Giovanni Sansavini

Laya Das, Mohammad Hossein Saadat, Blazhe Gjorgiev et al.

Deep learning-based object detection is a powerful approach for detecting faulty insulators in power lines. This involves training an object detection model from scratch, or fine tuning a model that is pre-trained on benchmark computer vision datasets. This approach works well with a large number of insulator images, but can result in unreliable models in the low data regime. The current literature mainly focuses on detecting the presence or absence of insulator caps, which is a relatively easy detection task, and does not consider detection of finer faults such as flashed and broken disks. In this article, we formulate three object detection tasks for insulator and asset inspection from aerial images, focusing on incipient faults in disks. We curate a large reference dataset of insulator images that can be used to learn robust features for detecting healthy and faulty insulators. We study the advantage of using this dataset in the low target data regime by pre-training on the reference dataset followed by fine-tuning on the target dataset. The results suggest that object detection models can be used to detect faults in insulators at a much incipient stage, and that transfer learning adds value depending on the type of object detection model. We identify key factors that dictate performance in the low data-regime and outline potential approaches to improve the state-of-the-art.

Laya Das, Blazhe Gjorgiev, Giovanni Sansavini

Deep neural networks (DNNs) are often coupled with physics-based models or data-driven surrogate models to perform fault detection and health monitoring of systems in the low data regime. These models serve as digital twins to generate large quantities of data to train DNNs which would otherwise be difficult to obtain from the real-life system. However, such models can exhibit parametric uncertainty that propagates to the generated data. In addition, DNNs exhibit uncertainty in the parameters learnt during training. In such a scenario, the performance of the DNN model will be influenced by the uncertainty in the physics-based model as well as the parameters of the DNN. In this article, we quantify the impact of both these sources of uncertainty on the performance of the DNN. We perform explicit propagation of uncertainty in input data through all layers of the DNN, as well as implicit prediction of output uncertainty to capture the former. Furthermore, we adopt Monte Carlo dropout to capture uncertainty in DNN parameters. We demonstrate the approach for fault detection of power lines with a physics-based model, two types of input data and three different neural network architectures. We compare the performance of such uncertainty-aware probabilistic models with their deterministic counterparts. The results show that the probabilistic models provide important information regarding the confidence of predictions, while also delivering an improvement in performance over deterministic models.

Laya Das, Blazhe Gjorgiev, Giovanni Sansavini

Inspection of insulators is important to ensure reliable operation of the power system. Deep learning is being increasingly exploited to automate the inspection process by leveraging object detection models to analyse aerial images captured by drones. A purely object detection-based approach, however, suffers from class imbalance-induced poor performance, which can be accentuated for infrequent and hard-to-detect incipient faults. This article proposes the use of anomaly detection along with object detection in a two-stage approach for incipient fault detection in a data-efficient manner. An explainable convolutional one-class classifier is adopted for anomaly detection. The one-class formulation reduces the reliance on plentifully available images of faulty insulators, while the explainability of the model is expected to promote adoption by the industry. A modified loss function is developed that addresses computational and interpretability issues with the existing model, also allowing for the integration of other losses. The superiority of the novel loss function is demonstrated with MVTec-AD dataset. The models are trained for insulator inspection with two datasets -- representing data-abundant and data-scarce scenarios -- in unsupervised and semi-supervised settings. The results suggest that including as few as five real anomalies in the training dataset significantly improves the model's performance and enables reliable detection of rarely occurring incipient faults in insulators.

Lorenzo Zapparoli, Blazhe Gjorgiev, Giovanni Sansavini

As the penetration of distributed energy resources increases, harnessing their flexibility becomes critical for power system operations. Virtual power plants (VPPs) offer a promising solution. However, existing VPP market scheduling tools exhibit a tradeoff between economic performance and tractability. Stochastic formulations provide probabilistically optimal decisions but are computationally intractable for large systems due to scenario explosion. Robust approaches are more tractable but often yield conservative decisions. This paper addresses this gap by proposing a stochastic multi-market VPP scheduling framework that represents uncertainty in the spectral domain via intrusive Polynomial Chaos Expansion (PCE). The resulting reformulation yields a low-dimensional deterministic spectral counterpart that preserves the stochastic structure and can be solved efficiently with standard optimization tools. The proposed spectral approach is demonstrated on a DER-based VPP operating on a realistic Swiss low-voltage grid and benchmarked against a state-of-the-art scenario-based solution. Results show that intrusive PCE achieves solution quality comparable to the scenario-based benchmark, with up to a 137 times reduction in computational effort, while yielding highly accurate bidding decisions. Finally, to facilitate adoption and reproducibility, we release an open-source, application-agnostic projection tool that automates the spectral reformulation for generic single- and two-stage stochastic programs.

Weijie Xia, James Ciyu Qin, Edgar Mauricio Salazar Duque et al.

Probabilistic power flow (PPF) is essential for quantifying operational uncertainty in modern distribution systems with high penetration of renewable generation and flexible loads. Conventional PPF methods primarily rely on Monte Carlo (MC) based power flow (PF) simulations or simplified analytical approximations. While MC approaches are computationally intensive and demand substantial data storage, analytical approximations often compromise accuracy. In this paper, we propose a novel analytical PPF framework that eliminates the dependence on MC-based PF simulations and, in principle, enables an approximation of the analytical form of arbitrary voltage distributions. The core idea is to learn an explicit and invertible mapping between stochastic power injections and system voltages using invertible neural networks (INNs). By leveraging the Change of Variable Theorem, the proposed framework facilitates direct approximation of the analytical form of voltage probability distributions without repeated PF computations. Extensive numerical studies demonstrate that the proposed framework achieves state-of-the-art performance both as a high-accuracy PF solver and as an efficient analytical PPF estimator.

Andrej Stankovski, Blazhe Gjorgiev, James Ciyu Qin et al.

Power system security assessments, e.g. via cascading outage models, often use operational set-points based on optimal power flow (OPF) dispatch. However, driven by cost minimization, OPF provides an ideal, albeit unrealistic, clearing of the generating units that disregards the complex interactions among market participants. In addition, existing market modeling tools often utilize economic dispatch and unit commitment to minimize total system costs, often disregarding the profit-driven behavior of market participants. The security of the system, therefore, may be overestimated. To address this gap, we introduce a social-welfare-based day-ahead market-clearing model. The security implications are analyzed using Cascades, a model for cascading failure analysis. We apply this model to the IEEE-118 bus system with three independent control zones. The results show that market dispatch leads to an increase in demand not served (DNS) of up to 80% higher than OPF, highlighting a significant security overestimation. This is especially pronounced in large-scale cascading events with DNS above 100MW. A key driver is the increased dispatch of storage and gas units, which can place the system in critical operating conditions. Operators can use this information to properly estimate the impact of the market on system security and plan efficient expansion strategies.

Colin Jüni, Mina Montazeri, Yi Guo et al.

Buildings account for approximately 40% of global energy consumption, and with the growing share of intermittent renewable energy sources, enabling demand-side flexibility, particularly in heating, ventilation and air conditioning systems, is essential for grid stability and energy efficiency. This paper presents a safe deep reinforcement learning-based control framework to optimize building space heating while enabling demand-side flexibility provision for power system operators. A deep deterministic policy gradient algorithm is used as the core deep reinforcement learning method, enabling the controller to learn an optimal heating strategy through interaction with the building thermal model while maintaining occupant comfort, minimizing energy cost, and providing flexibility. To address safety concerns with reinforcement learning, particularly regarding compliance with flexibility requests, we propose a real-time adaptive safety-filter to ensure that the system operates within predefined constraints during demand-side flexibility provision. The proposed real-time adaptive safety filter guarantees full compliance with flexibility requests from system operators and improves energy and cost efficiency -- achieving up to 50% savings compared to a rule-based controller -- while outperforming a standalone deep reinforcement learning-based controller in energy and cost metrics, with only a slight increase in comfort temperature violations.

Raffael Theiler, Leandro Von Krannichfeldt, Giovanni Sansavini et al.

Accurate and reliable energy forecasting is essential for power grid operators who strive to minimize extreme forecasting errors that pose significant operational challenges and incur high intra-day trading costs. Incorporating planning information -- such as anticipated user behavior, scheduled events or timetables -- provides substantial contextual information to enhance forecast accuracy and reduce the occurrence of large forecasting errors. Existing approaches, however, lack the flexibility to effectively integrate both dynamic, forward-looking contextual inputs and historical data. In this work, we conceptualize forecasting as a combined forecasting-regression task, formulated as a sequence-to-sequence prediction problem, and introduce contextually-enhanced transformer models designed to leverage all contextual information effectively. We demonstrate the effectiveness of our approach through a primary case study on nationwide railway energy consumption forecasting, where integrating contextual information into transformer models, particularly timetable data, resulted in a significant average mean absolute error reduction of 26.6%. An auxiliary case study on building energy forecasting, leveraging planned office occupancy data, further illustrates the generalizability of our method, showing an average reduction of 56.3% in mean absolute error. Compared to other state-of-the-art methods, our approach consistently outperforms existing models, underscoring the value of context-aware deep learning techniques in energy forecasting applications.

Lorenzo Zapparoli, Paul Fäth, Blazhe Gjorgiev et al.

As the penetration of distributed energy resources (DERs) increases, harnessing their flexibility becomes critical for power system operations. Virtual power plants (VPPs) offer a promising solution. However, most existing scheduling tools rely on simplified DER or grid models and largely overlook local flexibility procurement mechanisms such as dynamic network tariffs. This paper proposes a two-stage stochastic optimization framework for VPP multi-market scheduling that integrates detailed device-level constraints, network limitations, and operational and market uncertainties. Conditional value-at-risk is incorporated to represent risk preferences, and Benders decomposition ensures tractability with extensive scenario sets. The model jointly optimizes bidding across energy and reserve markets while explicitly accounting for local flexibility procurement through dynamic network tariffs. The results from a realistic case study show that both risk-neutral and risk-averse strategies exploit arbitrage opportunities. However, risk aversion reduces profit volatility through closer alignment with physical dispatch. Dynamic tariffs unlock local flexibility by shifting demand across the day, though strong tariff signals reduce expected profitability by up to 65% with limited additional flexibility gains.

Anna Varbella, Kenza Amara, Blazhe Gjorgiev et al.

Power grids are critical infrastructures of paramount importance to modern society and, therefore, engineered to operate under diverse conditions and failures. The ongoing energy transition poses new challenges for the decision-makers and system operators. Therefore, developing grid analysis algorithms is important for supporting reliable operations. These key tools include power flow analysis and system security analysis, both needed for effective operational and strategic planning. The literature review shows a growing trend of machine learning (ML) models that perform these analyses effectively. In particular, Graph Neural Networks (GNNs) stand out in such applications because of the graph-based structure of power grids. However, there is a lack of publicly available graph datasets for training and benchmarking ML models in electrical power grid applications. First, we present PowerGraph, which comprises GNN-tailored datasets for i) power flows, ii) optimal power flows, and iii) cascading failure analyses of power grids. Second, we provide ground-truth explanations for the cascading failure analysis. Finally, we perform a complete benchmarking of GNN methods for node-level and graph-level tasks and explainability. Overall, PowerGraph is a multifaceted GNN dataset for diverse tasks that includes power flow and fault scenarios with real-world explanations, providing a valuable resource for developing improved GNN models for node-level, graph-level tasks and explainability methods in power system modeling. The dataset is available at https://figshare.com/articles/dataset/PowerGraph/22820534 and the code at https://github.com/PowerGraph-Datasets.

Alfredo Oneto, Blazhe Gjorgiev, Giovanni Sansavini

Many data clustering applications must handle objects that cannot be represented as vectors. In this context, the bag-of-vectors representation describes complex objects through discrete distributions, for which the Wasserstein distance provides a well-conditioned dissimilarity measure. Kernel methods extend this by embedding distance information into feature spaces that facilitate analysis. However, an unsupervised framework that combines kernels with Wasserstein distances for clustering distributional data is still lacking. We address this gap by introducing a computationally tractable framework that integrates Wasserstein metrics with kernel methods for clustering. The framework can accommodate both vectorial and distributional data, enabling applications in various domains. It comprises three components: (i) an efficient approximation of pairwise Wasserstein distances using multiple reference distributions; (ii) shifted positive definite kernel functions based on Wasserstein distances, combined with kernel principal component analysis for feature mapping; and (iii) scalable, distance-agnostic validity indices for clustering evaluation and kernel parameter optimization. Experiments on power distribution graphs and real-world time series demonstrate the effectiveness and efficiency of the proposed framework.

Enrico Ampellio, Blazhe Gjorgiev, Giovanni Sansavini

Engineering design involves demanding models encompassing many decision variables and uncontrollable parameters. In addition, unavoidable aleatoric and epistemic uncertainties can be very impactful and add further complexity. The state-of-the-art adopts two steps, uncertainty quantification and design optimization, to optimize systems under uncertainty by means of robust or stochastic metrics. However, conventional scenario-based, surrogate-assisted, and mathematical programming methods are not sufficiently scalable to be affordable and precise in large and complex cases. Here, a multi-level approach is proposed to accurately optimize resource-intensive, high-dimensional, and complex engineering problems under uncertainty with minimal resources. A non-intrusive, fast-scaling, Kriging-based surrogate is developed to map the combined design/parameter domain efficiently. Multiple surrogates are adaptively updated by hierarchical and orthogonal decomposition to leverage the fewer and most uncertainty-informed data. The proposed method is statistically compared to the state-of-the-art via an analytical testbed and is shown to be concurrently faster and more accurate by orders of magnitude.