Laboratory: Unité Mixte de Recherche Espace-Dev, Energy Transition Team. Job type: Postdoctoral in automatic control and systems modelling applied to Smartgrid. Contract: The position is a fixed-term contract. Candidate Criteria The recruited candidate must hold a PhD in a field such as Electricity, Automation, and Energy Engineering applications for smart energy grids (Smartgrid and smart Microgrid) with strong skills on Machine Learning. He/she will have to demonstrate relevant and original research contributions in the required field, to show capabilities in both research methodology aspects and artificial intelligence applications to the Smartgrid. The candidate must have an excellent publication record. Recruitment will be based in French Guiana and will also involve occasional assignments at regional partner sites, as well as in France and Brazil. Key Responsibilities Description Context The stabilization and management of electrical grids, with intermittent sources, involves the use of storage systems (particularly hydrogen), and requires a deep knowledge of their various components' physics in order to contribute to the scientific problems resolution associated with the development of Smartgrid. Objectives Accordingly, the postdoc's mission is to develop an advanced energy management strategy based on the prediction of photovoltaic production, the cost of electricity, the stability of the power grid, and the state of charge of storage (batteries and hydrogen). The main objective is to produce hydrogen from photovoltaic panels production and minimize the number of START/STOP cycles of the electrolyzer. The strategy must be capable of maximizing panel output and electrolyzer lifetime as a function of the number of START/STOP cycles (minimum possible). Optimization methods and artificial intelligence will be used for this predictive energy management strategy. Key duties The postdoctoral aims to develop expertise in three areas: (i) modeling and simulation of the multi-source, multi-storage system; (ii) development of an innovative strategy for predictive energy management in a multi-source, multi-storage park; (iii) testing and validation of the strategy on a prototype in collaboration with other partners. In collaboration with two research laboratories, two pilot systems, one at UNICAMP in Brazil and the other at INESC-ID in Portugal, may enable the implementation of the developed energy management system strategy probably on experimental sites in French Guiana. In addition, implementation of the strategy described above requires power production forecasts, on a daily time horizon and measures that influence the system dynamics (e.g., state of charge, electricity cost, production stability). Machine Learning algorithms for time series forecasting will be developed by the post-doc. The recruited candidate may also participate in the laboratory research projects writing. Application Files To submit : - Detailed CV with biography and a list of the scientific publications, - Cover letter along with a synthetic presentation of the research works, - The dissertation summary and the defense report, - Reference letters are a plus. Exclusively, Applications must be sent by email in electronic format (.pdf files) to: - Detailed CV with biography and a list of the scientific publications, - Cover letter along with a synthetic presentation of the research works, - The dissertation summary and the defense report, - Reference letters are a plus.Exclusively, The President of the University of French Guiana, - recrutement@univ-guyane.fr - Application Deadline: 23/08/2025 Scientific contacts: abdellatif.elmouatamid@univ-guyane.fr - antoine.primerose@univ-guyane.fr