A5034315185- Water Systems and Optimization
- Energy Load and Power Forecasting
- Hydrological Forecasting Using AI
- Water resources management and optimization
- Hydrology and Watershed Management Studies
- Integrated Energy Systems Optimization
- Climate variability and models
- Urban Stormwater Management Solutions
- Flood Risk Assessment and Management
- Cryospheric studies and observations
- Meteorological Phenomena and Simulations
- Hydrology and Sediment Transport Processes
- Building Energy and Comfort Optimization
- Smart Grid Energy Management
- Water Quality Monitoring Technologies
- Water-Energy-Food Nexus Studies
- Fish Ecology and Management Studies
- Hydraulic flow and structures
- Smart Grid Security and Resilience
- High voltage insulation and dielectric phenomena
- Landslides and related hazards
- Climate Change Policy and Economics
- Atmospheric and Environmental Gas Dynamics
- Power Systems and Renewable Energy
- Water Treatment and Disinfection
Free University of Bozen-Bolzano
2018-2024
Eurac Research
2024
Aalborg University
2020
Study region: The region of interest is the South-Tyrol in southeastern Alps, Italy. A comparison meteorological forcing performed with reference to this while hydrological simulations are conducted Passirio river basin. focus: objective work evaluate suitability ERA5-Land reanalysis product as a dataset for modelling topographically complex Alpine regions. compared observational gridded dataset, obtained by means Kriging interpolation, available at two contrasting spatial resolutions coarse...
Abstract Short‐term forecasting of water demand is a crucial process for managing efficiently supply systems. This paper proposes to develop novel graph convolutional recurrent neural network (GCRNN) predict time series related some systems or district metering areas that belong the same geographical area. The aim build graph‐based model able capture dependence among different both in spatial and temporal terms. built on set graphs, its performance compared two methods, including...
Abstract Sustainable management of water resources is a key challenge nowadays and in the future. Water distribution systems have to ensure fresh for all users an increasing demand scenario related long-term effects due climate change. In this context, reliable short-term forecasting model crucial optimal resources. This study proposes novel deep learning based on long memory (LSTM) neural networks forecast hourly demand. Due limitations using multiple input sequences with different time...
Effective monitoring and forecasting of flood events are key aspects early warning systems, especially in areas susceptible to frequent floods. In this context, Artificial Intelligence (AI) techniques have proven be a strong tool for enhancing such systems because they can capture non-linear processes genesis. AI models make accurate predictions with minimum processing time, thus providing alternative nowcasting. However, the quality quantity stations black-box nature machine learning (ML)...
Preferential flow (PF) is a key hydrological process that influences water infiltration, soil moisture redistribution, and streamflow generation. In Mediterranean forested catchments, the dynamics of PF its controls remain largely underexplored. Here, we investigated mechanisms their impact on response in Re della Pietra experimental catchment (2 km²) Tuscan Apennines, central Italy. Two hillslope transects with sensors at shallow (15 cm) deep (35 layers were monitored for 34 18...
Proper hydraulic simulation models, which are fundamental to analyse a water distribution system, require calibration procedure. This paper proposes multi-objective procedure calibrate demands and pipe roughness in the context of an ill-posed problem, where number measurements is smaller than variables. The proposed methodology consists two-steps based on genetic algorithm. Firstly, several runs calibrator performed corresponding pressure flow-rates values averaged overcome non-uniqueness...
A reliable short-term forecasting model is fundamental to managing a water distribution system properly. This study addresses the problem of efficient development deep neural network for consumption in small-scale supply systems. These aqueducts experience significant fluctuations their due small number users, making them challenging task. To deal with this issue, proposes procedure develop an ensemble model. reinforce successfully weekly and yearly seasonality which affect these data, two...
Bridges are essential structures that connect riverbanks and facilitate transportation. However, bridge piers abutments can disrupt the natural flow of rivers, causing a rise in water levels upstream bridge. The levels, known as backwater or afflux, threaten stability service bridges riverbanks. It is postulated applications estimation models with more precise afflux predictions enhance safety flood-prone areas. In this study, eight machine learning (ML) were developed to estimate utilizing...
Abstract Water demand management is essential for water utilities, which have the critical task of supplying drinking from sources to end-users through distribution network. Therefore, utilities make decisions current and future functioning system. In this context, artificial intelligence approach with data-driven methods can be used develop powerful tools improve overall management. fact, model demands plenty tasks applications such as forecasting or anomaly detection. work, we propose...
It has been proved that the standard representation of water demand in a Water Distribution Network (WDN) leads to pipe head loss errors as well fully satisfied regardless pressure assumption is misleading. This follows different algorithms have developed order overcome these two drawbacks although separately and independently. Consequently, this paper introduces an alternative formulation Global Gradient Algorithm (GGA), referred UD-PD, which able solve uniformly distributed driven demands...
Developing data-driven models for bursts detection is currently a demanding challenge efficient and sustainable management of water supply systems. The main limit in the progress these lies large amount accurate data required. aim to present methodology generation reliable data, which are fundamental train anomaly set alarms. Thus, results proposed provide suitable consumption data. presented procedure consists stochastic modelling request hydraulic pipes simulation yield synthetic time...
The evolution of smart water grids leads to new Big Data challenges boosting the development and application Machine Learning techniques support efficient sustainable drinking management. These powerful rely on hyperparameters making models’ tuning a tricky crucial task. We hence propose an insightful analysis Artificial Neural Networks for demand forecasting. This study focuses layers nodes’ fitting different Network architectures through grid search method by varying dataset, prediction...
Drinking water demand modelling and forecasting is a crucial task for sustainable management planning of supply systems. Despite many short-term investigations, the medium-term problem needs better exploration, particularly analysis assessment meteorological data drinking demand. This work proposes to analyse suitability ERA5-Land reanalysis as weather input in modelling. A multivariate deep learning model based on long memory architecture used this study over prediction horizon ranging from...