Modern power grids are becoming increasingly complex with the integration of heterogeneous distributed energy resources, underscoring the need for accurate and efficient Power Flow Analysis to ensure stability, reliability, and market operations. Existing methods generally rely on iterative numerical techniques (INT) or machine learning (ML). While INT is physically consistent and highly accurate, it can be computationally expensive and vulnerable to slow or non-convergence. ML methods offer faster solutions but often require extensive data, suffer from limited extrapolation capabilities, and lack physical consistency. Physics-informed ML (PIML) bridges these gaps by embedding domain knowledge before, during, and after training. However, current PIML approaches typically do not leverage this full range of opportunities. In this paper, we propose a novel PIML framework for Power Flow Analysis that integrates physical insights at all three stages (pre, in, and post-processing) to achieve superior accuracy and efficiency. Notably, we introduce a new post-processing technique that partitions the power network into its mesh and radial components: the mesh portion is handled via PIML, while the radial portion is efficiently solved with a convex optimization approach informed by the PIML outputs. This approach is efficient with radial topologies, especially in power distribution networks where the radial part is predominant. Experiments on realistic power networks demonstrate that our method outperforms state-of-the-art approaches in both accuracy and computational performance.

Fast diesel engine models for real-time prediction in dynamic conditions are required to predict engine performance parameters, to identify emerging failures early on and to establish trends in performance reduction. In order to address these issues, two main alternatives exist: one is to exploit the physical knowledge of the problem, the other one is to exploit the historical data produced by the modern automation system. Unfortunately, the first approach often results in hard-to-tune and very computationally demanding models that are not suited for real-time prediction, while the second approach is often not trusted because of its questionable physical grounds. In this paper, the authors propose a novel hybrid model, which combines physical and data-driven models, to model diesel engine exhaust gas temperatures in operational conditions. Thanks to the combination of these two techniques, the authors were able to build a fast, accurate and physically grounded model that bridges the gap between the physical and data driven approaches. In order to support the proposal, the authors will show the performance of the different methods on real-world data collected from the Holland Class Oceangoing Patrol Vessel.

Reducing the noise impact of ships on the marine environment is one of the objectives of new propellers designs, since they represent the dominant source of underwater radiated noise, especially when cavitation occurs. Consequently, ship designers require new predictive tools able to verify the compliance with noise requirements and to compare the effectiveness of different design solutions. In this context, tools able to provide a reliable estimate of propeller noise spectra based just on the information available at design stage represent a fundamental tool to speed up the design process avoiding model scale tests. This work focus on developing such a tool, adopting methods coming from the world of Machine Learning and Deep Neural Networks, in order to create a model able to predict the cavitating marine propeller noise spectra. For this purpose authors will make use of a dataset collected by means of dedicated model scale measurements in a cavitation tunnel combined with the detailed flow characterization obtainable by calculations carried out with a Boundary Element Method. The performance of the proposed approaches are analyzed considering different definitions of the input and output variables used during the modelization.

The importance of reducing the noise impact of ships is being recognised worldwide. Consequently, the inclusion of this principle among the objectives and constraints of new designs is becoming a standard. For this reason, considerable attention is given to the propeller being often the dominant source of underwater radiated noise, especially when cavitation occurs, as it happens in most cases when a ship sails at design speed. The designers of quieter propulsion systems require the availability of predictive tools able to verify the compliance with noise requirements and to compare the effectiveness of different design solutions. In this context, tools able to provide a reliable estimate of propeller noise spectra based just on the information available during propeller design represent a fundamental tool to speed up the design process avoiding model scale tests. This work focuses on developing a tool able to predict the cavitating marine propeller generated noise spectra at design stage exploiting the most recent advances in Deep Learning, able to take advantage of both structured and unstructured data, and in hybrid modelling, able to exploit both data and physical knowledge about the problem. For this purpose authors will make use of a dataset collected by means of dedicated model scale measurements in a cavitation tunnel combined with the detailed flow characterisation obtainable by calculations carried out with a Boundary Element Method. The performance of the proposed approaches are analysed considering different scenarios and different definitions of the input and output variable used during the modelisation.

Shipping is responsible for approximately the 90% of world trade leading to significant impacts on the environment. As a consequence, a crucial issue for the maritime industry is to develop technologies able to increase the ship efficiency, by reducing fuel consumption and unnecessary maintenance operations. For example, the marine fouling phenomenon has a deep impact, since to prevent or reduce its growth which affects the ship consumption, costly drydockings for cleaning the hull and the propeller are needed and must be scheduled based on a speed loss estimation. In this work a data driven Digital Twin of the ship is built, leveraging on the large amount of information collected from the on-board sensors, and is used for estimating the speed loss due to marine fouling. A thorough comparison between the proposed method and ISO 19030, which is the de-facto standard for dealing with this task, is carried out on real-world data coming from two Handymax chemical/product tankers. Results clearly show the effectiveness of the proposal and its better speedloss prediction accuracy with respect to the ISO 19030, thus allowing reducing the fuel consumption due to fouling.

The maintenance of the several components of a Ship Propulsion Systems is an onerous activity, which need to be efficiently programmed by a shipbuilding company in order to save time and money. The replacement policies of these components can be planned in a Condition-Based fashion, by predicting their decay state and thus proceed to substitution only when really needed. In this paper, authors propose several Data Analysis supervised and unsupervised techniques for the Condition-Based Maintenance of a vessel, characterised by a combined diesel-electric and gas propulsion plant. In particular, this analysis considers a scenario where the collection of vast amounts of labelled data containing the decay state of the components is unfeasible. In fact, the collection of labelled data requires a drydocking of the ship and the intervention of expert operators, which is usually an infrequent event. As a result, authors focus on methods which could allow only a minimal feedback from naval specialists, thus simplifying the dataset collection phase. Confidentiality constraints with the Navy require authors to use a real-data validated simulator and the dataset has been published for free use through the OpenML repository.

The behavior and interaction of the main components of Ship Propulsion Systems cannot be easily modeled with a priori physical knowledge, considering the large amount of variables influencing them. Data-Driven Models (DDMs), instead, exploit advanced statistical techniques to build models directly on the large amount of historical data collected by on-board automation systems, without requiring any a priori knowledge. DDMs are extremely useful when it comes to continuously monitoring the propulsion equipment and take decisions based on the actual condition of the propulsion plant. In this paper, the authors investigate the problem of performing Condition-Based Maintenance through the use of DDMs. In order to conceive this purpose, several state-of-the-art supervised learning techniques are adopted, which require labeled sensor data in order to be deployed. A naval vessel, characterized by a combined diesel-electric and gas propulsion plant, has been exploited to collect such data and show the effectiveness of the proposed approaches. Because of confidentiality constraints with the Navy the authors used a real-data validated simulator and the dataset has been published for free use through the UCI repository.

Induction motors are fundamental components of several modern automation system, and they are one of the central pivot of the developing e-mobility era. The most vulnerable parts of an induction motor are the bearings, the stator winding and the rotor bars. Consequently, monitoring and maintaining them during operations is vital. In this work, authors propose an Induction Motors bearings monitoring tool which leverages on stator currents signals processed with a Deep Learning architecture. Differently from the state-of-the-art approaches which exploit vibration signals, collected by easily damageable and intrusive vibration probes, the stator currents signals are already commonly available, or easily and unintrusively collectable. Moreover, instead of using now-classical data-driven models, authors exploit a Deep Learning architecture able to extract from the stator current signal a compact and expressive representation of the bearings state, ultimately providing a bearing fault detection system. In order to estimate the effectiveness of the proposal, authors collected a series of data from an inverter-fed motor mounting different artificially damaged bearings. Results show that the proposed approach provides a promising and effective yet simple bearing fault detection system.

Crash stop maneuvering performance is one of the key indicators of the vessel safety properties for a shipbuilding company. Many different factors affect these performances, from the vessel design to the environmental conditions, hence it is not trivial to assess them accurately during the preliminary design stages. Several first principal equation methods are available to estimate the crash stop maneuvering performance, but unfortunately, these methods usually are either too costly or not accurate enough. To overcome these limitations, the authors propose a new data-driven method, based on the popular Random Forests learning algorithm, for predicting the crash stopping maneuvering performance. Results on real-world data provided by the DAMEN Shipyards show the effectiveness of the proposal.