The proliferation of distributed energy resources (DERs) and increasing electrification of the consumer energy space are progressively enabling end-consumers to play an active role in the smart grid sphere. Indeed, rather than merely consuming energy, active consumers (known as prosumers) have the ability to dynamically produce, store, and consume energy, enabling them to take up an unprecedented proactive role on various levels of the electricity value chain. Individually, or through the intermediary of aggregators, prosumers can actively make operational and investment decisions allowing them to reap financial benefits. This can be carried out through, e.g., participation in local and wholesale energy markets and involvement in collective energy communities, or through leveraging their flexibility by providing services to system operators. In fact, beyond direct control – using, e.g., conditional connection agreements or implemented control systems – flexibility service provision (i.e., possible volumetric adjustments and temporal shifts in load and generation) can be carried out by means of explicit participation in flexibility markets or implicit reactions to price signals and dynamic tariffs.

Hence, emerging distribution systems are rapidly evolving into interactive systems relying on distributed decisions taken autonomously by their various constituents; primarily by prosumers, aggregators, and operators. These constituents can naturally possess different individual objectives which can be at times conflicting – as, e.g., in competitive demand response and market settings – and at other times aligned – as, e.g., in collective investment agreements. In this distributed and interactive paradigm, the decision of one constituent has a direct effect not only on its own objective but also on the objective of other constituents as well as on the operation of the grid. As such, achieving the full potential of demand-side flexibility while ensuring the secure operation of the grid is contingent upon a complete understanding of the interactive decision making processes of the prosumers and, in general, of all grid constituents. To this end, beyond classical optimization methods, which traditionally consider a single decision maker, this work plans to build and advance game-theoretic models capable of capturing the underlying multi-player distributed decision making processes governing the participants’ behavior in demand side management and flexibility provision schemes.

However, game-theoretic models traditionally assume players to be fully rational. This implies, among different properties, that agents (e.g., prosumers) would always choose optimal decisions, have an objective perception of their environment and of the underlying uncertainties, and an objective perception of theirs and others’ decisions. However, as shown in different empirical observations, when faced with uncertainty and complexity, humans can deviate from those full rationality assumptions. Hence, to properly model, predict, and analyze prosumers’ decisions, this work aims to not only consider their interactive decisions making processes, but also to allow the incorporation of their potential bounded rationality (i.e. deviations from the full rationality assumptions). This, hence, enables capturing the subjective preferences, perceptions, and valuations of prosumers in this interactive decision making paradigm. This, as a result, allows forming a projection of their decision making processes capable to more closely reflect the practical decisions that prosumers make when reacting to price signals, participating in a flexibility market, interacting with aggregators/operators, or deciding to join a local energy community or collective investment agreement, among others.

The primary objective of this work is, hence, to identify and analyze the way in which strategic prosumers and aggregators make operational decisions in an inter-dependent setting and capture their effect on the operation of the grid and the provision of flexibility. This, as a result, enables a better design of consumer-centric demand side management mechanisms able to achieve the full potential of demand-side flexibility by accurately anticipating consumer-side reactions, incentives, preferences, and strategic decisions. Mathematically, this will be achieved by the development of advanced game-theoretic models, which explicitly consider bounded rationality principles (relying on well-established decision-theoretic and behavioral game-theoretic concepts).

Approach and Methodology:

The provisional steps to be taken within this PhD can be summarized as follows:

1. Conducting a literature review on the state of the art of mathematical models for demand side management schemes and the preliminary applications of game theory with bounded rationality within the scope of energy systems and demand side management, identifying available gaps and defining open research questions.
2. Identifying prosumers’ optimal reactions (investment and operational) to price signals and dynamic tariffs under settings where prosumers are price setters (to various degrees) considering prosumers’ bounded rationality 
3. Design of price signals, tariffs, and incentives by system operators for demand side management taking into consideration prosumers’ and aggregators (strategic) reactions while considering possible information asymmetry among the involved constituents.
4. Design of stable local energy communities and/or collective energy investment schemes taking into consideration the strategic positions of participants and their (possibly subjective) preferences. This step will also consider their ability to participate in demand response schemes or to provide flexibility as a collective unit.
5. Tentative: analyzing the links between strategic prosumers, aggregators, and operators while considering the strategic behavior and preferences of each constituent.
6. In each of the research questions, the overarching goal is to draw conclusions and tangible recommendations to the domain of application using in-depth mathematical models, numerical results, and case analyses.

Interested candidates are welcome to apply through the application link including:

1. A copy of the candidate’s CV
2. A motivation letter highlighting the candidate’s research interests
3. The candidate’s transcripts of grades (for all received degrees starting at bachelor’s level)
4. The names and contact information of up to three references (preferably included in the CV).

For more information, please contact dr. Anibal Sanjab: anibal.sanjab@vito.be

Expected start date of the selected candidate: May/June 2023.