• Energy Research
• 2021-2025close
• IT close
• CA close
• Sapienza University of Rome close

Wind energy represents one of the leading renewable energy sectors and is considered instrumental in the ongoing decarbonization process. Accurate forecasts are essential for a reliable large-scale wind power integration, allowing efficient operation and maintenance, planning of unit commitment, and scheduling by system operators. However, due to non-stationarity, randomness, and intermittency, forecasting wind power is challenging. This work investigates a multi-modal approach for wind power forecasting by considering turbine-level time series collected from SCADA systems and high-resolution Numerical Weather Prediction maps. A neural architecture based on stacked Recurrent Neural Networks is proposed to process and combine different data sources containing spatio-temporal patterns. This architecture allows combining the local information from the turbine’s internal operating conditions with future meteorological data from the surrounding area. Specifically, this work focuses on multi-horizon turbine-level hourly forecasts for an entire wind farm with a lead time of 90 h. This work explores the impact of meteorological variables on different spatial scales, from full grids to cardinal point features, on wind power forecasts. Results show that a subset of features associated with all wind directions, even when spatially distant, can produce more accurate forecasts with respect to full grids and reduce computation times. The proposed model outperforms the linear regression baseline and the XGBoost regressor achieving an average skill score of 25%. Finally, the integration of SCADA data in the training process improved the predictions allowing the multi-modal neural network to model not only the meteorological patterns but also the turbine’s internal behavior.

Wind energy represents one of the leading renewable energy sectors and is considered instrumental in the ongoing decarbonization process. Accurate forecasts are essential for a reliable large-scale wind power integration, allowing efficient operation and maintenance, planning of unit commitment, and scheduling by system operators. However, due to non-stationarity, randomness, and intermittency, forecasting wind power is challenging. This work investigates a multi-modal approach for wind power forecasting by considering turbine-level time series collected from SCADA systems and high-resolution Numerical Weather Prediction maps. A neural architecture based on stacked Recurrent Neural Networks is proposed to process and combine different data sources containing spatio-temporal patterns. This architecture allows combining the local information from the turbine’s internal operating conditions with future meteorological data from the surrounding area. Specifically, this work focuses on multi-horizon turbine-level hourly forecasts for an entire wind farm with a lead time of 90 h. This work explores the impact of meteorological variables on different spatial scales, from full grids to cardinal point features, on wind power forecasts. Results show that a subset of features associated with all wind directions, even when spatially distant, can produce more accurate forecasts with respect to full grids and reduce computation times. The proposed model outperforms the linear regression baseline and the XGBoost regressor achieving an average skill score of 25%. Finally, the integration of SCADA data in the training process improved the predictions allowing the multi-modal neural network to model not only the meteorological patterns but also the turbine’s internal behavior.

The Salcheto winery has undertaken a process of reduction of its primary energy consumption and the implementation of green energy technologies. They adopted solar photovoltaic, wood biomass, and geothermal energy sources. A horizontal ground source heat exchanger (GSHE) plant is used to cool a part of the pressed grapes and control the wine production temperature. The goal of this work was to investigate some technical issues of the plant and to increase the efficiency of the whole system. The first step was the evaluation of the actual operating conditions of the GSHE plant, by performing a thermal response test. The results allowed us to find the thermal diffusivity of 3.5 × 10−7 m2/s, and the calculation with the IGSHPA standard indicated a cooling performance of about 6 kW. A survey during the harvest highlighted a peak power of 6 kW. Therefore, to improve the plant, some modifications were proposed and analyzed. In the new layout, the geothermal plant serves the condenser of the refrigeration unit, allowing cooling of the all production lines, instead of only one. The peak power was evaluated as 32 kW, and the GSHE can fulfil this, up to 18 kW. For higher power, the evaporative tower will supply the remainder, covering a maximum of 45%. Furthermore, the refrigeration unit may cover the cooling requirements of the entire residential and office building, without other plant improvements.

The Salcheto winery has undertaken a process of reduction of its primary energy consumption and the implementation of green energy technologies. They adopted solar photovoltaic, wood biomass, and geothermal energy sources. A horizontal ground source heat exchanger (GSHE) plant is used to cool a part of the pressed grapes and control the wine production temperature. The goal of this work was to investigate some technical issues of the plant and to increase the efficiency of the whole system. The first step was the evaluation of the actual operating conditions of the GSHE plant, by performing a thermal response test. The results allowed us to find the thermal diffusivity of 3.5 × 10−7 m2/s, and the calculation with the IGSHPA standard indicated a cooling performance of about 6 kW. A survey during the harvest highlighted a peak power of 6 kW. Therefore, to improve the plant, some modifications were proposed and analyzed. In the new layout, the geothermal plant serves the condenser of the refrigeration unit, allowing cooling of the all production lines, instead of only one. The peak power was evaluated as 32 kW, and the GSHE can fulfil this, up to 18 kW. For higher power, the evaporative tower will supply the remainder, covering a maximum of 45%. Furthermore, the refrigeration unit may cover the cooling requirements of the entire residential and office building, without other plant improvements.

Abstract This paper presents a modified Jaya algorithm (MJaya) for optimizing the material costs and electric-thermal performance of an Underground Power Cable System (UPCS). Three power cables arranged in flat formation are considered. Three XLPE high voltage cables are situated in the thermal backfill layer for ensuring the optimal thermal performance of the cable system. The cable backfill dimensions, cable backfill material, and cable conductor area are selected as design variables in the optimization problem. In the study, the Finite Element Method model is validated experimentally. The Particle Swarm Optimization (PSO), Jaya, and MJaya algorithms are used for multiobjective optimization in order to design a cable system in such a way to minimize the cable backfill costs and maximize the allowable electric current flowing through the cables. For the case study, calculations performed using the Jaya algorithm indicated 1.7 mln Euro cable system costs while cable ampacity is equal to I = 1460 A. The calculations are performed for the objective function values equal to w1 = 0.5 and w2 = 0.5. Such an optimization parameters set allow obtaining low costs of UPCS alongside with reasonable cable line ampacity. What is more, the results of the optimization obtained by Jaya, MJaya, and PSO algorithms are compared. Therefore, Coverage and Hypervolume metrics are incorporated. It is concluded that both the Jaya and MJaya algorithms performed better when compared to the PSO algorithm.

Abstract This paper presents a modified Jaya algorithm (MJaya) for optimizing the material costs and electric-thermal performance of an Underground Power Cable System (UPCS). Three power cables arranged in flat formation are considered. Three XLPE high voltage cables are situated in the thermal backfill layer for ensuring the optimal thermal performance of the cable system. The cable backfill dimensions, cable backfill material, and cable conductor area are selected as design variables in the optimization problem. In the study, the Finite Element Method model is validated experimentally. The Particle Swarm Optimization (PSO), Jaya, and MJaya algorithms are used for multiobjective optimization in order to design a cable system in such a way to minimize the cable backfill costs and maximize the allowable electric current flowing through the cables. For the case study, calculations performed using the Jaya algorithm indicated 1.7 mln Euro cable system costs while cable ampacity is equal to I = 1460 A. The calculations are performed for the objective function values equal to w1 = 0.5 and w2 = 0.5. Such an optimization parameters set allow obtaining low costs of UPCS alongside with reasonable cable line ampacity. What is more, the results of the optimization obtained by Jaya, MJaya, and PSO algorithms are compared. Therefore, Coverage and Hypervolume metrics are incorporated. It is concluded that both the Jaya and MJaya algorithms performed better when compared to the PSO algorithm.

Open spaces (OSs), such as streets, squares, and green areas, in existing built environments (BEs) are key places in disaster risk management. The seismic risk in the OSs is strictly related to BE characteristics. Scientific literature mainly focuses on extrinsic factors affecting risk, which are related to BE elements on the OSs frontier (e.g., buildings) that could cause indirect effects on the OSs. Conversely, just a few risk assessment studies consider intrinsic factors, which are related to OS elements that could suffer direct effects. Moreover, synoptic studies on such factors are still missing. Through literature-based research, the paper identifies specific factors influencing seismic risk in the OSs, focusing notably on intrinsic vulnerability. The literature review methodology includes both a systematic review from Scopus databases and a traditional bibliographic search using snowball analysis. According to the final selected papers, risk factors are classified into five categories of OS characteristics: morpho-typology; physical; construction; use and users; and context. Statistical analysis of the categories’ recurrence in the final papers firstly allows current literature gaps to be defined. The results also provide a preliminary OSs risk index weighting each category in terms of such recurrences, thus representing a first useful step to support non-expert stakeholders in a preliminary assessment of priorities to define the seismic risk of Oss.