4 PhDs on AI-driven applications for Smart Distribution System Operation

Disruptive innovations are needed in managing and operating distribution grids. Are you our next PhD researchers in exploring disruptive innovations in managing and operating distribution grids?

Irène Curie Fellowship

No

Department(s)

Electrical Engineering

Institutes and others

EIRES - Eindhoven Institute for Renewable Energy Systems, EAISI - Eindhoven Artificial Intelligence Systems Institute

Reference number

V36.7489

The role of the distribution system operator (DSO) is changing from a passive maintainer of electricity networks to an active coordinator in the edge of the energy system. At the same time, customers become enabled to change from passive energy users to active participants in the local electricity system. Maintaining privacy and grid (cyber) security levels are part of the challenge to face.

A digital transformation at the edge of the distribution grid and at connected customers is unfolding. This opens possibilities for deploying distributed intelligence to enable smart network operations by collecting and processing data while preserving high levels of privacy for the customers. Exploring AI models for smart System Operation (AISO) is a collaboration project in which DSO Alliander will work together with the TU/e departments of Electrical Engineering (Electrical Energy Systems group) and Mathematics & Computer Science (Interconnected Resource-aware Intelligent Systems, and Stochastic Operations Research) to realize these innovations.

The project will be part of the TU/e’s Eindhoven AI Systems Institute (EAISI) and Eindhoven Institute for Renewable Energy Systems (EIRES) programs and therefore share, learn, and disseminate within the EAISI and EIRES communities and through the TU/e master programs Data Science and AI, Medical Engineering and AI Engineering Systems, and educational activities from the TU/e Electrical Energy Systems group and Math & Computer Science department.

If you are eager to work with a multi-disciplinary team focusing on AI-driven applications to support the DSO then this is the right position for you.   

Job Description

The project focuses on synthetical data generation, AI-driven state estimation, stochastic modelling and reliability assessment, and grid-edge optimal solutions. These models will be combined with the AI-driven state estimations to enhance network observability and grid monitoring. Additionally, integration with the stochastic modelling and reliability assessment process will provide valuable insights into the impact of uncertainties on grid reliability. Finally, in conjunction with the developed edge intelligence, these advancements will enable optimal solutions for the electricity grids in the Netherlands and e.g. the rest of Europe, while maintaining user privacy.

The research results will be immediately utilized by Alliander for congestion estimation and flexibility procurement. To achieve this, it is part of this project that all the developed (AI-driven) models and algorithms are also implemented in production-ready open-source packages.

The four main research tracks (RTs) of AISO are as follows:

RT1: Synthetical data generation using multivariate models

This research will focus on a scalable approach by using multivariate models to capture uncertainties in spatial-temporal correlations. It is expected to combine both top-down and bottom-up approaches into a scalable and stable solution. More specifically, the research will include:

• Requirement specifications for both real-time state estimation and long-term power flow analysis
• Design algorithms which combine the aggregated data and anonymized individual data
• Validation of the developed algorithms in the simulation environment with grid use-case scenarios
• Integrate and validate solutions in the virtual grid environment.

RT2: AI-driven state estimation and prediction

This research aims to combine physics-based models, i.e. state estimation based on WLS, with physics-aware neural network structures to improve network observability and grid monitoring capability. Besides a normal estimation of system state, it should be also possible to determine anomalous events using AI-driven techniques. More specifically, the research will include:

• Development of semi real-time measurement solutions
• Uncertainty modeling of (synthetic) LV load/generation profiles
• Increasing network observability with physics-aware neural network algorithms
• Anomaly detection and mitigation solutions against anomalous attacks
• Integrate and validate solutions in the virtual grid environment.

RT3: Stochastic modelling and reliability assessment

This research will perform a microscopic bottom-up approach to develop a thorough understanding of how various component affect overall network reliability. To this end, we will develop detailed agent-based probabilistic models to examine various vulnerability assessments, like the event of unacceptable voltage fluctuations, or degradation acceleration of temporary exceeding thermal limits, relevant to the daily operations of power systems. More specifically, the research will include:

• Development of multivariate uncertainty models describing random user behavior (e.g. arrival patterns of electric vehicles and user preferences (either behavioral or utility-based)
• Development of physical models of distribution grids, in particular for voltages: tractable linearized distflow models or less tractable but more realistic models
• Integration of component degradation models along with the grid models
• Development of stochastic models considering the uncertainties in load and weather forecasting
• Integrate and validate solutions in the virtual grid environment.

RT4: Grid-edge optimal solutions

Along with the new generation of smart meters and the emerging development of virtual grid concept, there is a clear trend for system operation to push intelligence towards the edge devices.

To this end, within this research track, the successful candidate will work on designing privacy-aware accurate, fast, and efficient semi/self-learning AI models that can deliver results comparable with cloud-based fully supervised AI models. More specifically, the research will include:

• Designing self-learning edge AI models based on the privacy/security-by-design principle
• Tackling AI model generalization
• Optimization of distributed and centralized AI model design and load balancing
• Integration and validation of simulation-based solutions in the virtual grid environment

Job Requirements for RT1 & RT2

• A MSc degree related to modeling and analysis of power systems and distributed energy resources.
• Having a good understanding of distribution grid planning and operation.
• Experience in data-driven modeling, probabilities, stochastic optimization solutions is an advantage.
• Skills in scientific programming and/or numerical computing in languages like Python, Julia, or MATLAB are advantages.
• Excellent modeling skills in simulation tools such as PowerFactory or other equivalent open-source power system packages. Having experience in connecting such simulation tools with the programming environment of MATLAB and/or Python is an advantage.
• Enthusiasm in open-source and motivated to learn basic skills of scientific software engineering.
• Ability to work in an interdisciplinary team and interested in collaborating with industrial partners.
• Motivated to develop your teaching skills and coach MSc and BSc students.
• Fluent in spoken and written English (C1 level).
• Dutch language skill is an advantage.

Job Requirements for RT3 & RT4

• A MSc degree in Computer Science, Data Science, or related fields
• A strong background in deep learning, distributed ML, and AI model optimization
• Good scientific programming skills and experience inlanguages such as Python, C++, Julia, etc.
• Enthusiasm in open-source and motivated to learn basic skills of scientific software engineering.
• Strong analytical, implementation, and experimentation skills
• Ability to work in an interdisciplinary team and be a team player
• Motivated to develop your teaching skills and coach MSc and BSc students
• Fluent in spoken and written English (C1 level)

A meaningful job in a dynamic and ambitious university, in an interdisciplinary setting and within an international network. You will work on a beautiful, green campus within walking distance of the central train station. In addition, we offer you:

• Full-time employment for four years, with an intermediate evaluation (go/no-go) after nine months. You will spend 10% of your employment on teaching tasks.
• Salary and benefits (such as a pension scheme, paid pregnancy and maternity leave, partially paid parental leave) in accordance with the Collective Labour Agreement for Dutch Universities, scale P (min. €2,770 max. €3,539).
• A year-end bonus of 8.3% and annual vacation pay of 8%.
• High-quality training programs and other support to grow into a self-aware, autonomous scientific researcher. At TU/e we challenge you to take charge of your own learning process .
• An excellent technical infrastructure, on-campus children's day care and sports facilities.
• An allowance for commuting, working from home and internet costs.
• A Staff Immigration Team  and a tax compensation scheme (the 30% facility) for international candidates. 

About us

Eindhoven University of Technology is an internationally top-ranking university in the Netherlands that combines scientific curiosity with a hands-on attitude. Our spirit of collaboration translates into an open culture and a top-five position in collaborating with advanced industries. Fundamental knowledge enables us to design solutions for the highly complex problems of today and tomorrow. 

Curious to hear more about what it’s like as a PhD at TU/e? You can navigate here .

Information and application

This general job posting covers 4 different vacancies. To apply for a particular vacancy, please click on the link of the vacancy you are interested in and therefore not on the apply button on this page for this overall vacancy.

PhD 1 / RT1: Synthetical data generation using multivariate models .

PhD2 / RT2: AI-driven state estimation and prediction .

PhD3 / RT3: Stochastic modelling and reliability assessment .

PhD4 / RT4: Grid-edge optimal solutions .

Visit our website for more information about the application process or the conditions of employment. You can also contact [email protected] .

Are you inspired and would like to know more about working at TU/e? Please visit our career page .