Modern distribution systems with high penetration of distributed energy resources face multiple sources of uncertainty. This transforms the traditional Optimal Power Flow problem into a problem of sequential decision-making under uncertainty. In this framework, the solution concept takes the form of a \textit{policy}, i.e., a method of making dispatch decisions when presented with a real-time system state. Reasoning over the future uncertainty realization and the optimal online dispatch decisions is especially challenging when the number of resources increases and only a small dataset is available for the system’s random variables. In this paper, we present a data-driven distributed policy for making dispatch decisions online and under uncertainty. The proposed policy is guaranteed to satisfy the system’s constraints while experimentally shown to achieve a performance close to the optimal-in-hindsight solution, significantly outperforming state-of-the-art policies based on stochastic programming and plain machine-learning approaches.

In modern power systems, small distributed energy resources (DERs) are considered a valuable source of flexibility towards accommodating high penetration of Renewable Energy Sources (RES). In this paper we consider an economic dispatch problem for a community of DERs, where energy management decisions are made online and under uncertainty. We model multiple sources of uncertainty such as RES, wholesale electricity prices as well as the arrival times and energy needs of a set of Electric Vehicles. The economic dispatch problem is formulated as a multi-agent Markov Decision Process. The difficulties lie in the curse of dimensionality and in guaranteeing the satisfaction of constraints under uncertainty. A novel method, that combines duality theory and deep learning, is proposed to tackle these challenges. In particular, a Neural Network (NN) is trained to return the optimal dual variables of the economic dispatch problem. By training the NN on the dual problem instead of the primal, the number of output neurons is dramatically reduced, which enhances the performance and reliability of the NN. Finally, by treating the resulting dual variables as prices, each distributed agent can self-schedule, which guarantees the satisfaction of its constraints. As a result, our simulations show that the proposed scheme performs reliably and efficiently.

Along with high penetration of Electric Vehicles (EVs), charging stations are required to service a large amount of charging requests while accounting for constraints on the station’s peak electricity consumption. To this end, a charging station needs to make online charging scheduling decisions often under limited future information. An important challenge relates to the prioritization of EVs that have unknown valuations for different levels of charging services. In this paper, we take into consideration the inability of EV users to express these valuations in closed-form utility functions. We consider a paradigm where a menu of possible charging schedules and corresponding prices is generated online. By letting the EV users pick their most preferable menu option, the proposed algorithm commits on each EV’s charging completion time upon its arrival, achieves a near optimal total weighted charging completion time, and prevents the users from strategically misreporting their preferences, while offering a practical and implementable solution to the problem of EVs - charging station interaction.

In modern smart grids, the focus is increasingly shifted towards distributed energy resources and flexible electricity assets owned by prosumers. A system with high penetration of flexible prosumers, has a very large number of variables and constraints, while a lot of the information is local and non-observable. Decomposition methods and local problem solving is considered a promising approach for such settings, particularly when the implementation of a decomposition method features a market-based analogy, i.e. it can be implemented in a Transactive Energy fashion. In this paper we present an auction-theoretic scheme for a setting with non-convex prosumer models and resource constraints. The scheme is evaluated on a particular case study and its scalability and efficiency properties are tested and compared to an optimal benchmark solution. A game-theoretic analysis is made with respect to how an intelligent agent, that bids on behalf of a prosumer can try to strategize within the auction, in order to make itself better-off. Our simulations show that there is an alignment of incentives, i.e., when the prosumers try to strategize, they actually improve the auction's efficiency.

The proliferation of distributed energy assets necessitates the provision of flexibility to efficiently operate modern distribution systems. In this paper, we propose a flexibility market through which the DSO may acquire flexibility services from asset aggregators in order to maintain network voltages and currents within safe limits. A max-min fair formulation is proposed for the allocation of flexibility. Since the DSO is not aware of each aggregator's local flexibility costs, we show that strategic misreporting can lead to severe loss of efficiency. Using mechanism design theory, we provide a mechanism that makes it a payoff-maximizing strategy for each aggregator to make truthful bids to the flexibility market. While typical truthful mechanisms only work when the objective is the maximization of Social Welfare, the proposed mechanism lets the DSO achieve incentive compatibility and optimality for the the max-min fairness objective.