Abstract

Abstract
The increasing integration of distributed energy generation into the electrical grid has led to changes in the structure and organization of energy markets over the past years. Market trading has become increasingly demanding due to the different types of production profiles. A forecast of the total production of all assets is made to bid for energy. Whenever there are differences between the forecast and the actual produced energy, a deviation occurs, which is assigned to the agent responsible for its settlement. This article proposes the application of a linear regression algorithm supported by a clustering method to forecast energy production. Based on the historical production profile of the installations in each cluster, it is possible to predict the production pattern for a period with no available data, thus standardizing this data for other assets belonging to the same cluster. © 2025 IEEE.

Abstract
Polarization-based fiber sensors rely on the dynamics of the Stokes vector at the output of the optical fiber to probe stimuli that induce polarization variations. However, these sensors often suffer from limitations in sensitivity, precision, and reproducibility. In this work, we address these challenges by incorporating concepts from the Mueller matrix formalism to enhance the capabilities of such sensors. Specifically, we measure the Mueller matrix in the polarization basis that describes how the polarization evolves inside the optical fiber. Leveraging this formalism, we configure the system as a precise sensor to detect deformations along the fiber. By utilizing the Fisher Information framework, we significantly improve accuracy and resolution, enabling the detection of subtle perturbations with greater precision. This study introduces a novel approach for precise polarization control and advanced fiber-based sensing applications.

Abstract
[No abstract available]

Abstract
Capricious data streams, marked by random emergence and disappearance of features, are common in practical scenarios such as sensor networks. In existing research, they are mainly handled based on linear classifiers, feature correlation or ensemble of trees. There exist deficiencies such as limited learning capacity and high time cost. More importantly, the concept drift problem in them receives little attention. Therefore, drifting capricious data streams are focused on in this paper, and a new algorithm DCFHT (online learning from Drifting Capricious data streams with Flexible Hoeffding Tree) is proposed based on a single Hoeffding tree. DCFHT can achieve non-linear modeling and adaptation to drifts. First, DCFHT dynamically reuses and restructures the tree. The reusable information includes the tree structure and the information stored in each node. The restructuring process ensures that the Hoeffding tree dynamically aligns with the latest universal feature space. Second, DCFHT adapts to drifts in an informed way. When a drift is detected, DCFHT starts training a backup learner until it reaches the ability to replace the primary learner. Various experiments on 22 public and 15 synthetic datasets show that it is not only more accurate, but also maintains relatively low runtime on capricious data streams.

Abstract
Capricious data streams make no assumptions on feature space dynamics and are mainly handled based on feature correlation, linear classifier or ensemble of trees. There exist deficiencies such as limited learning capacity, high time cost and low interpretability. To enhance effectiveness and efficiency, capricious data streams are handled through a single tree in this paper, and the proposed algorithm is named OCFHT (Online learning from Capricious data streams with Flexible Hoeffding Tree). OCFHT does not rely on the correlation pattern among features and can achieve non-linear modeling. Its performance is verified by various experiments on 18 public datasets, showing that it is not only more accurate than state-of-the-art algorithms, but also runs faster.