Loss Of Vacuum Accident Research Articles

Employing adaptive fuzzy computing for RCP intelligent control and fault diagnosis

Loss of coolant accident (LOCA), loss of fluid accident (LOFA), and loss of vacuum accident (LOVA) are the most severe accidents that can occur in nuclear power reactors (NPRs). These accidents occur when the reactor loses its cooling media, leading to uncontrolled chain reactions akin to a nuclear bomb. This article is focused on exploring methods to prevent such accidents and ensure that the reactor cooling system remains fully controlled. The reactor coolant pump (RCP) has a pivotal role in facilitating heat exchange between the primary cycle, which is connected to the reactor core, and the secondary cycle associated with the steam generator. Furthermore, the RCP is integral to preventing catastrophic events such as LOCA, LOFA, and LOVA accidents. In this study, we discuss the most critical aspects related to the RCP, specifically focusing on RCP control and RCP fault diagnosis. The AI-based adaptive fuzzy method is used to regulate the RCP’s speed and torque, whereas the neural fault diagnosis system (NFDS) is implemented for alarm signaling and fault diagnosis in nuclear reactors. To address the limitations of linguistic and statistical intelligence approaches, an integration of the statistical approach with fuzzy logic has been proposed. This integrated system leverages the strengths of both methods. Adaptive fuzzy control was applied to the VVER 1200 NPR-RCP induction motor, and the NFDS was implemented on the Kori-2 NPR-RCP.

Numerical Simulation of Airflow and Dust Migration in Vacuum Vessel During Loss of Vacuum Accident for Updated CFETR

A loss of vacuum accident (LOVA) occurs during in-vessel component failure and air ingress. The airflow characteristics of a LOVA are determined by many factors like initial pressure, location of a break, and size of a break and have a great impact on dust migration, which could cause a serious explosion with incoming air and H2. In this paper, a computational fluid dynamics method is adopted, and the k-ε Shear Stress Transport model for airflow and the Discrete Phase Model for dust are used to simulate a LOVA with the updated Chinese Fusion Engineering Test Reactor (CFETR) tokamak device. The effects of initial pressure, break size, and break location on airflow during the LOVA are discussed, and the effects of dust size, break size, and break location on dust migration during the LOVA are investigated as well. The results indicate that the initial pressure and size of a break have a greater impact on airflow of a LOVA than the location of the break and that both the dust size and the characteristics of the airflow have a greater impact on the distribution of the dust. A break located in the upper port has even more dust chaos. This research is the basis for the safety analysis of the CFETR device, and it provides a reference for subsequent studies on dust removal, mitigation of dust explosions, and radioactive substances.

In vitro analysis of the cytotoxic effect of two different sizes ITER-like tungsten nanoparticles on human dermal fibroblasts

BackgroundBased on the current configuration of the International Thermonuclear Experimental Reactor, tungsten (W) was chosen as the armour material. Nevertheless, during operation, the expected power and temperature of plasma can trigger the formation of W dust in the plasma chamber. According to the scenario for a Loss Of Vacuum Accident (LOVA), in the case of confinement failure dust is released, which can lead to occupational or accidental exposure. MethodsFor a first evidence of potential risks, fusion devices relevant W dust has been produced on purpose, using a magnetron sputtering gas aggregation source. We aimed to assess the in vitro cytotoxicity of synthesized tungsten nanoparticles (W-NPs) with diameters of 30 and 100 nm, on human BJ fibroblasts. That was systematically investigated using different cytotoxic endpoints (metabolic activity, cellular ATP, AK release and caspase-3/7 activity) and by direct observation with optical and scanning electron microscopy. ResultsIncreasing concentrations of W-NPs of both sizes induced cell viability decrease, but the effect was significantly higher for large W-NPs, starting from 200 μg/mL. In direct correlation with the effect on the cell membrane integrity, high concentrations of large W-NPs appear to increase AK release in the first 24 h of treatment. On the other hand, activation of the cellular caspase 3/7 was found significantly increased after 16 h of treatment solely for low concentrations of small W-NPs. SEM images revealed an increased tendency of agglomeration of small W-NPs in liquid medium, but no major differences in cells development and morphology were observed after treatment. An apparent internalization of nanoparticles under the cell membrane was also identified. ConclusionThese results provide evidence for different toxicological outputs identified as mechanistic responses of BJ fibroblasts to different sizes of W-NPs, indicating also that small W-NPs (30 nm) display lower cytotoxicity compared to larger ones (100 nm).

Highly underexpanded helium jet and its effect on dust resuspension under in-vessel loss of coolant accident of fusion reactor

During an in-vessel loss of coolant accident (LOCA) of a helium-cooled blanket in a fusion reactor, the high-pressure helium gas ingress into the vacuum vessel (VV) will produce highly underexpanded jet, resulting in the resuspension and migration of the radioactive dust deposited in the VV. In this study, the flow field in the VV and the influence of friction velocity were simulated and analyzed by using ANSYS Fluent code. In addition, the differences between in-vessel LOCA and loss of vacuum accident (LOVA) were compared. The results showed that the pressure of the VV reaches the pressure relief limit of the VV pressure suppression system rapidly after a LOCA. With the development of a LOCA, the Mach disk continued to move towards the equatorial break of the VV, its shape continued to shrink and the number of structures gradually increased. Compared with the LOVA, the LOCA had a higher friction velocity, resulting in a greater probability of dust resuspension and radioactive risk.

SHS Synthesis, SPS Densification and Mechanical Properties of Nanometric Tungsten

Recent studies have shown that low grain sizes are favorable to improve ductility and machinability in tungsten, as well as a resistance to ablation and spallation, which are key properties for the use of this material in a thermonuclear fusion environment (Tokamaks such as ITER). However, as one of the possible incidents during Tokamak operation is the leakage of air or water from the cooling system inside the chamber, resulting in the so-called loss of vacuum accident (LOVA), extensive oxidation may arise on tungsten components, and the use of an alloy with improved oxidation resistance is therefore highly desirable. As current production routes are not suitable for the fabrication of bulk nanostructured tungsten or tungsten alloys samples, we have proposed a new methodology based on powder metallurgy, including the powder synthesis, the densification procedure, and preliminary mechanical testing, which was successfully applied to pure tungsten. A similar study is hereby presented on tungsten-chromium alloys with up to 6 wt.% Cr. Results show that full tungsten densification may be obtained by SPS at a temperature lower than 1600 °C. The resulting morphology strongly depends on the amount of the alloying element, presenting a possible second phase of chromium oxide, but always keeps a partial nanostructure inherited from the synthesized powders. Such microstructure had previously been identified as being favorable to the use of these materials in fusion environments and for improved mechanical properties, including hardness, yield strength and ductility, all of which is confirmed by the present study.

Numerical simulations of radioactive dust particle releases during a Loss Of Vacuum Accident in a nuclear fusion reactor

A loss of vacuum accident is one of the possible phenomena that may be triggered by off-normal events in nuclear fusion reactors, such as disruptions. After a break of the isolation system and the formation of a penetration line, the air flows inside the Tokamak due to the large pressure gradient, resuspending toxic and radioactive dust. At the end of the event, an over pressurisation of the vacuum vessel may occur due to inertia effects and the heating of the air through the first wall, that may have still high temperatures. Thus, air will flow from the inside to the outside, dragging the dust particles and involving a radiological and toxic release. This work aims to provide, by numerical simulations, an analysis of particle release as a function of main parameters that characterise the release: diameter of the penetration line, wall temperature and particle sizes. The numerical simulation is based in a one-way coupling. Starting from the simulation of the velocity, pressure and density fields of the continuous phase(air) by a monophasic compressible CFD model, the motion of the particles inside the vessel is simulated.

Evaluation of different models for turbulent combustion of hydrogen-air mixtures. Large Eddy Simulation of a LOVA sequence with hydrogen deflagration in ITER Vacuum Vessel

• LES with TFM model was validated vs H2 experiments. • Validated model was used to simulate ITER VV LOVA sequence with H2 combustion. • In LOVA sequence burning speed increased 1–3.5 times due to flame acceleration. • Maximum average VV pressure was 4–5.5 times to the initial one. The modelling of premixed combustion in the presence of a turbulent field in three-dimensional (3-D) confined scenarios was studied in this work, and applied to hydrogen combustion within ITER vacuum vessel (VV). Two different combustion approaches were tested with Large Eddy Simulation: a Flamelet Progress Variable (LES-FPV) and a Thickened Flame Model (LES-TFM). For the case of LES-TFM modelling, Dynamic Adaptive Chemistry (DAC) with a detailed kinetic mechanism for hydrogen combustion and in-situ adaptive tabulation (ISAT) methods were employed. Moreover, an adaptive meshing technique was used with the aim of tracking the flame front to ensure an adequate spatial resolution in this region. Experimental validation was performed to assess the ability of the different studied approaches to predict the flame burning speed, flame acceleration, and pressure evolution for lean H 2 -Air volume percent mixtures from 16 to 28 % propagating within a turbulent field. Results revealed that both approaches led to accurate predictions in terms of flame burning speed. When considering DAC and ISAT methods with detailed chemistry, LES-TFM model was found to be a cost-efficient solution. This model was used to analyse two loss of vacuum accident (LOVA) sequences within ITER VV. Results showed that turbulence might increase the flame burning speed by a factor of up to 3.5 for the case of big breaches (0.15 m 2 ) but it would not affect in case of breaches of 0.02 m 2 . Besides, results showed that autoignition with 2 kg of H 2 within the VV at 13.35 kPa might degenerate in detonation with average wall pressure levels ∼70 kPa.

Particle behavior close to the lower wall of a tokamak type geometry during a loss of vacuum accident

The behavior of tungsten and beryllium particles close to the wall of a tokamak type geometry (i.e. a toroidal vacuum vessel) during a loss of vacuum accident (LOVA) is explored in this paper.Values of the flow field and temperature of the air surging into the torus were calculated by Computational Fluid Dynamics (CFD) simulations from an initial low pressure of 1000 Pa by Gelain et al., Fusion Engineering and Design, 100:87–99, 2015. A new CFD calculation is performed here for an initial low pressure of 500 Pa. The aerodynamic forces on particles in rarefied flow are derived from the calculated friction velocity and temperature in the lowest wall region. The focus is on spherical particles, but these values of forces are complemented for comparison with an estimate of the force on an elongated particle following Sentman (1961). Typical particles with diameters of 1, 2, 5, 10, 20 μm are considered. The possibility for particles which are detached from the lower wall to be entrained away the wall is explored. A particle may be detached by the flow field provided it escapes the adhesion force with the wall, which occurs only for a rough wall and large particles. It is then entrained away from the lower part of the wall provided its weight is smaller than the aerodynamic force, which occurs even at low pressure for all sizes of studied beryllium particles and only for specific sizes and a pressure above 1500 Pa for tungsten particles. This key influence of particle weight on the particle dynamics in rarefied gas flow is the focus of the paper. As a result, only particles of intermediate diameters may be resuspended and transferred away from the wall during the whole pressurisation sequence of the vessel. Using the data for the initial low pressure of 500 Pa, these particle diameters are 5, 10 μm for tungsten and 5, 10, 20 μm for beryllium.

CFD modelling of particle resuspension in a toroidal geometry resulting from airflows during a loss of vacuum accident (LOVA)

During a loss of vacuum accident (LOVA), dusts that will be present in the future tokamak ITER are likely to be resuspended. Such dusts may present a risk of explosion and airborne contamination. Hence, an assessment of the amount of particles which may be resuspended in case of LOVA is a major safety issue in ITER and more widely for all the fusion reactors. This article presents the implementation of the Rock’n’Roll model for the particle resuspension in ANSYS CFX and its coupling with a validated particles transport and deposition model.After reminding the main phenomena contributing to the particle resuspension and their equations, this article proposes in a first time a checking of the good implementation of the resuspension model in ANSYS CFX. To do that, numerical results are compared to analytical calculations and experimental results acquired on facilities dedicated to the study of the resuspension.After this first step, this article proposes an application for a LOVA scenario in a toroidal vessel in order to demonstrate the feasibility of this kind of calculation in conditions similar to those encountered in a tokamak type geometry and the robustness of the coupling between the particle transport model and the resuspension model.The final application for modelling a LOVA in a toroidal geometry with different particle sizes shows a good behaviour of the coupled resuspension and transport models and a good agreement of the results with the expected physics.However, it must be reminded that this numerical application is a first step towards the analysis of dust resuspension and transport in a tokamak like ITER and the interpretation of quantitative results presented here should remain in this specific context.

Design of model tokamak particles for future toxicity studies: Morphology and physical characterization

During ITER operation, it is expected that the large panel of plasma-wall interactions triggers the production of dust particles from the tungsten and beryllium first wall. These particles can be loaded with tritium (T) used as a fuel for the fusion reaction; T inventory is of prime importance for the safety assessment, particularly when considering dispersible matters that could be released in case of a Loss Of Vacuum Accident (LOVA),. The impact of accidental inhalation of such particles has therefore to be evaluated. Yet, particles properties in terms of tritium inventory and general behaviour are strongly dependent on their physical and chemical characteristics, and large uncertainties remain on those parameters; It is therefore important to obtain and characterize relevant W dust samples for toxicity studies before ITER operation starts. We produced model tungsten (W) Nano-Particles (NPs) by two different methods with characteristics closer to the plasma-wall interaction processes than grinding (usually used to produce commercial NPs): (i) magnetron plasma sputtering followed by gas condensation as well as (ii) laser ablation. The two types of tungsten NPs obtained exhibit very different properties investigated by a large characterization techniques panel; but both sets show similarities with samples collected in tokamak or observed in dusty plasmas setups, therefore could potentially be expected in ITER. Producing samples with strong differences is an asset to lead a first global study, highlighting which parameters are key, whether it affects the tritium inventory or the toxicity for lung cells during in vitro studies. In a second phase, it will help precise the impact of ITER particles when their definition will clarify. The aim of this paper is therefore to study and describe the characteristics of different types of W dusts, this knowledge being mandatory for the next steps of the project dealing with tritium inventory in the described W dusts and their suspension in various liquid media for cell exposure for toxicity studies.

Numerical analysis of loss of vacuum accident (LOVA) and preliminary discussion about dust resuspension for CFETR

Loss of vacuum accident (LOVA) is the air ingress into the vacuum vessel (VV) due to the failure of components. The high speed air jet flow induced by the pressure difference between the ambient condition and the in-vessel condition may cause the dust migration, re-suspension bring risk to the integrity of components and radioactive release. The present work deals with the computational fluid dynamics (CFD) modeling of airflow characteristics in different scenarios during the LOVA for CFETR. Then, forces applied on the deposited particles are investigated, based on which the preliminary discussion are carried out to analyze the possibility of dust resuspension in various LOVA scenarios. The results show that the break location, break size and the operational status during the LOVA have significant influence on the key parameters related to airflow forces such as the friction velocity. Furthermore, the airflow force gradually plays the dominate role as the particle diameter increases. It preliminarily demonstrates the possibility of dust resuspension as the particle diameter changing. The results can provide references to study the dust re-suspension and implement accident mitigation measures.

A scaling law of pressurisation time in the case of Loss Of Vacuum Accidents (LOVAs): Theoretical and experimental analysis

The Loss Of Vacuum Accident is one of the hazardous events that may occur in nuclear fusion power plants which work with the magnetic confinement technology. During these accidents, the intake of air may re-suspend toxic, explosive and radioactive dust, involving a hazardous release in the external environment. Several studies have been conducted in scaled facilities, in order to investigate and understand how the phenomenon works. In the previous studies, the authors replicated these events in several different conditions by means of the scaled facility called “STARDUST-Upgrade”, providing a good description of the phenomenon. A Computational Fluid Dynamic (CFD) code able to replicate these events has been developed and validated. Once a scaled experiment is realized, a specific analysis must be performed to understand if the phenomenon can be scaled and in what conditions. A result obtained in STARDUST-Upgrade, that is much different in terms of size and shape from a real TOKAMAK, must be modified by an adequate scaling law. The aim of this work is to analyse how to deduce a scaling law (for large scale flow) by a theoretical analysis based on the Buckingham π theorem. The authors will present also the experimental analysis provided to validate some features of this model. The conclusion of the work will be the use of the scaling law to obtain technical information (pressurisation time) about a LOVA inside an ITER-like reactor and further fusion power plants.

Experimental measurements of pressure, temperature and dust velocities in case of LOVA: Comparisons with a multiphase numerical model

The production of dust inside the nuclear fusion power plants is one of the safety issues of this technology. Dust is generated because of plasma-material interactions and subsequently it deposits in the bottom regions of the TOKAMAK. In case of a Loss Of Vacuum Accident (LOVA), the dust may be resuspended, threatening the functioning and the safety of these reactors. A deep study of this phenomenon is required to develop countermeasures and to improve the safety of this promising way to produce energy. The authors have studied the fluid dynamics of these accidents with a scaled experiment, called STARDUST-Upgrade. Optical techniques have been implemented to measure dust resuspension and diffusion properties, such as velocity vectors and resuspension rate. This work shows the results obtained with a new numerical model able to take into account also the dispersed phase (dust). The software uses the Euler-Euler approach, a Schiller-Naumann resistance model, and a k-ε turbulence model. The dust used is tungsten dust, that has been placed close to the inlet valve in both cases (numerical and experimental). The numerical results are analysed and compared with the experimental ones and the main agreements and differences are highlighted. The results show good accordance with the velocity vectors of dust, while the resuspension rate is overestimated in the numerical case because of the absence of adhesion and cohesion forces between dust particles and walls. This analysis is the starting point for the evolution and completion of a numerical model suitable for dust resuspension in case of LOVAs.

Investigation of vibrations caused by the steam condensation at sub-atmospheric condition in VVPSS

Fusion technology deployment passes through the design of safety systems aimed to protect the Vacuum Vessel (VV) from pressurizing accidents event such as LOVA (Loss Of Vacuum Accident) or the failure of the Tokamak Water Cooling System (TWCS) causing the LOCA (Loss Of Coolant Accident). One of important safety systems of the ITER plant is the Vacuum Vessel Pressure Suppression System (VVPSS), which is designed to protect the VV by the steam condensation at sub-atmospheric pressure conditions in Suppression Tanks.The aim of this study is to investigate vibrations phenomenon, originated during the steam-water direct contact condensation (DCC) at sub-atmospheric conditions, and determine any correlations between the steam jet dynamic and water temperature (TW), steam mass flux (GS) and downstream pressure (PW).According to the thermal hydraulic conditions characterising the DCC, vibrations may evolve because of the latent heat of the phase change (large amount of heat transmitted quickly to the water).In the paper also presents the numerical investigations performed by means of proper FEM code, the set-up procedure and the experimental activity carried out at Lab. B. Guerrini of DICI- University of Pisa. This latter allowed correlating the different steam condensation regimes (chugging, condensation oscillation, etc.) to the acceleration and force, to be in turn used to determine the strength capacity of the VVPSS.

Experimental Apparatus, Analyses and Comparisons of the Feature Matching and Lucas-Kanade Software to Measure Dust Particle Velocities

The mobilisation of dust during Loss of Vacuum Accidents is crucial to provide detailed answers to security and safety issues of many nuclear and industrial plants. The experimental apparatus STARDUST-Upgrade can replicate Loss of Vacuum Accidents in different conditions and the fluid dynamics related to these experiments has been deeply investigated. In this work, the authors provide a detailed description of the optical experimental setups and software developed and used to track dust particle velocities. These techniques are based on the particle tracking velocimetry. Two different experimental setups are used to take into account the different optical properties of dust, each image obtained during the experiments has been analysed with customized software. The algorithms used to develop the software are the feature matching algorithm and the Lucas-Kanade algorithm. In this paper, the two algorithms and their results are discussed separately and then a comparison of the data obtained is presented.

Numerical simulation of airflow characteristics during the loss of vacuum accident of CFETR

During a loss of vacuum accident (LOVA) of China Fusion Engineering Test Reactor (CFETR), the high velocity airflow will result in the migration, re-suspension even explosion of the radioactive dust deposited in the vacuum vessel (VV) during plasma burning. In addition, large amounts of hydrogen may be produced from the exothermic reaction between tungsten and water/steam. In order to minimize the risk of explosion and the leakage of radioactivity, the airflow characteristics must be well studied at the first step. In this paper, a break due to a failed component is assumed to take place at the equatorial port. The numerical simulation has been performed by using ANSYS CFX code with a simplified VV model of CFETR. The airflow field during LOVA has been studied in detail under different initial pressure and first wall (FW) temperature. The results show that the FW temperature significantly affects the pressurization process. Besides, the friction velocity is also greatly affected by the FW temperature and initial pressure. From the 3D airflow distribution, the velocity near the lower part of torus can reach up to 87.4 m/s. This study provides indispensable basis for the research about dust resuspension, migration, dust/hydrogen explosion and the radiological safety of CFETR.

3D Simulation of a Loss of Vacuum Accident (LOVA) in ITER (International Thermonuclear Experimental Reactor): Evaluation of Static Pressure, Mach Number, and Friction Velocity

ITER (International Thermonuclear Experimental Reactor) is a magnetically confined plasma nuclear reactor. Inside it, due to plasma disruptions, the formation of neutron-activated powders, which are essentially made out of tungsten and beryllium, occurs. As many windows for diagnostics are present on the reactor, which operates at very low pressure, a LOVA (Loss of Vacuum Accident) could be possible and may lead to dust mobilisation and a toxic and radioactive fallout inside the plant. This study is aimed at reproducing numerically the first seconds of a LOVA in ITER, in order to get information about the dust resuspension risk. This work has been carried out by means of a CFD (Computational Fluid Dynamics) simulation of the beginning of the pressurisation transient inside the whole Tokamak. It has been found that the pressurization transient is extremely slow, and that the friction speed on the walls is very high, and therefore a high mobilization risk of the dust is expected on the entire internal surface of the reactor. It has been observed that a LOVA in a real-scale reactor is more severe than the one reproduced in reduced-scale facilities, as STARDUST-U, because the speeds are higher, and the dust resuspension capacity of the flow is greater.

DNS Study of Dust Particle Resuspension in a Fusion Reactor Induced by a Transonic Jet into Vacuum

This paper reports on a two-phase flow Direct Numerical Simulation (DNS) aimed at analyzing the resuspension of solid particles from a surface hit by a transonic jet inside a low pressure container. Conditions similar to those occurring in a fusion reactor vacuum vessel during a Loss of Vacuum Accident (LOVA) have been considered. Indeed, a deep understanding of the resuspension phenomenon is essential to make those reactors safe and suitable for a large-scale sustainable energy production. The jet Reynolds and Mach numbers are respectively set to 3300 and 1. The Thornton and Ning impact/adhesion model is adopted and improved. An advanced resuspension model, which takes into account the dynamics (rolling and slipping) of particles at the wall, is implemented. The use of this model combined with a DNS represents a great novelty in simulating the particle resuspension process. The particles initially deposited at the wall have constant density, whereas their diameters are drawn according to a log-normal distribution, with parameters obtained from experimental data. It has been found that the flow induced motion of wall deposited particles is highly linked with the instantaneous fluid structures and the resuspension phenomenon predominantly affects particles with the largest diameters. Moreover, the jet-deposit interaction is mostly confined within a circumference around the jet of radius approximately equal to the jet diameter.

Imaging of dust re-suspension in case of LOVA

Dust re-suspension and mobilization is one of the more challenging safety issues in industries and fusion reactors (Tokamaks). In Tokamaks, plasma-material interactions generate dust by several different mechanisms called Plasma-Material Interactions. In these reactors, Loss of Vacuum Accidents (LOVA) may occur and imply dust re-suspension and mobilization. These events must be avoided since dust could involve reactor function problematics, toxic and radiological releases and dust explosions. The Quantum Electronics and Plasma Physics (QEP) research group is studying the dust mobilization through an experimental device, Small Tank for Aerosol Removal and DUST Upgrade (STARDUST-Upgrade), able to reproduce LOVAs. An optical experimental apparatus has been implemented on STARDUST-U in order to measure dust mobilization and re-suspension. In this work, the authors show the results of dust re-suspension measurements in case of LOVA reproduced in mid-height of the vessel (valve E). The experimental results are argued referring to numerical data from a monophase Computational Fluid Dynamics (CFD) and they show the meaning of dust initial position and dust properties in these kinds of events. The authors discuss the impossibility to adopt simple numerical models, such as 1-way coupling model, to describe, accurately, the phenomenon.

Optical measures of dust velocities and direction during loss of vacuum accidents in confined environment and correlation between dust positions and properties with the resuspension degrees and the velocity modules

Dust explosions are dangerous events that still today represent a risk to all the industries that produce and/or handle combustible dust like the agro-alimentary, pharmaceutical and energy ones. When a dust cloud is dispersed in an oxidant gas, like air, it may reach the explosive concentration range. A model to predict the dust critical conditions, that can cause explosions, is a key factor for safety of operators and the security of the plants. The key point to predict this dust resuspension is to measure the velocity vectors of dust under the accidental conditions. In order to achieve this goal the authors have developed an experimental facility, STARDUST-U, which allow to obtain different conditions of temperature and pressurization rates characteristic of accidents in confined environment. The authors have developed also optical methods and software to analyse different dust resuspension phenomena under different conditions in confined environment. In this paper, the author will present how they measure the dust velocity vectors in different experimental conditions (and for different type of dusts) and how they have related the dust characteristics and positions inside STARDUST-U with the resuspension degree and the velocity values.