Original papers
Evaluation of droplets size distribution and velocity pattern using Computational Fluid Dynamics modelling

https://doi.org/10.1016/j.compag.2019.104886Get rights and content

Highlights

  • We highlighted the significance of the horizontal distance from the nozzle on the vertical distribution of spray and droplet sizes.

  • The significance of horizontal distance from the nozzle on distribution of droplet velocity.

  • Low liquid flow rate had finer droplet size spectrum compared to high liquid flow rate.

  • The motion of the sprayer affects the spatial distribution of the droplets.

Abstract

Synchronizing airflow output in air atomized spray to achieve effective spraying and minimal crop damage is a big challenge in controlled cultivation. The environmental conditions inside these structures presents several limitations in spray application including limited sprayer-target distance. A Computational Fluid Dynamics (CFD) modeling of droplet sizes and droplets velocity was developed and applied to evaluate the distribution pattern of spray and spraying effectiveness at varied sprayer settings for a new model greenhouse air blast sprayer. The evaluated sprayer settings were: three horizontal distances from the nozzle (0.3, 0.5 and 0.7 m), two nozzle discharge rates (0.40 and 0.65 l min−1) and two ground speeds (0 and 0.36 m s−1). The airflow velocity at the nozzle outlet was maintained at 24.3 m s−1. Particles were tracked according to the Lagrangian approach taking into account the spray atomization. To complement the simulation, droplet sizes were determined using the Malvern Mastersizer 2000. The model predicted droplet sizes considerably well with a root mean square error, ERMS of 8.29 μm. The results indicated that both distance and nozzle’s flow rate influenced droplets size distribution. At low flow rate, the spray corresponded to a finer droplet size spectrum compared to at higher flow rate. The motion of the sprayer significantly affected the trajectories of the droplets hence their spatial location.

Introduction

The size of droplets is among the factors that may influence the biological efficacy of the applied pesticides spray (Nuyttens et al., 2007). Yet, nearly all designs of sprayers in the market polydisperse sprays (Sirignano, 1999) of diverse droplet sizes from very fine to very coarse. Once a spray is injected into the environment, the big droplets traverse at higher velocities than the small droplets (Nuyttens et al., 2009) but the later accelerate or decelerate more rapidly (Sirignano, 1999). Delele et al. (2007) pointed out that tiny droplets are prone to drift and hardly reach the targets while big droplets maintain readily higher momentum and penetrate plant canopy better. More to this, the atmospheric temperature difference between wet and dry bulb temperature during spraying influence the lifetime of droplets (Lebeau et al., 2011). Logically, these aspects direct or indirectly affect the instantaneous intensity and position of droplets in a spray cloud especially in poly-dispersed spray.

The distribution of droplet sizes is contingent to the liquid flow rate (Vallet and Tinet 2013) and the spray’s travel distance (Dekeyser et al., 2013). Dekeyser et al. (2013) further highlighted a correlation between distribution of liquid and air flow pattern for air assisted spray. Therefore, failure to quantify the interdependence of the above mentioned factors and appropriately incorporate their effects during greenhouse sprayer development may lead to continual risk of heterogeneous and off-target spray deposits. Musiu et al. (2019) attributed non-uniform spray deposits on varied vertical zones of artificial plants to variation in spray intensity in the spray cloud.

Over the years the performance of new sprayer designs has been evaluated through experiments. However, experiments alone may be insufficient because of the complexity of fluid flows including the tendency of spray to destabilize in gaseous environment (Sirignano, 1999). Computational Fluid Dynamics (CFD) modeling has successfully been applied in numerous studies (Bartzanas et al., 2013, Musiu and Qi, 2016). Delele et al. (2005) modelled the pattern of airflow for a cross flow air sprayer relative to ground speed and fan speed. The authors noted an air flow velocity decay coefficient with linear proportionality to ground speed and jet outlet velocity ratio of 0–0.13. Endalew et al. (2010) modeled the characteristics of airflow through pear canopies and noted a link between the design of the sprayer and the distribution of the airflow. Numerous fans in the sprayer ameliorated homogeniety of airflow distribution. Dekeyser et al. (2013) combined experiments and simulation to compare three unique sprayers with distinctive air discharge units and found a correlation between the architecture of the air discharge unit and air flow pattern.

At present, several CFD models have been developed and applied. Brown and Sidahmed (2001) modeled the horizontal travel distance of droplets released from a forestry air blast sprayer. Sidahmed et al. (2005) proposed a model for predicting the behavior of droplets from a flat fan hydraulic nozzle in a spray cloud. Duga et al. (2017) developed a CFD model for predicting spray drift from orchard sprayers. Moa’ath et al. (2011) proposed a mathematical model for predicting the movement of droplets on a virtual leaf surface. Da Silva et al. (2006) proposed a lagrangian model for predicting spray deposition on vine canopies. Their model overlooked evaporation of spray as well as the transport between the sprayer and the canopy. Moreover, the authors assumed that the spray cloud composed of homogeneous droplet sizes with equal mass of liquid.

Most published studies on CFD modeling have explored orchard or open field sprayers. This therefore limits the applicability of these models to sites and conditions similar to open fields where the data used to develop them were collected. Of these models, none was devoted to greenhouse air blast sprayers. Greenhouses have totally different environmental conditions to open fields. Hence, the applicability of models developed using data collected for orchard sprayers and open field may be unsuitable for greenhouse sprayers. Therefore the objective of this study was to develop a CFD model that can predict the movement of droplets from a greenhouse air blast sprayer while taking into account downstream airflow velocity decay, the geometry of the nozzle and liquid atomization. Three downstream horizontal distances from the nozzle (0.3, 0.5 and 0.7 m), and two liquid flow rates (0.40 and 0.65 l min−1) were considered. Two sprayer modes: stationery and moving (at a ground speed of 0.36 m s−1) were evaluated. The simulated results were compared against the measured results.

Section snippets

Air assisted greenhouse sprayer

The modeling was done for a new design greenhouse air blast sprayer fitted with a motor driven axial fan (Fig. 1a). The droplets were generated from a single circular nozzle (Shenzhen Long Rui Co Ltd, China) whose geometry included six fixed identical blades (Fig. 1b) which served to impose an angular velocity component to the exiting air hence redirect the flow. Details on the geometry of the blades are outlined in Musiu et al. (2019). The central axis of the nozzle was located at 0.6 m above

Effects of nozzle’s flow rate and distance on volumetric droplets size distribution

Fig. 3 shows predicted distribution of droplet sizes for flow rates of 0.40 and 0.65 l min−1. At same outlet pressure (2.67 bar) and 0.3 m from the nozzle, the simulation indicated a finer droplets size spectrum at a flow rate of 0.40 l min−1 than at 0.65 l min−1. A similar observation was made by Vallet and Tinet (2013). This observation was because of the ratio of air to liquid during atomization. Given that the amount of atomizing air determines the liquid break up intensity, at constant air

Conclusion

The CFD model proposed allowed us to predict well; the distribution of droplet sizes and droplets velocity at varied nozzle discharge rates; distance from the nozzle and ground speed for a spray from a greenhouse sprayer. The model did not take into account temperature gradient which may have affected the dynamics of surface tension and viscosity of the liquid. Consequently, this may have influenced the size of droplets generated hence their trajectories and therefore need to be accounted in

Acknowledgment

This work was supported by Research and Development Project, China (2017YFD0701400).

References (37)

Cited by (19)

  • Evaluation of UAV spraying quality based on 1D-CNN model and wireless multi-sensors system

    2024, Information Processing in Agriculture
    Citation Excerpt :

    Several prediction models for droplet deposition based on different conditions have been studied. The computational fluid dynamics (CFD) model is a common method for predicting droplet deposition [14]. This model can effectively simulate various parameter states during spraying with high prediction accuracy.

  • Analysis of droplet size uniformity and selection of spray parameters based on the biological optimum particle size theory

    2022, Environmental Research
    Citation Excerpt :

    Elizabeth M. Musiu conducted atomization experiments on nozzles with different spraying heights and equivalent outlet diameters. She observed the droplet sizes using particle tracking technology and found that the spray height and equivalent outlet diameter affected the droplet size distribution (EMMA et al., 2019). In a wind tunnel environment, Ru et al. conducted atomization experiments on fan-shaped nozzles with different equivalent outlet diameters.

  • Real-time measurement of droplet size and its distribution of an air-induced air-assisted electrostatic nozzle

    2022, Journal of Electrostatics
    Citation Excerpt :

    It will also give a relationship between the droplet size and applied high voltage to charging electrode of an air-induced air-assisted electrostatic nozzle. Numerous researchers have adopted various strategies to measure the droplet size for different kinds of spray generated by various spraying methods and used the techniques ranging from water sensitive paper based measurement to laser diffraction based measurement [27–32]. A portable scanning system to measure the spray coverage and deposition has been developed by Zhu [33].

View all citing articles on Scopus
View full text