Limiting Operating Conditions Research Articles

Improving the out-of-sample generalization ability of data-driven chiller performance models using physics-guided neural network

Modeling of the chiller performance is essential for the implementation of optimal energy-efficient control strategies in a heating, ventilation, and air conditioning (HVAC) system. Though classical data-driven chiller performance models are widely adopted in the industry, they generally suffer from poor out-of-sample generalization abilities, which refers to the model's capability to extrapolate for new data outside the range of the training dataset. In practice, however, the available chiller operation data for model development are often insufficient or collected from a few limited operating conditions, such that extrapolation is unavoidable after the model is applied for control purposes. To deal with this issue, this paper proposed a physics-guided neural network (PGNN) to model the energy performance of chillers. By adopting a new neural network architecture, modifying the loss function, and adding limited out-of-sample data, the PGNN incorporates domain knowledge into the data-driven model to achieve better out-of-sample generalization performance. Meanwhile, the convexity and monotonicity between the dependent and independent variables in the PGNN are properly addressed. The proposed PGNN is applied to model the chiller serving a high-rise building, and results show that PGNN performs much better in extrapolation than classical models and the multi-layer perceptron model. The research demonstrated the usefulness and effectiveness of the PGNN in modeling HVAC equipment.

A Novel Fault Diagnosis Method of Rolling Bearings Combining Convolutional Neural Network and Transformer

Efficient and accurate fault diagnosis plays an essential role in the safe operation of machinery. In respect of fault diagnosis, various data-driven methods based on deep learning have attracted widespread attention for research in recent years. Considering the limitations of feature representation in convolutional structures for fault diagnosis, and the demanding requirements on the quality of data for Transformer structures, an intelligent method of fault diagnosis is proposed in the present study for bearings, namely Efficient Convolutional Transformer (ECTN). Firstly, the time-frequency representation is achieved by means of short-time Fourier transform for the original signal. Secondly, the low-level local features are extracted using an efficient convolution module. Then, the global information is extracted through transformer. Finally, the results of fault diagnosis are obtained by the classifier. Moreover, experiments are conducted on two different bearing datasets to obtain the experimental results showing that the proposed method is effective in combining the advantages of CNN and transformer. In comparison with other single-structure methods of fault diagnosis, the method proposed in this study produces a better diagnostic performance in the context of limited data volume, strong noise, and variable operating conditions.

Considerations about the gas fraction in two-phase electrochemical reactors by using a rigorous hydrodynamic model

The experimental evidence that a limiting value of gas fraction is achieved in a gas-liquid dispersion as the gas volumetric flow rate is increased is explained by means of a rigorous hydrodynamic model. The mathematical treatment, based on the Eulerian concept applied to each phase, considers a population balance with a range of bubble sizes allowing the coalescence and breakup between them; and it is implemented in the OpenFOAM framework. It is concluded that the limiting gas fraction is a consequence of the increase of the gas velocity and the constancy of the drag term of the interfacial exchange moment as the volumetric gas flow is enlarged. A simplified hydrodynamic model is also reported that closely matches both the rigorous one and the experimental results under limiting operating conditions.

Inverse Curves - Research on Two Quondam Inversor Mechanisms

The field of research considered in this work is that of mechanisms generating inverse curves. We were motivated to approach this field, proposing variants of mechanisms based on Artobolevsky’s inversor, in blocking conditions, due to the lack of recent applications related to inversor mechanisms that draw curves other than straight lines. Based on Crawford’s inversor, Artobolevsky’s inversor is structurally and kinematically analysed in what follows. Hence, we carried out the kinematic analysis and simulation of four inversor mechanisms derived from Artobolevsky’s inversor, which draw different initial curves (a circle, a straight line, an ellipse and an Archimedes’ spiral) were performed. Inverse curves were generated under limited operating conditions due to geometrical dimensions and kinematic parameters. Given the researchers’ increasing interest in the development of applications related to inversor mechanisms (in fields such as: robotics, machine tools, visual arts and symbolic languages), as well as expanding databases of planar and spatial mechanisms, it is suggested that there is a need to create new mechanisms similar to those studied in this paper and to analyse their mode of operation.

A new control parameter to predict micro-warping-induced job failure in LPBF of TI6AL4V titanium alloy

Laser powder bed fusion (LPBF) includes a few printing techniques widely used, in recent years, concerning the additive manufacturing of Ti6Al4V alloys. These produced parts, typically utilized in sectors such as aerospace and biomedical, are characterized by very high added value. It is therefore fundamental to identify the influence of process parameters typical of LPBF technology on the occurrence of warping leading to process failure. This study deals with the characterization of single-track and “micro-scale” level warping phenomena which may lead to protrusion of material over the powder bed and process failure before normal termination. This phenomenon was investigated as a function of process parameters, referring also to the strength and ductility characteristics of the manufactured samples. With this purpose, several samples were printed using variable process parameters both in terms of line energy density (LED) values and in terms of laser power and speed combinations such as to guarantee constant LED values. For the samples that did not show significant micro-warping phenomena, in addition to the transversal and longitudinal geometric characterization of the single track, tensile tests were performed to determine both the resistance of the material and the ductility characteristics. The single tracks, for given process parameters, were printed on a homogeneous material substrate. For every single track, a microstructural and morphological transverse and longitudinal characterization has been carried out and the measured geometrical features were correlated to the process parameters. The obtained results allowed the identification of a new threshold parameter, indicating the limit operating conditions beyond which significant warping phenomena and process failure occur.

Long-term durability test of highly efficient membrane electrode assemblies for anion exchange membrane seawater electrolyzers

Direct seawater electrolysis faces fundamental electrochemical challenges, such as suppression of chlorine adsorption and/or evolution reactions that poison the current state-of-the-art anode catalysts and accelerate the degradation of the anion exchange membrane (AEM). The technically favored solution path for addressing these issues includes development of novel selective catalysts and implementation of limited operating conditions. Therefore, development of highly active, selective, cost-effective, and stable seawater-splitting catalysts is required for a successful commercialization of the AEM water electrolyzers (AEMWEs). Herein, we report for the first-time an extended operation for over 1000 h of advanced AEMWEs, operating directly with seawater. To fabricate the membrane electrode assemblies (MEAs), catalyst coated electrodes (CCEs) are assembled with Sustainion® X37-50 grade T anion exchange membranes. The anode CCEs are fabricated by spraying an ink prepared from nanostructured NiFe-layered double hydroxide (LDH) catalyst that are synthesized by the solvothermal method and deposited directly onto platinized titanium porous transport layers (PTLs). The cathode CCEs are prepared by spraying an ink prepared from commercial Renay-Nickel catalyst onto a nickel fiber felt. As prepared AEM electrolyzer cells operated with seawater and demonstrated better than previously reported in the literature performance of over 1000 h of operation at a constant current density of 300 mA cm−2.

An evolving learning-based fault detection and diagnosis method: Case study for a passive chilled beam system

Traditional fault detection and diagnosis (FDD) methods learn from training data obtained under limited operating conditions, after which they stop learning. In this study, we developed an evolving learning-based FDD method for HVAC systems, which learns as the performance of a building system and its components changes. Specifically, an evolving learning algorithm—growing Gaussian mixture regression—is used to construct both a data-driven model representing normal performance and a transfer function for fault diagnosis. The evolving learning-based FDD method was demonstrated for detecting and diagnosing common faults of passive chilled beam systems. We employ generalized performance indices, such as the deviations between predictions (expectations) and measurements, the differences between two parameters, and other features extracted from parameters. A novel feature selection method was developed for selecting fault signatures. An uncertainty threshold determining whether a performance index was within the range of normal operation influences false alarm rates. By increasing the uncertainty thresholds from zero to two standard deviations, false alarm rates for normal operations were reduced from 14.8% to 1.3% and the percentage of normal operation data categorized as an unknown operation was reduced from 25% to 0%. Eight known faults were detected and diagnosed with an accuracy of 100%. A new fault was first categorized as an unknown fault before evolving. After evolving the transfer function by updating the key parameters of the Gaussian components, the unknown fault was also accurately diagnosed. The evolving learning-based FDD method and novel feature selection method can be employed for detecting and diagnosing common faults of other systems or subsystems in the built environment.

Flow characteristics and flow rate prediction of pulverized coal through a regulation valve

• The optimum valve sweeping gas fraction was determined. • Control characteristics of the regulation valve were studied. • Limit operating conditions of the regulation valve were defined. • A semi-empirical model was established for pulverized coal flow rate prediction. In this paper, experiments of dense-phase pneumatic conveying of pulverized coal were carried out in an industrial-scale system to study the control characteristics of the regulation valve and to predict the solid mass flow rate. Firstly, effects of valve sweeping gas on conveying stability and solid mass flow rate were investigated and the optimum valve sweeping gas was determined. Second, effects of valve opening on pressure distribution and solid mass flow rate were investigated by conducting experiments at different conveying pressure drops and different valve openings. A good linear relationship between the valve pressure drop ratio and the valve opening was found, and as the valve opening increased from 13 % to 70 % the solid mass flow rate increased gradually. Limit operating conditions of the regulation valve including flow blockage and control failure were consequently determined and analyzed. Finally, a robust model was established to predict the solid mass flow rate by introducing the valve sensitivity coefficient into the traditional pressure drop ratio model. The model can predict the solid mass flow rate well by providing errors mostly within ± 10 %. This study will provide certain reference for solid mass flow rate regulation in the dry coal gasification process.

Retrieval methodology for similar NPP LCO cases based on domain specific NLP

Nuclear power plants (NPPs) have technical specifications (Tech Specs) to ensure that the equipment and key operating parameters necessary for the safe operation of the power plant are maintained within limiting conditions for operation (LCO) determined by a safety analysis. The LCO of Tech Specs that identify the lowest functional capability of equipment required for safe operation for a facility must be complied for the safe operation of NPP.There have been previous studies to aid in compliance with LCO relevant to rule-based expert systems; however, there is an obvious limit to expert systems for implementing the rules for many situations related to LCO. Therefore, in this study, we present a retrieval methodology for similar LCO cases in determining whether LCO is met or not met. To reflect the natural language processing of NPP features, a domain dictionary was built, and the optimal term frequency-inverse document frequency variant was selected. The retrieval performance was improved by adding a Boolean retrieval model based on terms related to the LCO in addition to the vector space model. The developed domain dictionary and retrieval methodology are expected to be exceedingly useful in determining whether LCO is met.

One-Step Synthesis of Selective Nife-Layered Double Hydroxide Anode Catalyst for Highly Efficient and Stable Anion Exchange Membrane Water Electrolyzers, Operating with Seawater

Seawater electrolysis faces unique and fundamental chemical challenges, such as the suppression of highly detrimental halogen chemistries [1], that poison the current state-of-the-art anode catalyst and accelerate the degradation of the membrane [2]. The technically favored solution path is to address these issues by developing a new selective catalyst and possibly by limiting operating conditions [3]. Therefore, developing a facile and cost-effective way to design highly active, selective, and and stable seawater-splitting catalysts is of great importance for research and industry [4]. In this work, hexagonal NiFe-layered double hydroxide (LDH) nanosheets were synthesized by one-step solvothermal reactions. As synthesized NiFe-LDH catalyst exhibits high activity, selectivity, and stability towards the oxygen evolution reaction (OER) in alkaline electrolyte, delivering current densities of 10 mA/cm2 at low overpotentials of 359 mV, with no significant degradation observed during RDE testing at constant voltage of 1.57 V for 72 hours in 1.0 M KOH electrolyte. In addition, in order to fabricate catalyst coated electrode (CCE), the nanostructured NiFe-LDH catalyst was homogeneously sprayed onto a platinized titanium porous transport layer (PTL) using an ultrasonic spray coating system. As prepared CCE is used as an anode and commercially available Raney nickel catalyst deposited on a nickel fiber frit is used as a cathode to assemble a MEA with Sustainion® X37-50 grade T anion exchange membrane (AEM) that was tested in alkaline seawater. Employing local seawater, the AEM electrolyzer cell demonstrated stable performance for over 1000 hours at a constant current density of 300 mA cm -2. Reference: [1] S.-C. Ke, R. Chen, G.-H. Chen, X.-L. Ma, Energy Fuels 2021, 35, 12948.[2] S. Dresp, F. Luo, R. Schmack, S. Kühl, M. Gliech, P. Strasser, Energy Environ. Sci. 2016, 9, 2020.[3] S. Dresp, F. Dionigi, S. Loos, J. Ferreira de Araujo, C. Spöri, M. Gliech, H. Dau, P. Strasser, Advanced Energy Materials 2018, 8, 1800338.[4] H. Koshikawa, H. Murase, T. Hayashi, K. Nakajima, H. Mashiko, S. Shiraishi, Y. Tsuji, ACS Catal. 2020, 10, 1886.

Post-oxidation phenomena enhancement with scavenging and secondary air injection in exhaust manifold of a turbocharged GDI engine

A detailed investigation of post-oxidation phenomena by individual and combined effects of scavenging (VVT tuning) and secondary air injection (SAI) was performed. The 1-D simulation including a post-oxidation model developed by Stuttgart University as an international collaboration was used for investigation which includes the main exhaust gas species as CO, H2, and O2-based chemical reactions. Then, experimental validation was conducted on a 4-cylinder turbocharged gasoline direct injection (GDI) engine. From the results, it was noted that the post-oxidation can be actuated at limited operating conditions as higher overlap in moderated speed and load only when scavenging phenomena are considered. However, at lower overlap, it is restricted due to lower O2 scavenging even though the exhaust temperature meets the post-oxidation requirement. Also, the in-homogeneity observed at higher overlap that restricts the significant post-oxidation before the turbocharger upstream. On the other hand, the SAI mechanism can actuate the post-oxidation even at lower overlap if enough O2 concentration, exhaust temperature, and adequate mixing are attained. Hence, the post-oxidation zone can be extended to lower speed-load and overlap if both parameters as scavenging and SAI introduced together. This can possibly lead to better turbo-performance along with lower emissions. However, thermal efficiency needs to be compromised to some extent. It was also found that the effective post-oxidation can be actuated by SAI compared to scavenging-based phenomena if the same concentration of the O2 and temperature are maintained by both mechanisms. This appeared due to the fresh air continuously injected at the exhaust port even at the time of exhaust valve opening duration in SAI mechanism that allows the better mixing of O2 and hot unburned gas species. However, in scavenging-based phenomena, firstly, hot unburned gas passed through the exhaust manifold and then scavenged air follows which restricts mixing between scavenged air and unburned gas species.

Experimental performance evaluation of a convective thermoelectric building envelope for building heating and cooling

The thermoelectric building envelope (TBE) integrates thermoelectric materials with the building envelope for active space heating and cooling. The advantage of TBE heating and cooling includes its significantly low-profile design and no refrigerant use. Although there are existing studies evaluating TBE performance, they were based on limited operating conditions. The study aims to experimentally evaluate the heating and cooling performance of a TBE prototype under various operating conditions. The TBE prototype was installed between two psychrometric chambers, which simulated indoor and outdoor conditions. The prototype was tested at an indoor temperature of around 22.35–23.58 °C and outdoor temperatures from −7.35 °C and 16.99 °C for heating and from 28.36 °C to 40.95 °C for cooling, with varied power inputs and fan conditions. The maximum coefficient of performance (COP) of TBE in heating mode is 3.2. The average heating COP of TBE with a current of 1.5 A in four winter scenarios is 1.37. The average heating COP of TBE operating with the current of 0.3–1.5 A at an outdoor temperature of 12 °C is 2.27. The TBE system demonstrates a better heating efficiency than an auxiliary electric heater for the heat pump system. The experimental results and evaluation obtained provide critical guidance for the deployment of TBE applications.

Mass transport limitations in polymer electrolyte water electrolyzers using spatially-resolved current measurement

This work utilizes spatially-resolved current measurements to provide insight into mass transport limitations in electrolyzers that are not observable from traditional polarization measurement. In this study, two types of flow-fields (parallel and triple-serpentine) and two types of diffusion media (patterned porous thin titanium foil LGDLs and Ir-coated titanium felt PTLs) were examined. A non-uniform current distribution dominated by mass transport limitations was observed to be instigated by the restriction of liquid water transport to catalyst sites. Additionally, conditions are revealed which yield similar polarization performance but dissimilar current distributions. In such cases, the transport limitations for different architectures and porous media affect polarization in different regions of the active area. Furthermore, the triple-serpentine flow-field used in this study performs better than the parallel flow-field under mass transport limited operating conditions. This indicates that the parallel flow-field used in this study is more susceptible to starvation than the triple-serpentine flow-field for the electrolyzer studied.

Research on wet steam condensation flow characteristics of steam turbine last stage under zero output condition

The zero output transformation of the low-pressure cylinder can fully exploit the peak shaving capacity of the unit and solve the problem of the mismatch between power generation peak shaving and heating supply. However, the friction between the low-pressure rotor and the internal working fluid will cause steam ventilation loss, and the flow field in the last stage is extremely unstable. It is necessary to study the condensed flow characteristics and flow field stability of wet steam under small flow rate conditions in the low-pressure stage group to determine the zero-output operating conditions and the steam ventilation limit operating conditions, which can guide the flexible transformation of cogeneration units. The flow field distribution and condensation characteristics of the last stage of the steam turbine under different flow conditions are obtained through numerical calculations by taking the last stage cascade of a 300 MW steam turbine as the research object. The results show that the vortex inside the last stage of the steam turbine is mainly distributed in the rotor cascades. As the flow rate decreases, the flow instability increases. When the flow rate is lower than 20% turbine heat acceptance (THA), vortex flow first appears at the root of the rotor cascade, and then the height of the vortex and separation increase. When the flow rate is lower than 10% THA, a vortex appears in the clearance between the rotor and stationary cascades, and finally, the vortex fills the entire space. The zero output condition appears when the flow rate is 57,000 kg/h, and the steam ventilation limit condition appears when the flow rate is 10,000 kg/h. At this time, the maximum temperature of the rotor cascade reaches 368.15 K. When the flow rate decreases, the condensation strength and the nucleation rate decrease. When the flow rate is lower than 60,000 kg/h, the Markov loss coefficient in the rotor and stationary cascades increases significantly. Based on the analysis of wet steam flow characteristics, condensation characteristics, steam ventilation conditions, and Markov loss coefficient, it is recommended that the steam turbine's last stage inlet steam flow is not less than 57,000 kg/h. The research results can provide a reference for the steam flow design of steam turbine zero-output retrofitting.

Obstacle Avoidance and Profile Ground Flight Test and Analysis for Plant Protection UAV

In recent years, with the further development of agricultural aviation technology, the plant protection UAV has been widely used, especially in some agricultural environments with limited operating conditions due to its advantages of high efficiency, environmental protection and safety guarantee. A plant protection UAV generally flies at low altitude during operation. However, the low altitude operation environment, such as farmland and mountainous areas, is relatively complex, and is faced with many types of obstacles, proposing higher requirements for obstacle avoidance and the profiling system of a plant protection UAV. In order to test the obstacle avoidance and profiling performance of the commercialized plant protection UAV at this stage and explore the performance boundary of obstacle avoidance and profiling of the UAV, EAVISION E-A2021 and XAG P80, the flagship models of the plant protection UAV manufacturer on the market, were hereby selected as the experimental test objects in the paper. Firstly, the obstacle avoidance and profiling test scheme of plant protection UAVs is designed; then, the above two UAVs are adopted for corresponding tests, and the test data are discussed based on the analysis of software and hardware technology; finally, the practical application status of different obstacle avoidance and profiling technologies of plant protection UAVs is clarified, and the shortcomings of obstacle avoidance and profiling technology of plant protection UAVs on the market are summarized, providing a reliable reference for the future development of plant protection UAVs.

Analysis of oil shedding and ligaments formation on bearing rotary elements

For designing efficient lubrication system of an aeroengine bearing chamber, sufficient knowledge on oil/air two-phase flow characteristics is required. When analyzing bearing chamber two-phase flow, the essential prerequisite is quantifying the oil ligaments, which are detached from bearing rotary elements and shed into the bearing chamber. Related investigations are mainly targeted at liquid shedding on the rotating disk as opposed to the bearing rotary elements. Moreover, the research based on bearing rotary elements is conducted by experiment. Due to the limited operating conditions, experimental studies cannot guide engineering applications. To overcome these limitations, a theoretical model is established in this paper, for revealing the mechanism of oil shedding from bearing rotary elements and quantifying the shedding ligaments. The theoretical model is validated against experimental results from classical studies. In addition, the correlation for shedding ligaments number based on aeroengine bearing structural and operational parameters is obtained via theoretical analysis. The analytical results demonstrate that oil shedding and ligaments formation appear at the edge of bearing inner race outer-periphery. The number of shedding ligaments increases with the rise of shaft rotational speed while decreases with the growth of oil viscosity.

Comparison of Heat Transfer Theory, CFD and Experimental Results in the Design Process of High-Power Fiber Laser Cooling Plate

For the stabilization of laser output power and wavelength of the high power fiber laser, the cooling plate must be properly taken into account. In this study, three analyzing methods which are heat transfer theory, CFD and experiment are used to analyze cooling plate performance by measuring pump Laser Diode(LD) temperature. Under limited operating conditions of a cooling plate, the internal flow of cooling plate is transitional flow so that the internal flow is assumed to be laminar and turbulence flow and conducted theoretical calculation. Through CFD, temperature of pump LD and characteristics of the internal flow were analyzed. By the experiment, temperature of pump LD was measured in real conditions and the performance of the cooling plate was verified. The results of this study indicate that three analyzing methods are practically useful to design the cooling plate for the high power fiber laser or similar things.

Thermophilic biodigestion of fermented sugarcane molasses in high-rate structured-bed reactors: Alkalinization strategies define the operating limits

Energetic exploitation of sugarcane-derived byproducts via anaerobic digestion (AD) has recently been highlighted as an alternative to the conventional ethanol-producing approach. In this context, the thermophilic (55 °C) methane production from fermented molasses was assessed in the long-term (>200 d) continuous operation of two anaerobic structured-bed reactors. Two types of alkalinization strategies, namely, NaHCO3 dosing (RM1) and NaOH dosing coupled to effluent recirculation (RM2) were compared, as well as different levels of organic loading rate (OLR) and hydraulic retention time (HRT) were applied to reach limiting operating conditions. Both alkalinization strategies provided an equivalent buffer control up to an OLR of 7.5 kg-CODt m−3 d−1, from which unwanted accumulation of organic acids was observed in RM2. NaHCO3 dosing enabled doubling the OLR (15.0 kg-CODt m−3 d−1) without performance deterioration, i.e., organic matter removal > 80.0% and maximum methane yield values. A detailed assessment of the food-to-microorganism ratio demonstrated the maintenance of stable operating conditions at excess substrate availability (>3.0 g-CODt g−1VSS d−1) in the feeding chamber (FDC) of the reactors, in which substrate conversion exceeded 70.0%. Organic overloads occurred only when biomass retention reached a saturation level in the FDC, hampering the uptake of volatile organic acids, regardless of the operating condition. In any case, the wide range of operating conditions (OLR = 2.5–15.0 kg-CODt m−3 d−1 and HRT = 24.0–36.0 h) associated with maximized methane production characterized a highly flexible process, which enables varied design arrangements in full-scale AD plants.

Dispersion Effects of Particulate Lead (Pb) from the Stack of a Lead Battery Recycling Plant

The contribution of emissions from the stack of a lead battery recycling plant to atmospheric lead concentrations and, eventually, to the topsoil of the surrounding area, were studied. A Gaussian dispersion model, of the American Meteorological Society/United States Environmental Protection Agency Regulatory Model, (AERMOD) was used to determine atmospheric total suspended particulate lead dispersion, which originated from stack emissions, over the wider study area. Stack emission parameters were obtained from online measurements of the industry control sensors. AERMOD simulated two scenarios for four calendar years, 2015 to 2018, one for the typical stack measured operating conditions and one for the legal limit operating conditions (emissions from the stack set by legislation to 0.5 mg m−3). Deposition fluxes modeled the input of atmospheric total suspended particulate Pb to the topsoil of the area. X-ray fluorescence (XRF) analyses were used to determine lead concentrations in the topsoil. The modeling results were compared with topsoil of six inhabited locations downwind from the stack in the direction of the prevailing winds to estimate the influence of lead deposition on topsoil near the industrial area.

Measurement of the porosity of a pseudo-spherical nanostructured particle deposit formed by filtration

In many applications involving nanoparticles, the question of controlling the properties of the films formed (particularly in terms of porosity) remains incomplete. These applications cover an extensive array of fields ranging from the manufacture of nanofilms to the prediction of the pressure drop of nanoparticle films formed on the surface of air purification filters. A review of the literature shows that the porosity of nanometric particle deposits has mainly been studied by numerical simulation. The few experimental studies conducted have focused on micron-sized particles or nanoparticle aggregates under limited operating conditions. The objective of this study is to confirm the porosity results of nanometric particle deposits obtained by the previous simulations and to correlate the porosity with relevant dimensionless numbers. The changes in porosity according to the various operating conditions confirm those obtained during the numerical simulations. All the experimental porosity values could be correlated to the Stokes number alone.