We present EUV-OVL-SYN, an open and fully reproducible synthetic benchmark dataset for EUV lithography. Overlay error, the misalignment between consecutive patterning layers, is a primary yield-limiting factor at advanced nodes, yet the absence of publicly available, physically realistic datasets impedes systematic progress in data-driven correction research. EUV-OVL-SYN is built on documented scanner physics and replicates the signal hierarchy of real production data. The dataset captures three distinct levels of physical structure: a dominant inter-field Brink/van den Brink polynomial fingerprint (cross-wafer R2 ≥ 0.93); four lot-level temporal drift scenarios modeling linear scanner heating, reticle alignment drift, multi-parameter process drift, and chiller-cycling oscillation; and a structured nonlinear intra-field residual analytically verified to lie outside the Brink polynomial basis. The dataset consists of 4 lots of 50 wafers each, measured at 420 metrology targets per wafer (284 000 measurements in total), formatted identically to industrial automatic process control exports for direct compatibility with existing correction pipelines. The nonlinear residual (1σ ≈ 1.1–1.4 nm per axis) is well above the 0.15 nm measurement noise floor yet non-trivially masked by the dominant Brink fingerprint, providing a detectable but genuinely challenging correction target. Full ground-truth decomposition is provided for every signal component, permitting exact, method-agnostic evaluation of any correction approach. The complete dataset is generated by a single reproducible Python script and released under an open license.

Edge AI deployment demands neural architectures that are simultaneously accurate, computationally efficient, and hardware-deployable - a challenge addressed by hardware-aware Neural Architecture Search (NAS). While recent works incorporate quantization directly into the NAS loop, these approaches expand search complexity and tightly couple architecture and quantization design. The simpler post-search quantization strategy has received little analytical attention: the effects of Post-Training Quantization (PTQ) on the NAS-discovered Pareto structure remain uncharacterised, and no framework combines quantized architecture mapping onto reconfigurable accelerators with automated hardware exploration. This paper addresses both gaps. First, a three-stage pipeline is proposed: a hardware-agnostic Pareto rank surrogate frontend on NAS-Bench-201, a quantization bridge with Pareto-aware filtering and feedback control, and an evolutionary Domain Space Exploration (DSE) backend on CGRA4ML for optimal hardware mapping. Second, an empirical study characterises how INT4 PTQ perturbs the NAS-Bench-201 Pareto space through formal stability metrics on ground-truth data for all 15,625 architectures, and demonstrates that an FP32 zero-shot surrogate outperforms a dedicated INT4-trained surrogate in Pareto space coverage across two standard search strategies.

This paper introduces a Model Context Protocol (MCP)-based framework that enables simplified management and orchestration of UERANSIM containerized deployments for 5G network simulation. Leveraging the standardized MCP interface, the proposed architecture provides specialized tools that handle the complete lifecycle of simulated gNodeB (gNB) and User Equipment (UE) containers, by enabling users to express intents to generative AI models. The framework implements dynamic configuration management with automatic IP address detection and runtime YAML updates, eliminating manual intervention while maintaining precise control over network parameters. By abstracting complex operations behind user-friendly AI chats, the system reduces operational complexity and enables rapid deployment of 5G test environments. Experimental evaluations demonstrate the framework's effectiveness through a practical deployment scenario integrating Open5GS core network with UERANSIM components, validating both functional correctness and operational efficiency. Performance analysis reveals significant improvements in deployment time and configuration accuracy compared to traditional manual approaches, while maintaining flexibility for diverse testing scenarios.

The ever-increasing need for energy-efficient implementation of AI algorithms has driven the research community towards the development of many hardware architectures and frameworks for AI. A lot of work has been presented around FPGAs, while more sophisticated architectures like CGRAs have also been at the center. However, AI ecosystems are isolated and fragmented, with no standardized way to compare different frameworks with detailed Power-Performance-Area (PPA) analysis. This paper bridges the gap by presenting a unified, fully open-source hardware-aware AI acceleration pipeline that enables seamless deployment of neural networks on both FPGA and CGRA architectures. Built around the Brevitas quantization framework, it supports two distinct backend flows: FINN for high-performance dataflow accelerators and CGRA4ML for low-power coarse-grained reconfigurable designs. To facilitate this, a model translation layer from QONNX to QKeras is also introduced. To demonstrate its effectiveness, we use an autoencoder model for anomaly detection in wind turbines. We deploy our accelerated models on the AMD's ZCU104 and benchmark it against a Raspberry Pi. Evaluation on a realistic cyber-physical testbed shows that the hardware-accelerated solutions achieve substantial performance and energy-efficiency gains-up to 10x and 37x faster inference per flow and over 11x higher efficiency-while maintaining acceptable reconstruction accuracy.

Edge intelligence, i.e., the execution of Machine Learning (ML) algorithms in computing resources at the edge, provides unprecedented benefits for applications in different verticals regarding data privacy, bandwidth, costs, and latency. Non-Intrusive Load Monitoring (NILM) is an application in the smart grid technology domain that could benefit from the advancements in edge intelligence to ensure consumer data privacy and decrease implementation costs. This paper proposes a federated learning-based transformer architecture for the NILM of energy-intensive residential devices, i.e., Heat Pumps (HP). We evaluate the architecture on an open-source dataset, showcasing that the performance does not deteriorate significantly compared to the centralized and is robust against the distribution shifts between the training and inference datasets and the increasing heterogeneity level between clients in the training data.

This paper introduces a novel framework that integrates agentic Artificial Intelligence (AI) with Intent-Based Networks (IBN) to enable autonomous management, configuration, and optimization of mobile network services and resources. Leveraging the advanced reasoning and natural language processing capabilities of an Large Language Model (LLM), the proposed architecture translates high-level user intents into precise network actions, facilitating user-friendly and scalable network orchestration. The framework employs a distributed multi-agent system, where specialized agents collaborate to decompose user intents, provide computational infrastructure, and deploy services using industry-standard Infrastructure-as-Code (IaC) tools. By supporting natural language interactions, the system reduces operational complexity and enhances accessibility for users with varying technical expertise. Experimental evaluations demonstrate significant improvements in task completion rates, response accuracy, and operational efficiency compared to traditional manual methods, particularly for complex network management tasks. In essence, this work creates an intelligent network orchestration framework that adapts to user needs by automatically configuring network and computing resources while operating with minimal human intervention.

An accurate prediction of imbalance prices is crucial for making well-informed decisions within short-term energy markets. This study proposes a two-stage probabilistic framework for the prediction of system imbalance and imbalance price. In the first stage, the quantiles of system imbalance are predicted employing the Quantile Regression Forest algorithm. In the second stage, the quantiles are leveraged to predict the imbalance prices, given the observed correlation between the system imbalance and the imbalance prices. The corresponding imbalance prices for each 15-minute interval are then predicted utilizing a novel methodology that combines Gradient Boosting with Markov switching models. The proposed methodology was evaluated using real data from the Greek balancing market. Comparative analysis with alternative models including a persistence model, Quantile Linear Regression, Long Short-Term Memory, and Quantile Regression Forest reveals that the proposed approach consistently outperforms the rest in terms of forecast accuracy. Specifically, the results demonstrate that the proposed approach outperforms all alternative methods, achieving an 9% improvement in nMAE and a 10.8% reduction in CRPS with respect to the second best method (Quantile Regression Forest). The findings suggest that this approach can significantly improve the decision-making process for market participants, helping mitigate the financial risks associated with imbalance prices.

In the photolithography process of integrated circuit (IC) manufacturing, overlay (OL) control is a key factor for successful exposure. Overlay control is achieved by creating models that can estimate the expected overlay error, so that this can be corrected before the wavefront reaches the wafer surface. Such models consist of basis functions, influenced by application-controllable variables such as process settings, tool characteristics, and field-related factors with their corresponding model parameters. The process of tuning the model parameters involves several time-consuming sensor measurements on markers distributed across the wafer surface and significantly impacts the throughput performance of the exposure system. For this reason, a strategic selection of wafer markers is necessary. In this paper, we propose a methodology to improve the overlay modeling process by exploiting Surface Reconstruction (SR). SR is used as an intermediate step, during the parameter estimation process, to generate additional data from a strategically selected set of markers that are spatially uniform and provide maximum information gain. The proposed method reconstructs the wafer surface by incorporating spatially interpolated estimates derived from the physical insights of existing measurements. This augmented data set, comprised of measured and synthetic overlay data, serves as a comprehensive input for the parameters’ tuning process leading to more accurate overlay modeling. The proposed method is evaluated using real-industry data from a semiconductor process of 300mm diameter wafers. The results demonstrate a significant reduction in the overlay residuals in both x and y directions.

This study introduces a simulation-based method for generating realistic overlay error patterns in photolithography, leveraging Zernike polynomial representations to model optical aberrations alongside spatially varying intra-field distortions. The framework accounts for both global wafer-level effects and localized field-dependent deviations, capturing key contributors to overlay inaccuracies such as lens misalignments, mechanical instabilities, and thermal-induced deformations. Per measurement marker, displacement vectors are synthesized to reflect these influences, with their magnitudes modulated by position within the exposure field. To emulate real-world process noise, the model incorporates heavy-tailed stochastic perturbations. A key outcome of this work is a curated dataset comprising four distinct lithography layers, each representing a different process scenario with unique tool characteristics and error signatures. For each layer, six wafers are simulated, capturing a range of interfield and intra-field variation patterns. This dataset serves as a valuable resource for validating overlay correction techniques, assessing metrology system performance, and training datadriven models for overlay prediction and diagnostics.

The decentralization of the energy grid, coupled with advanced communication technologies (LoRaWan, LTE -4G/5G- etc.), improves its operation and enhances its cyber-physical characteristics, thereby increasing the need for cyber resilience. This paper reviews the benefits of an IoT-based Local Energy Market (LEM), located at the grid's edge, suitable to capitalize on the reduced energy costs and increased flexibility offered by the heavy penetration of Distributed Energy Resources (DERs). LEM utilizes IoT-enabled devices such as DERs, Electric Vehicles (EVs), and smart meters to optimize energy services within user preferences and grid constraints. However, the increase in IoT devices introduces vulnerabilities to cyber-attacks, potentially compromising the LEM's operation and financial stability. We analyze the effects of cyber-risks through three attack scenarios involving False Data Injection on smart meters, Behind The Meter appliances, and photovoltaics, demonstrating possible outcomes like voltage violations and increased operational costs. To improve cyber-resilience, we propose a nodal Distribution Locational Marginal Pricing (DLMP) market architecture that addresses voltage and congestion issues and uses abnormal nodal prices as early attack indicators for swift intrusion detection. The IoT-based nodal market approach secures a robust local energy market structure, highlighting the essential role of cyber-security in modern power grids, and offers uninterrupted operations and enhanced resilience in decentralized energy systems.

Advances in the field of edge computing and the emergence of edge-cloud has enabled the monitoring and management of one of the most complex existing systems, the electric grid, which is undergoing a decentralized and ‘smart transition’ towards ‘smart grid’. Advanced Monitoring Infrastructure (AMI) and sensing devices (PMUs, RTUs etc.) combined with advanced communications networking allow the monitoring of the grid parameters, while edge computing capabilities (including hardware acceleration engines, reconfigurable hardware, etc.) enable real-time Transient State Estimation (TSE) of the grid nodes and therefore enabling other smart grid applications, such as preventive maintenance, control and data analytics. In this work, we focus on the processing requirements of TSE and we present an edge-cloud architecture for efficient TSE application. The architecture leverages low-cost hardware acceleration engines at the edge (devices enhanced with FPGA resources) in order to solve the TSE equations effectively and in tight time thresholds, while reducing the transmission of traffic towards the cloud and thus saving bandwidth for more sophisticated smart grid applications.

The concept of Network-as-a-Service is based on the automatic deployment and dynamic reconfiguration of next-generation networks in order to meet the needs of the respective stakeholders. Towards that path, orchestrating both telco and computational resources across the different network domains, imposes a great challenge even for the network experts. This paper explores an intent-based management framework designed to simplify this procedure. By receiving high-level network requirements through descriptive text and supplementary images representing the desired outcome, the framework ingests and translates them into network configuration files. These files facilitate the seamless deployment of a open-source 5G Core, without any human intervention, tailored to the stakeholders’ needs. The proposed framework achieves this by employing multimodal Generative AI models, particularly Large Language Models, to bridge the gap between the user’s intent and network configuration.

As the energy sector undergoes a profound digital transformation, demanding a fusion of resilience, efficiency, and cutting-edge technology, 5G technology emerges as a beacon, promising not just enhanced connectivity but a holistic transformation of how we conceive and manage energy infrastructure. This work aims to provide an in-depth exploration for experts in the energy domain, unraveling the innovative aspects of 5G through the demonstration of important achievements and results of the Horizon 2020 5G Infrastructure Public Private Partnership (5G-PPP) Phase-3 5G-VICTORI’s Project and its trial results on the impact of 5G technology in an energy facility located in the city of Patras, Greece.

The emerging concept of delivering Network-as-a-Service (NaaS) foresees the deployment and reconfiguration of the next-generation networks, such as 6G, in a dynamic and elastic manner, tailored to the respective stakeholder’s intention. Taking this into account, the efficient management and orchestration of both telecommunication and computational resources across the network domains, i.e. access, transport and core presents a considerable challenge, even for network experts. To tackle this complexity, this paper explores the implementation of an intent-based management framework. The framework receives a high-level description of the desired network capabilities along with supplementary files, e.g. deployment descriptors, and translates them into configuration files consumable by the network itself. In order to achieve this, the paper establishes a translation pipeline that leverages the employment of emerging multimodal generative artificial intelligence (GenAI) models, specifically Large Language Models (LLMs), and open industry-ready standard templates. The adoption of those two emerging technologies offers high dynamicity on the interpretation process of the user’s intent, while ensuring that its outcome is compatible with every orchestrator or next-generation Operating Support System (Next-gen OSS) that adheres to those standards.

The extreme ultraviolet (EUV) photolithography process is a cornerstone of semiconductor manufacturing and operates under demanding precision standards realized via nanometer-level overlay (OVL) error modeling. This procedure allows the machine to anticipate and correct OVL errors before impacting the wafer, thereby facilitating near-optimal image exposure while simultaneously minimizing the overall OVL error. Such models are usually high dimensional and exhibit rigorous statistical phenomena such as collinearities that play a crucial role in the process of tuning their parameters. Ordinary least squares (OLS) is the most widely used method for parameters tuning of overlay models, but in most cases it fails to compensate for such phenomena. In this paper, we propose the usage of ridge regression, a widely known machine learning (ML) algorithm especially suitable for datasets that exhibit high multicollinearity. The proposed method was applied in perturbed data from a 300 mm wafer fab, and the results show reduced residuals when ridge regression is applied instead of OLS.

The energy sector is undergoing a transformative shift, driven by advancements in Distributed Energy Resources (DERs), the digitization of the energy supply chain and decarbonization policy objectives across the world. This paradigm shift has led to the emergence of Local Energy Markets (LEMs), which enable small-scale prosumers to actively participate in the energy market, trade power, and leverage their flexible resources. To ensure the success and acceptance of LEMs, this paper proposes a cooperative game-theoretic approach that fosters prosumer engagement and fair profit allocation. We utilize prospect theory from behavioural economics to examine the decision-making process of prosumers and incorporate their preferences for changes in wealth status. By adopting a cooperative game structure, prosumers can pool their resources, reduce transaction costs, and enhance data utilization. The paper introduces a novel pricing algorithm inspired by prospect theory that incentivizes prosumer participation and accounts for the uncertainty involved in LEM operations. Additionally, a computationally efficient method for profit allocation based on the variation of the Shapley value is proposed to ensure scheme stability. A use case evaluation is conducted on a real-world low-voltage network, demonstrating the effectiveness of the proposed approach in terms of economic efficiency and market characteristics. The results highlight the benefits of the consumer-centric LEM, including improved local trading dynamics, fair profit distribution, and enhanced grid stability. Overall, this research contributes to the design and development of LEMs that prioritize prosumer engagement, community cooperation, financial inclusion and democratization of the energy market.

In the realm of power systems engineering, heightened attention has been directed towards the concept of Transient State Estimation (TSE) due to its potential for enhancing system analysis within smart grids. This paper presents an efficient approach to address challenges in power grid monitoring by introducing adaptive architectures for real-time power system estimation, focusing on the pivotal role of TSE in enhancing grid resilience and efficiency. Leveraging the Vitis High-level Synthesis (HLS) tool and the Xilinx Zynq UltraScale+ MPSoC ZCU104 platform, the study proposes four distinct FPGA architectures. These architectures are systematically optimized for latency and resource utilization, emphasizing the delicate balance between these factors. The achieved goal through the paper is to demonstrate the feasibility of supporting real-time TSE for a representative network via an FPGA implementation.

In photolithography process, nanometer level precise, wavefront aberration models enable the machine to be able to meet the overlay (OVL) drift and critical dimension (CD) specifications. Software control algorithms take as input these models and correct any expected wavefront imperfections before reaching the wafer. In such way a near optimal image is exposed on the wafer surface. Optimizing the parameters of these models though, involves several time costly sensor measurements which reduce the throughput performance, in terms of exposed wafers per hour, of the machine. In that case, photolithography machines come across the trade-off between throughput and quality. Therefore one of the most common Optimal Experimental Design (OED) problems in photolithography machines (and not only) is how to choose the minimum amount of sensor measurements that will provide the maximum amount of information. Additionally, each sensor measurement corresponds to a point on the wafer surface and therefore we must measure uniformly around the wafer surface as well. In order to solve this problem, we propose a Sensor Marks Selection Algorithm which exploits Genetic Algorithms. The proposed solution first selects a pool of points that qualify as candidates to be selected in order to meet the uniformity constraint. Then, the point that provides the maximum amount of information, quantified by the Fisher based criteria of G, D and A-Optimality, is selected and added to the measurement scheme. This process though is considered "greedy", and for this reason Genetic Algorithms (GA) are exploited to further improve the solution. By repeating in parallel the "greedy" part several times we get an initial population that will be the input to our GA. This meta-heuristic approach outperforms the "greedy" approach significantly. The proposed solution is applied in a real life semiconductors industry use case and achieves interesting industry and academical results as well.

The high penetration of distributed energy resources, especially weather-dependent sources, even at the edge of the distribution grids, has increased the power system uncertainties and drastically shifted the operational status quo for the system operators. For the operators to ensure the uninterrupted electricity supply of the end-consumers, the fast and accurate response to fault events is of critical importance. This paper proposes a data-driven fault location identification and types classification application based on the continuous wavelet transformation and convolutional neural networks optimally configured through Bayesian optimization. This application leverages the proliferation of high-resolution measurement devices in distribution networks. It can locate the exact position of the short-circuit faults and classify them into eleven different types. Its intrinsic models grasp the spatial characteristics and the converted in frequency domain temporal ones of the three-phase voltage and current timeseries measurements stemming from the field devices, thus increasing the operators’ visibility of their networks in real-time. We conduct simulations through synthetic data, which we provide in an open-source repository, that replicate a wide range of fault occurrence scenarios with eleven different types, with the resistance ranging from 50Ω to 2kΩ and with duration from 20ms to approximately 2s, under noise conditions injected by devices and load variability. The results showcase the efficacy of the proposed method reaching an accuracy of 91.4% for fault detection, 93.77% for correct branch identification, 94.93% for fault type classification, and RMSE value of 2.45% for location calculation.

Electricity price forecasting (EPF) has become an essential part of decision-making for energy companies to participate in power markets. As the energy mix becomes more uncertain and stochastic, this process has also become important for industrial companies, as their production schedules are greatly impacted by energy costs. Although various approaches have been tested with varying degrees of success, this study focuses on predicting day-ahead market (DAM) prices in different European markets and how this directly affects the optimal production scheduling for various industrial loads. We propose a fuzzy-based architecture that incorporates the results of two forecasting algorithms; a random forest (RF) and a long short-term memory (LSTM). To enhance the accuracy of the proposed model for a specific country, electricity market data from neighboring countries are also included. The developed DAM price forecaster can then be utilized by energy-intensive industries to optimize their production processes to reduce energy costs and improve energy-efficiency. Specifically, the tool is important for industries with multi-site production facilities in neighboring countries, which could reschedule the production processes depending on the forecasted electricity market price.