Abstract

Power quality (PQ) issues, including weak grids, voltage transients, harmonics, notches, current imbalance, and voltage sags, are critical challenges in the textile industry. Even a brief power interruption can halt industrial processes, leading to substantial financial losses. This paper proposes a Model Predictive Control (MPC)-based Unified Power Quality Conditioner (UPQC) as a robust solution to mitigate these PQ disturbances in textile industry-integrated distribution grids. The proposed UPQC is designed to enhance voltage stability, suppress harmonics, regulate reactive power, and correct current imbalance, ensuring uninterrupted industrial operation. A key contribution of this work is the realistic modeling of a textile industry’s electrical network, replicating actual industry ratings to evaluate system performance. The proposed MPC-based UPQC is assessed through five case studies, addressing weak vs. strong grids, voltage transients, current imbalance, and voltage sags—the most significant PQ challenges in textile applications. Simulation results demonstrate that the UPQC significantly improves voltage profiles, reduces harmonic distortion, and effectively compensates for current imbalance. Compared to conventional Proportional-Integral (PI) controllers, the MPC-based UPQC exhibits superior performance in dynamic PQ disturbance mitigation and grid stabilization. These findings underscore the proposed system’s suitability for large-scale industrial deployment, offering a cost-effective and robust solution to enhance operational efficiency and grid reliability in the textile sector.

Abstract
The Iliad Digital Twins of the Ocean is a European Green Deal Project which aims at the development of an architecture and set of components, tools and services for the creation of digital twins of the ocean. The approach aims to support the emerging European Digital Twins of the Ocean (EDITO) initative with associated projects like EDITO Infra and EDITO Model lab and the overall Destination Earth (DestinE) initiative and also taking advantage of the evolving European Common Data Spaces including the Green Deal Data Space, the Copernicus Data Space and the EOSC cross domain Data Space. The paper presents the final version of the Iliad digital twin interoperability architecture based on four steps of a digital twin pipeline from Data Acquisition/Collection to Digital Twin Data Representation to Digital Twin Hybrid and Cognitive/AI Analytics Models and further to Digital Twin Visualisation and Control, which are presented together with associated Digital twin components and services.

Abstract

Abstract
Reconfigurable Intelligent Surfaces (RISs) are expected to play a pivotal role in future indoor ultra high data rate wireless communications as well as highly accurate three-dimensional localization and sensing, mainly due to their capability to provide flexible, cost- and power-efficient coverage extension, even under blockage conditions. However, when considering beyond millimeter wave frequencies where there exists GHz-level available bandwidth, realistic models of indoor RIS-parameterized channels verified by field-trial measurements are unavailable. In this article, we first present and characterize three RIS prototypes with unit cells of half-wavelength intercell spacing, which were optimized to offer a specific nonspecular reflection with 1-, 2-, and 3-bit phase quantization at 304 GHz. The designed static RISs were considered in an indoor channel measurement campaign carried out with a 304 GHz channel sounder. Channel measurements for two setups, one focusing on the transmitter-RIS-receiver path gain and the other on the angular spread of multipath components, are presented and compared with both state-of-the-art theoretical models as well as full-wave simulation results. The article is concluded with a list of challenges and research directions for RIS design and modeling of RIS-parameterized channels at THz frequencies.

Abstract
Deep learning models are widely used in multivariate time series forecasting, yet, they have high computational costs. One way to reduce this cost is by reducing data dimensionality, which involves removing unimportant or low importance information with the proper method. This work presents a study on an explainability feature selection framework composed of four methods (IMV-LSTM Tensor, LIME-LSTM, Average SHAP-LSTM, and Instance SHAP-LSTM) aimed at using the LSTM black-box model complexity to its favour, with the end goal of improving the error metrics and reducing the computational cost on a forecast task. To test the framework, three datasets with a total of 101 multivariate time series were used, with the explainability methods outperforming the baseline methods in most of the data, be it in error metrics or computation time for the LSTM model training.

Abstract