Search

Before a droplet can contact a surface during impact, it must first displace the air beneath it. Over a wide range of impact velocities, the air is compressed into a thin film, slowing the progress of the liquid toward the surface. Below a critical impact velocity, the air film remains intact; the droplet rebounds off of the air film without making contact. Above this critical impact velocity, the droplet always makes contact. The initiation of contact formation occurs via a topological transition, whereby the initially connected gas domain is ruptured, and a liquid capillary bridge forms, binding the droplet to the surface. Here we probe this transition in detail around the critical impact velocity using calibrated total internal reflection (TIR) microscopy to monitor the air film profile at high speed during the impact process. Two air film rupture modalities are observed: nucleated contacts, which are isolated and don't correspond to the global minimum air film thickness, and spontaneous contacts, that occur always on a ring centered upon the impact axis where the air film reaches its global minimum. Our measurements show that for impact velocities exceeding the critical velocity for contact initiation, the air film ruptures at a nearly identical height h min ~ 20 nm, for two fluids - silicone oil and a water glycerol mixture. The height and time duration of the air film prior to contact, analyzed for over 180 droplet impact experiments, are consistent with a linear instability driven by van der Waals forces. Impact events of water solution droplets show different statistics for contact nucleation than the silicon oil; nevertheless, a critical impact velocity above which contact always occurs is identified for both liquids.

In this paper, we investigate the feasibility and physical consequences of cyber attacks against energy management systems (EMS). Within this framework, we have designed a complete simulation platform to emulate realistic EMS operations: it includes state estimation (SE), real-time contingency analysis (RTCA), and security constrained economic dispatch (SCED). This software platform allowed us to achieve two main objectives: 1) to study the cyber vulnerabilities of an EMS and understand their consequences on the system, and 2) to formulate and implement countermeasures against cyber-attacks exploiting these vulnerabilities. Our results show that the false data injection attacks against state estimation described in the literature do not easily cause base-case overflows because of the conservatism introduced by RTCA. For a successful attack, a more sophisticated model that includes all of the EMS blocks is needed; even in this scenario, only post-contingency violations can be achieved. Nonetheless, we propose several countermeasures that can detect changes due to cyber-attacks and limit their impact on the system.

This report describes what was done during a Master’s semester project lasting 14 weeks and concerns the calibration of Frustrated Total Internal Reflection (FTIR) and optimizing the setup for measuring the thin layer of air trapped between an impacting droplet and a solid surface before total wetting occurs. For this, a LED light source illuminating a prism is used, and a high speed camera reads the reflected image which can be analysed and can give the thickness of the layer or gap between the droplet and the solid surface, as the reflected light intensity read by the camera is directly a function of this gap thickness. To calibrate the setup a spherical lens is placed on top of the prism and its elastically deformed profile (under its own weight) is measured thanks to the FTIR setup, and compared to the analytical solution for the deformed profile provided by Hertzian solid contact theory. The measures are investigated with the light source being polarized and therefore the differences between the results of s− and p−polarized light are directly compared. It is found that s−polarized light gives a higher quality image with less error and noise and therefore s−polarized light is used for the rest of the calibration. A new intensity normalization method is investigated and appears to yield very promising results with low error considering the nanometer scale of deformation. After completion of the calibration, a droplet impact was tested and the image revealed that a pulsing LED would yield better results given that the very high velocity at which the layer of air disappears as the liquid completely wets the solid surface requires a very high framerate.