Abstract
Solar photovoltaic technology has grown significantly as a renewable energy, with unmanned aerial vehicles equipped with thermal infrared cameras effectively inspecting solar panels. However, long-distance capture and low-resolution infrared cameras make the targets small, complicating feature extraction. Additionally, the large number of normal photovoltaic modules results in a significant imbalance in the dataset. Furthermore, limited computing resources on unmanned aerial vehicles further challenge real-time fault classification. These factors limit the performance of current fault classification systems for solar panels. The multi-scale and multi-branch Reparameterization of convolutional neural networks can improve model performance while reducing computational demands at the deployment stage, making them suitable for practical applications. This study proposes an efficient framework based on reparameterization for infrared solar panel fault classification. We propose a Proportional Balanced Weight asymmetric loss function to address the class imbalance and employ multi-branch, multi-scale convolutional kernels for extracting tiny features from low-resolution images. The designed models were trained with Exponential Moving Average for better performance and reparameterized for efficient deployment. We evaluated the designed models using the Infrared Solar Module dataset. The proposed framework achieved an accuracy of 83.8% for the 12-Class classification task and 74.0% for the 11-Class task, both without data augmentation to enhance generalization. The accuracy improvements of up to 16.4% and F1-Score gains of up to 18.7%. Additionally, we achieved an inference speed that is 3.4 times faster than the training speed, while maintaining high fault classification performance. © 2025 Elsevier Ltd

Abstract
The increasing occurrence of extreme weather events has severely compromised the resilience of power distribution systems, resulting in widespread outages and substantial economic losses. This paper proposes a novel solution to enhance the resilience of distribution networks without the need for significant infrastructure upgrades. We introduce a bilevel optimization framework that integrates Demand Response Programs (DRPs) to strategically manage electricity consumption and mitigate the impact of system disruptions. The approach fosters collaboration between Distribution System Operators (DSOs) and Demand Response Aggregators (DRAs), optimizing both operational resilience and economic efficiency. To solve the bilevel problem, we employ a Mathematical Program with Equilibrium Constraints (MPEC), transforming the bilevel model into a single- level problem by utilizing the Karush-Kuhn-Tucker (KKT) conditions. This method is applicable when the lower-level problem is convex with linear constraints. The model also incorporates Long Short-Term Memory (LSTM) neural networks for wind generation forecasting, enhancing decision-making precision. Furthermore, we conduct multiple case studies under varying severities of incidents to evaluate the method's effectiveness. Simulations performed on the IEEE 33-bus test system using GAMS and Python validate that the proposed method not only improves system resilience but also encourages active consumer participation, making it a robust solution for modern smart grid applications. The simulation results show that by performing DRP to handle the contingencies in a high-impact incident, the resilience of the system can be improved by 5.3%.

Abstract
In real-world scenarios, a wide variety of datasets contain inconsistencies. One example of such inconsistency is missing data (MD), which refers to the absence of information in one or more variables. Missing imputation strategies emerged as a possible solution for addressing this problem, which can replace the missing values based on mean, median, or Machine Learning (ML) techniques. The performance of such strategies depends on multiple factors. One factor that influences the missing value imputation (MVI) methods is the presence of noisy instances, described as anything that obscures the relationship between the features of an instance and its class, having an adversarial effect. However, the interaction between MD and noisy instances has received little attention in the literature. This work fills this gap by investigating missing and noisy data interplay. Our experimental setup begins with generating missingness under the Missing Not at Random (MNAR) mechanism in a multivariate scenario and performing imputation using seven state-of-the-art MVI methods. Our methodology involves applying a noise filter before performing the imputation task and evaluating the quality of the imputation directly. Additionally, we measure the classification performance with the new estimates. This approach is applied to both synthetic data and 11 real-world datasets. The effects of noise filtering before imputation are evaluated. The results show that noise preprocessing before the imputation task improves the imputation quality and the classification performance for imputed datasets.

Abstract
This paper contributes to developing a Method for Creating Persona Templates (MCPT), addressing a significant gap in user-centred design methodologies. Utilising qualitative data collection and analysis techniques, MCPT offers a systematic approach to developing robust and context-oriented persona templates. MCPT was created by applying the Design Science Research (DSR) methodology, and it incorporates multiple iterations for template refinement and validation among project stakeholders; all of the proposed steps of this method were based on theoretical contributions. Furthermore, MCPT was tested and refined within a real-life R&D project focusing on developing a digital platform e-marketplace for short agrifood supply chains in two iteration cycles. MCPT fills a critical void in persona research by providing detailed instructions for each step of template development. By involving the target audience, users, and project stakeholders, MCPT adds rigour to the persona creation process, enhancing the quality and relevance of personae casts. This paper contributes to the body of knowledge by offering an initial proposal of a comprehensive method for creating persona templates within diverse projects and contexts. Further research should explore MCPT’s adaptability to different settings and projects, thus refining its effectiveness and extending its utility in user-centred design practices. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

Abstract
Missing data is characterized by the presence of absent values in data (i.e., missing values) and it is currently categorized into three different mechanisms: Missing Completely at Random, Missing At Random, and Missing Not At Random. When performing missing data experiments and evaluating techniques to handle absent values, these mechanisms are often artificially generated (a process referred to as data amputation) to assess the robustness and behavior of the used methods. Due to the lack of a standard benchmark for data amputation, different implementations of the mechanisms are used in related research (some are often not disclaimed), preventing the reproducibility of results and leading to an unfair or inaccurate comparison between existing and new methods. Moreover, for users outside the field, experimenting with missing data or simulating the appearance of missing values in real-world domains is unfeasible, impairing stress testing in machine learning systems. This work introduces mdatagen, an open source Python library for the generation of missing data mechanisms across 20 distinct scenarios, following different univariate and multivariate implementations of the established missing mechanisms. The package therefore fosters reproducible results across missing data experiments and enables the simulation of artificial missing data under flexible configurations, making it very versatile to mimic several real-world applications involving missing data. The source code and detailed documentation for mdatagen are available at https://github.com/ArthurMangussi/pymdatagen.

Abstract
Image inpainting is a crucial technique in computer vision, particularly for reconstructing corrupted images. In medical imaging, it addresses issues from instrumental errors, artifacts, or human factors. The development of deep learning techniques has revolutionized image inpainting, allowing for the generation of high-level semantic information to ensure structural and textural consistency in restored images. This paper presents a comprehensive review of 53 studies on deep image inpainting in medical imaging, analyzing its evolution, impact, and limitations. The findings highlight the significance of deep image inpainting in artifact removal and enhancing the performance of multi-task approaches by localizing and inpainting regions of interest. Furthermore, the study identifies magnetic resonance imaging and computed tomography as the predominant modalities and highlights generative adversarial networks and U-Net as preferred architectures. Future research directions include the development of blind inpainting techniques, the exploration of techniques suitable for 3D/4D images, multiple artifacts, and multi-task applications, and the improvement of architectures.