Prediction of the droplet spreading dynamics on a solid substrate at irregular sampling intervals: Nonlinear Auto-Regressive eXogenous Artificial Neural Network approach (NARX-ANN)

Heidari, Elham; Daeichian, Abolghasem; Sobati, Mohammad Amin; Movahedirad, Salman

doi:10.1016/j.cherd.2020.01.033

Cited by 19 publications

(11 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The drying time of a single droplet (event I) determines the lifespan of a binder droplet before the granule–droplet collision [ 34 ] and spreading. [ 35 ] If the drying rate of the droplet is lower than the spreading rate, the binder droplet spreads to the equilibrium radius before drying, which has a positive effect on the aggregation rate. The drying time of a spherical droplet depends not only on the initial diameter of the droplet and the relative velocity between air and droplet but also on the air humidity, temperature, and the physical properties of the liquid binder.…”

Section: Resultsmentioning

confidence: 99%

Mechanistic models for liquid binder evaporation in wet granulation

2022

Self Cite

View full text Add to dashboard Cite

In this study, binder evaporation as a significant microscale phenomenon in fluidized bed wet granulation has been investigated in three situations: a single droplet, a sessile droplet, and a liquid bridge. Single droplet evaporation has been analytically modelled by mass transfer analysis from a spherical droplet subjected to the relative velocity of the medium and considering evaporation rate variations versus droplet diameter. Then, the sessile droplet evaporation model has been used from the literature. Finally, the liquid bridge evaporation has been modelled, and the rupture time of the bridge has been computed. The local temperature dependence of air physicochemical properties has been considered in all models. The single and sessile droplet evaporation rates have been successfully validated by the experimental data of the air–water system. The effects of operational conditions and liquid/particle properties on each evaporation event have been evaluated and quantified. The results indicate that both the higher relative velocity between the air and a single droplet and the smaller equilibrium contact angle in a constant volume of sessile droplets increase the evaporation rate. Also, increasing the length of the binder bridge reduces the rupture time.

show abstract

Section: Resultsmentioning

confidence: 99%

Mechanistic models for liquid binder evaporation in wet granulation

2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…As described in the previous section, the vectorΘ parameters can be calculated, and then the V T will be approximated. For a better understanding, we obtain the matrix P for this problem, which is given in (24).…”

Section: Simulation Results Of the Proposed Methods For Estimating Jmentioning

confidence: 99%

“…The described system is a form of NARX system but in this paper, we call it the Nonlinear Feed-Forward Memory-Less (NFFML) model. A wide range of the nonlinear dynamic system can be described with the input u and the output y in the NARX structure using the following equation [22][23][24][25] :…”

Section: The Proposed Nffml Modelmentioning

confidence: 99%

“…A wide range of the nonlinear dynamic system can be described with the input u and the output y in the NARX structure using the following equation 22‐25 :

\begin{matrix} trueleft \\ y (k + 1) = f (R (k)) \\ R (k) = {(][,,,,,, y ()(, k), \dots, y ()(, k - n_{y} + 1), u ()(, k), \dots, u ()(, k - n_{u} + 1))}^{T} \end{matrix}

where y ( k + 1) refers to the predicted output at the future time ( k + 1) and R ( k ) is a regressor vector that includes finite numbers of the past inputs and outputs. Dynamic order of this system is determined by the number of n u and n y lags.…”

Section: The Proposed Nffml Modelmentioning

confidence: 99%

See 1 more Smart Citation

A nonlinear feed‐forward memory‐less model to fast prediction of threshold voltage in junction‐less double‐gate MOSFETs

Annabestani

Nasserian

Hasanzadeh

et al. 2020

Int J Numerical Modelling

View full text Add to dashboard Cite

Decreasing Drain-Induced-Barrier-Lowering (DIBL) is one of the nondesirable short-channel effects, causes the threshold voltage of the transistor to be reduced by increasing the drain voltage. DIBL makes it impossible for engineers to consider V T as a constant, and it is necessary to calculate V T as a function of the drain voltage. Therefore, to consider the DIBL effect in the design of ICs, a large computational burden is imposed on the system, which slows down the simulation process in circuit-level simulators. Accordingly, a Nonlinear Feed-Forward Memory-Less (NFFML) model using the Gram-Schmidt orthogonalization approach is proposed, which calculates the V T of the new generation of MOSFETs, that is, Junctionless Double-Gate MOSFETs (JL-DG-MOSFETs), with high precision and a significant speed-up in the computation of the model. It is shown that the proposed numerical method is 313 times faster than the state-of-the-art analytical model. The normalized MSE is 0.435% on average, showing that the proposed approach can be a fast and accurate candidate for replacing the analytical modeling.

show abstract

“…By using neural networks that can be trained employing training data gathered from high-fidelity numerical simulations, Um et al 30 introduced a data-driven method for simulating liquid splash. Besides, a technique based on the nonlinear autoregressive exogenous ANN model (NARX-ANN) methodology for forecasting droplet spreading dynamics on a solid substrate was presented by Heidari et al 31 Furthermore, their model was trained in a constrained manner, with limited application considering the We, Re, and contact angle (θ).…”

Section: Introductionmentioning

confidence: 99%

Uncovering the Effect of Physical Conditions and Surface Roughness on the Maximum Spreading Factor of Impinging Droplets Using a Supervised Artificial Neural Network Model

Choudhury,

Nasir,

Kaiser

et al. 2023

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

Prediction of the droplet spreading dynamics on a solid substrate at irregular sampling intervals: Nonlinear Auto-Regressive eXogenous Artificial Neural Network approach (NARX-ANN)

Cited by 19 publications

References 38 publications

Mechanistic models for liquid binder evaporation in wet granulation

Mechanistic models for liquid binder evaporation in wet granulation

A nonlinear feed‐forward memory‐less model to fast prediction of threshold voltage in junction‐less double‐gate MOSFETs

Uncovering the Effect of Physical Conditions and Surface Roughness on the Maximum Spreading Factor of Impinging Droplets Using a Supervised Artificial Neural Network Model

Contact Info

Product

Resources

About