, microelectronic devices, heating elements and fabrication methods have been developed over all these years in the SALICO instrument. These to be tested in proximity at each component during the execution of experiment. Automatic environmental monitoring equipment provides a unique opportunity to monitor the presence of different biological indicator species like pike in rivers and lakes. Because it is an invasive species in fresh water and easily prey upon the small community fish species and destroy. The SALICO cassette with microfluidic chip is designed in a way of performing the molecular assay of Loop mediated Isothermal Amplification (LAMP), Nucleic acid sequence-based amplification (NASBA) together with extraction of DNA/RNA and could be used in the fields (i) the aquaculture industry, (ii) molecular eDNA environmental monitoring and (iii) home-based primary health monitoring of humans and animals. The source of samples are water, mucosa and salvia and have been tested inside the SALICO cassettes with different methods such as LAMP and NASBA. Viruses, bacteria, salmon, and pike have been detected both manually and automatically in the prototype technology. Automatic purification of DNA and RNA from all these samples has been repeatedly demonstrated in our laboratory. Another inconvenience in field monitoring is, the bio reagents could not be stored at room temperature, due to enzyme degradation.

The first objective of the project is to synthesize the lyophilized beads with primers and probes of pike-mitochondria Cyt b gene, confirm the beads are bio active and testing the performance for long term. This lyophilized bead could be stored in on site, where resources are limited and does not require cold chain for transportation.  The suitable stabilizing agent and percentage is determining factor in protecting the bioactivity of enzyme and primers in lyophilized beads. Further, the bioactive beads will be tested on the new prototypes on instruments and Lab on Chip platform. The gene selection and primer designing is the vital step in setting the experiment. The second objective is to pretreat the eDNA sample in appropriate method (filtration), before entering the Lab on Chip. This work is performed in coordination with a consortium. The amplification rate of DNA and specificity are analyzed with the development of primers and probes.

Currently, there is no equipment available for on site, real time, environmental monitoring of eDNA. So, the output of this project will be better preventive in environmental technology.

About Me: I am PhD candidate at Department of Microsystems, University of South-Eastern Norway, Vestfold campus. I received a graduation in Master of technology - Biotechnology (six year integrated program), from Bharathidasan University, India. After an education, I got a research assistant position in one of the renowned research institutes named, CSIR- Central Electrochemical Research Institute, India and earned a research experience in various stream of approach. I took the opportunity to work with team members and publish 6 research papers, since I was engaged in Industry project.

Project Title: Utilizing Big data within early phase of the New Product Development Process (NPD).

Brief Description: The phd project strives to utilize feedback data in terms of failure data in the early phase product development process to enhance data-driven decision-making. The project also seeks to use conceptual modeling and data analysis to guide and support each other in an iterative and recursive manner.

Background: Haytham B. Ali is employed as PhD research fellow and Assistant Professor at the University of South-Eastern Norway (USN). He is working on connecting engineering with science, focusing on mathematics. Haytham focuses on his PhD at using a combination of conceptual modeling and data analysis to enhance the early design phase in the product development process. He holds a Master of Science in Systems Engineering with Industrial Economics degree and a Bachelor's degree in Mechanical Engineering with a specialization in Product Development, both from USN. 

Project title: Effective Utilization of Virtual Reality (VR) affordances in Maritime Simulator Training

Brief description: Virtual Reality (VR) has emerged as a potential alternative form of technology which could assist the contemporary training goals of full-mission maritime simulators. VR has already been used as a training intervention in different levels of training in several safety-critical domains (aviation, maritime, health, education etc.). The integration of immersive technologies in Maritime Education and Training (MET) could pave new ways for skill training for the operators working with manned, unmanned, remote or future autonomous technologies in differing maritime workplaces constituting complex socio-technical systems. Research to create, validate and improve VR training intervention will open doors to more innovative solutions to be integrated in MET practices.

The aim of the PhD project is two-fold. During the first phase, the current state-of-the-art of VR simulators will be explored in maritime education and training contexts. A literature review has been conducted to get an overview of VR simulators for education and differing skill training purposes. In addition, a hybrid Multi-Criteria Decision-Making method (MCDM) have been utilized to evaluate the current state of maritime VR simulators compared to others.

On the second phase, novel methods for training and assessment will be proposed to increase the efficiency of VR simulators. Data collection methods include surveys, interviews, and simulated experiments. Artificial Neural Network (ANN) modelling will be utilized to develop a novel assessment framework for the trainees in VR. Exploring differing constructs such as motivation, self-efficacy and technology acceptance of the maritime trainees to get better insights into their immersive learning process are some of the potential scopes of the project. The remaining duration for this PhD project is three (03) years.

Project Title: Intelligent Diagnosis of Multiple Faults of Permanent magnet Synchronous drives

Project Description: In today's fast-paced industrial world, cutting-edge technology like highly efficient electric machines and the Internet of Things (IoT) are becoming increasingly prevalent in complex applications such as wind turbines, electric vehicles, centrifugal pumps, and compressors. As a result, ensuring the reliable and efficient operation of these applications is more important than ever. That's why our team is working on developing intelligent fault diagnosis schemes for Permanent Magnet Synchronous Motors using state-of-the-art machine learning and data-driven methods. Our goal is to detect faults in industrial applications at an early stage, allowing for more reliable operation and reduced production downtime. With our cutting-edge research, we're confident that we can help companies in a variety of industries to stay ahead of the curve and achieve greater success.

Background: Mahmoud Sayed, a PhD research fellow based in Norway. He received his BSc in Mechatronics engineering from Helwan University, Cairo, Egypt, in 2013, followed by his master's degree in advanced mechanical and robotics engineering from Ritsumeikan University, Kyoto, Japan, in 2019. Mahmoud is currently pursuing his PhD studies at the Top Research Center Mechatronics (TRCM), Department of engineering and sciences, University of Agder, Norway, where he is exploring his passion for data-driven fault diagnosis, electric power trains, electromagnetic modelling and artificial intelligence. With a strong background as a Project Engineer and Research Assistant, as well as experience as an Assistant Lecturer at the Mechatronics Research centre, (MRC), Helwan University, Cairo, Egypt (2019-2021), Mahmoud's expertise and knowledge make him a valuable member of the team.

Title of the project: Fabrication of Single-Atom Catalyst (SAC) based Metal Covalent Organic Frameworks (MCOFs) with CO2 reduction and water splitting applications.

Project Description: The recent trends in the world’s energy consumption patterns indicate an expected 48% surge in energy demand by the year 2040. The major donor to accommodate these growing energy projections is fossil fuel combustion which will still provide a 78% contribution to total energy generation by 2040. However, this anthropogenic combustion of carbon-based fossil fuels leads to the release of greenhouse gases in the atmosphere which is responsible for universal issues such as global warming, environmental pollution, and climate change. Therefore, it is the need of the hour to reduce greenhouse gases by decreasing the reliance on fossil fuels in tandem with somehow fulfilling the ginormous energy demand. This precedence has led to an increasing interest among the scientific community in seeking solutions for these global issues. One of the prospected solutions is the attractive hydrogen economy that revolves around the generation of hydrogen mainly by water-splitting, followed by its storage and then employing it as a fuel to close off the cycle. Another perspective is to utilize the greenhouse gases for their available CO2 content to convert them into meaningful products such as energy-dense fuels and other useful chemicals such as formic acid, methanol, ethanol, etc. Therefore, the goal of this project work would be to synthesize novel single atom-based metal covalent organic frameworks (MCOFs) with a particular focus on electrocatalysis for CO2 - utilization and water splitting applications. These materials possess unique properties such as ginormous surface area, open porosity, mesoporous structures, metal utilization sites, tuneable structures, enhanced stability, and cost-effectiveness. The promise of similar MOF materials is already visible as numerous industries around the world such as Svante (Canada) are already applying this kind of materials for CO2 capture applications. Moreover, if these materials are utilized with attractive techniques such as pulsed electrocatalysis, there is a possibility of achieving unique efficiencies and selectivities. These materials may well be the future of electrocatalysis in various environmental and renewable energy applications.

Background: I am currently a Ph.D. student (in first year of my Ph.D. program) at USN Norway (Department of microsystems, Vestfold Campus). Moreover, I am a trained and qualified individual with a Masters's degree in Advanced Energy and System Engineering from University of Science and Technology (UST, Republic of Korea) following a background in Materials Science. In addition to working with fabrication of materials for CO2 utilization and hydrogen splitting as a part of my PhD program, I have around 3 years of research experience in the fabrication and application of Solid Oxide Fuel Cells (SOFCs) working at the Korea Institute of Energy Research (KIER).

Project Title: Learning Multimodal Intermediate Video and Language representations in Deep Networks for Descriptive Object Identification and Tracking in Urban Environments.

Brief Description: Touseef started his PhD position in September 2021 at University of Agder, Grimstad Norway. Our primary contribution is in the field of big data analytics for multi-modal information, which aims to advance the development of technologies, applications, and services that enhance resilience. The purpose is to improve safety and security in smart districts with a particular emphasis on content analysis in the absence of metadata for the detection and reconstruction of abnormal activities with preventative, operational or intelligence objectives. 

Urban areas are facing rapid growth and require efficient management of resources such as energy, water, and transportation to improve citizens' quality of life while also increasing society's resilience. In this PhD project, we investigate the fusion of video and language modalities through deep learning models to enhance descriptive object identification and tracking in urban environments. Our goal is to bridge the semantic gap between visual and language modalities, which have different statistical properties, and to focus on relevant semantics for correlating visual context and language content in joint embedding spaces. Our investigation into the fusion of video and language modalities through deep learning models has promising applications in various smart city domains. For example, the enhanced descriptive object identification and tracking can be used for object retrieval in videos using text descriptions, which can aid in road traffic monitoring, crowd monitoring and person activity recognition in urban environments. These findings have potential applications in various smart city domains, including intelligent transportation systems, emergency services, and other related areas, to optimize resource management and enhance citizens' quality of life.

Title of the project: Evaluating 5G Core Network in the Context of Cybersecurity for Critical Communication

Project Description: In Norway, various critical communication services, including those of the police, healthcare, fire, and rescue departments, rely on Nødnett, which is built on TETRA network technology.  However, The Norwegian Directorate for Civil Protection (DSB) has plans to transition away from this system and adopt a 4G/5G-enabled Next Generation Critical Communication (NGCC) ( Nødnett in Norway ) As part of this transition. The day-to-day operations and maintenance of Nødnett will be decommissioned by the end of 2026. The Norwegian government has determined that the Next Generation Nødnett(NGN) will be implemented on commercial mobile networks utilizing 4G and 5G technology.

The RAKSHA project aims to conduct research on the integration of threat modelling techniques with a cyber range approach for simulating potential 5G-specific risks to the NGCC Network and evaluating the effectiveness of countermeasures.

Project Partners:  The consortium consists of two groups – research and industry. The research group consists of SINTEF Digital (SINTEF), Simula Metropolitan (SimulaMet), University of South-Eastern Norway (USN), and University of Oslo (UiO). The industry group consists of Norwegian Communications Authority (Nkom), Nasjonal Sikkerhetsmyndighet (NSM), and Telenor ASA (Telenor).

This Ph.D. research project aims to focus on the security of Work Package 4(WP4) in collaboration with SIMULAMet and SINTEF. The primary objectives of WP4 include:

• Hardware-assisted authentication methods that were proposed in WP4 will be further developed.
• By conducting real-world demonstrations in WP3 and WP4, we hope to increase awareness of the importance and societal significance of cyber security in cellular networks, which may help to make it a more attractive career choice.
• The ethical hacking courses offered on our 5G cyber range platform in WP4 will produce highly qualified candidates for positions in this field.
• The project aims to improve the cyber resiliency of our digital society by creating new authentication and attack detection mechanisms to protect critical infrastructures.
• We plan to develop energy-efficient and privacy-focused methods for protecting mobile and IoT devices, autonomous vehicles, and drones against wireless attacks using the lightweight on-device privacy protection demonstration developed in WP4.
• Using the proposed 5G eSIM techniques WP4, we will develop new end-to-end authentication approaches for services accessed through mobile networks.

Project title: Analytics for Wind Farm Asset Management

Project description: A significant portion of wind turbines, currently in operation, will reach the end of their designed life in the coming years. These turbines need regular maintenance, which requires extensive planning and expensive replacements. This makes the Prognostics and health management of wind farms essential for the efficient operation of these assets. Initially the health status of wind energy systems would be quantified at component and turbine levels using intelligent learning algorithms, which would be then extended for the estimation of remaining useful life of the wind turbines. Finally, we develop a reinforcement learning based control for maximizing both the power output as well as the remaining useful life of the wind turbines.

Project Title: Design of soft sensors based on context aware data-fusion and machine learning techniques.

Brief description: Soft sensor design has emerged as a vital engineering field that can make use of the recent advancements in the fields such as Oil and Gas industry, Automotive industry, water networks etc. However, for industrial plants in petrochemical and refining production, it is still important to improve methods for soft sensor designs, in order to reach better estimation accuracy, prediction and estimation of products quality. Under uncertain conditions, it is connected with the overcoming of difficulties such as accounting for the variable time of output measurement and a lack of information about the model's structure. The objective of my Ph.D. work is to develop real-time data- driven machine learning (ML) algorithms to improve the efficiency and performance of context aware Soft sensors. In particular, graph topology identification will be proposed as a tool aimed at understanding the spatio-temporal interactions among various parameters associated to the process variables to estimate. The work is funded and supported by Sentrene for forskningsdrevet innovasjon (SFI) under work package 6 (Data Analytics, IT Integration and Big Data).

Project title: Machine Learning for Ultra Precision Process Control and Optimization

Brief description: In this project, we are looking into improving ultra-precision polishing process efficiency by reducing time, material waste, and ultimately the cost of manufacturing. The goal is to apply machine learning that will improve the process determinism and understanding.

Background: Michael obtained a first-class BEng degree in Computer Systems Engineering from Bangor University (Wales) in 2018. He was awarded the Data Lab scholarship (Scotland's Innovation Centre for AI), and in 2020 he received an MSc degree with distinction in Artificial Intelligence from the University of Aberdeen (Scotland). 

During the summer of 2017, he worked as a control systems intern for Zeeko Ltd at National Facility for Ultra-Precision surfaces (UK), where he was involved in developing an automated system for optics manufactury. Since 2020 he is involved with the Ultra-Precision Surfaces research group based at the University of Huddersfield (England), where he is responsible for the process automation, prototyping, and software implementation for the Swing Arm Profilometer project. 

His current research focuses on the application of machine learning for ultra-precision process control and optimization. He is particularly interested in AI and ML applied to the engineering and manufacturing sectors as well as concepts such as Industry 4.0, cyber-physical systems, automation, and digitalisation.

Project Title: Dilemmas and Choices in games: From machine learning to artificial virtues in intelligent systems

Brief Description: A need to ensue, safe and ethical atificial intelligence arises due to the rapid rise of these technologies in this centrury. Apart from the research on ethical usage of AI, simultaneous effort is required to study and design inherent vitues in AL. This work investigates virtues within atificial agents by training these agents to solve a role-playing game riddled with moral dilemmas. Using highly effective frameworks such as deep reinforcement learming to train agents, we analyse the decisions using state-of-the-at explainable AI (XAI) techniques to interpret and explain them through a virtue ethical lens, to truly understand whether the agent has developed virtues. This way, a platfom to test a variety of artificial agents is developed to compare AI algorithms in tems of their inherent virtues. On top of this system we explore the prediction of virtues of other agents using techniques such as inverse reinforcement learning and Bayesian inference, and finally, to train an agent to behave virtuously to achieve favourable outcomes, given the uncertainty surrounding of virtues of another agent. Overall, this work aims to develop antificial virtuous agents.

Project Title: Benchmarking Product Development Practices for De-Risking the Green Transition

Brief Description: The project aims at medium and large product development organizations in Denmark and Norway. The purpose is to understand how to accommodate for the uncertainties introduced by the green transition in the organizations product/services portfolio management practices. What are the uncertainties and how are they interdependent of each other? What can be done to resolve part of the inbound complexity of those systems where the uncertainty is related? The project is following the Design Research Mythology with high level of case study research.

Project title: Systems approach to human-autonomy interaction (HAI), case of Trust in Automation.

Brief description: My Ph.D. is focused on modeling human-automation interaction for Maritime Autonomous Surface Ships (MASS). More specifically, I am engaged in modeling “trust in automation” to understand how an operator’s trust mediates reliance on automation, which may affect the overall system’s performance. I am applying the principles of systems engineering and system dynamics modeling to address such complexity as an inherent characteristic of today’s socio-technical systems. Moreover, I aim to explore the psychophysiological measurements to validate my proposed model via an eye-tracking experimental study.

Background: I am a doctoral candidate at the Department of Maritime Operations (IMA), University of South-Eastern Norway (USN). I have completed my master’s degree in System Dynamics from the University of Bergen (UiB) and a bachelor’s degree in Industrial and System engineering from Azad University (IAU).