Physics informed neural networks for surrogate modeling of accidental scenarios in nuclear power plants

Antonello Federico,Buongiorno Jacopo,Zio Enrico 한국원자력학회 2023 Nuclear Engineering and Technology Vol.55 No.9

Licensing the next-generation of nuclear reactor designs requires extensive use of Modeling and Simulation (M&S) to investigate system response to many operational conditions, identify possible accidental scenarios and predict their evolution to undesirable consequences that are to be prevented or mitigated via the deployment of adequate safety barriers. Deep Learning (DL) and Artificial Intelligence (AI) can support M&S computationally by providing surrogates of the complex multi-physics high-fidelity models used for design. However, DL and AI are, generally, low-fidelity ‘black-box’ models that do not assure any structure based on physical laws and constraints, and may, thus, lack interpretability and accuracy of the results. This poses limitations on their credibility and doubts about their adoption for the safety assessment and licensing of novel reactor designs. In this regard, Physics Informed Neural Networks (PINNs) are receiving growing attention for their ability to integrate fundamental physics laws and domain knowledge in the neural networks, thus assuring credible generalization capabilities and credible predictions. This paper presents the use of PINNs as surrogate models for accidental scenarios simulation in Nuclear Power Plants (NPPs). A case study of a Loss of Heat Sink (LOHS) accidental scenario in a Nuclear Battery (NB), a unique class of transportable, plug-and-play microreactors, is considered. A PINN is developed and compared with a Deep Neural Network (DNN). The results show the advantages of PINNs in providing accurate solutions, avoiding overfitting, underfitting and intrinsically ensuring physics-consistent results

Climate Change Effects on the Suitability of an Agricultural Area to Maize Cultivation: a New Land Evaluation Hybrid System for Maize

Antonello Bonfante,Angelo Basile,Silvia Maria Alfieri,Eugenia Monaco,Francesca De Lorenzi 한국토양비료학회 2014 한국토양비료학회 학술발표회 초록집 Vol.2014 No.6

Assessing Climatic Influences on Rodent Density : A Neural Network Modelling Approach and a Case Study in Central Italy

Antonello Pasini,Germana Szpunar,Giovanni Amori,Rocco Langone,Mauro Cristaldi 한국기상학회 2009 Asia-Pacific Journal of Atmospheric Sciences Vol.45 No.3

In a climate change regime (observed in the past and forecasted for the future), it becomes more and more important to assess the role of the climatic environment in influencing the density of some species of interest. In this paper, using density data from Capture-Mark-Recapture (CMR) of Yellow-necked mouse (Apodemus flavicollis), we perform non-linear analyses and apply a neural network (NN) model (conceived for recognising links in complex systems) in order to establish which climatic parameters represent the driving forcings of rodent density in Central Italy. We discover direct ad indirect climatic effects on this variable and, finally, a satisfying reconstruction of Yellow-necked mouse densities over the last 20 years is achieved by a NN model. A further “hindcast” in the past (back to 1965) is performed by the model trained on a recent period and clear trends can be recognised in the record of reconstructed densities.

Tests of a dual-readout fiber calorimeter with SiPM light sensors

Antonello, M.,Caccia, M.,Cascella, M.,Dunser, M.,Ferrari, R.,Franchino, S.,Gaudio, G.,Hall, K.,Hauptman, J.,Jo, H.,Kang, K.,Kim, B.,Lee, S.,Lerner, G.,Pezzotti, L.,Santoro, R.,Vivarelli, I.,Ye, R.,Wig Elsevier 2018 Nuclear Instruments & Methods in Physics Research. Vol.899 No.-

<P><B>Abstract</B></P> <P>In this paper, we describe the first tests of a dual-readout fiber calorimeter in which silicon photomultipliers are used to sense the (scintillation and Čerenkov) light signals. The main challenge in this detector is implementing a design that minimizes the optical crosstalk between the two types of fibers, which are located very close to each other and carry light signals that differ in intensity by about a factor of 60. The experimental data, which were obtained with beams of high-energy electrons and muons as well as in lab tests, illustrate to what extent this challenge was met. The Čerenkov light yield, a limiting factor for the energy resolution of this type of calorimeter, was measured to be about twice that of the previously tested configurations based on photomultiplier tubes. The lateral profiles of electromagnetic showers were measured on a scale of millimeters from the shower axis and significant differences were found between the profiles measured with the scintillating and the Čerenkov fibers.</P>

Monitoring an iconic heritage structure with OMA: the Main Spire of the Milan Cathedral

Antonello Ruccolo,Carmelo Gentile,Francesco Canali 국제구조공학회 2021 Smart Structures and Systems, An International Jou Vol.27 No.2

One of the most remarkable structural elements characterizing the Milan Cathedral is its Main Spire, built in Candoglia marble and completed in 1769. The Main Spire, reaching the height of about 108 m and supporting the statue of the Virgin Mary, is about 40 m high and stands on the octagonal <i>tiburio</i> erected around the main dome. The structural arrangement of the spire includes a central column which is connected through a spiral staircase to 8 perimeter columns and each column is stiffened by inverse flying buttress. Metallic clamps and dowels connect the marble blocks and metallic rods connect the perimeter columns to the central core. A large monitoring system was recently installed in the Milan Cathedral, including seismometers and temperature sensors at 3 levels of the Main Spire as well as a weather station at the top of the spire. After a concise historic background on the Main Spire and the description of the sensing devices installed in this structure, the paper focuses on the dynamic characteristics of the spire and their evolution during a time span of about 16 months. The presented results highlight that: (a) a high density of vibration modes is automatically detected in the frequency range 1.0-7.0 Hz; (b) the lower identified modes correspond to global modes of the cathedral; (c) the normal evolution in time of the resonant frequencies is characterized by clear fluctuations induced by the environmental effects (temperature and wind); (d) especially the dependence of resonant frequencies on temperature is very distinctive and reveals the key role of the metallic elements in the overall dynamic behavior; (e) notwithstanding the remarkable effects exerted by the changing environment on the resonant frequencies, output-only removal of environmental effects and novelty analysis allow an effective monitoring of the structural condition.

PCE : A Knowledge Base of Semantically Disambiguated Contents

Georgia Sanna,Antonello Angius,Giulio Concas,Dino Manca,Filippo Eros Pani 한국정보기술융합학회 2015 JoC Vol.6 No.2

With the Semantic Web people participate more actively in the building of information and take on an active role in enriching collaborative knowledge bases with user-generated contents. Moreover, new online resources have quickly gained impact on information management, such as the social bookmarking systems, which enable users to store, manage and share tagged Web content through folksonomies. These social tools allow users to associate free chosen keywords (tags) with Web content with the aim of providing an efficient navigation system through the heterogeneous amount of information available on the Web. However, the association of tags is totally arbitrary so that the use of synonyms, homonyms and new slang terms often hampers the Web information retrieval. Our project proposes the development of an innovative social bookmarking system based on a priori classification with semantic tags and structured categories extracted from online linguistic resources, which aims at disambiguating Web Content. According to this approach, each content will be described through two axes: a vertical one, concerning a hierarchical taxonomic classification and a horizontal one, dealing with a folksonomic classification. The final aim is to create a collaborative knowledge base of semantically disambiguated contents and to achieve more precise results in Web retrieval.

Measurement of the neutrino velocity with the ICARUS detector at the CNGS beam

ICARUS Collaboration,Antonello, M.,Aprili, P.,Baiboussinov, B.,Baldo Ceolin, M.,Benetti, P.,Calligarich, E.,Canci, N.,Centro, S.,Cesana, A.,Cieslik, K.,Cline, D.B.,Cocco, A.G.,Dabrowska, A.,Dequal, D. North-Holland Pub. Co 2012 Physics letters: B Vol.713 No.1

At the end of the 2011 run, the CERN CNGS neutrino beam has been briefly operated in lower intensity mode with ∼10<SUP>12</SUP> p.o.t./pulse and with a proton beam structure made of four LHC-like extractions, each with a narrow width of ∼3 ns, separated by 524 ns. This very tightly bunched beam allowed a very accurate time-of-flight measurement of neutrinos from CERN to LNGS on an event-by-event basis. The ICARUS T600 detector (CNGS2) has collected 7 beam-associated events, consistent with the CNGS collected neutrino flux of 2.2x10<SUP>16</SUP> p.o.t. and in agreement with the well-known characteristics of neutrino events in the LAr-TPC. The time of flight difference between the speed of light and the arriving neutrino LAr-TPC events has been analysed. The result δt=0.3+/-4.9(stat.)+/-9.0(syst.) ns is compatible with the simultaneous arrival of all events with speed equal to that of light. This is in a striking difference with the reported result of OPERA (OPERA Collaboration, 2011) [1] claiming that high energy neutrinos from CERN arrive at LNGS ∼60 ns earlier than expected from luminal speed.

3D reconstruction of two-phase random heterogeneous material from 2D sections: An approach via genetic algorithms

Pizzocri, D.,Genoni, R.,Antonello, F.,Barani, T.,Cappia, F. Korean Nuclear Society 2021 Nuclear Engineering and Technology Vol.53 No.9

This paper introduces a method to reconstruct the three-dimensional (3D) microstructure of two-phase materials, e.g., porous materials such as highly irradiated nuclear fuel, from two-dimensional (2D) sections via a multi-objective optimization genetic algorithm. The optimization is based on the comparison between the reference and reconstructed 2D sections on specific target properties, i.e., 2D pore number, and mean value and standard deviation of the pore-size distribution. This represents a multi-objective fitness function subject to weaker hypotheses compared to state-of-the-art methods based on n-points correlations, allowing for a broader range of application. The effectiveness of the proposed method is demonstrated on synthetic data and compared with state-of-the-art methods adopting a fitness based on 2D correlations. The method here developed can be used as a cost-effective tool to reconstruct the pore structure in highly irradiated materials using 2D experimental data.

Cyclic behaviour of beam-to-column welded

Luis Calado,Elena Mele,Antonello De Luca 국제구조공학회 2001 Steel and Composite Structures, An International J Vol.1 No.3

In this paper the results of an experimental program devoted to the assessment of the cyclic behaviour of full scale, European type, beam-column subassemblages with welded connections are presented. Six tests (five cyclic and one monotonic) have been carried out on three different series of specimens, encompassing a total of eighteen tests. The three specimen series have been designed with the aim of defining the effect of the column size on the connection behaviour, under different applied loading histories. The tests have evidenced the effect of the column size and panel zone design and of the applied loading history on the cyclic behaviour and failure modes of the connections.

Stress and strain state in the segmental linings during mechanized tunnelling

Do, Ngoc-Anh,Oreste, Pierpaolo,Dias, Daniel,Antonello, Croce,Djeran-Maigre, Irini,Livio, Locatelli Techno-Press 2014 Geomechanics & engineering Vol.7 No.1

The application of the mechanized tunnelling has been extended in recent years. There are at present different approaches that are used in the design of segmental tunnel linings supported in mechanized tunnels. Even though segmental lining is utilized for mechanized tunnels, its behaviour is still quite unclear under in situ stress and there is a lack of data regarding the distribution of stresses inside segmental linings. So far no single effective calculation method exists for segmental lining design. The lack of clear solutions makes the use of segmental lining to be more expensive due to the adoption of greater safety factors. Therefore, a particular attention must be given in order to obtain data from monitored tunnels which permits to validate design methods. In this study, strain measurements, which were conducted during the construction of twin tunnels in the Bologna-Florence railway line, have been presented. The behaviour of segmental lining during the excavation and the influence of a new tunnel excavation on an existing tunnel have been shown through the measured data. The data are then compared with the results obtained with Einstein and Schwartz's method and Duddeck and Erdmann's method, which permits to highlight the fact that the two analytical methods underestimate structural forces induced in the segmental lining and then must be used with caution.