Où docteurs et entreprises se rencontrent
Menu
Connexion

Vous avez déjà un compte ?

Nouvel utilisateur ?

Integration of a machine learning module for on line optimal adjustment in an energy management system

ABG-96408 Sujet de Thèse
03/03/2021 Contrat doctoral
Centrale Lille
Villeneuve d'Ascq - Les Hauts de France - France
Integration of a machine learning module for on line optimal adjustment in an energy management system
  • Sciences de l’ingénieur
  • Informatique
smart grids, énergie renouvelable, ajustement temps réel, réseau de neurones, apprentissage, Intelligence artificielle, planification automatique, Self-optimisation

Description du sujet

From the predictions of electricity demand and the production of generators based on intermittent renewable energy (mainly PV and wind power), the research team has developed deterministic and stochastic optimization methods allowing to plan, the day before for the next day, the hourly profiles of the set points of controllable conventional generators in an urban micro-grid. This allows:

- the minimization of operating costs (fuel, etc.),

- the minimization of emissions linked to the use of fuels,

- the planning of a power reserve in the event of the appearance of uncertainties.

The objective of this PhD is to apply auto adaptive methods for updating power references according to observed variations in real time by applying machine learning techniques derived from artificial intelligence. More specifically, two methods will be explored: one using a cost function with variable parameters in a specified domain and one based on the integration of the cost function in the learning algorithm of an artificial neural network for mapping the corrective control function. Orthogonal ANN and Support Vector Machines will be considered as AI technologies.

Prise de fonction :

04/10/2021

Nature du financement

Contrat doctoral

Précisions sur le financement

Présentation établissement et labo d'accueil

Centrale Lille

Thèse délivré par Centrale Lille Institut

Directeur de thèse : FRANCOIS Bruno

Laboratoire d’Electrotechnique et d’Electronique de Puissance de Lille : L2EP (France) – Centrale Lille

équipe Réseaux électriques

 

Intitulé du doctorat

Doctorat, spécialité Génie Electrique

Pays d'obtention du doctorat

France

Etablissement délivrant le doctorat

Centrale Lille

Ecole doctorale

ed SPI72

Profil du candidat

Expected profile :

Candidates should have a Master degree in Electrical Engineering or in Artificial Intelligence. The candidate with the following knowledge will be preferred:

-  Knowledge about machine learning algorithms hardware integration.

· Fundamental knowledge about the power system operation, control and analysis

· Good knowledge in optimization theory and stochastic problems, familiar with one of the optimization software, such as Cplex or Conopt.

· Knowledge/experience about distributed networks, energy systems modelling and operation

- Strong capability of coding using Python and Matlab

 

Skills :

The PhD-position's main objective is to qualify for work in research positions, a past experience related to research activities will be appreciated. The candidate must have the ability to work independently and to well organize himself. Good communication and writing skills in English are mandatory. The following tests can be used with following minimum scores:

  •  TOEFL: 600 (paper-based test), 92 (Internet-based test).
  •  IELTS: 7.0, with no section lower than 6.5 (only Academic IELTS test accepted).
  •  CAE/CPE: grade B or A.
26/03/2021
Partager via
Postuler
Fermer

Vous avez déjà un compte ?

Nouvel utilisateur ?

Besoin d'informations ?

Vous souhaitez recevoir nos infolettres ?

item1
Ils nous font confiance