Uncovering the effect of physical conditions and surface roughness on the maximum spreading factor of impinging droplets using a supervised artificial neural network model

The phenomenon of liquid droplet impingement on solid surfaces is particularly important in industrial applications related to spray coating, thermal spraying, inkjet printing, spray cooling, and powder generation industries. Atomized liquid metal droplet impact over surfaces where impingement on both stationary and rotating surfaces, such as rotating disks, can be used to carefully control droplet sizes. Furthermore, several other aspects, such as liquid properties (especially its surface tension), falling height, surface roughness, and wettability, play a vital role in controlling characteristics that not only affect droplet size but also influence droplet trajectories and spread. These parameters were studied in fine detail in a previous article, where a series of experiments were conducted to investigate the phenomenon of transient liquid spreading under varying conditions. In this paper, we further extend the previous study by demonstrating the effect of surface roughness, r_a, the droplet Reynolds, and Weber numbers and the contact angle by fitting 342 data points to obtain a high-fidelity model using an artificial neural network (ANN) for predicting β_max, the dimensionless spreading diameter. By comparing the obtained model with ten models in the literature, the authors demonstrated the development of a more precise neural network-based predictive model and its accuracy using a large set of experimental data. It is shown that the spreading is strongly affected in an inverse manner by the impinged surface roughness, which the ANN modeling well captures along with the complex interaction of the other independent factors.