PhD#4 at Mines Paris in Data Science & Energy: "Flexibility-aware forecasting of local energy demand"

Title:  "Flexibility-aware forecasting of local energy demand"

Context and background:

Short-term forecasts of energy demand (electricity, heating/cooling, gas) at local level, ranging from a single household up to a group of buildings, a district, a node of the grid or a microgrid, become more and more necessary in the context of smart grids. Several new business models emerge, where the involved actors require such forecasts (together with information on the associated uncertainty) for a few minutes to days ahead in order to manage the corresponding energy systems. The objective may be auto-consumption (when generation and/or storage capabilities are available), provision of flexibility services to the grid, energy exchanges with other members of an energy community a.o. Although the literature on electricity forecasting at a national level is broad and the accuracy of existing models is very high, this is not the case for demand at local level. The existing models are not adapted to the ongoing transformation and digitalization of energy networks and the electrification of new usages that results in increasing demand by consumers (i.e. electric vehicles charging). Furthermore, the integration of high shares of renewables (wind, solar) is a challenge for grid operators. It becomes more and more necessary to adopt solutions that permit to adapt consumption to the variable renewable generation. For that, they deploy technologies that enable more flexibility of the consumption (e.g. load shifting, EV charging in zones with lower grid operational constraints, etc.). The activation of such flexibility options becomes more frequent and concerns more and more consumers. This induces spatio-temporal modifications in energy consumption patterns. As digital information has become central in the organization and dynamic behavior of both rural and urban territories, an efficient treatment of contextual information regarding the expected use of energy at the local scale is needed.

Scientific objectives:

The overall objective of the thesis is to develop a forecasting approach for local energy demand that is flexibility-aware, i.e. that can adapt to flexibility activations from energy networks (electricity, heat/cold, gas). The forecasting approach will be able to integrate contextual information relative to local situations, e.g. traffic, environmental conditions, weather conditions, news and social media.  

Methodology and expected results:  

The first step of this thesis will be to characterize and define typical patterns of energy demand at local level as a function of the aggregation of consumers (from single households up to tens or hundreds of consumers). Then existing load forecasting methods will be applied to estimate the reference performance that can be achieved. Several sources of input data, like smart meters data, weather forecasts, EV charging information etc. will be considered. The sensitivity of the existing methods on missing data will be studied together with the contribution of spatiotemporal information (i.e. measurements from neighbor smart homes with similar consumption profiles). It will be studied what are the requirements of the considered and future applications (use cases) for forecasting models that have to be taken into account in the design of such models. I.e. if they have to operate autonomously (i.e. at the “edge” or at automats with minimal maintenance and tuning requirements), if they have to be adaptive to structural changes of the consumption (i.e. addition of new usages), their robustness on missing data, their sensitivity on personal information a.o. In a next step, the contribution of NLP (natural language processing methods) methods will be assessed. Existing models developed at PERSEE for prediction at regional and national level will be considered. These methods will extract explanatory variables from text and audiovisual sources that are potentially informative for local energy consumption. Language data generated by AI will be integrated into the NLP methodology, which requires potentially AI detection and ethics analysis. Representative case studies will be provided by ongoing projects and will prioritize open-source datasets. A second step will be dedicated to the dynamic adaptation of the forecasting approach to massive penetration of flexibility in energy networks, e.g. from the electric distribution and transmission systems. This means to develop a prediction of the evolution of local demand before, during and after flexibility activations of different characteristics, durations and amplitudes.

The PERSEE Center is one of the 18 research centers of MINES Paris. Its field of expertise concerns New Energy Technologies and Renewable Energy Sources (RES). Its research strategy is based on a "micro/macro" approach ranging from (nano)materials to energy systems. It is built around three structuring themes: i) materials and components for energy, ii) sustainable energy conversion and storage processes and technologies, and iii) renewable energies and smart energy systems.

This late is developped by one of the three groups of the Center, ERSEI, which stands for “Renewable Energies and Smart Energy Systems”. The ERSEI group develops methods and tools allowing the optimal integration of decentralized sources, including RES, storage devices, electric vehicles, active demand and other technologies, in energy systems and electricity markets. The research activity of the group is divided into three main themes. The first is based on the development of advanced short-term forecasting methods for different applications in power systems (i.e. forecasting of RES production, demand, dynamic line rating, market quantities, etc.). The second concerns the control and predictive management of energy systems. The aim is to design innovative approaches to optimise the operation (from real-time to days ahead) of different types of systems (smart-homes, microgrids, virtual power plants, energy communities, hybrid RES/storage plants, distribution grids multi-energy systems a.o.) considering uncertainties. The third axis concern planning and prospective studies that aim to optimise the design of future energy systems, generate furture scenarios, optimise investements etc.  

The PERSEE Center is located within the scientific parc of Sophia-Antipolis, near the cities of Nice, Cannes and Antibes in the south of France. Its workforce is around 55 people.

PROFILE:

Engineer and / or Master of Science degree (candidates may apply prior to obtaining their master's degree. The PhD will start though after the degree is succesfully obtained).

Good level of general and scientific culture. Good analytical, synthesis, innovation and communication skills. Qualities of adaptability and creativity. Motivation for research activity. Coherent professional project. Skills in programming (eg R, Python, Julia,…).  A succesful candidate will have a solid background in three or more of the following competencies:

• applied mathematics, statistics and probabilities
• data science, machine learning, artificial intelligence
• energy forecasting
• power system management, integration of renewables
• optimization

Expected level in french : Good level 

Expected level in english : Proficiency

Desired starting date: as soon as possible in 2024. Duration 36 months. 

Full-time paid position.

For more information please contact Prof. Georges Kariniotakis and Dr Simon Camal (emails below)