Where PhDs and companies meet
Menu
Login

Distributed learning for Adaptive Control of Local Batteries in electrical networks with high intermittency

ABG-130310 Thesis topic
2025-04-01 Public funding alone (i.e. government, region, European, international organization research grant)
Centrale Lille
- Les Hauts de France - France
Distributed learning for Adaptive Control of Local Batteries in electrical networks with high intermittency
  • Energy
  • Engineering sciences
Distributed computing, intermittent renewable energy integration, Supervised and unsupervised learning, power systems, adaptive prediction, local area balancing, optimal operation control, decentralized control, multiagent system

Topic description

Context:

Frequency control was historically managed through power reserves that are provided solely by synchronous machines. The significant penetration of DERs, connected to networks by power electronics, tends to reduce the number of synchronous machines. Power reserves must therefore be partly provided by these DERs. But, their controllability is much more difficult because PV and wind generation are intermittent, and highly dependent on external conditions (wind, sun, etc.).

Scientific objective:

More adaptive controls are required to connect renewable, intermittent and uncertain power sources.

Balancing of power systems in time scales of one day ahead to instantaneous time involves contextual information coming from predictions (load demand, PV production, …) and state measurements (grid voltages, currents, ….) that contain various types of uncertainties. The edge computing and the increase in computing power have enabled the development of learning techniques, which improve local control algorithms in on line operation. Distributed learning techniques offer the advantage of being able to control electrical networks more resiliently than current centralized techniques.

The aim is to investigate the design of a distributed grid control system with ANN that will improve the local real time balancing and not require complex physics based models of power systems. This research project is based on developed works by the research group “Power systems”. The considered use case will be a local energy community, which includes several types of each DER (batteries, PV, water heater, controllable electrical vehicle charger, a small synchronous generator) and tests will be performed onto the experimental demonstration EPMLAB of the L2EP lab.

Starting date

2025-09-01

Funding category

Public funding alone (i.e. government, region, European, international organization research grant)

Funding further details

2300 brut/mois

Presentation of host institution and host laboratory

Centrale Lille

The Laboratory of Electrical Engineering and Power electronics (L2EP) (EA 2697) brings together all the electrical engineering research activities from Centrale Lille Institute, the University of Lille, Arts et Métiers, and Junia. The laboratory currently employs 107 people, including 36 professors and researchers, and 14 administrative and technical staff. Around forty junior researchers enrolled in doctoral thesis work at the laboratory, as well as around twenty postdoctoral fellows and engineers on fixed-term contracts.

Centrale Lille is a public higher education and research institution that has been training high-level engineers and researchers for 170 years. It is comprised of four engineering schools (Ecole Centrale de Lille, ENSCL, IG2I, and ITEEM), offers 14 master's programs, and awards doctorates (three doctoral schools). The institute also co-supervises 7 research laboratories in the Lille metropolitan area.

PhD title

Doctorat en Génie Electrque

Country where you obtained your PhD

France

Institution awarding doctoral degree

Centrale Lille

Graduate school

Sciences pour l'ingénieur (SPI)

Candidate's profile

- A Master in Computing Science or Electrical Engineering with a focus on process control 

- Knowledge in power systems, control engineering, artificial neural networks, distributed and parallel computing will be appreciated 

- Strong analytical and programming skills, experience in at least one software platform: Python, MATLAB 

- Research experience/publications in project related areas 

- The candidate must have the ability to work independently, good analytical, synthesis and innovation skills 

- Good communication and writing skills in English 

- Experience in on board implementation of software as OPAL-RT, DSpace, Spherea, Typhoon, Arduino, Raspberry, … are added values

2025-05-02
Partager via
Apply
Close

Vous avez déjà un compte ?

Nouvel utilisateur ?