José Paulos

Abstract

Abstract
The European Union's Artificial Intelligence (AI) Act defines robustness, resilience, and security requirements for high-risk sectors but lacks detailed methodologies for assessment. This paper introduces a novel framework for quantitatively evaluating the robustness and resilience of reinforcement learning agents in congestion management. Using the AI-friendly digital environment Grid2Op, perturbation agents simulate natural and adversarial disruptions by perturbing the input of AI systems without altering the actual state of the environment, enabling the assessment of AI performance under various scenarios. Robustness is measured through stability and reward impact metrics, while resilience quantifies recovery from performance degradation. The results demonstrate the framework's effectiveness in identifying vulnerabilities and improving AI robustness and resilience for critical applications.

Abstract
With the growing need for AI-driven solutions in power grid management, this work addresses the challenge of creating realistic synthetic operating scenarios essential for developing, testing, and validating AI-based decision-making systems. It uses spatial-temporal noise functions, predefined patterns, and optimal power flow to model renewable energy and conventional power plant generation, load, and losses. Quantitative and visual key performance indicators are proposed to evaluate the quality of the generated operating scenarios, and the validation highlights the framework's ability to emulate diverse and practical operating scenarios, bridging gaps in AI-driven power system research and real-world applications.

Abstract
Distributed energy resources (DERs) offer untapped potential to meet the flexibility needs of power systems with a high share of non-dispatchable renewable generation, and local flexibility markets (LFMs) can be effective mechanisms for procuring it. In LFMs, energy communities (ECs) can aggregate and offer flexibility from their members' DERs to other parties. However, since flexibility prices are only known after markets clear, flexibility bidding curves can be used to deal with this price uncertainty. Building on previous work by the authors, this paper employs a two-stage methodology to calculate flexibility bids for an EC participating in an LFM, including not only batteries and photovoltaic panels, but also cross-sector (CS) flexible assets like thermal loads and electric vehicles (EVs) to assess their impact. In Stage 1, the EC manager minimizes the energy bill without flexibility to define its baseline. In Stage 2, it computes the optimal flexibility to be offered for each flexibility price to build the flexibility bidding curve. Case examples allow to assess the impact of CS flexible assets on the final flexibility offered.

Abstract
Planning Energy communities (ECs) requires engaging members, designing business models and governance rules, and sizing distributed energy resources (DERs) for a cost-effective investment. Meanwhile, the growing share of non-dispatchable renewable generation demands more flexible energy systems. Local flexibility markets (LFMs) are emerging as effective mechanisms to procure this flexibility, granting ECs a new revenue stream. Since sizing with flexibility becomes a highly complex problem, we propose a 2-stage methodology for estimating DERs size in an EC with collective self-consumption, flexibility provision and cross-sector (CS) assets such as thermal loads and electric vehicles (EVs). The first stage computes the optimal DER capacities to be installed for each member without flexibility provision. The second stage departs from the first stage capacities to assess how to modify the initial capacities to profit from providing flexibility. The impact of data clustering and flexibility provision are assessed through a case study.